[Federal Register: June 10, 2008 (Volume 73, Number 112)]
[Notices]
[Page 32833-32935]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr10jn08-114]
[[Page 32833]]
-----------------------------------------------------------------------
Part III
Department of Justice
-----------------------------------------------------------------------
Antitrust Division
-----------------------------------------------------------------------
United States v. Abitibi-Consolidated Inc. et al.; Response to Public
Comment on the Proposed Final Judgment; Notice
[[Page 32834]]
DEPARTMENT OF JUSTICE
Antitrust Division
United States v. Abitibi-Consolidated Inc. et al.; Response to
Public Comment on the Proposed Final Judgment
Pursuant to the Antitrust Procedures and Penalties Act, 15 U.S.C.
16(b)-(h), the United States hereby publishes the public comment
received on the proposed Final Judgment in United States of America v.
Abitibi-Consolidated Inc. et al., Civil Action No. 1:07-cv-1912 and the
response to the comment. On October 23, 2007, the United States filed a
Complaint alleging that the merger between Abitibi-Consolidated Inc.
(``Abitibi'') and Bowater Inc. (``Bowater'') violated Section 7 of the
Clayton Act, 15 U.S.C. 18. The proposed Final Judgment, filed on
October 23, 2007, requires the combined company to divest Abitibi's
Snowflake, Arizona paper mill. Public comment was invited within the
statutory 60-day comment period. Copies of the Complaint, proposed
Final Judgment, Competitive Impact Statement, Public Comment and the
United States' Response to the Comment and other papers are currently
available for inspection in Suite 1010 of the Antitrust Division,
Department of Justice, 450 5th Street, NW., Washington, DC 20530,
telephone: (202) 514-2481 and the Office of the Clerk of the United
States District Court for the District of the District of Columbia, 333
Constitution Ave., NW, Washington, DC 20001. Copies of any of these
materials may be obtained upon request and payment of a copying fee.
J. Robert Kramer II,
Director of Operations, Antitrust Division.
In the matter of: United States of America, Plaintiff, v. Abitibi-
Consolidated Inc. and Bowater Inc., Defendants.
Case No: [1:07-cv-01912]
Judge: Collyer, Rosemary M.; Deck type: Antitrust.
Response of Plaintiff United States to Public Comments on the Proposed
Final Judgment
Pursuant to the requirements of the Antitrust Procedures and
Penalties Act (``APPA'' or ``Tunney Act''), 15 U.S.C. 16(b)-(h), the
United States hereby files the Comment received from members of the
public concerning the proposed Final Judgment in this case and the
Response by the United States to the Comment. The United States will
move the Court for entry of the proposed Final Judgment after the
Comment and this Response have been published in the Federal Register,
pursuant to 15 U.S.C. 16(d).
The United States filed a civil antitrust Complaint under Section
15 of the Clayton Act, 15 U.S.C. 25, on October 23, 2007, alleging that
the merger of Abitibi-Consolidated Incorporated (``Abitibi'') and
Bowater Incorporated (``Bowater'') would violate Section 7 of the
Clayton Act, 15 U.S.C. 18. Simultaneously with the filing of the
Complaint, the United States filed a proposed Final Judgment and an
Asset Preservation Stipulation and Order (``Stipulation'') signed by
plaintiff and defendants consenting to the entry of the proposed Final
Judgment after compliance with the requirements of the Tunney Act.
Pursuant to those requirements, the United States filed a Competitive
Impact Statement (``CIS'') in this Court on October 23, 2007, published
the proposed Final Judgment and CIS in the Federal Register on November
8, 2007, see United States v. Abitibi-Consolidated Inc. and Bowater
Inc., 72 FR 63187 (November 8, 2007); and published summaries of the
terms of the proposed Final Judgment and CIS, together with directions
for the submission of written comments relating to the proposed Final
Judgment, in The Washington Post for seven days beginning on November
18, 2007, and ending on November 24, 2007. The 60-day period for public
comments ended on January 7, 2008, and one comment was received as
described below and attached hereto.
I. Background: The United States' Investigation and the Proposed
Resolution
On January 29, 2007, Abitibi and Bowater announced plans to merge
into a new company to be called AbitibiBowater Incorporated
(``AbitibiBowater''). Over the next nine months, the United States
Department of Justice (the ``Department'') conducted an extensive,
detailed investigation into the competitive effects of the proposed
transaction. As part of this investigation, the Department obtained
substantial documents and information from the merging parties and
issued 37 Civil Investigative Demands to third parties. In response,
the Department received and considered more than 150,000 pages of
material. The Department conducted more than 60 interviews with
customers, competitors and other individuals with knowledge of the
industry. The sole commenter here, the Newspaper Association of America
(the ``NAA''), represents newspaper publishers in the United States.
During the course of the Department's investigation into the proposed
merger, the NAA shared with the investigative staff its concerns about
the impact of the proposed merger on competition; the investigative
staff carefully analyzed its concerns and submissions, as well as the
data, market facts and opinions of other knowledgeable parties.
The Department concluded that the combination of Abitibi and
Bowater likely would lessen competition in the North American newsprint
market. Newspapers are printed on newsprint, the lowest quality and
generally the least expensive grade of groundwood paper. Newspaper
publishers, who buy more than 80 percent of all newsprint sold in the
United States, have no close substitutes to use for printing newspapers
because of newsprint's price and physical characteristics. Because
publishers' newsprint presses are optimized to use newsprint, switching
to another grade of paper would be costly. A small but significant
increase in price likely would not cause customers to switch sufficient
newsprint tonnes to other products or otherwise curtail their newsprint
usage so as to render the increase unprofitable.
As explained more fully in the Complaint and CIS, the merger of
Abitibi and Bowater would substantially increase concentration and
lessen competition in the production, distribution and sale of
newsprint in North America. After conducting a detailed analysis of the
merger, the Department filed its Complaint alleging competitive harm in
the newsprint market in North America and sought a remedy that would
ensure that such harm is prevented.
The proposed Final Judgment in this case is designed to preserve
competition in the production, distribution and sale of newsprint in
North America. It requires the divestiture of a newsprint mill that
manufactures newsprint for sale in North America. Specifically, the
proposed Final Judgment directs a sale of Abitibi's Snowflake, Arizona,
newsprint mill (``Snowflake,'' or the ``Snowflake mill'') to a
purchaser acceptable to the United States.
In the Department's judgment, divestiture of the Snowflake mill to
a qualified purchaser would remedy the violation alleged in the
Complaint because the Snowflake mill, located in northeastern Arizona,
is one of the most efficient and profitable newsprint mills in North
America. Plans to improve the mill's efficiency in coming years with
investments in energy and machinery are already underway. Snowflake's
size and cost position ensure that its divestiture to a competitor of
the
[[Page 32835]]
merged firm will preserve competition in the North American newsprint
market. Although entry of the proposed Final Judgment would terminate
this action, the Court would retain jurisdiction to construe, modify,
or enforce the provisions of the proposed Final Judgment and punish
violations thereof. \1\
---------------------------------------------------------------------------
\1\ The merger closed on October 29, 2007. In keeping with the
United States' standard practice, neither the Stipulation nor the
proposed Final Judgment prohibited closing the merger. See ABA
Section of Antitrust Law, Antitrust Law Developments 406 (6th ed.
2007) (noting that ``[t]he Federal Trade Commission (as well as the
Department of Justice) generally will permit the underlying
transaction to close during the notice and comment period''). Such a
prohibition could interfere with many time-sensitive deals and
prevent or delay the realization of substantial efficiencies. In
consent decrees requiring divestitures, it is also standard practice
to include a ``preservation of assets'' clause in the decree and to
file a stipulation to ensure that the assets to be divested remain
competitively viable. That practice was followed here. Proposed
Final Judgment Sec. IV(K). In addition, the Stipulation entered by
the Court in this case required AbitibiBowater to hold separate the
Snowflake newsprint mill, pending the divestiture contemplated by
the proposed Final Judgment.
---------------------------------------------------------------------------
II. Standard of Judicial Review
Upon the publication of the Comment and this Response, the United
States will have fully complied with the Tunney Act and will move for
entry of the proposed Final Judgment as being ``in the public
interest.'' 15 U.S.C. 16(e), as amended.
The Tunney Act states that, in making that determination, the Court
shall consider:
(A) the competitive impact of such judgment, including
termination of alleged violations, provisions for enforcement and
modification, duration of relief sought, anticipated effects of
alternative remedies actually considered, whether its terms are
ambiguous, and any other competitive considerations bearing upon the
adequacy of such judgment that the court deems necessary to a
determination of whether the consent judgment is in the public
interest; and
(B) the impact of entry of such judgment upon competition in the
relevant market or markets, upon the public generally and
individuals alleging specific injury from the violations set forth
in the complaint including consideration of the public benefit, if
any, to be derived from a determination of the issues at trial.
15 U.S.C. 16(e)(1)(A)-(B); see generally United States v. SBC Commc'ns,
Inc., 489 F. Supp. 2d 1, 11 (D.D.C. 2007) (concluding that the 2004
amendments ``effected minimal changes'' to scope of review under Tunney
Act, leaving review ``sharply proscribed by precedent and the nature of
Tunney Act proceedings'').\2\
---------------------------------------------------------------------------
\2\ The 2004 amendments substituted ``shall'' for ``may'' in
directing relevant factors for court to consider and amended the
list of factors to focus on competitive considerations and to
address potentially ambiguous judgment terms. Compare 15 U.S.C.
16(e) (2004), with 15 U.S.C. 16(e)(1) (2006).
---------------------------------------------------------------------------
As the United States Court of Appeals for the District of Columbia
Circuit has held, under the APPA a court considers, among other things,
the relationship between the remedy secured and the specific
allegations set forth in the government's complaint, whether the decree
is sufficiently clear, whether enforcement mechanisms are sufficient,
and whether the decree may positively harm third parties. See United
States v. Microsoft Corp., 56 F.3d 1448, 1458-62 (D.C. Cir. 1995). With
respect to the adequacy of the relief secured by the decree, a court
may not ``engage in an unrestricted evaluation of what relief would
best serve the public.'' United States v. BNS, Inc., 858 F.2d 456, 462
(9th Cir. 1988) (citing United States v. Bechtel Corp., 648 F.2d 660,
666 (9th Cir. 1981)); see also Microsoft, 56 F.3d at 1460-62. Courts
have held that:
[t]he balancing of competing social and political interests
affected by a proposed antitrust consent decree must be left, in the
first instance, to the discretion of the Attorney General. The
court's role in protecting the public interest is one of insuring
that the government has not breached its duty to the public in
consenting to the decree. The court is required to determine not
whether a particular decree is the one that will best serve society,
but whether the settlement is ``within the reaches of the public
interest.'' More elaborate requirements might undermine the
effectiveness of antitrust enforcement by consent decree.
Bechtel, 648 F.2d at 666 (emphasis added) (citations omitted). Cf. BNS,
858 F.2d at 464 (holding that the court's ``ultimate authority under
the [APPA] is limited to approving or disapproving the consent
decree''); United States v. Gillette Co., 406 F. Supp. 713, 716 (D.
Mass. 1975) (noting that, in this way, the court is constrained to
``look at the overall picture not hypercritically, nor with a
microscope, but with an artist's reducing glass''). See generally
Microsoft, 56 F.3d at 1461 (discussing whether ``the remedies [obtained
in the decree are] so inconsonant with the allegations charged as to
fall outside of the `reaches of the public interest' ''). In making its
public interest determination, a district court ``must accord deference
to the government's predictions about the efficacy of its remedies, and
may not require that the remedies perfectly match the alleged
violations'' because this may only reflect underlying weakness in the
government's case or concessions made during negotiation. SBC Commc'ns,
489 F. Supp. 2d at 17; see also Microsoft, 56 F.3d at 1461 (noting the
need for courts to be ``deferential to the government's predictions as
to the effect of the proposed remedies''); United States v. Archer-
Daniels-Midland Co., 272 F. Supp. 2d 1, 6 (D.D.C. 2003) (noting that
the court should grant due respect to the United States' prediction as
to the effect of proposed remedies, its perception of the market
structure, and its views of the nature of the case).
Court approval of a consent decree requires a standard more
flexible and less strict than that appropriate to court adoption of a
litigated decree following a finding of liability. ``[A] proposed
decree must be approved even if it falls short of the remedy the court
would impose on its own, as long as it falls within the range of
acceptability or is `within the reaches of public interest.' '' United
States v. Am. Tel. & Tel. Co., 552 F. Supp. 131, 151 (D.D.C. 1982)
(citations omitted) (quoting United States v. Gillette Co., 406 F.
Supp. 713, 716 (D. Mass. 1975)), aff'd sub nom. Maryland v. United
States, 460 U.S. 1001 (1983); see also United States v. Alcan Aluminum
Ltd., 605 F. Supp. 619, 622 (W.D. Ky. 1985) (approving the consent
decree even though the court would have imposed a greater remedy). To
meet this standard, the United States ``need only provide a factual
basis for concluding that the settlements are reasonably adequate
remedies for the alleged harms.'' SBC Commc'ns, 489 F. Supp. 2d at 17.
Moreover, the Court's role under the APPA is limited to reviewing
the remedy in relationship to the violations that the United States has
alleged in its complaint, and does not authorize the Court to
``construct [its] own hypothetical case and then evaluate the decree
against that case.'' Microsoft, 56 F.3d at 1459. Because the ``court's
authority to review the decree depends entirely on the government's
exercising its prosecutorial discretion by bringing a case in the first
place,'' it follows that ``the court is only authorized to review the
decree itself,'' and not to ``effectively redraft the complaint'' to
inquire into other matters that the United States did not pursue. Id.
at 1459-60. As this Court recently confirmed in SBC Communications,
courts ``cannot look beyond the complaint in making the public interest
determination unless the complaint is drafted so narrowly as to make a
mockery of judicial power.'' SBC Commc'ns 489 F. Supp. 2d at 15.
In its 2004 amendments, Congress made clear its intent to preserve
the
[[Page 32836]]
practical benefits of utilizing consent decrees in antitrust
enforcement, adding the unambiguous instruction ``[nlothing in this
section shall be construed to require the court to conduct an
evidentiary hearing or to require the court to permit anyone to
intervene.'' 15 U.S.C. 16(e)(2). The language wrote into the statute
what the Congress that enacted the Tunney Act in 1974 intended, as
Senator Tunney then explained: ``[t]he court is nowhere compelled to go
to trial or to engage in extended proceedings which might have the
effect of vitiating the benefits of prompt and less costly settlement
through the consent decree process.'' 119 Cong. Rec. 24,598 (1973)
(statement of Senator Tunney). Rather, the procedure for the public
interest determination is left to the discretion of the court, with the
recognition that the court's ``scope of review remains sharply
proscribed by precedent and the nature of Tunney Act proceedings.'' SBC
Commc'ns, 489 F. Supp. 2d at 11.\3\
---------------------------------------------------------------------------
\3\ See United States v. Enova Corp., 107 F. Supp. 2d 10, 17
(D.D.C. 2000) (noting that the ``Tunney Act expressly allows the
court to make its public interest determination on the basis of the
competitive impact statement and response to comments alone'');
United States v. Mid-Am. Dairymen, Inc., 1977-1 Trade Cas. (CCH) ]
61,508, at 71,980 (W.D. Mo. 1977) (``Absent a showing of corrupt
failure of the government to discharge its duty, the Court, in
making its public interest finding, should * * * carefully consider
the explanations of the government in the competitive impact
statement and its responses to comments in order to determine
whether those explanations are reasonable under the
circumstances.''); S. Rep. No. 93-298, 93d Cong., 1st Sess., at 6
(1973) (``Where the public interest can be meaningfully evaluated
simply on the basis of briefs and oral arguments, that is the
approach that should be utilized.'').
---------------------------------------------------------------------------
III. Summary of the Comment and Response
During the 60-day comment period, the United States received one
Comment, from the NAA. That Comment is attached to this memo. After
reviewing the Comment, the United States continues to believe that the
proposed Final Judgment is in the public interest. The Comment includes
concerns relating to whether the proposed Final Judgment adequately
remedies the harms alleged in the Complaint. The United States
addresses these concerns below and explains how the remedy is
appropriate.
A. Summary of Comment Submitted by the NAA
The NAA is an association whose members include daily and Sunday
newspapers in the United States who purchase a significant proportion
of North America's newsprint production. In its Comment of January 2,
2008, the NAA expressed concerns relating to whether the proposed Final
Judgment adequately remedies the alleged harms. The NAA argued in its
Comment that the Court should not enter the proposed Final Judgment
without a hearing for two reasons: (1) the newly merged AbitibiBowater,
despite its agreement to divest the Snowflake mill, ``has already begun
to exercise the market power created by the merger to anticompetitively
raise newsprint prices to North American newsprint customers''; and (2)
the United States ``has not provided the Court with any factual or
economic analysis to demonstrate that the proposed remedy will
eliminate the incentive for AbitibiBowater to reduce industry capacity
and raise prices to North American newsprint customers.'' (NAA Comment
at 2.)
1. The NAA's Argument That AbitibiBowater Has Already Begun To Exercise
Market Power and Anticompetitively Raise Newsprint Prices
The NAA notes that a little more than five weeks following the
merger that created AbitibiBowater, the combined firm announced that it
would remove 600,000 metric tonnes of newsprint capacity from the North
American market and would raise newsprint prices by $60 per metric
tonne, to be implemented in three $20 price increases. The NAA further
notes that ``[m]ost'' North American newsprint manufacturers not only
joined AbitibiBowater's price increase but also implemented a
``previously stalled'' price increase of $25 per metric tonne. The NAA
estimated that, taken together, these two price increases constitute a
15 percent price increase as compared to the pre-merger, October 2007,
price for newsprint. The NAA also noted that, at the time
AbitibiBowater announced the removal of 600,000 metric tonnes of
newsprint capacity from the North American market, it also announced
that ``more mills could close in Canada later [in 2008].'' (Comment at
7.)
The NAA claims that these post-merger actions by AbitibiBowater
demonstrate that the United States ``severely underestimated the risk
that the merger posed to competition in the North American newsprint
market and severely underestimated the incentive and ability of the
merged firm to remove capacity from the market to raise the price of
newsprint well above competitive levels.'' (Comment at 7.) Accordingly,
the NAA contends that a ``significantly larger divestiture'' than the
Snowflake mill is required to prevent ``the substantial anticompetitive
price increases that are already occurring and will continue to occur
as a result of the merger.'' (Comment at 7.)
2. The NAA's Argument That the United States Has Not Provided Adequate
Factual or Legal Analysis Upon Which To Base a Public Interest
Determination
The NAA concedes that in the Complaint, the United States
``correctly identifies the competitive harm produced by the merger.''
(Comment at 9.) The NAA argues, however, that the United States has not
provided the Court with a factual or legal analysis to demonstrate that
the divestiture of the Snowflake mill will ``eliminate the incentive to
reduce industry capacity and raise prices to North American newsprint
customers,'' and thus has provided the Court with no basis by which to
determine if the proposed remedy is in the public interest. (Comment at
9.) Specifically, the NAA argues that, other than noting that Snowflake
is ``among the largest and most profitable mills in the United
States,'' the United States ``provided no further explanation for its
decision that Snowflake was both a sufficient remedy and the best
solution, no detail regarding under what `circumstances' this
conclusion was reached, and no scale against which it measured
Snowflake as the best alternative.'' (Comment at 17.)
The NAA contends that the proposed Final Judgment should not be
entered because the United States has not explained to the Court ``why
the remedy it proposes restores or preserves competition.'' (Comment at
19.) In particular, the NAA criticizes the United States for failing to
reference in the Complaint or CIS what the NAA describes as historical
anticompetitive behavior of Abitibi and Bowater, and it contends that
absent such references, it is impossible for the Court to determine if
and how much of a factor such conduct played in the United States'
evaluation and settlement of the merger. The NAA also criticizes the
United States for failing to discuss the anticipated effects of
alternative remedies actually considered.
B. Response of the United States to the NAA's Comment
The divestiture of the Snowflake mill adequately remedies the harm
alleged in the Complaint. In negotiating this remedy, the United States
carefully considered the capabilities and economic viability of the
Snowflake mill as well as other assets of the merging parties; the
extent of industry excess capacity; the history of declining demand for
newsprint, and the forecasts for that decline to continue; the costs of
[[Page 32837]]
production of all newsprint mills in North America; and the financial
viability of the merging parties and their competitors. After
considering these issues, the United States analyzed the merger using a
comprehensive data set of prices, sales, production volumes and costs,
capacities and forecasts of North American newsprint demand. In its
analysis, which drew upon non-public information unavailable to the
NAA, the United States concluded that the divestiture of the Snowflake
mill to a viable qualified purchaser will adequately redress the
competitive harm alleged in the Complaint and restore competition to
the market for the sale of newsprint in North America.
The United States and the NAA employed the same general economic
model to examine the competitive effects of the merger. Accurate data
about prices, manufacturing costs, the elasticity of demand and other
factors can allow economists to model whether merging firms have an
added incentive to exercise market power by reducing capacity after a
merger. The United States and the NAA both attempted to determine
whether the merger will cause the combined AbitibiBowater to eliminate
newsprint capacity earlier than Abitibi and Bowater would have if they
had remained independent competitors.
Although the United States and the NAA used a similar framework to
model competition, the results differed significantly because of
several important differences in the data. First, the United States had
more complete and accurate data. Unlike the NAA, the United States was
able to use a compulsory process to gather information. See, e.g., 15
U.S.C. 1311-14 (empowering the Antitrust Division to subpoena documents
and take oral testimony). In this case, the United States had access to
extensive and mill-by-mill data on sales (including exports),
production volumes, capacities and costs. The NAA, on the other hand,
had to rely on less accurate and publicly available information
relating to mill capacities, prices and costs in assessing the
profitability of and competitors' likely response to a post-merger
price increase. Second, the United States conducted its own analysis of
the effect of price changes on the demand for newsprint, using
confidential information, in addition to considering estimates provided
by others. Based upon its analysis, the United States believes that the
estimate used by NAA understates the sensitivity of newsprint
consumption to changes in price. In other words, the United States
believes that if the price for newsprint rose, customers would purchase
less newsprint than the NAA estimates. Third, the United States and the
NAA viewed 2007 differently. While the NAA assumed that the newsprint
market in 2007 was in equilibrium--which would allow that year's prices
to be used as a reference point from which to measure future changes--
the United States' investigation revealed that much of 2007 was a
period of instability. Unexpectedly large declines in demand for
newsprint created excess capacity and caused prices to fall
dramatically. The fact that AbitibiBowater and other firms responded to
declining demand for newsprint by closing mills that were consistently
losing money is discussed in further detail in the following section.
The United States is confident that at the time it negotiated the
proposed Final Judgment the divestiture of the Snowflake mill was in
the public interest, based upon the best information available at that
time. The United States remains confident that the divestiture of the
Snowflake mill is in the public interest and adequately remedies the
harms alleged in the Complaint.
1. AbitibiBowater's Recently Announced Decision To Reduce Excess
Newsprint Capacity, and Industry-Wide Price Increases, Do Not Mean That
the Parties Have Exercised Market Power
The NAA's argument, that the Snowflake mill divestiture is
insufficient to prevent the combined firm from exercising market power
by shutting additional capacity in order to raise prices, assumes that
the combined firm's post-merger capacity reductions are the result of
the merger. The NAA's suggestions to the contrary events since the
filing of the proposed Final Judgment appear to be unrelated to any
exercise of market power. The ongoing sharp decline in demand for
newsprint in North America, increases in the prices of key inputs into
the production of newsprint, and the continued decline in the value of
the United States dollar all have disrupted the supply and demand
equilibrium for newsprint. Industry observers expect disruptions to
continue as North American demand for newsprint declines. Manufacturers
will respond by intermittently closing capacity, which will cause the
market price to lurch from one equilibrium to another as it adjusts to
these shocks to supply. Thus, in a market with declining demand, prices
can be expected to fall when the decline in demand creates excess
supply and increase when unprofitable capacity is closed in response to
that decline in demand. In the remainder of this section, we will
discuss the effects of these trends on the newsprint market and show
that a careful analysis suggests that the NAA's claims are unfounded.
Demand for newsprint in the North American market ``has declined
over the last several years at a rate of approximately 5 to 10 percent
per year because of a significant decline in demand for newspapers. * *
* This decline in the demand for newsprint is projected to continue,
and the resulting excess newsprint capacity will likely lead Defendants
and their competitors to close, idle or convert more newsprint mills.''
(Complaint at ] 17; see also CIS at 5.) As North American demand
continues to decline, notwithstanding the merger, all firms, including
AbitibiBowater, will eventually have to close inefficient newsprint
capacity. In its Comment, the NAA ignores the possibility that
AbitibiBowater's post-merger decision to close some of its inefficient
capacity was a natural reaction to the continued decline in demand for
newsprint and may in fact be perfectly consistent with a competitive
market.
The pressure to close inefficient capacity also intensified in 2007
because the prices of key production inputs--specifically, recycled
fiber, wood pulp and energy--rose sharply. This increase in input costs
has raised the costs of all producers and put upward pressure on the
price of newsprint. Further, the United States dollar has lost value
relative to the Canadian dollar, which has the effect of raising the
costs of Canadian producers of newsprint--the bulk of North American
newsprint capacity is located in Canada--and hence the price of
newsprint.
Finally, the adjustment of the newsprint market to these disruptive
market conditions will not be instantaneous or smooth. Because
newsprint mills have very significant fixed costs and relatively
smaller incremental costs, newsprint manufacturers may not be able to
respond to declining demand by gradually withdrawing capacity. The
market therefore can be expected to swing between periods of
overcapacity and shortage as companies retire paper machines or entire
paper mills. As these swings occur, there will not be smooth changes to
the industry's overall capacity or its price levels. For example, while
the price of newsprint has risen in the past six months, it is at the
time of this filing at or below its lowest level in 2006 when input
prices were lower. Further, the United States' investigation
[[Page 32838]]
has found that the price is so low that many newsprint producers' mills
do not cover their costs. Indeed, the three mills that AbitibiBowater
closed after the merger were unprofitable.
In summary, the NAA's conclusion that recent newsprint capacity
closures and price increases necessarily are anticompetitive actions
driven by the merger is misguided and fails to account for significant
market facts affecting the supply and demand equilibrium of the North
American newsprint market.
2. The United States Has Provided Sufficient Explanation of Why the
Proposed Divestiture Is an Adequate Remedy to the Harm Alleged in the
Complaint, and Entry of the Proposed Final Judgment Will Be in the
Public Interest
The proposed Final Judgment provides an effective and appropriate
remedy for the antitrust violation alleged in the Complaint, and its
entry, therefore, will be in the public interest. The purpose of Tunney
Act review is not for the Court to engage in an ``unrestricted
evaluation of what relief would best serve the public,'' BNS, 858 F.2d
at 462 (citing Bechtel Corp., 648 F.2d at 666) or to determine the
relief ``that will best serve society,'' Bechtel Corp., 648 F.2d at
666. Instead, the purpose of Tunney Act review is simply to determine
whether the divestiture of the Snowflake mill is within the reaches of
the public interest, ``even if it falls short of the remedy the court
would impose on its own.'' AT&T, 552 F. Supp. at 151. In other words,
the purpose of Tunney Act review is to determine whether the
divestiture is a ``reasonably adequate'' remedy for the harms alleged
in the Complaint. SBC Commc'ns, 489 F. Supp. 2d at 17.
Subsections (A) and (B) of 15 U.S.C. 16(e)(1) set forth a number of
factors for courts to consider when assessing the competitive impact of
proposed final judgments. Many of those factors are not at issue
here.\4\ Instead, the second argument in the NAA's Comment focuses on
the competitive considerations relevant to the proposed Final Judgment,
the divestiture it requires and the alternatives the United States
considered.
---------------------------------------------------------------------------
\4\ The NAA does not contest several factors listed for courts
to consider under subsection (A). For instance, with respect to
``provisions for enforcement and modification,'' 15 U.S.C.
16(e)(1)(A), the proposed Final Judgment contains the standard
provisions that have been effective in numerous other cases brought
by the United States. In particular, the proposed Final Judgment
provides that the Court retains jurisdiction over this action, and
the parties may apply to the Court for any order necessary or
appropriate for the modification, interpretation, or enforcement of
the Final Judgment. With respect to ``duration of relief sought,''
id., the proposed divestiture is permanent. Finally, with respect to
``whether its terms are ambiguous,'' id., no term in the proposed
Final Judgment is ambiguous.
---------------------------------------------------------------------------
The NAA questions whether the United States has adequately
demonstrated to this Court that the divestiture eliminates
AbitibiBowater's post-merger incentive to reduce capacity and raise
prices to North American newsprint customers. It has. As explained
previously, the United States conducted an extensive investigation and
compiled comprehensive data on market shares, costs of production,
estimations of rest-of-industry newsprint capacity and future
reductions in newsprint demand gathered from public and non-public
sources. This data was used in an economic model to determine if the
merger would cause an anticompetitive increase in newsprint prices.\5\
The United States concluded that a merger between Abitibi and Bowater,
without a divestiture, would allow the merged firm to ``close its
capacity strategically, allowing the merged firm to raise newsprint
prices and recoup its lost profits on the combined output.'' (CIS at
8.) But, as the United States concluded in the CIS, ``[d]ivesting
Snowflake * * * will reduce the capacity over which the merged firm
could profit to a level at which it would not have the ability to close
capacity strategically.'' (Id.) In other words, the United States'
investigation found that without Snowflake, AbitibiBowater did not have
enough newsprint capacity to benefit sufficiently from the post-merger
price increase to offset the costs associated with shutting down
profitable newsprint capacity.
---------------------------------------------------------------------------
\5\ To raise prices above competitive levels, the merged firm
must create an artificial shortage by shutting down profitable
newsprint mills. The merged firm has the incentive to follow this
strategy when the costs of this strategy, which are the profits the
merged firm forgoes by prematurely shutting down profitable
newsprint mills, are less than its benefits, which are the increased
prices the merged firm can expect to recoup across its remaining
newsprint capacity. After completing its investigation, the United
States concluded that without a divestiture AbitibiBowater would
have the incentive to follow this strategy, that is, to create an
artificial shortage by shutting down otherwise-profitable newsprint
mills.
---------------------------------------------------------------------------
The NAA further contends that the United States ``has left the
Court entirely in the dark with absolutely no basis for making a
meaningful comparison between a Snowflake-only divestiture and any
alternative course of action, including a full trial on the merits.''
(Comment at 18.) This is incorrect; in the CIS the United States
addressed both alternatives. (CIS at 10-11.) As the United States noted
in the CIS, a full trial on the merits would require significant time
and expense, and the outcome would be uncertain. In light of such
uncertainty, the United States' decision to take an adequate and
available remedy and forgo the risk of trial is well within ``the
reaches of the public interest.'' See SBC Commc'ns, 489 F. Supp. 2d at
23 (``Success at trial was surely not assured, so pursuit of that
alternative may have resulted in no remedy at all. While a trial may
have created an even greater evidentiary record, that benefit may not
outweigh the possible loss of the settlement remedies. * * *'').
Similarly, the United States need not rehearse every permutation of
possible divestiture in order to demonstrate to this Court that the
divestiture of Snowflake would adequately address the competitive harm
alleged in the Complaint. The competitive harm that the United States
alleged--and that the NAA acknowledges--is AbitibiBowater's incentive
and ability to raise newsprint prices above competitive levels in the
North American market. Any divestiture that removes either the combined
firm's incentive or its ability to raise prices above competitive
levels would therefore be an adequate remedy. Given AbitibiBowater's
ownership of all or part of 19 paper mills in the United States and
Canada (see Complaint ]] 7 & 8), the United States could have selected
different mills, individually or in combination, to remove the merged
firm's ability and incentive to raise prices anticompetitively. In this
instance, considering all the factors--including the inherent
advantages of settlement and avoidance of the risk and uncertainty of
litigation \6\--the United States reasonably chose to require the
divestiture of one of ``the largest and most profitable newsprint mills
in the United States,'' which its analysis determined would deprive the
merged firm of the scale needed to recoup its lost profits. (See CIS at
6, 11.) As discussed above, given the continuing decline in demand for
newsprint, the United States anticipated that AbitibiBowater would
continue to close inefficient newsprint capacity. (See Complaint at ]
17, CIS at 5.) The United States determined that, coupled with the exit
from the market of such inefficient capacity, the divestiture of
[[Page 32839]]
the Snowflake mill will be sufficient to prevent AbitibiBowater from
engaging in an anticompetitive closure of efficient capacity. Abitibi
and Bowater, even before the merger, had the incentive to close money-
losing mills. The question therefore is whether the merger somehow gave
them the incentive to close profitable mills in order to raise prices
above competitive levels. The United States determined that
AbitibiBowater was not likely to have that incentive once it divested
Snowflake.
---------------------------------------------------------------------------
\6\ As noted previously, when making its public interest
determination, this Court ``must accord deference to the
government's predictions about the efficacy of its remedies, and may
not require that the remedies perfectly match the alleged violations
because this may only reflect underlying weakness in the
government's case or concessions made during negotiation.'' SBC
Commc'ns, 489 F. Supp. 2d at 17.
---------------------------------------------------------------------------
Finally, the NAA suggests that the proposed Final Judgment should
not be entered because Abitibi and Bowater previously had engaged in
anticompetitive conduct of the sort alleged in the Complaint, which it
alleges the United States did not properly account for in negotiating
the proposed Final Judgment. This suggestion is misplaced for two
reasons. First, as mentioned earlier, the United States spoke with a
number of market participants, including the NAA, and examined
historical data on prices and costs in the course of its investigation.
The evidence does not support the NAA's claims that the parties' prior
behavior was in fact anticompetitive. Second, the NAA's allegations
about the parties' prior behavior are irrelevant because the prior
behavior does not address whether, after Snowflake is divested,
AbitibiBowater will have the incentive and ability to unilaterally
raise price above competitive levels. (And as the United States has
already explained, the answer to this question is likely to be ``no.'')
Ultimately, in making its public interest determination, the
district court ``must accord deference to the government's predictions
about the efficacy of its remedies.'' See SBC Commc'ns, 489 F. Supp. 2d
at 17. As already has been demonstrated, the United States' analysis
supports the conclusion that divestiture of the Snowflake mill is an
appropriate remedy to the harms alleged in the Complaint.
IV. Conclusion
The issues raised in the NAA's public Comment were among the many
considered during the United States' extensive and thorough
investigation. The United States has determined that the proposed Final
Judgment as drafted provides an effective and appropriate remedy for
the antitrust violations alleged in the Complaint, and is therefore in
the public interest. The United States will move this Court to enter
the proposed Final Judgment after the Comment and Response are
published.
Respectfully Submitted,
Dated: April 18, 2008,
Karl D. Knutsen,
Ryan Danks,
Rebecca Perlmutter,
Michelle Seltzer (D.C. Bar No. 475482).
Trial Attorneys. United States Department of Justice, Antitrust
Division, Litigation I Section, 1401 H St., N.W., Suite 4000,
Washington, DC 20530, Telephone: (202) 514-0976, Facsimile: (202)
307-5802.
Certificate of Service
I hereby certify that on April 18, 2008, I caused a copy of the
foregoing Response of Plaintiff United States to Public Comments on
The Proposed Final Judgment in this matter to the following
individuals by electronic mail:
Counsel for Defendant Abitibi-Consolidated Inc.
Joseph J. Simons, Esq., Paul, Weiss, Rifkind, Wharton & Garrison
LLP, 1615 L Street, NW., Suite 1300, Washington, DC 20036-5694,
Telephone: (202) 223-7370, Facsimile: (202) 223-7470, E-mail:
jsimons@paulweiss.com.
Counsel for Defendant Bowater Incorporated
R. Hewitt Pate, Esq., Hunton & Williams, 1900 K Street, NW.,
Washington, DC 20006, Telephone: (202) 955-1921, Facsimile: (202)
857-3894, E-mail: hpate@hunton.com.
Counsel for the Newspaper Association of America
Alan L. Marx, Esq., King and Ballow, 1100 Union Street Plaza, 315
Union Street, Nashville, TN 37201, Telephone: (615) 726-5455,
Facsimile: (615) 726-5413, E-mail: amarx@kingballow.com.
-----------------------------------------------------------------------
Karl D. Knutsen.
Comments of the Newspaper Association of America Regarding Proposed
Final Judgment in United States of America v. Abitibi-Consolidated,
Inc. and Bowater, Incorporated
In its Explanation of Consent Decree Procedures, the Justice
Department requests the Court to enter the proposed Final Judgment
settling United States of America v. Abitibi-Consolidated, Inc. and
Bowater, Incorporated without a hearing ``provided that the Court
concludes that the Final Judgment is in the public interest.'' \1\ The
main provision of the proposed Final Judgment is the requirement that
the defendants divest Abitibi-Consolidated's Snowflake, Arizona
newsprint mill in order to settle the Justice Department's Complaint
\2\ enjoining the proposed merger of Abitibi-Consolidated, Inc.
(``Abitibi'') and Bowater, Incorporated (``Bowater'').\3\ Shortly after
the settlement agreement, Abitibi and Bowater completed their merger.
The merged firm is named AbitibiBowater.\4\
---------------------------------------------------------------------------
\1\ Plaintiff United States' Explanation of Consent Decree
Procedures filed with the Court on October 23, 2007 at ] 6.
\2\ The Complaint and proposed Final Judgment were filed with
the Court on October 23, 2007.
\3\ Proposed Final Judgment at pages 5-8.
\4\ Abitibi and Bowater completed their merger on October 29,
2007. AbitibiBowater press release, October 29, 2007.
---------------------------------------------------------------------------
The Newspaper Association of America (``NAA'') is an association
whose membership includes most of the daily and Sunday newspaper
publishers in the United States. NAA represents the newsprint customers
most significantly affected by the merger of Abitibi and Bowater and
the provisions of the proposed Final Judgment.
In its Competitive Impact Statement, the Justice Department asserts
that the divestiture of the Snowflake mill ``would adequately address
the likelihood that the proposed merger substantially would reduce
competition for newsprint in the United States.'' \5\ In its filings on
this matter, including the Competitive Impact Statement and proposed
Final Judgment, the Justice Department provides no information or
analysis to the Court to support or justify this assertion.
---------------------------------------------------------------------------
\5\ Competitive Impact Statement at page 6. The Competitive
Impact statement was also filed with the Court on October 23, 2007.
---------------------------------------------------------------------------
In these Comments, the NAA makes two separate but related arguments
explaining why it believes the Court should reject the Justice
Department's request to approve the proposed Final Judgment without a
hearing. (1) The newly merged AbitibiBowater, despite its agreement to
divest the Snowflake mill, has already begun to exercise the market
power created by the merger to anticompetitively raise newsprint prices
to North American newsprint customers. This post-settlement exercise of
market power by AbitibiBowater shows that the proposed Final Judgment
is not in the public interest. (2) Even without the post-settlement
evidence of anticompetitive conduct by AbitibiBowater, there would
still be ample grounds to reject the proposed remedy. The Justice
Department has not provided the Court with any factual or economic
analysis to demonstrate that the proposed remedy will eliminate the
incentive for AbitibiBowater to reduce industry capacity and raise
prices to North American newsprint customers (the injury charged in the
Complaint). Each argument, standing on its own, provides sufficient
grounds for the
[[Page 32840]]
rejection by the Court of the Justice Department's request to enter the
proposed Final Judgment without a hearing.
If the proposed Final Judgment is entered without modification, the
newly merged AbitibiBowater will have the ability and incentive to
unilaterally engage in anticompetitive conduct to raise newsprint
prices above competitive levels to U.S. daily newspapers and other
North American newsprint customers. The Court should reject the Justice
Department's request to enter the proposed Final Judgment and conduct a
hearing into this matter to determine a remedy sufficient to prevent
the harm to competition and the economic harm to U.S. daily newspapers
and other North American newsprint customers that will otherwise result
from the merger and from the inadequate divestiture remedy as contained
in the proposed Final Judgment.
Analysis of the Competitive Impact of the Merger and the Adequacy of
the Divestiture of the Snowflake Mill
On November 8, 2007, the Justice Department published in the
Federal Register the Proposed Final Judgment resolving a Complaint
filed by the United States to enjoin the merger of Abitibi and Bowater.
The Complaint describes the acquisition as creating a newsprint
producer ``three times larger than the next North American newsprint
producer'' that ``will have the incentive and ability to withdraw
capacity and raise newsprint prices in the North American newsprint
market.'' \6\ Prior to the merger, Abitibi was the largest producer
with 25 percent of the North American newsprint capacity.\7\ With
Bowater's second place share of 16 percent, the combined firm would own
``over 40'' percent of the North American newsprint capacity.\8\ The
Complaint seeks to enjoin the transaction because it will ``provide the
merged firm with an incentive to close capacity sooner than it
otherwise would to raise prices and profit from the higher margins on
its remaining capacity.'' \9\
---------------------------------------------------------------------------
\6\ Complaint at ] 2.
\7\ Complaint at ] 7, 16.
\8\ Complaint at ] 8, 16.
\9\ Complaint at ] 19.
---------------------------------------------------------------------------
Newspaper publishers do not have alternatives to newsprint to turn
to when newsprint prices rise. The Complaint states that ``newspaper
publishers have no close substitutes to use for printing newspapers,''
\10\ and that ``demand for newsprint is highly inelastic to changes in
price.'' \11\ Consequently, if North American newsprint manufacturers
attempted to exercise market power by raising newsprint prices above
competitive levels, U.S. newspaper publishers and other North American
newsprint buyers could not successfully resist that exercise of market
power.\12\ Furthermore, U.S. newspaper publishers and other North
American newsprint buyers would not be able to count on other suppliers
to produce more newsprint or entry by new suppliers to roll back the
price increase. According to the Complaint, ``neither supply responses
nor entry will defeat the exercise of market power.'' \13\
---------------------------------------------------------------------------
\10\ Complaint at ] 10.
\11\ Complaint at ] 11-12.
\12\ In Section 0.1 of the Horizontal Merger Guidelines, the
Justice Department defines the exercise of market power by a seller
or sellers as ``the ability profitably to maintain prices above
competitive levels for a significant period of time.'' 1992
Horizontal Merger Guidelines, U.S. Department of Justice and Federal
Trade Commission, Issued April 2, 1992 and revised April 8, 1997
(``Horizontal Merger Guidelines'' or ``Guidelines''). Available at
http://www.usdoj.gov/atr/public/guidelines/hmg.htm.
\13\ Complaint at ] 20-26.
---------------------------------------------------------------------------
In recent years, the U.S. newspaper industry has experienced
declining circulation and advertising revenue. As a result, North
American demand for newsprint has also declined, leading to excess
newsprint capacity. The decline in newsprint demand is projected to
continue.\14\ In such circumstances, newsprint prices would ordinarily
be expected to also decline. According to the Complaint, however, the
merger will give the merged firm both the incentive and ability to
strategically close enough capacity to raise newsprint prices above
competitive levels.\15\ The Complaint also concludes that absent the
merger, neither Abitibi nor Bowater as separate firms would have the
incentive or ability to strategically close capacity to raise newsprint
prices.\16\ In the words of the Justice Department, the ``merger will
substantially lessen competition in the production and sales of
newsprint,'' with the result that ``prices charged for newsprint in
North America likely will increase.'' \17\
---------------------------------------------------------------------------
\14\ Complaint at ] 17.
\15\ Complaint at ] 2-3, 16.
\16\ Complaint at ] 18.
\17\ Complaint at ] 3, 16, 28(c).
---------------------------------------------------------------------------
In order to remedy the anticompetitive effects that the Justice
Department concluded would otherwise result from the merger, the
Department obtained the agreement of Abitibi and Bowater to divest
Abitibi's Snowflake, Arizona newsprint mill.\18\ In the Competitive
Impact Statement, the Justice Department asserts that ``[w]ithout
Snowflake's capacity, the merged firm would not be of sufficient size
to be able to recoup the losses from such strategic closures through
increases in prices on its remaining newsprint production. The
divestiture of Snowflake would adequately address the likelihood that
the proposed merger substantially would reduce competition for
newsprint in the United States.'' \19\ The Snowflake mill accounts for
about 3 percent of North American newsprint capacity.\20\ Thus, the
Justice Department is claiming that with a newsprint capacity share of
about 40 percent, the merged firm would have the incentive and ability
to unilaterally exercise market power to raise newsprint prices above
competitive levels but that with a slightly smaller capacity share of
37 percent the merged firm would not have the incentive and ability to
unilaterally exercise market power. The Justice Department provides the
Court with no data or analysis in support of these assertions.
---------------------------------------------------------------------------
\18\ Proposed Final Judgment at pp. 5-8, Competitive Impact
Statement at pp. 8-11.
\19\ Competitive Impact Statement at p. 6.
\20\ Neither the Proposed Final Judgment nor the Competitive
Impact Statement provides the North American newsprint capacity
share of the Snowflake mill. At page 2, the Competitive Impact
Statement states that the annual newsprint capacity of the Snowflake
mill is 375,000 metric tonnes, which would be about 3 percent of
current annual North American newsprint capacity of about 11.7
million metric tonnes based on November 2007 newsprint statistics
provided by the Pulp and Paper Products Council.
---------------------------------------------------------------------------
The Justice Department's prediction that the Snowflake divestiture
would be sufficient to eliminate the incentive and ability of the
merged firm to exercise market power by strategically removing
newsprint capacity from the market to raise the price of newsprint has
already been proven wrong. North American newsprint producers,
including Abitibi and Bowater, had been trying to implement a $25 per
tonne price increase since September of this year. Until November,
newspaper publishers were successful in resisting the price
increase.\21\ On November 29, a little more than five weeks after the
agreement to divest the Snowflake mill, the newly combined
AbitibiBowater announced that it would remove about 600,000 metric
tonnes of newsprint capacity from the North American market,
representing about 5 percent of North American newsprint capacity.\22\
[[Page 32841]]
In conjunction with the capacity closures, AbitibiBowater initiated a
newsprint price increase of $60 per metric tonne to be implemented in
three $20 per metric tonne monthly increments beginning in January
2008. Most North American newsprint manufacturers quickly joined the
$60 per metric tonne price initiated by AbitibiBowater.\23\ Also, as a
result of AbitibiBowater' s announced newsprint capacity closures of
600,000 metric tonnes, the previously stalled $25 per metric tonne
price hike has been successfully implemented by North American
newsprint manufacturers. As described in the trade press, ``[p]ublisher
resistance to $25/tonne North American newsprint increase collapse[d]''
and the price hike went in ``like a hot knife through butter,'' \24\
Combined, these two price increases will raise the price of newsprint
by $85 per metric tonne or about 15 percent over the October 2007 price
of $560 per metric tonne.\25\ As RISI economist Kevin Conley concluded,
``AbitibiBowater's capacity closures will obviously provide the upward
pressure for an extended price recovery in 2008, as operating rates
soar past the magic 95% threshold generally needed for prices to
rise.'' \26\
---------------------------------------------------------------------------
\21\ Publisher resistance to $25/tonne North American newsprint
increase collapses; producers looking to fast track recovery, 29
Pulp & Paper Week 48 (Dec. 17, 2007) at 1.
\22\ AbitibiBowater plans to shut down one million tonnes/yr of
capacity in 1Q; expects more closures could follow in 2Q, 29 Pulp &
Paper Week 46 (Dec. 3, 2007) at 1. A capacity closure of 600,000
metric tonnes would be about 5 percent of current annual North
American newsprint capacity of about 11.7 million metric tonnes
based on November 2007 newsprint statistics provided by the Pulp and
Paper Products Council. In addition to announcing the removal of
600,000 metric tonnes of newsprint capacity from the market,
AbitibiBowater also announced the closure of about 400,000 metric
tonnes of commercial printing paper capacity.
\23\ Most North American newsprint makers join $60/tonne 1Q 2008
hike, 29 Pulp & Paper Week 46 at 2.
\24\ 29 Pulp & Paper Week 48 at 1.
\25\ Generally, if a merger creates market power resulting in a
price increase of 5 percent or more, that price increase is
considered to be ``significant.'' In Section 1.11 of its Merger
Guidelines, the Justice Department states that in defining the
relevant markets affected by a merger in most contexts it ``will use
a price increase of five percent lasting for the foreseeable
future.'' Horizontal Merger Guidelines at Sec. 1.11. The October
2007 North American newsprint price is from 29 Pulp & Paper Week 45
(Nov. 19, 2007) at 3.
\26\ Newsprint giant AbitibiBowater embraces industry
leadership, eyes $200/tonne North American newsprint price increase,
29 Pulp & Paper Week 47 at 5.
---------------------------------------------------------------------------
The combined AbitibiBowater is seeking to ``leverage the North
American (newsprint) price up to the price in Europe and not the other
way around,'' according to AbitibiBowater President and CEO David
Paterson.\27\ If AbitibiBowater is successful in ``leveraging'' the
North American newsprint price up to the price of newsprint in Europe,
that will result in a $200 per metric tonne price increase or about 36
percent over the North American price of $560 per metric tonne in
October 2007.\28\ At the time AbitibiBowater announced the removal of
600,000 metric tonnes of newsprint capacity from the market, it also
announced that ``more mills could close in Canada later [in 2008].''
\29\ Based on these statements and other statements by AbitibiBowater
executives and past and current actions by AbitibiBowater and its
predecessor companies, it is very likely that AbitibiBowater will close
additional capacity in 2008 to ``leverage'' the North American
newsprint price up to the newsprint price in Europe.
---------------------------------------------------------------------------
\27\ 29 Pulp & Paper Week 47 at 1.
\28\ Id. at 1, ``Newsprint prices in Europe were close to $200/
tonne higher than in the USA in November.''
\29\ 29 Pulp & Paper Week 46 at 1.
---------------------------------------------------------------------------
These post-settlement actions by AbitibiBowater show that the
Justice Department severely underestimated the risk that the merger
posed to competition in the North American newsprint market and
severely underestimated the incentive and ability of the merged firm to
remove capacity from the market to raise the price of newsprint well
above competitive levels. It is evident that a significantly larger
divestiture is required to prevent the substantial anticompetitive
price increases that are already occurring and will continue to occur
as a result of the merger.
NAA Represents the Newsprint Customers Most Significantly Affected by
AbitibiBowater's Exercise of Market Power
These comments are timely submitted pursuant to the Antitrust
Procedures and Penalties Act, 15 U.S.C. 16(b)-(e) (known as the
``Tunney Act''), on behalf of the Newspaper Association of America
(``NAA''). NAA members are the primary purchasers of newsprint. NAA has
approximately 2,000 members, representing a broad range of newspaper-
related companies ranging from independent, small market, and family
owned publishers to the large newspaper chains. These members account
for approximately 90 percent of the paid daily and Sunday newspaper
circulation in the United States. U.S. daily newspapers are the primary
purchasers of newsprint produced by North American newsprint mills and
account for about 80 percent of the newsprint consumed in the U.S. and
about 70 percent of the newsprint consumed in North America.
Newsprint is an essential and irreplaceable input for newspapers.
Because newsprint is second only to labor as a cost for newspapers,
higher newsprint prices have a direct impact on the ability of
newspaper companies to serve their customers, newspaper readers and
newspaper advertisers. When confronted with newsprint price increases,
newspapers are forced to restrict their use of newsprint by reducing
their circulation, withdrawing from more distant geographic areas,
ending editions, and reducing the size and number of pages published.
The impact of these changes adversely impacts the interest of the
public, with less news available in print to the millions of newspaper
readers and less information available in print for the electorate. At
price levels equal to the prevailing prices in Europe, $200 per tonne
above the pre-settlement October 2007 price, some newspapers will be
unprofitable and at risk of failure.
This memorandum and the attached Economic Analysis \30\ are
submitted as a comment on the Justice Department's Competitive Impact
Statement and proposed Final Judgment settling the proposed merger of
Abitibi and Bowater. The Economic Analysis addresses, in particular,
the inadequacy of the Snowflake divestiture to prevent the competitive
harm from the merger that is identified in both the Complaint and
Competitive Impact Statement. The attached Economic Analysis references
``An Economic Analysis of Competitive Effects of the Proposed Abitibi-
Bowater Merger'' (``White Paper'') and two Supplements to the White
Paper, which were provided to the Justice Department during its
investigation of the merger. The White Paper and two Supplements, which
are attached to the Economic Analysis, address the recent history of
anticompetitive conduct by Abitibi and Bowater and explain why a merger
of Abitibi and Bowater, if permitted, would lead to a continuation of
that anticompetitive conduct. Also cited throughout the Comment are
trade press articles relating to post-settlement newsprint capacity
removals announced by Abitibi-Bowater and resulting price increases,
which are attached to this Comment.\31\
---------------------------------------------------------------------------
\30\ See ``An Economic Analysis of the Adequacy of the Snowflake
Divestiture in the Settlement of United States of America v.
Abitibi-Consolidated, Inc. and Bowater, Incorporated.''
\31\ See Attachment A: Trade Press Articles Relating to Post-
Settlement Newsprint Capacity Removals Announced by AbitibiBowater
and Resulting Newsprint Price Increases.
---------------------------------------------------------------------------
NAA members are the primary victims that the Complaint identifies
as suffering competitive injury from the transaction and on whose
behalf the Government seeks relief. NAA agrees with the Justice
Department that the alleged harm to competition identified in the
Complaint is accurate,
[[Page 32842]]
demonstrable, and unless adequately remedied, will cause significant
economic harm to the U.S. newspaper industry. Indeed, NAA and its
members produced documents, economic analyses, and other information to
the Justice Department demonstrating the recent anticompetitive pricing
and output history of the North American newsprint industry resulting
from the joint dominant firm behavior of Abitibi and Bowater and
showing how the proposed transaction would permit a merged
AbitibiBowater to continue to strategically close capacity to raise
newsprint prices well above competitive levels.
But while the Complaint correctly identifies the competitive harm
produced by the merger, the remedy in the proposed Final Judgment fails
to satisfy even the most deferential standard for Tunney Act review.
The Justice Department has not provided the Court with any factual or
economic analysis to demonstrate that the proposed remedy will
eliminate the incentive to reduce industry capacity and raise prices to
North American newsprint customers (the injury charged in the
Complaint). Recent events have already proven that the remedy set forth
in the proposed Final Judgment is woefully inadequate to prevent the
injury charged in the Complaint. Hence, reviewing the remedy ``in
relationship to the violations that the United States has alleged in
its Complaint,'' \32\ and deferring to the Justice Department to
whatever extent is required by law, the remedy does not provide any
basis to allow the Court to find that it will ameliorate the harm
alleged in the Complaint. This is not a case in which there is a debate
as to whether the Justice Department inappropriately narrowed the
alleged harm. Rather, this is the case in which the economics and
recent history of the newsprint industry, along with the Justice
Department's conclusions regarding the competitive harm created by the
consolidation, compel the conclusion that the remedy is not a
``reasonably adequate remed[y] for the alleged harms.'' \33\
---------------------------------------------------------------------------
\32\ This is the standard the Justice Department claims is ``the
Court's role under the APPA.'' Competitive Impact Statement, at
Section VII.
\33\ This is the standard that the Justice Department contends
it must meet for approval of the decree: ``the United States `need
only provide a factual basis for concluding that the settlements are
reasonably adequate remedies for the alleged harms.' '' id., citing
SBC Commc'ns, 489 F. Supp. 2d at 17.
---------------------------------------------------------------------------
The Proper Standard of Review for the Justice Department's Proposed
Remedy for This Merger
``The antitrust laws [* * *] were enacted for the protection of
competition, not competitors.'' \34\ This means that antitrust remedies
are designed to restore competition to the market, not to ensure
profits to the competitors in that industry.\35\ Since the Supreme
Court accepted this notion first proposed by Congress, antitrust law
enforcement has been guided by this principle. Since these Supreme
Court decisions and Congressional mandates, antitrust law and its
regulators have sought to preserve competition ``in the public
interest.'' \36\ The divestiture of the Snowflake mill is a remedy that
fails to preserve competition in the North American newsprint market
and is, therefore, not in the public interest.
---------------------------------------------------------------------------
\34\ Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477,
488; 97 S. Ct. 690,712; 50 L. Ed. 2d 701 (1977), citing Brown Shoe
Co. v. United States, 370 U.S. 294, 320 (1962).
\35\ See Brunswick, 429 U.S. at 487-88 (In that case, the court
would not grant relief to Respondents for profits that the
Respondents would have gained had the acquired party exited the
industry).
\36\ ``In the public interest'' is the standard for entry of
proposed Final Judgments under the Tunney Act. 15 U.S.C. Sec. 1
6(e)(1). Congress mandated considerations for determining whether a
decree is in the public interest, but never defined the term, ``in
the public interest'' itself. NAA believes that it is safe to assume
that achieving the goals of the antitrust laws--including preserving
competition--is ``in the public interest.''
---------------------------------------------------------------------------
As is discussed in the attached Economic Analysis, the economic
model appropriate to evaluate the current merger as well as prior
anticompetitive conduct by Abitibi and Bowater is the dominant firm
model.\37\ The description of the anticompetitive effects of the merger
contained in both the Complaint and the Competitive Impact Statement
suggests that the Justice Department applied the dominant firm model in
its analysis of the merger.\38\ The Merger Guidelines Commentary of the
Justice Department and the Federal Trade Commission describes the
dominant firm model as follows:
---------------------------------------------------------------------------
\37\ See Section B.1., ``Unilateral Effects and the Dominant
Firm Model,'' and Appendix A, ``Merger Analysis, Unilateral Effects,
and the Dominant Firm Model.''
\38\ Complaint at ] 16-19 and Competitive Impact Statement at
pp. 5-6.
[The dominant firm] model posits that all competitors but one in
an industry act as a ``competitive fringe,'' which can economically
satisfy only part of total market demand. The remaining competitor
acts as a monopolist with respect to the portion of total industry
demand that the competitive fringe does not elect to supply. This
model might apply, for example, in a homogeneous product industry in
which the fringe competitors are unable to expand output
significantly.\39\
---------------------------------------------------------------------------
\39\ Commentary on the Horizontal Merger Guidelines, U.S.
Department of Justice and Federal Trade Commission, March 2006
(``Guidelines Commentary''), at p. 25.
In the Competitive Impact Statement, the Justice Department claims that
the divestiture of the Snowflake mill will be sufficient to eliminate
the incentive for AbitibiBowater to act as a dominant firm.\40\
However, the large post-settlement capacity closures accompanied by a
large price increase initiated by AbitibiBowater shortly after the
Justice Department's settlement demonstrate that AbitibiBowater has the
incentive and ability to act as a dominant firm and will likely retain
that incentive and ability for future strategic capacity closures.
---------------------------------------------------------------------------
\40\ Competitive Impact Statement at p. 6.
---------------------------------------------------------------------------
One consequence of AbitibiBowater's incentive and ability to act as
the dominant firm in the North American newsprint market is that the
merged firm will likely close at least some capacity that is more
efficient than some of the capacity of the fringe firms.\41\ The nature
of the dominant firm model is that in closing capacity to raise the
industry operating rate and newsprint prices, the dominant firm allows
the fringe firms to operate at full capacity enjoying the price
increasing benefits of AbitibiBowater's dominant firm behavior. Indeed,
once they are at full capacity, the fringe firms would have no
incentive to do anything other than to follow the price leadership of
the dominant firm. Thus, in a declining market, such as the North
American newsprint market, it is likely that some inefficient fringe
firm capacity is preserved, which, in the absence of dominant firm
behavior, would otherwise have to close as the price of newsprint
dropped below the cash costs of operating the inefficient fringe
capacity.
---------------------------------------------------------------------------
\41\ During the period 2002 to 2006, very little newsprint
capacity was removed from the market by fringe firms as Abitibi and
Bowater were responsible for the great majority of the North
American newsprint capacity closures during this period. See the
discussion of Abitibi's and Bowater's prior joint anticompetitive
conduct below and in the attached Economic Analysis, Section B.2,
``Abitibi and Bowater Engaged in Joint Dominant Firm Behavior to
Raise NA Newsprint Prices Significantly above Competitive Levels
2002 to 2006,'' which also contains references to the relevant
portions of the White Paper and the Supplements to the White Paper.
---------------------------------------------------------------------------
For instance, Pulp & Paper Week reported that newsprint industry
analyst Claudia Shank of JP Morgan believes that AbitibiBowater's
announced capacity closures for the first quarter of 2008 ``together
with Abitibi-Bowater's
[[Page 32843]]
indication that it could cut more capacity in mid-2008, provided
second- and third-tier producers some additional `breathing room' and
limit closures from the broader industry before the second half of next
year.'' \42\ According to RISI economist Kevin Conley, ``[w]ithout
AbitibiBowater's bold move [to remove 600,000 metric tonnes of
newsprint capacity from the market] operating rates and prices would
have continued to languish at low levels until the highest-cost mills
could no longer survive, eventually leading to the inevitable closures
needed to balance the North American market.'' \43\ Even after the
competitive ``balancing'' of the North American newsprint market,
however, the prevailing newsprint price would be the competitive price,
not the much higher anticompetitive prices resulting from
AbitibiBowater's current and likely future strategic newsprint capacity
closures.
---------------------------------------------------------------------------
\42\ 29 Pulp & Paper Week 46 at 5.
\43\ 29 Pulp & Paper Week 47 at 5.
---------------------------------------------------------------------------
On the other hand, if the Justice Department had successfully
blocked this merger, a separate Abitibi and Bowater would likely have
considerably less incentive and ability to engage in joint dominant
behavior than the current merged AbitibiBowater.\44\ The principal
effect of the merger is that U.S. newspaper publishers and other North
American newsprint customers directly bear the cost of the dominant
firm behavior in the form of significantly higher newsprint prices. As
a secondary effect of the merger, it is likely some inefficient fringe
firm capacity may be preserved by AbitibiBowater's dominant firm
behavior. The misallocation of resources that likely results imposes a
social cost on the economy that is inconsistent with the goals of the
antitrust laws.
---------------------------------------------------------------------------
\44\ According to p. 6 of the CIS, ``But for the merger, neither
Defendant acting alone would be of sufficient size to profitably
increase the price of newsprint by reducing its own output through
strategically closing, idling, or converting its capacity.''
---------------------------------------------------------------------------
The basic premise of the antitrust laws is to protect competition
and consumers, not competitors. As interpreted by the courts and by
Congress, the antitrust laws are not intended to protect inefficient
suppliers to a market. Because the Justice Department is asking the
Court to enter a proposed consent decree that would provide no remedy
for the customer-victims of AbitibiBowater's dominant firm behavior and
that would likely permit the survival of inefficient capacity of fringe
firm competitors that would otherwise be forced to close down in a
competitive newsprint market, the Justice Department has an obligation
to explain the basis for its decision to the Court. By asking the Court
to accept with no further analysis or explanation the Department's
claim that the Snowflake divestiture will remedy the competitive harms
alleged in the Complaint, the Department puts the Court in the position
of having no basis upon which to determine if the proposed remedy (a)
is adequate to address these competitive problems, (b) is consistent
with the Justice Department's own prior positions, or (c) is in
accordance with the well established standards of the antitrust laws,
all of which are relevant to the determination of ``public interest.''
The Complaint and Competitive Impact Statement Ignore Abitibi's and
Bowater's Recent History of Anticompetitive Conduct Prior to Their
Merger Announcement
As is discussed above, shortly after Abitibi and Bowater reached
their agreement with the Justice Department in October to divest the
Snowflake mill and settle the case, the newly merged firm proceeded to
announce significant capacity closures and to initiate a substantial
price increase. Most other North American newsprint manufacturers
quickly matched AbitibiBowater's announced price increase.
During and immediately prior to the period when the merger was
being reviewed by the Justice Department,\45\ newsprint prices steadily
declined from $675 per metric tonne to $560 per metric tonne, a decline
of about 17 percent. Also, during this time, Abitibi and Bowater did
not take strategic actions to raise the price of newsprint. As
discussed immediately below, Abitibi and Bowater had engaged in joint
dominant firm behavior to strategically close capacity to raise the
price of newsprint well above competitive levels over the period 2002
to 2006. There are two plausible explanations as to why Abitibi and
Bowater did not continue their joint dominant firm behavior during and
immediately prior to the Department's merger review: (1) Abitibi and
Bowater determined, due to the extent of previous capacity closures
that occurred between 2002 and 2006, that their ability and incentive
to jointly engage in dominant firm behavior had been significantly
diminished, thus leading to their decision to merge; and (2) Abitibi
and Bowater decided it would be imprudent to attempt to exercise market
power during the merger review period as it might adversely affect the
outcome of that review.
---------------------------------------------------------------------------
\45\ Abitibi and Bowater announced their merger on January 29,
2007. Presumably, the Justice Department began their review of the
merger shortly after the merger announcement and continued their
investigation until the filing of the Complaint, Competitive Impact
Statement, and proposed Final Judgment on October 23, 2007.
---------------------------------------------------------------------------
Between 2002 and 2006, the pricing analysis in the White Paper
demonstrates that Abitibi and Bowater jointly acted as a dominant firm,
strategically removing newsprint capacity from the market to
significantly raise the newsprint industry operating rate, and, thus,
increasing the price of newsprint above competitive levels. Due to
these strategic capacity closures, the price of newsprint during that
period increased by a total of 49 percent despite a steady decline in
consumption by North American newsprint customers. The economic White
Paper and the two Supplements, presented to the Justice Department
during the course of its investigation, extensively document and
analyze this joint dominant firm behavior by Abitibi and Bowater.\46\
The prior anticompetitive actions of Abitibi and Bowater to close
capacity strategically during this four-year period are identical to
the anticompetitive strategic behavior alleged in ] 2 and ] 19 of the
Complaint and described on page 6 of the Competitive Impact Statement.
Since the Complaint and Competitive Impact Statement contain no
references to this prior anticompetitive conduct by Abitibi and
Bowater, it is impossible for the Court to determine if and how much of
a factor the prior anticompetitive conduct played in the Justice
Department's evaluation and settlement of this merger.
---------------------------------------------------------------------------
\46\ See Section B.2. of the attached Economic Analysis.
---------------------------------------------------------------------------
Earlier mergers in the North American newsprint industry,
especially the Abitibi-Donohue merger in 2000 and the Bowater-Alliance
merger in 2001, created both the incentive and ability for Abitibi and
Bowater to jointly engage in this anticompetitive conduct. Economic
analysis in papers and presentations by representatives of NAA and the
U.S. newspaper industry submitted to the Justice Department in 2000 and
2001 forecasted that these two mergers, if not challenged, would have
significant anticompetitive results. The Justice Department took no
action against either of these two earlier mergers and, as predicted by
the economic analyses submitted to the Department, the two mergers
enabled Abitibi and Bowater to engage in the anticompetitive conduct
that occurred
[[Page 32844]]
between 2002 and 2006. As a result, U.S. newspapers and other North
American newsprint customers incurred significantly higher newsprint
prices.
Prior anticompetitive conduct is a highly relevant factor in most
merger investigations, according to the Guidelines Commentary:
Facts showing that rivals in the relevant market have
coordinated in the past are probative of whether a market is
conducive to coordination. Guidelines Sec. 2.1. Such facts are
probative because they demonstrate the feasibility of coordination
under past market conditions. Other things being equal, the removal
of a firm via merger, in a market in which incumbents already have
engaged in coordinated behavior, generally raises the risk that
future coordination would be more successful, durable, or
complete.\47\
---------------------------------------------------------------------------
\47\ Guidelines Commentary at p. 22.
The Complaint, Competitive Impact Statement, and Proposed Final
Judgment do not contain any explanation by the Justice Department as to
what, if any, consideration was given to the evidence of Abitibi's and
Bowater's prior joint anticompetitive conduct. Before determining
whether the proposed relief ``is in the public interest,'' the Court is
entitled to know whether the Justice Department considered evidence of
prior anticompetitive conduct and if not, why not. By failing to
provide that evidence in its Court filings, the Justice Department has
deprived the Court of information vital to its review of the adequacy
of the proposed divestiture.
The Competitive Impact Statement and Proposed Final Judgment Fail To
Address the Congressional Mandates of the Tunney Act
As previously noted, the Tunney Act requires that a court determine
whether entry of the proposed Final Judgment ``is in the public
interest.'' \48\ As the Justice Department outlines more thoroughly in
its Competitive Impact Statement, the Court is required to consider
certain factors in making that determination.\49\ Among those
considerations mandated by Congress are: (1) ``The competitive impact
of such judgment, including * * * anticipated effects of alternative
remedies actually considered,'' and (2) the ``impact of entry of such
judgment upon competition in the relevant market or markets.'' \50\
While evidence of other factors upon which the Court is asked to base
its decision are certainly lacking, these two points are noticeably
deficient.
---------------------------------------------------------------------------
\48\ 15 U.S.C. Sec. 16(e)(1).
\49\ Competitive Impact Statement at VII, citing 15 U.S.C. Sec.
16(e)(1)(A)-(B), United States v. SBC Commc'ns, Inc., 489 F. Supp.
2d 1, 11 (D.D.C. 2007).
\50\ 15 U.S.C. 16(e)(1)
---------------------------------------------------------------------------
The Anticipated Competitive Effects of Alternative Remedies Actually
Considered by the Justice Department
The Justice Department lists two alternative remedies to the one it
chose: (1) A full trial on the merits, and (2) ``a number of
divestiture alternatives.'' \51\ After considering other options, the
Justice Department ``determined that divestiture of the Snowflake mill,
under the circumstances, was the best solution given the size and
efficiency of the Snowflake mill.'' \52\ Other than noting that
Snowflake is ``among the largest and most profitable mills in the
United States,'' the Justice Department provided no further explanation
for its decision that Snowflake was both a sufficient remedy and the
best solution, no detail regarding under what ``circumstances'' this
conclusion was reached, and no scale against which it measured
Snowflake as the best alternative. The Justice Department leaves the
Court entirely in the dark as to what other divestitures it considered
and why those were inferior to the divestiture of Snowflake. The
Justice Department also failed to note why Snowflake alone--without an
additional divestiture--was sufficient. While a detailed rank or
scoring of each of the remedies the Justice Department considered may
not be necessary, the Justice Department here has left the Court
entirely in the dark with absolutely no basis for making a meaningflul
comparison between a Snowflake-only divestiture and any alternative
course of action, including a full trial on the merits.
---------------------------------------------------------------------------
\51\ Competitive Impact Statement at VI.
\52\ Id.
---------------------------------------------------------------------------
Critically, the Justice Department also failed to account for the
actual ``anticipated effects'' of the alternatives. Determining
``anticipated effects,'' such as whether a transaction will result in
one firm having the unilateral power to profitably raise prices or
close capacity without being restrained by other competitors in the
market, or whether a transaction will result in the market becoming
more conducive to competitors coordinating on price, is the essential
element of any merger investigation. Yet, here, even though the Court
is required to consider it, the Justice Department remains silent. How
can the Court determine if the Justice Department chose an acceptable
alternative as opposed to one so weak as to provide no meaningful
relief? Is the Court expected to take on faith that this alternative is
a viable one? The Court is given no support that would assist it in
reaching a conclusion that the Justice Department's chosen alternative
is in the public interest. If the recent actions by AbitibiBowater are
placed on the scale, the Justice Department's silence fails to meet any
reasonable burden of proof to establish that its chosen alternative is
sufficient to meet the standard that the proposed remedy is ``in the
public interest.''
The Impact of the Proposed Final Judgment in the Relevant Market
The divestiture required under the proposed Final Judgment fails to
restore the competition lost by the combination of North America's two
largest newsprint producers.
The Justice Department has an obligation to explain to the Court
why the remedy it proposes restores or preserves competition. The
formal policy guidance of the Antitrust Division regarding merger
remedies is contained in the Antitrust Division Policy Guide to Merger
Remedies.\53\ In this policy statement, the Antitrust Division sets
forth broad principles that it says guide its decisions to seek
remedies to offset potential harms to competition from mergers. A
controlling policy principle is that ``restoring competition is the
`key to the whole question of antitrust remedy.' '' \54\
---------------------------------------------------------------------------
\53\ Antitrust Division Policy Guide to Merger Remedies, U.S.
Department of Justice, Antitrust Division, October 2004. Available
at http://www.usdoj.gov/atr/public/guidelines/205108.htm.
\54\ Id., citing United States v. E.I. du Pont de Nemours & Co.,
366 U.S. 316, 326 (1961). Ford Motors Co. v. United States, 405 U.S.
562, 573 (1972) (``relief in an antitrust case must be effective to
redress the violations and `to restore competition' * * * '').
---------------------------------------------------------------------------
The Horizontal Merger Guidelines ``describe the analytical
framework and specific standards normally used by the [Justice
Department] in analyzing mergers.'' \55\ While the Complaint and
Competitive Impact Statement do not directly reference the Guidelines,
absent a disclaimer from the Justice Department, the Court can fairly
assume the Department followed its own Guidelines in its investigation
of this merger.
---------------------------------------------------------------------------
\55\ 1992 Horizontal Merger Guidelines, U.S. Department of
Justice and Federal Trade Commission, Issued April 2, 1992 and
revised April 8, 1997 (''Guidelines''). Available at http://
www.usdoj.gov/atr/public/guidelines/hmg.htm. The Guidelines also say
that, ``By stating its policy as simply and clearly as possible, the
[Justice Department] hopes to reduce the uncertainty associated with
enforcement of the antitrust laws in this area.'' Id. The ``unifying
theme of the Guidelines,'' like the Merger Remedy Policy noted
above, ``is that mergers should not be permitted to create or
enhance market power or to facilitate its exercise.'' Id at Sec.
0.1.
---------------------------------------------------------------------------
The Guidelines identify two analytical frameworks for assessing
[[Page 32845]]
whether a merger between competing firms may substantially lessen
competition. Those frameworks require the Justice Department to ask
whether the merger may increase market power by facilitating
coordinated interaction among rival firms (``coordinated effects'') and
whether the merger may enable the merged firm to raise price
unilaterally or otherwise exercise market power (``unilateral
effects'').\56\ Though the Justice Department provides the Court with
no indication of what framework it applied or why, the allegations in
the Complaint appear to be consistent with the application of the
unilateral effects framework.
---------------------------------------------------------------------------
\56\ Guidelines Commentary at p. 17.
---------------------------------------------------------------------------
A merger may diminish competition because the ``merging firms may
find it profitable to alter their behavior unilaterally following the
acquisition by elevating price and suppressing output.'' \57\ How a
merger generates anticompetitive unilateral effects is relatively
straightforward: ``The merger provides the merged firm a larger base of
sales on which to enjoy the resulting price rise and also eliminates a
competitor to which customers otherwise would have diverted their
sales.'' \58\
---------------------------------------------------------------------------
\57\ Merger Guidelines at Sec. 2.2.
\58\ Merger Guidelines at Sec. 2.22.
---------------------------------------------------------------------------
The Complaint states that the combined post-merger share of
newsprint held by AbitibiBowater is ``over 40 percent.'' \59\ The
Complaint also states that ``neither supply responses nor entry will
defeat an exercise of market power.'' \60\ The Complaint further states
that ``[t]he proposed transaction would combine Defendants' large share
of newsprint capacity, thereby expanding the quantity of newsprint
sales over which the merged firm would benefit from a price increase.
This would provide the merged firm with an incentive to close capacity
sooner than it otherwise would to raise prices and profit from the
higher margins on its remaining capacity.'' \61\
---------------------------------------------------------------------------
\59\ Complaint at ] 16.
\60\ Complaint at ] 20-26.
\61\ Complaint at ] 19.
---------------------------------------------------------------------------
Given these market circumstances, which are highly conducive to the
unilateral exercise of market power, the Justice Department fails to
explain to the Court why the divestiture of just the Snowflake mill
will be sufficient to prevent the merged firm from exercising market
power. As noted above, the Snowflake mill represents only 3 percent of
North American newsprint capacity. The divestiture of the Snowflake
mill would reduce AbitibiBowater's North American newsprint capacity
share from about 40 percent to about 37 percent. The Justice Department
fails to explain to the Court how reducing AbitibiBowater's capacity
share from 40 percent to a slightly smaller share of 37 percent, a
difference of 3 percent, will be sufficient to restore the market to
competitive conditions. In the absence of a convincing explanation, the
Court should reach the conclusion that the Justice Department's
assertion that the divestiture of the Snowflake mill will be sufficient
to prevent unilateral anticompetitive conduct by AbitibiBowater is
simply wrong.
A Previous Application of the Guidelines by the Justice Department to a
Comparable Paper Industry Merger Resulted in a Much Larger Divestiture
Than the Department Has Proposed for This Merger
In the Justice Department's November 2000 challenge to Georgia-
Pacific's proposed acquisition of Fort James Corporation, the two
parties were the two largest producers of ``away-from-home'' tissue
products. Georgia-Pacific's capacity share of ``away-from-home'' parent
tissue rolls was 11 percent and Fort James' capacity share was 25
percent. The combined share of the two companies in the ``away-from-
home'' parent tissue roll market would have been 36 percent. The
Justice Department challenged the merger using the same basic theory
applied here--unilateral effects. The Justice Department's
investigation revealed that the industry was operating at nearly full
capacity, that the capacity could not be quickly expanded, and that
demand for parent rolls was relatively inelastic with respect to price.
These factors combined to create the likelihood that, after the merger,
Georgia-Pacific would act as a dominant firm by restricting output of
parent rolls and thereby forcing up prices for away-from-home tissue
products. As a result, the Justice Department settled the case by a
consent decree requiring the complete divestiture of Georgia-Pacific's
parent tissue roll capacity share of 11 percent.\62\
---------------------------------------------------------------------------
\62\ See Competitive Impact Statement describing DOJ's Complaint
and settlement of the proposed Georgia-Pacific/Fort James merger at
pp. 8-10. For copies of the DOJ's Complaint and Competitive Impact
Statement in this matter see the Justice Department Web site at
http://www.usdoj.gov/atr/cases/indx276.htm.
---------------------------------------------------------------------------
Nothing in the Competitive Impact Statement for the AbitibiBowater
merger explains or even suggests to the Court why a divestiture
comparable to that in the Georgia-Pacific/Fort James merger is not
required for this merger.\63\ AbitibiBowater's post-merger actions have
already shown that the divestiture remedy proposed by the Justice
Department for this merger will not prevent the exercise of market
power.
---------------------------------------------------------------------------
\63\ The complete divestiture of Georgia-Pacific's pre-
acquisition capacity share reduced Georgia-Pacific's post-
acquisition parent tissue roll capacity share to 25 percent. With
respect to the Abitibi-Bowater merger, a comparable divestiture
would reduce the combined pre-merger newsprint capacity share of
``over 40 percent'' to 25 percent.
---------------------------------------------------------------------------
The Justice Department's action in the Georgia-Pacific/Fort James
merger strongly suggests that significantly more capacity needs to be
divested by AbitibiBowater to ensure that the merged firm will not have
the incentive and ability to unilaterally exercise market power.
Conclusion
U.S. newspaper publishers, the primary victims who will bear the
cost of the conduct challenged in the Complaint and the inadequate
Snowflake mill divestiture, see the proposed divestiture as ineffective
and inadequate. The Justice Department has not provided the Court with
sufficient information with which the Court can enter an informed
judgment that the remedy proposed by the Justice Department is ``in the
public interest.'' Furthermore, events subsequent to the Justice
Department's settlement of the Abitibi-Bowater merger have already
demonstrated that the proposed Final Judgment does not remedy the
public interest harms presented to the Court in the Complaint.
The Court should not enter the proposed Final Judgment. NAA
requests that the Court conduct a hearing to determine the amount of
divestiture sufficient to prevent the anticompetitive effects that will
otherwise result from this merger and the inadequate proposed Final
Judgment.
Submitted on behalf of the Newspaper Association of America by Alan
L. Marx, King & Ballow, Union Street Plaza 1100, 315 Union Street,
Nashville, TN 37201, (615) 259-3456, amarx@kingballow.com. January 2,
2008.
[[Page 32846]]
Attachment A--Trade Press Articles Relating to Post-Settlement
Newsprint Capacity Removals Announced by AbitibiBowater and Resulting
Newsprint Price Increases
Pulp & Paper Week
Dec. 3, 2007 [bond] Vol. 29, No. 46
AbitibiBowater Plans to Shut Down One Million Tonnes/yr of Capacity in
1Q; Expects More Closures Could Follow in 2Q
AbitibiBowater unveiled the first phase of its long-awaited
post-merger rationalization plan and announced the closure of four
money-losing mills in Canada in the first quarter 2008. A total of
600,000 tonnes/yr of newsprint capacity and 400,000 tonnes/yr of
commercial printing papers will be removed.
AbitibiBowater said more mills could close in Canada later next
year, and added that it wanted to reopen its Canadian union
contracts to ``explore ways to reduce overall labor costs and
provide enhanced flexibility in the workplace.'' Salaried employees
would also be asked to take cuts.
Under what it called ``phase one of an action plan to address
company challenges,'' AbitibiBowater will permanently close its
Belgo mill in Shawinigan, QC, and Dalhousie, NB, mill, and
indefinitely idle its Donnacona, QC, and Mackenzie, BC, paper mills.
Additionally, the company will permanently close its previously
idled Fort William mill in Thunder Bay, ON, and Lufkin, TX, paper
mills, as well as paper machine 3 at its Gatineau, QC, mill. The
previously idled operations run total capacity of about 650,000
tonnes/yr.
Execution is key. ``(AbitibiBowater) has done what I expect them
to do and be really aggressive, but the issue is going to be
execution,'' said one newsprint buyer contact with a major U.S.
publishing group. ``It is going to be impossible to take out 600,000
tonnes on Jan 1 and people will be looking to see how much comes out
in February and March. That will be the test.''
The reaction from Wall Street analysts was broadly favorable.
Citibank analyst Chip Dillon said the newsprint capacity reduction
figure was double his expectations. JPMorgan's Claudia Shank said
that while she believes another 300,000 tonnes/yr would need to come
out next year, the closures, together with AbitibiBowater's
indication that it could cut more capacity in mid 2008, provided
second- and third-tier producers some additional ``breathing room''
and limit closures from the broader industry before the second half
of next year.
``AbitibiBowater will probably say `We've done our part' to get
ahead of the curve and gain momentum on the pricing front,'' an
analyst in Canada said. ``But the market is looking for a million
tonnes (of newsprint reductions) year-over-year so more capacity
will have to be taken out if the market is going to be in balance in
2008.''
While AbitibiBowater did not disclose the number of jobs that
would be lost by its restructuring, the Communication, Energy and
Paperworkers Union of Canada (CEP) estimated at least 1,000 workers
could be eliminated in Canada.
CEP wants forestry ``summit.'' The CEP called for an emergency
summit of union and industry leaders in the forestry sector.
``Today's 1,000 or more victims in the mills in Dalhousie,
Shawinigan, Donnacona, and Mackenzie bring the job losses in the
sector to over 20,000 in the past two to three years,'' said CEP
pres Dave Coles.
AbitibiBowater pres/CEO David Paterson said management had been
very transparent with employees about their mills.
Under phase two of the plan, which starts immediately,
AbitibiBowater will continue reviewing all operations.
More Canadian mills at risk. Company chmn John Weaver said
several mills in eastern Canada were under particular pressure from
high fiber, energy, and labor costs, and the company planned to
involve government, communities, and labor to make the mills
competitive at dollar parity. Decisions would be taken in the second
quarter of 2008 and closures could start by mid-2008, he said.
AbitibiBowater has increased its merger synergies target to $350
million from $250 million. It is also targeting another $500 million
in asset sales, which could include overseas mills, non-core
facilities, U.S. timberlands, and its Snowflake, AZ, newsprint mill,
which it agreed to divest in return for U.S. Dept of Justice
approval of the Abitibi-Consolidated/Bowater merger.
Proceeds from the sales will go towards the company's three-
year, $1-billion debt-reduction target.
Citing rising costs and ``difficuit market
conditions,'' AbitibiBowater told customers that it would increase
prices on its AbiBow high-bright product line by $65/ton effective
Jan. 1. The increase applies to all basis weights, calipers, and
finishes of Book, Book Cream, Select, Sert, and Form products.
Separately, Blue Heron announced a $35/tonne ($31.75/ton) high-
bright increase for its reBrite product range, also effective Jan.
1.
Newsprint
Most North American Newsprint Makers Join $60/Tonne 1Q 2008 Hike
U.S. daily newspaper publishers face a New Year's perfect storm,
with producers who account for more than 80% of North American
production slating $60/tonne first quarter price hikes and
AbitibiBowater closing 600,000 tonnes/yr of newsprint capacity,
contacts said last week.
The price increases will be phased in monthly increments of $20/
tonne in January, February and March.
AbitibiBowater, which with 5.7 million tonnes/yr of capacity
accounts for about 45% of all North American newsprint production,
initiated the hike.
Among companies that contacts said would keep prices consistent
with AbitibiBowater are White Birch, Kruger, SP Newsprint, Catalyst,
Tembec, and Blue Heron. Other producers are still considering a
price hike, contacts said last week.
In addition to the 1Q 2008 hike almost all North American
newsprint producers will seek this month to implement a $25/tonne
fall increase that many producers have been trying to apply since
September.
Publishers start to panic. ``There is a general panic in the
market right now. Supply has tightened up and (producers) are really
pushing this December hike. I'm sure there are (publishers) who have
been particularly aggressive in the past that are going to get stuck
and be told to pay or buy somewhere else,'' said one publisher
contact.
One contact with a large supplier said the $25 hike had managed
to gain traction in November. ``Things happened in the back half of
the month'' buying sources conceded, saying that newsprint producers
did have the strength to move November's price ``a little bit.''
Pulp & Paper Week's November Price Watch had showed newsprint
prices on U.S. East and West coasts holding flat at $560/tonne.
Suppliers are in dire need for higher prices given the current
10.4% year-to-date decline in North American demand, strong Canadian
dollar and high input costs.
``I've never before seen such a confluence of bad things on this
side of the business. To save a dollar on production is a Herculean
task,'' said one producer contact in Canada.
Sign of modest improvement. According to the latest Pulp and
Paper Products Council data, the North American supply-demand
balance improved modestly in October, with production falling almost
in line with overall demand.
The biggest barometer for newsprint consumption, the U.S.
dailies, showed an 11.4% fall. But adjusting for four Sundays in
October 2007 compared with five in October last year, the decline
was closer to 7-8%.
More significantly, overall inventories fell to 1.13 million
tonnes, their lowest level since December 1979, after a two-month
242,000 tonnes or 18% plunge. Exports rose 29.0% in October, but
those extra 49,000 tonnes were more than offset by a 69,000 tonnes
drop in domestic shipments.
Gloomy economic outlook. With the economy sagging and the
outlook for newspaper advertising looking increasingly gloomy,
contacts say capacity cuts remain the only answer if mills are going
achieve the 95% operating rates that historically lead to higher
prices.
RISI economists say that despite higher exports, North American
mills will have to shut 800,000 tonnes/yr of capacity by the end of
next year (relative to third quarter 2007) if they are to push
operating rates above the 95% mark in 2008.
With plans to eliminate 38,000 tonnes of newsprint
production, Catalyst Paper last week extended the shutdown of PM 1
at its Elk Falls newsprint mill in Campbell River, BC, and keep the
PM down for the entire first quarter because of a shortage of fiber.
PM 1 was shut in September due to a fiber shortage. The company said
the mill has been hurt by a coastal fiber strike that recently ended
and a weak U.S. lumber market, Canadian Press said. In addition, the
mill's kraft pulp line and white-top linerboard PM will also shut 18
days between Dec. 16 and Jan. 2--and could be shut for longer
periods depending on fiber availability. PMs 2 and 5 will be shut
Dec. 23, and restart Jan. 2 and Jan. 6, respectively.
Japan's Oji Paper plans to hike the price of newsprint
exports by $50/tonne effective
[[Page 32847]]
with December orders, citing higher energy and raw material prices
that will add $460 million to its costs in the current financial
year. The company will also hike the price of other export grades,
ranging from $30/tonne for coated and uncoated products to $80/tonne
for kraft paper.
Germany's Palm Paper received planning permission to
construct a 400,000 tonnes/yr recycled newsprint mill at King's Lynn
in eastern England, which would expand Palm's UK production to
550,000 tonnes/yr. Ecco Newsprint, which has plans for a recycled
mill of its own at Middlesbrough in the north of the country, also
has planning permission but has not yet begun construction. The UK
currently imports about 1.2m tonnes of newsprint and exports 1.5m
tonnes of waste paper annually.
BILLING CODE 4410-11-M
[GRAPHIC] [TIFF OMITTED] TN10JN08.000
BILLING CODE 4410-11-C
Dec. 10, 2007 [bond] Vol. 29, No. 47
Newsprint Giant AbitibiBowater Embraces Industry Leadership, Eyes $200/
Tonne North American Newsprint Price Increase
Any doubts about AbitibiBowater's determination to regain
profitability and retire a billion dollars in debt within three
years were dispelled last week when pres/CEO David Paterson told
analysts at the Citi Investment Research Basic Materials Symposium:
``Our need is to leverage the North American (newsprint) price up to
the price in Europe and not the other way around.''
Newsprint prices in Europe were close to $200/tonne higher than
in the USA in November.
AbitibiBowater, the worlds largest newsprint maker, accounts for
about 45% of all North American newsprint production capacity.
Paterson said the company's $25/tonne fall price increase was in
place, and he anticipated that the company's recently announced $60/
tonne first quarter hike would be implemented entirely.
A presentation slide showed the effect of a $25/tonne increase
was an additional $126.8 million in operating income.
The benefit to AbitibiBowater's bottom line from shuttering
loss-making Canadian newsprint capacity was explained by CFO William
Harvey, who said production costs for the entire 600,000 tonnes/yr
slated for closure were $60/tonne higher than the company average.
Most North American producers expect the closures to save the
struggling North American newsprint industry, and have joined
AbitibiBowater's call for a $60/tonne increase in the first quarter
of 2008 implemented in three $20/tonne monthly increments.
Upward price pressure. ``AbitibiBowater's capacity closures will
obviously provide the upward pressure for an extended price recovery
in 2008, as operating rates soar past the magic 95% threshold
generally needed for prices to rise,'' said senior RISI economist
Kevin Conley. ``Without AbitibiBowater's bold move, operating rates
and prices would have continued to languish at low levels until the
highest-cost mills could no longer survive, eventually leading to
the inevitable
[[Page 32848]]
closures needed to balance the North American market.''
European producers are also addressing overcapacity, and
Europe's largest newsprint producer, Norske Skog, said that it would
decide by Feb. 7 how to permanently close 300,000-400,000 tonnes/yr
of newsprint capacity.
``We now see 1.4 to 1.5 million tonnes of announced capacity
removals in Europe and North America in just the past 10 weeks,''
said Citi analyst Chip Dillon, who told investors in a research note
that he expected a recovery in U.S. newsprint prices to close almost
all of the gap with European prices over the next 12-18 months.
Dismaying prospect for publishers. What a $60/tonne hike and
imminent closure of 5% of North American newsprint capacity portends
for U.S. daily newspapers had publishers shaking their heads.
``There is a sense of inevitability that seems to be recognized
by most on the publishing side. There's a sense of resignation in
their voices that hasn't been there before,'' said one contact with
a major metropolitan daily.
``It's a very different world from just a month ago. We
certainly did not have these kind of increases in our plans for
2008, so if they are implemented we would have to find ways to use
less newsprint,'' said another contact with a major publishing
group.
Newspaper publishers have their own business issues which have
largely brought about the decline in North American newsprint demand
to under nine million tonnes in 2007, from a peak of slightly more
than 13 million tonnes in 1999 and more than ten million tonnes as
recently as 2005.
Only 2.9% of the 11.4% drop in North American newsprint demand
this year is due to lighter basis weights and reduced web widths,
according to the Pulp and Paper Products Council. Of the rest, 2.2%
is attributed to falling circulation and 6.3% to lost advertising.
The bulk of the lost advertising is in real estate and automotive
sectors, neither of which show signs of a rebound anytime soon.
2008 a challenging year. ``From a fiscal standpoint 2008 will be
a challenging year almost without precedent for publishers. It's an
alignment of circumstances and realities that none of us have ever
seen before,'' said one publishing source.
But while the domestic market for newsprint is undeniably
shrinking, the global market is still growing. Industry consultant
Dave Allan told RISI's 2nd annual Latin American Pulp & Paper
Outlook Conference in Sao Paulo, Brazil, last week that world demand
showed flat growth in 2007 only because of North America's 10%
plunge.
Allan said he expected North American demand decline would slow
to 2.5% by the end of 2008, and that global demand would see a 2%/yr
upturn and grow at close to 1.0 million tonnes/yr in 2008 and 2009.
AbitibiBowater, which like some other North American producers
is growing its overseas exports, sees its key destinations as
Europe, Latin America and the Middle East and India. Chmn John
Weaver said last week that because the company's Canadian export
mills were located on ocean ports, the cost of bulk shipments to
Europe were comparable with shipments to North American
destinations.
Members of Canada's largest pulp and paper union, the
Communications, Energy & Paperworkers union (CEP) want to go to the
bargaining table a year earlier than scheduled to tackle the issue
of mill closures and job losses. The measure was adopted last week
by delegates representing AbitibiBowater paper workers and will go
to a conference of eastern Canada union Locals early next year. The
CEP opposes reopening negotiated contracts to cut wages and benefits
but says there are ways the union could help cut costs that do not
involve concessions.
Dec. 17, 2007\Vol. 29, No. 48
Publisher Resistance to $25/Tonne North American Newsprint Increase
Collapses; Producers Looking To Fast Track Recovery
Trenchant publisher resistance to a $25/tonne fall newsprint
price increase that persisted as late as mid-November vanished
toward the end of the month, and the hike went in ``like a hot knife
through butter'' in December, sources said last week.
Contacts said the market was tightening and order books filling
up due to some newspaper buyers trying to stock up ahead of next
year's fresh round of price increases and some commercial printers
switching to newsprint because of a shortage of specialty grades.
The price of 30-lb standard newsprint on the U.S. East and West
Coasts increased to $585/tonne this month, up $15 from a revised
$570/tonne in November, according to Pulp & Paper Week. The revised
November level represented a $10/tonne increase. The price of 27.7
lb newsprint was $625/tonne in December, up from $610/tonne in
November.
Newspaper publishers' rapid change of heart came after the
combination of AbitibiBowater's larger than expected 600,000 tonnes/
yr of newsprint capacity cuts along with a $60/tonne first quarter
price increase, contacts said. Analysts believe the closures remove
sufficient newsprint capacity to match North American market
demand--at least temporarily--in the first quarter.
70% 1Q price recovery? AbitibiBowater accounts for about 45% of
all North American newsprint capacity, and producers that account
for almost all the rest also announced $60/tonne hikes. If these are
successfully implemented, by the end of March suppliers will have
recovered $85 of the past year's $115/tonne price drop.
``You've got to take your hat off to this guy. He's determined
to show value to his shareholders and gained the upper hand very
quickly, while we are going to be fighting for our lives,'' remarked
one publisher contact, referring to AbitibiBowater CEO David
Paterson.
Both buyers and sellers expected that 2008 would bring higher
newsprint prices and many contacts believed suppliers would seek a
second price hike later in the year.
Three years of increases? ``If you are not building 10% price
increases into your budget for the next three years you are foolish.
Suppliers are pretty cocky right now and there's no sympathy for
publishers,'' commented a buyer contact with a major U.S. newspaper
group.
``I think what is going to drive (AbitibiBowater's) decisions is
their income statements and balance sheets, and I think they would
tell you they have been too deferential to their customers
historically--to their own detriment,'' said one contact.
``There's 800,000 tonnes compared to 2007 that will be closed
and I'd say the odds are 50-50 or better that we will get north of
$700/tonne in 2008, because even with a $150 increase Canadian mills
are not going to make money with the dollar at parity,'' said a
producer contact in Canada.
Consumption will be key. ``When you start hearing big numbers
thrown out, there is a tendency by some publishers to panic, but my
concerns are how many tonnes are really coming out and will
consumption continue to fall at the same rate we have seen this past
year,'' said one big U.S. newsprint buyer. ``Seeing a company like
Kruger that rarely takes downtime closing 100,000 tonnes will curl
your toes, but how much consumption is going to fall is more
important from my point of view.''
Publishers in Canada would be hurt less by higher newsprint
prices because the stronger Canadian dollar has shrunk their
newsprint costs to the lowest level in almost two decades.
``AbitibiBowater has shown what should be done to get the price
up to a level where they can make a dollar or two, but at the same
time I don't think U.S. publishers can afford to pay the price,''
said a contact with a major Canadian publisher. ``I am pretty sure
they will cut the size (of U.S. newspapers) and at the end of the
day demand is going to go down big time--another million tonnes I'm
sure.''
Dailies will shrink page size. Supplier sources also said they
anticipated consumption cutbacks, but said that given the 6% demand
drop in 2006 and near 11% drop in 2007, producers would have
difficulty increasing conservation significantly in the first
quarter.
``I think it's a given that everybody will go to 44-in. webs as
quickly as they can, cut out what they can from editorial, and make
the standard U.S. newspaper page 11 inches. That will cut demand 6-
8%,'' said one supplier contact.
Still, AbitibiBowater has said it is ready to shutter more mills
in eastern Canada if they cannot be made competitive.
``Their goal is to align capacity with demand, and whatever that
entails in terms of demand decline they are committed to matching
that,'' noted one U.S. publisher source.
But although suppliers are desperate to push prices higher, some
producers are wary of them going too high.
AbitibiBowater's weight and the world. ``There has to be an
upper limit. At $550/tonne, or even $600 or $625, we don't have any
issues with imports. But at $675, $700, or $725 we will see Chinese
tonnes here. It's one thing for AbitibiBowater to carry the North
American market on its back, but it's another to carry the whole
world,'' remarked one producer contact in Canada.
Europe's Holman Paper intends to close 150,000 tonnes/
yr. of standard newsprint
[[Page 32849]]
capacity at its 795,000 tonnes/yr. Hallsta mill at Hallstavik,
Sweden.
Norway's Norske Skog, the world's second-largest
newsprint producer behind AbitibiBowater, may spin off its Asian
operations. The company said it has been looking into a separate
stock market listing for its South Korean, Chinese, and Thai mills,
which run capacity of 1.6 million tonnes/yr. or a quarter of the
company's total. Some of Norske's investors want the company to sell
its Asian operations to reduce debt, but Norske has ruled out
selling the mills outright, saying the price would not reflect their
value in a market currently suffering from significant overcapacity,
according to a Financial Times report.
Economists Incorporated
An Economic Analysis of the Adequacy of the Snowflake Divestiture
in the Settlement of United States of America v. Abitibi-
Consolidated, Inc. and Bowater, Incorporated
Submitted on Behalf of the NAA
John H. Preston, Kent W. Mikkelsen, PhD, Economists Incorporated,
Washington, DC. January 2, 2008.
Table of Contents
Section A. Introduction
Section B. Economic Analysis
1. Unilateral Effects and the Dominant Firm Model
2. Abitibi and Bowater Engaged in Joint Dominant Firm Behavior to
Raise NA Newsprint Prices Significantly Above Competitive Levels
2002 to 2006
3. While the Proposed Merger of Abitibi and Bowater Was Under Review
by DOJ, Abitibi and Bowater Suspended Their Dominant Firm Behavior
and, As a Result, NA Newsprint Prices Declined Significantly
4. AbitibiBowater Resumed the Dominant Firm Behavior in November
2007 Following the October 23, 2007 Settlement Agreement With DOJ to
Divest the Snowflake Mill
5. DOJ Required a Much More Significant Divestiture to Settle a
Comparable Paper Industry Merger in 2000
Section C. Conclusion
Appendix A--Merger Analysis, Unilateral Effects, and the Dominant Firm
Model
Attachments
Attachment A Curricula Vitae of John H. Preston and Kent W.
Mikkelsen, Ph.D.
Attachment B White Paper by Economists Incorporated, Submitted on
Behalf of the NAA to DOJ on April 11, 2007
Attachment C Supplement 1 to the White Paper by Economists
Incorporated, Submitted on Behalf of the NAA to DOJ on July 9, 2007
Attachment D Supplement 2 to the White Paper by Economists
Incorporated, Submitted on Behalf of the NAA to DOJ on July 20, 2007
A. Introduction
On January 29, 2007, Abitibi-Consolidated, Inc. (``Abitibi'')
and Bowater Incorporated (``Bowater'') announced that they had
reached an agreement to merge the two companies.\1\ Following an
investigation of the merger, the U.S. Department of Justice
(``DOJ'') filed a civil antitrust complaint (``Complaint'') with the
United States District Court for the District of Columbia
(``Court'') on October 23, 2007 seeking to enjoin the merger.\2\
Paragraphs 2, 3 and 19 of the Complaint explain why DOJ was
challenging the proposed merger.
---------------------------------------------------------------------------
\1\ At the time of the merger announcement, newsprint accounted
for about 48 percent of the value of the combined sales of the two
companies. Other products produced by the two companies include
coated papers, uncoated papers, market pulp and wood products.
Source: The presentation accompanying the merger announcement,
``AbitibiBowater: Creating a Global Leader in Paper and Forest
Products,'' January 29, 2007, page 10.
\2\ The Complaint is captioned United States of America v.
Abitibi-Consolidated, Inc. and Bowater, Incorporated.
---------------------------------------------------------------------------
2. Abitibi and Bowater are the two largest newsprint producers
in North America. The combination of these two firms will create a
newsprint producer three times larger than the next largest North
American newsprint producer. After the merger, the combined firm
will have the incentive and ability to withdraw capacity and raise
newsprint prices in the North American newsprint market.
3. Unless the proposed transaction is enjoined, Defendants'
merger will substantially lessen competition in the production and
sale of newsprint, in violation of Section 7 of the Clayton Act, 15
U.S.C. Sec. 18.
19. The proposed transaction would combine Defendants' large
share of newsprint capacity, thereby expanding the quantity of
newsprint sales over which the merged firm would benefit from a
price increase. This would provide the merged firm with an incentive
to close capacity sooner than it otherwise would to raise prices and
profit from the higher margins on its remaining capacity.
At the same time the Complaint was filed, DOJ also filed a
proposed Final Judgment (``PFJ'') which, if approved by the Court,
would settle DOJ's case against defendants Abitibi and Bowater. As a
condition of the settlement, the defendants are required to sell
Abitibi's Snowflake, Arizona newsprint mill (``Snowflake mill'') to
an acquirer acceptable to DOJ.\3\ Following the filing of the
Complaint and PFJ, Abitibi and Bowater completed their merger on
October 29, 2007.\4\ The newly merged company is named
AbitibiBowater.
---------------------------------------------------------------------------
\3\ See the PFJ, Section IV.A.
\4\ AbitibiBowater press release, October 29, 2007.
---------------------------------------------------------------------------
Prior to the completion of the merger, Abitibi's share of North
American (``NA'') newsprint capacity was about 25 percent and
Bowater's share was about 16 percent.\5\ According to the Complaint,
the post-merger share of the combined company would be ``over 40
percent.'' The NA newsprint capacity share of the Snowflake mill is
about 3 percent.\6\ Thus, the divestiture of the Snowflake mill
would reduce the combined NA newsprint capacity share of the merged
firm from about 40 percent to about 37 percent.
---------------------------------------------------------------------------
\5\ See the Complaint, paragraph 16.
\6\ The annual newsprint capacity of the Snowflake mill is
375,000 metric tonnes, according to page 2 of the CIS. However, none
of the documents filed by DOJ with the court in this case provides
the NA newsprint capacity share of the Snowflake mill nor the amount
of total NA newsprint capacity that would be necessary to calculate
that share. Based on total NA newsprint production and operating
rates for November 2007, current total annual NA newsprint capacity
is about 11.7 million metric tonnes, which would give the Snowflake
mill a NA newsprint capacity share of about 3 percent. Source: The
November 2007 North American Newsprint Flash Report (``Flash
Report''), published by the Pulp and Paper Products Council
(``PPPC''). The members of the PPPC are NA pulp and paper
manufacturers, including most if not all NA newsprint manufacturers.
---------------------------------------------------------------------------
In its Competitive Impact Statement (``CIS''),\7\ DOJ explains
why it believes the divestiture of the Snowflake mill will be an
adequate remedy to prevent anticompetitive conduct by the merged
firm.
\7\ The CIS was also filed with the Court on October 23, 2007.
---------------------------------------------------------------------------
The combination enhances Defendants' incentives to exercise
market power because the merged firm will control a greater base of
capacity over which the merged firm would benefit from an increase
in newsprint prices after strategically closing, idling, or
converting some of its capacity. Without Snowflake's capacity, the
merged firm would not be of sufficient size to be able to recoup the
losses from such strategic closures through increases in prices on
its remaining newsprint production. The divestiture of Snowflake
would adequately address the likelihood that the proposed merger
substantially would reduce competition for newsprint in the United
States.\8\
---------------------------------------------------------------------------
\8\ See the CIS, page 6. The CIS does not specifically define
the terms ``strategically closing, idling, or converting some of its
capacity'' or ``strategic [capacity] closures.'' However from the
context of the paragraph on page 6 of the CIS quoted above, it is
evident that a newsprint manufacturer with a relatively large
capacity share will, acting by itself, have the incentive and
ability to ``strategically'' close capacity if the newsprint
manufacturer expects to recoup the losses from the capacity closure
through increases in prices on the manufacturer's remaining
newsprint production. The larger the newsprint manufacturer's
capacity share, the more likely the manufacturer will have the
incentive and ability to engage in such unilateral strategic
behavior. Newsprint manufacturers with relatively small capacity
shares will likely have neither the incentive nor ability to
strategically close capacity.
---------------------------------------------------------------------------
It is evident that DOJ has concluded that with a capacity share
of about 40 percent, the merged firm would have the incentive and
ability to unilaterally engage in anticompetitive conduct to raise
the price of newsprint but that with a slightly smaller capacity
share, about 37 percent, the merged firm would lose that incentive
and ability. DOJ provides no information or analysis in the CIS or
any other document it filed with the Court to support this claim.
We have been asked by the Newspaper Association of America
(``NAA'') and its attorneys to provide an economic antitrust
analysis of the Snowflake divestiture to determine whether that
divestiture will likely be sufficient to eliminate the
anticompetitive effects that would otherwise result from the
[[Page 32850]]
merger. The purpose of this analysis \9\ is to assist the Court in
its evaluation of the adequacy of the Snowflake divestiture under
the Antitrust Procedures and Penalties Act (known as the ``Tunney
Act''). During the course of DOJ's investigation of the proposed
merger of Abitibi and Bowater, we also submitted to DOJ an economic
White Paper and two Supplements to the White Paper on behalf of the
NAA. These submissions to DOJ are attached to this analysis.\10\
---------------------------------------------------------------------------
\9\ The authors of this analysis, John H. Preston and Dr. Kent
W. Mikkelsen, are both Senior Vice Presidents at Economists
Incorporated, an economic consulting firm headquartered in
Washington, DC and specializing in the economic analysis of
antitrust and regulation matters for over 25 years. Many economists
at Economists Incorporated, including Mr. Preston and Dr. Mikkelsen,
worked at DOJ as economists before joining Economists Incorporated.
The curricula vitae of Mr. Preston and Dr. Mikkelsen are attached to
this analysis as Attachment A.
\10\ The White Paper and the two Supplements to the White Paper
are attached to this analysis as Attachment B (``White Paper,''
submitted to DOJ on April 11, 2007), Attachment C (``Supplement 1,''
submitted to DOJ on July 9, 2007), and Attachment D (``Supplement
2,'' submitted to DOJ on July 20, 2007.) The White Paper is titled
``An Economic Analysis of the Competitive Effects of the Proposed
Abitibi-Bowater Merger,'' Supplement 1 to the White Paper is titled
``Response to Issues Raised at Our Meeting With the DOJ Staff on
April 20, 2007,'' and Supplement 2 is titled ``Revision to the July
9, 2007 Response.'' In addition, we met with the DOJ staff on four
occasions and participated in a number of conference calls with the
DOJ staff, including calls with newsprint buyers for newspapers, to
discuss the competitive issues raised by the proposed merger.
---------------------------------------------------------------------------
The NAA is an association whose membership includes newspaper
chains of all sizes and independent, small market, and family-owned
newspaper publishers. The NAA is headquartered in Arlington,
Virginia. NAA members account for nearly 90 percent of the daily
newspaper circulation in the U.S.\11\ U.S. daily newspapers are the
primary purchasers of newsprint produced by NA newsprint mills
accounting for about 80 percent of the newsprint consumed in the
U.S. and about 70 percent of the newsprint consumed in NA.\12\ If
the divestiture of the Snowflake mill proves to be inadequate to
eliminate the anticompetitive effects of the merger in the NA
newsprint market, NAA member newspapers and other purchasers of
newsprint in NA will bear the cost of that inadequacy in terms of
higher newsprint prices.
---------------------------------------------------------------------------
\11\ Source: NAA Web site.
\12\ Source: November 2007 Flash Report. Newsprint is also used
in the printing of nondaily newspapers and certain advertising
materials such as newspaper inserts and grocery store flyers.
---------------------------------------------------------------------------
As discussed in more detail below, less than six weeks after its
agreement to divest the Snowflake mill, AbitibiBowater announced
plans to remove a large amount of capacity from the newsprint market
and, at about the same time, initiated a significant newsprint price
increase. Additional AbitibiBowater capacity closures leading to
further price increases appear likely in 2008. The CIS claims that
``[w]ithout Snowflake's capacity, the merged firm would not be of
sufficient size to be able to recoup the losses from such strategic
closures through increases in prices on its remaining newsprint
production.'' \13\ This recent unilateral price-increasing action by
AbitibiBowater shows that DOJ has seriously misjudged the incentive
and ability of the merged firm to engage in strategic behavior to
raise the industry operating rate and the price of newsprint. This
misjudgment will likely cost U.S. newspapers and other U.S.
newsprint customers billions of dollars in coming years.
---------------------------------------------------------------------------
\13\ See the CIS, p. 6.
---------------------------------------------------------------------------
Even without this recent price-increasing action by
AbitibiBowater, there already existed substantial evidence that the
merger would likely provide AbitibiBowater with significant market
power and that the divestiture of just the Snowflake mill would be
unlikely to prevent AbitibiBowater from exercising that market
power. As documented and analyzed in the White Paper and in
Supplement 1 to the White Paper, Abitibi and Bowater jointly acted
as a dominant firm over the period 2002 to 2006 to strategically
remove newsprint capacity from the market to raise the price of
newsprint, the same type of anticompetitive strategic behavior
alleged in Paragraphs 2 and 19 of the Complaint and described on
page 6 of the CIS. Neither the Complaint nor the CIS, however,
mentions this prior anticompetitive behavior. In our opinion, a
history of prior anticompetitive conduct in the market affected by a
merger is relevant to merger analysis in two main respects: (1) It
provides both support and a justification for the filing of the
Complaint; and (2) in cases that are settled with a consent decree,
it allows the Court and other interested parties to more accurately
evaluate the adequacy of a proposed remedy. By failing to mention
the prior anticompetitive conduct of Abitibi and Bowater in the
North American newsprint market, DOJ has deprived the Court of
information highly relevant to an evaluation of the adequacy of the
Snowflake divestiture.
The Complaint and CIS also ignore the significant decline in
newsprint prices during the period the proposed merger was under
review by DOJ, a period of approximately 9 months. Abitibi and
Bowater did not engage in strategic behavior during this period or
in the months leading up to their merger announcement. It is
plausible that Abitibi and Bowater suspended their strategic
capacity closures to maximize the likelihood of a favorable merger
review by avoiding conduct that DOJ would likely find
anticompetitive. It is also plausible that the incentive and ability
of AbitibiBowater to jointly engage in strategic behavior had been
significantly weakened by previous capacity closures over the period
2002 to 2006, which led to the decision to merge. The decline in
newsprint prices during the merger review period is also information
highly relevant to an evaluation of the Snowflake divestiture,
information which DOJ did not provide in any of the documents it
filed with the Court.
To summarize, from 2002 to 2006, Abitibi and Bowater jointly
engaged in strategic dominant firm behavior causing newsprint prices
to rise significantly above competitive levels. During DOJ's review
of the proposed merger, Abitibi and Bowater suspended their joint
strategic dominant firm behavior and, as a result, newsprint prices
declined significantly. Shortly after Abitibi and Bowater agreed to
divest the Snowflake mill, the newly merged AbitibiBowater resumed
the dominant firm behavior by announcing significant newsprint
capacity closures and initiating significant newsprint price
increases. This resumption of strategic dominant firm behavior was
made possible by the merger and was not deterred by the Snowflake
divestiture.
B. Economic Analysis
1. Unilateral Effects and the Dominant Firm Model
The type of anticompetitive effect alleged in Paragraphs 2 and
19 of the Complaint and described on page 6 of the CIS is called a
``unilateral effect.'' That is, a unilateral effect results if the
merger provides the merged firm with the incentive and ability to
unilaterally engage in anticompetitive conduct without the need to
coordinate with non-merging firms in the market.
A dominant firm model is a model of unilateral conduct often
applied in circumstances where the product is relatively homogeneous
and where there is a single dominant firm with a relatively large
capacity share and a ``competitive fringe'' consisting of a number
of firms with relatively small capacity shares. These
characteristics apply to the newsprint industry.
While we have no direct knowledge of the model or models used by
DOJ to analyze the competitive effects of the proposed merger of
Abitibi and Bowater, the allegations in Paragraphs 2 and 19 of the
Complaint and described on page 6 of the CIS are consistent with an
application of the dominant firm model. See Appendix A below for
additional discussion of merger analysis, unilateral effects, and
the dominant firm model.
The method by which AbitibiBowater could unilaterally raise
newsprint prices is straightforward. In the newsprint industry,
newsprint prices increase at industry operating rates of about 95
percent and above. At industry operating rates below 95 percent,
newsprint prices are likely to remain constant or decline.\14\ If
there is a significant amount of excess capacity, as has recently
been the case in the newsprint industry, then newsprint prices are
unlikely to increase unless enough capacity is removed from the
market to raise the operating rate above 95 percent. Newsprint
customers are beneficiaries of the lower prices that result from the
excess capacity.
---------------------------------------------------------------------------
\14\ See the White Paper, Section F, pages 83 to 87, and Section
11, pages 94-105, for a discussion and analysis of the relationship
between the newsprint operating rate and the price of newsprint.
---------------------------------------------------------------------------
A firm with a sufficiently large capacity share would have the
incentive and ability to unilaterally remove capacity from the
market to raise the price of newsprint if the increased profit from
the price increase on its remaining capacity exceeds the loss in
profit from the closed capacity. DOJ's Complaint and CIS are
evidently based on the theory that a merger creating a firm with
about a 40
[[Page 32851]]
percent newsprint capacity share would enable that firm to
profitably remove capacity from the market in order to raise the
industry operating rate to a high enough level to also raise the
price of newsprint.
2. Abitibi and Bowater Engaged in Joint Dominant Firm Behavior to
Raise NA Newsprint Prices Significantly above Competitive Levels
2002 to 2006
An argument that the merger will provide AbitibiBowater with the
incentive and ability to strategically close capacity to raise the
price of newsprint is not based solely on a theoretical model. The
White Paper and Supplement 1 to the White Paper submitted to DOJ
document and analyze prior anticompetitive conduct of Abitibi and
Bowater that occurred between the third quarter of 2002 and the
third quarter of 2006. See the following sections of the White Paper
for this analysis:
Section F: Evidence from Presentations to Investment Analysts and
Other Public Information That Abitibi and Bowater Have Used Their
Control Over Newsprint Capacity and the Newsprint Industry Operating
Rate to Significantly Raise the Price of Newsprint 2002 to 2006 (pp.
73-87)
Section G: An Analysis of Permanent Newsprint Capacity Reductions
Between 2002 and 2006 (pp. 88-93)
Section H: Four Articles by Two Newsprint Industry Experts
Describing the Abitibi-Bowater Strategy to Raise Prices by Closing
Capacity (pp. 94-105)
See also the following section from Supplement 1 to the White
Paper:
Section C: Additional Evidence that Abitibi and Bowater Exercised
Market Power Over the Period 2002 to 2006 (pp. 16-23)
As explained in these analyses, Abitibi and Bowater jointly
acted as a dominant firm to strategically remove newsprint capacity
from the NA market to raise the price of newsprint to NA customers
significantly above competitive levels during this four-year period.
During this four-year period of strategic capacity closures, NA
newsprint prices steadily increased by an aggregate of 49 percent
between the third quarter of 2002 and the third quarter of 2006
despite a steady decline in the consumption of newsprint by U.S.
newspapers. These newsprint price increases were far in excess of
the price increases for closely-related uncoated groundwood
specialty grades during this period.\15\
---------------------------------------------------------------------------
\15\ See the White Paper, Section J: A Comparison of Newsprint
Prices with the Prices of Uncoated Groundwood Specialty Grades 3Q
1999 to 4Q 2006 (pp. 109-119).
---------------------------------------------------------------------------
Earlier mergers in the NA newsprint industry, especially the
Abitibi-Donohue merger in 2000 and the Bowater-Alliance merger in
2001, created both the incentive and ability for Abitibi and Bowater
to jointly engage in this anticompetitive conduct. In papers and
presentations to the DOJ staff submitted on behalf of the NAA and
the U.S. newspaper industry, Economists Incorporated explained in
2000 and 2001 that these two mergers, if not challenged, would have
significant anticompetitive results. DOJ took no action against
either of these two earlier mergers and, as predicted by Economists
Incorporated, the two mergers enabled Abitibi and Bowater to engage
in the anticompetitive conduct that occurred between 2002 and
2006.\16\ U.S. newspapers and other NA newsprint customers bore the
cost of DOJ's inaction in the form of significantly higher newsprint
prices.
---------------------------------------------------------------------------
\16\ The implementation of strategic capacity closures by
Abitibi and Bowater following their mergers was likely delayed by
the U.S. economic recession in 2001 and the economic aftermath of
the events of 9/11. During this time, U.S. newspapers suffered a
significant decline in the sale of newspapers and newspaper
advertising, resulting in a significant decline in the demand for
newsprint by U.S. newspapers.
---------------------------------------------------------------------------
Despite its obvious relevance to an evaluation of the adequacy
of DOJ's settlement with Abitibi and Bowater, this prior history of
anticompetitive conduct by Abitibi and Bowater is not mentioned in
the CIS, Complaint or PFJ. This is surprising since the
documentation of prior anticompetitive conduct would strengthen the
grounds for DOJ's challenge of the merger.
The Commentary on the Horizontal Merger Guidelines (``Merger
Guidelines Commentary''), jointly published by DOJ and the Federal
Trade Commission, explains why evidence of prior anticompetitive
effects by finns in a relevant market is probative to the agencies'
evaluation of a merger of two firms in that market.
Facts showing that rivals in the relevant market have
coordinated in the past are probative of whether a market is
conducive to coordination. Guidelines Sec. 2.1. Such facts are
probative because they demonstrate the feasibility of coordination
under past market conditions. Other things being equal, the removal
of a firm via merger, in a market in which incumbents already have
engaged in coordinated behavior, generally raises the risk that
future coordination would be more successful, durable, or
complete.\17\
---------------------------------------------------------------------------
\17\ See Merger Guidelines Commentary, p. 22.
Two DOJ cases are cited to illustrate the significance of prior
anticompetitive conduct in DOJ's merger analysis and, in each of
these cases, the anticompetitive conduct was described in the
complaint challenging the merger.\18\ While these two cases
identified in the Merger Guidelines Commentary were challenged on a
coordinated interaction theory,\19\ evidence of prior
anticompetitive conduct should logically also be highly relevant to
the agencies' analysis of mergers based on a unilateral effects
theory.
---------------------------------------------------------------------------
\18\ The two cited DOJ examples are Premdor-Masonite (2001) and
Suiza-Broughton (1999).
\19\ On page 22, the Merger Guidelines Commentary describes an
increase in the likelihood of ``coordinated interaction'' that might
result from a merger as follows: ``A horizontal merger is likely to
lessen competition substantially through coordinated interaction if
it creates a likelihood that, after the merger, competitors would
coordinate their pricing or other competitive actions, or would
coordinate them more completely or successfully than before the
merger.'' See Appendix A for additional discussion of the
distinctions between unilateral effects theories and coordinated
interaction theories.
---------------------------------------------------------------------------
3. While the Proposed Merger of Abitibi and Bowater Was Under
Review by DOJ, Abitibi and Bowater Suspended Their Dominant Firm
Behavior and, as a Result, NA Newsprint Prices Declined
Significantly
Abitibi and Bowater began their merger discussions in June 2006,
which culminated in their joint merger announcement on January 29,
2007.\20\ The four-year run-up in newsprint prices described in the
previous section reached a peak of $675 per metric tonne in May
2006. That price prevailed through September 2006. Between September
2006 and December 2006, the NA newsprint price declined slightly to
$660 per metric tonne.\21\ Between December 2006 and October 2007,
the price of newsprint dropped by $100 to $560 per metric tonne, a
decline of about 15 percent.\22\ The $115 per metric tonne decline
in the NA price of newsprint between September 2006 and October 2007
was about 17 percent.
---------------------------------------------------------------------------
\20\ See Supplement 1 to the White Paper, page 11.
\21\ Source: the following editions of Pulp & Paper Week; June
19, 2006, p. 3; September 18, 2006, p. 3; November 20, 2006, p. 3;
February 19, 2007, p. 3; and November 19, 2007, page 3.
\22\ Between September 2006 and October 2007, the NA price of
newsprint dropped $115 per metric tonne, a decline of about 17
percent.
---------------------------------------------------------------------------
Between September 2006 and October 2007, Abitibi and Bowater did
not engage in joint dominant firm behavior despite a decline in NA
newsprint prices of about 17 percent. It is plausible that Abitibi
and Bowater suspended their joint dominant firm behavior during this
period for two reasons: (1) Abitibi and Bowater wanted to maximize
their chances of a favorable merger review by DOJ by avoiding
conduct that DOJ would likely construe as anticompetitive; and (2)
their ability and incentive to jointly engage in strategic capacity
closures had been significantly weakened by their previous strategic
capacity closures over the period 2002 to 2006. It is also plausible
that a weakened incentive and ability to engage in joint dominant
firm behavior led to the decision to merge.\23\
---------------------------------------------------------------------------
\23\ The continued decline in NA newsprint demand likely also
contributed to the decision to merge. A continued decline in NA
newsprint demand would require continued strategic capacity closures
in order to maintain high newsprint industry operating rates and
increasing newsprint prices. By merging, Abitibi and Bowater
increased their incentive and ability to strategically close
capacity in the face of declining demand.
---------------------------------------------------------------------------
From the trade press commentary during the merger review period,
it is apparent that newsprint industry analysts and newsprint
competitors of Abitibi and Bowater were waiting for the merger to be
completed in anticipation that a merged AbitibiBowater would
increase NA newsprint prices by shutting down enough newsprint
capacity to create a tight market. It is also apparent that these
same analysts and competitors believed that Abitibi and Bowater
would not take any significant actions to remove capacity from the
market until after their merger review was completed.\24\
---------------------------------------------------------------------------
\24\ See Section B.3., ``Newsprint Industry Analysts and
Competitors of Abitibi and Bowater Do Not Expect Abitibi and Bowater
to Take Any Significant Action to Remove Newsprint Capacity from the
Market Until After They Have Merged,'' in Supplement 1 to the White
Paper, pp. 13-15.
---------------------------------------------------------------------------
[[Page 32852]]
The following quotation is typical of comments that appeared in
the trade press during the merger review period. ``No one will close
any capacity because they figure AbitibiBowater will do it for them.
And Abitibi kind Bowater will figure they can't be too aggressive on
pricing or close capacity until their deal closes, said one
contact.'' \25\ For other similar trade press commentary see
Supplement 1 to the White Paper, pp. 13-14.
---------------------------------------------------------------------------
\25\ ``Market abuzz over merger: concerns center on pricing and
customer relationships,'' Pulp & Paper Week, February 5, 2007, p.
11.
---------------------------------------------------------------------------
4. AbitibiBowater Resumed the Dominant Firm Behavior in November
2007 Following the October 23, 2007 Settlement Agreement With DOJ
to Divest the Snowflake Mill
Less than five weeks after the filing of the Complaint and MFJ,
AbitibiBowater announced the removal of about 600,000 metric tonnes
of capacity from the NA newsprint market \26\ amounting to about a 5
percent reduction in total NA newsprint capacity. These capacity
closures will occur during the first quarter of 2008. At
approximately the same time, AbitibiBowater initiated a $60 per
metric tonne newsprint price increase. This price increase will also
take place during the first quarter of 2008. Most other NA newsprint
manufacturers quickly joined AbitibiBowater in this $60 price
increase.\27\
---------------------------------------------------------------------------
\26\ Source: Press release on AbitibiBowater Web site, November
29, 2007.
\27\ Source: Pulp & Paper Week, Dec. 3, 2007, pp. 1, 2, and 5.
---------------------------------------------------------------------------
In addition, AbitibiBowater's announced capacity closures have
permitted the successful implementation of a previously announced
$25 per metric tonne price increase. \28\ Newsprint manufacturers,
including Abitibi and Bowater, had previously been unable to
successfully implement this price increase, originally scheduled for
September 2007, because of excess NA newsprint industry
capacity.\29\ Combined, these two price increases will raise the
price to NA newsprint customers by $85 per metric tonne, which is
about a 15 percent price increase over the October 2007 price of
$560 per metric tonne.\30\
---------------------------------------------------------------------------
\28\ Source: Pulp & Paper Week, Dec. 17, 2007, pp. 1 and 11.
\29\ On p. 9 in its October 22, 2007 edition, published the day
before DOJ's settlement agreement with Abitibi and Bowater, Pulp &
Paper Week reported on the failure of NA newsprint producers to
implement the September price increase in an article titled ``North
American newsprint hikes lack market traction, price declines $5/
tonne more.''
\30\ Source for October 2007 newsprint price: Pulp & Paper Week,
Nov. 19, 2007, p. 3.
---------------------------------------------------------------------------
These post-settlement events are captured in headlines from the
trade press newsletter Pulp & Paper Week during the first three
weeks of December 2007 following the capacity closure announcement
of AbitibiBowater on November 29, 2007 and the $60 per metric tonne
newsprint price increase initiated by AbitibiBowater.\31\
---------------------------------------------------------------------------
\31\ Pulp & Paper Week is published by RISI, which describes
itself as ``the leading source of global news for the forest
products industry.'' These articles are attached as Attachment A to
the Comments of the Newspaper Association of America.
---------------------------------------------------------------------------
``AbitibiBowater plans to shut down one million tonnes/yr of
capacity in 1Q; expects more closures could follow in 2Q,'' December
3, 2007, p. 1.\32\
---------------------------------------------------------------------------
\32\ The capacity reduction announced by AbitibiBowater totaled
about 600,000 metric tonnes of newsprint capacity and 400,000 metric
tonnes of commercial printing papers according to the Pulp & Paper
Week article.
---------------------------------------------------------------------------
``Most North American newsprint makers join $60/tonne 2008
hike,'' December 3, 2007, p. 2.\33\
---------------------------------------------------------------------------
\33\ The Pulp & Paper Week article states that the $60 per
metric tonne increase was initiated by AbitibiBowater.
---------------------------------------------------------------------------
``Newsprint giant AbitibiBowater embraces industry leadership,
eyes $200/tonne North American newsprint price increase,'' December
10, 2007, p. 1.
``Publisher resistance to $25/tonne North American newsprint
increase collapses; producers looking to fast track recovery,''
December 17, 2007, p. 1.
In comments reported in Pulp & Paper Week, RISI economist Kevin
Conley explains the cause and effect between AbitibiBowater's
capacity closures and the increase in newsprint prices.
``AbitibiBowater's capacity closures will obviously provide the
upward pressure for an extended price recovery in 2008, as operating
rates soar past the magic 95% threshold generally needed for prices
to rise. Without AbitibiBowater's bold move [to remove 600,000
metric tonnes of newsprint capacity from the market] operating rates
and prices would have continued to languish at low levels until the
highest-cost mills could no longer survive, eventually leading to
the inevitable closures needed to balance the North American
market.'' \34\
---------------------------------------------------------------------------
\34\ Source: Pulp & Paper Week, Dec. 10, 2007, p. 5.
---------------------------------------------------------------------------
The combined AbitibiBowater is seeking to ``leverage the North
American (newsprint) price up to the price in Europe and not the
other way around,'' according to AbitibiBowater President and CEO
David Paterson.\35\ If AbitibiBowater is successful in
``leveraging'' the North American newsprint price up to the price of
newsprint in Europe, that will result in a $200 per metric tonne
price increase or about 36 percent over the North American price of
$560 per metric tonne in October 2007.\36\ At the time
AbitibiBowater announced the removal of 600,000 metric tonnes of
newsprint capacity from the market, it also announced that ``more
mills could close in Canada later [in 2008].'' \37 \ Based on these
statements and other statements by AbitibiBowater executives and
past and current actions by AbitibiBowater and its predecessor
companies, it is very likely that AbitibiBowater will close
additional newsprint capacity in 2008 to ``leverage'' the North
American newsprint price up to the newsprint price in Europe.
---------------------------------------------------------------------------
\35\ Source: Pulp & Paper Week, Dec. 10, 2007, p. 1.
\36\ ``Newsprint prices in Europe were close to $200/tonne
higher than in the USA in November.'' Source: Pulp & Paper Week,
Dec. 10, 2007, p. 1.
\37\ Source, Pulp & Paper Week, Dec. 3, 2007, p. 1.
---------------------------------------------------------------------------
In the CIS, DOJ asserts that ``[w]ithout Snowflake's capacity,
the merged firm would not be of sufficient size to be able to recoup
the losses from such strategic closures through increases in prices
on its remaining newsprint production.'' These strategic closures
announced by AbitibiBowater less than five weeks after the filing of
the Complaint, CIS, and PFJ show that DOJ seriously misjudged the
incentive and ability of the merged firm to strategically close
capacity despite the agreement to divest the Snowflake mill.
Furthermore, based on comments by AbitibiBowater, additional
strategic capacity closures will likely occur later in 2008.\38\ In
the absence of a significantly larger divestiture, DOJ's misjudgment
will likely cost U.S. newspapers and other U.S. newsprint customers
billions of dollars in coming years.\39\
---------------------------------------------------------------------------
\38\ Source: Pulp & Paper Week Dec. 3, 2007, pp. 1 and 5.
\39\ Based on the November 2007 Flash Report, current annual
U.S. newsprint consumption is about 7.8 million metric tonnes. The
$85 per metric tonne price increase resulting from AbitibiBowater's
recently announced capacity closures will increase the aggregate
cost of newsprint to U.S. newsprint customers by about $663 million
per year. If the NA newsprint price rises by a total of $150 per
metric tonne due to continued strategic behavior by AbitibiBowter
(an increment of $65 per metric tonne over the current price
increase of $85 per metric tonne), the cost to U.S. newsprint
consumers would be about $1.2 billion on an annual basis. If
AbitibiBowater is able to ``leverage the North American (newsprint)
price up to the price in Europe,'' as David Paterson, President and
CEO of AbitibiBowater, is apparently seeking to do, the annual cost
to U.S. newsprint consumers resulting from the $200 per metric tonne
price increase (an increment of $115 per metric tonne over the
current price increase of $85 per metric tonne) would be about $1.6
billion. These calculations are based on the assumption that the
U.S. consumption of newsprint remains at the November 2007 level. In
practice, U.S. newsprint consumption will likely continue to
decline, as discussed above. Therefore, the magnitudes of the
aggregate cost increases to U.S. newsprint customers calculated in
this footnote would be reduced somewhat by a continued decline in
consumption. Regardless, the aggregate cost increases to U.S.
consumers will be substantial.
---------------------------------------------------------------------------
5. DOJ Required a Much More Significant Divestiture To Settle a
Comparable Paper Industry Merger in 2000
In August 2000, Georgia-Pacific announced plans to acquire Fort
James. At the time of the acquisition Georgia-Pacific was a broadly-
based forest products company and Fort James was the largest
manufacturer of tissue paper in the United States. Both companies
operated paper mills that produced parent tissue rolls used to make
tissue products sold to commercial customers (known as ``away-from-
home'' tissue products). At the time of the proposed acquisition,
Fort James and Georgia-Pacific were the two largest producers of
parent tissue rolls in NA. Fort James' share of NA parent tissue
role capacity was 25 percent and Georgia-Pacific's share was 11
percent for a combined capacity share of 36 percent.
On November 21, 2000, DOJ filed a complaint challenging the
merger in the
[[Page 32853]]
parent tissue roll market. At the same time, DOJ filed a proposed
final judgment requiring the divestiture of all of Georgia-Pacific's
parent tissue roll capacity.\40\
---------------------------------------------------------------------------
\40\ For an explanation of the allegations in the complaint and
the provisions of the proposed final judgment, as well as background
information relating to the merger, see the Georgia-Pacific/Fort
James competitive impact statement, dated January 25, 2001. DOJ also
required the divestiture of certain downstream tissue converting
capacity.
---------------------------------------------------------------------------
As described in the competitive impact statement for the
Georgia-Pacific/Fort James merger, the theory DOJ relied upon to
challenge the proposed acquisition of Fort James by Georgia-Pacific
in the NA tissue parent roll market merger appears to be based on
the same basic theory of unilateral anticompetitive conduct DOJ used
in its challenge of the Abitibi-Bowater merger.
Georgia-Pacific has approximately 11 percent of North American
capacity for the production of AFH tissue, and Fort James has
approximately 25 percent. Hence, the acquisition would result in
Georgia-Pacific accounting for approximately 36 percent of available
North American AFH parent roll capacity. This increase in industry
capacity controlled by Georgia-Pacific would give it sufficient
capacity to profit from the increase in price caused by a unilateral
reduction in output after this merger.\41\
---------------------------------------------------------------------------
\41\ Georgia-Pacific/Fort James competitive impact statement, p.
7.
It is evident that DOJ concluded that the combination of firms
with a 26 percent capacity share and an 11 percent capacity to
create a firm with a 36 percent capacity share would give Georgia-
Pacific the incentive and ability to unilaterally exercise market
power in the NA parent tissue roll market. It is also evident that
DOJ concluded that the divestiture of Georgia-Pacific's entire 11
percent of its NA parent tissue roll capacity share was necessary to
eliminate Georgia-Pacific's incentive and ability to engage in
unilateral strategic behavior. The divestiture left Georgia-Pacific
with a capacity share of 25 percent in the NA parent tissue roll
market.
Nothing in the Abitibi-Bowater CIS explains the great disparity
between the divestiture required to settle the Abitibi-Bowater
merger and the divestiture required to settle the Georgia-Pacific/
Fort James merger. The prior recent and well-documented unilateral
anticompetitive conduct of Abitibi and Bowater (unacknowledged by
DOJ in the Complaint and CIS) makes this disparity all the more
puzzling.
If the former Bowater's newsprint capacity, which accounts for
16 percent of NA newsprint capacity according to the Complaint, were
divested, the merged firm (AbitibiBowater) would have a NA newsprint
capacity share of 25 percent. This divestiture would be comparable
to the divestiture DOJ required to settle the Georgia-Pacific/Fort
James merger, which left Georgia-Pacific with a 25 percent capacity
share in the NA parent roll tissue market.
C. Conclusion
Based on the economic analysis contained in this memorandum and
the economic analyses we have previously submitted to DOJ, we
conclude that the Snowflake divestiture will not be sufficient to
eliminate the anticompetitive effects of the merger and that a
substantially larger divestiture is needed to ensure that
AbitibiBowater no longer has the incentive and ability to engage in
the type of anticompetitive conduct alleged in Paragraphs 2 and 19
of the Complaint and described on page 6 of the CIS.
Appendix A-Merger Analysis, Unilateral Effects, and the Dominant Firm
Model
In determining the competitive effects of a merger, DOJ utilizes
the analytical framework set out in the U.S. Department of Justice
and Federal Trade Commission (``FTC'') Horizontal Merger Guidelines
(``Merger Guidelines'').\42\ In March 2006, DOJ and the FTC jointly
issued a Commentary on the Merger Guidelines (``Merger Guidelines
Commentary'') to provide interested parties with a greater
understanding of how the agencies apply the Merger Guidelines to the
investigation of specific mergers.
---------------------------------------------------------------------------
\42\ The Merger Guidelines were issued on April 2, 1992 and
revised on April 8, 1997.
---------------------------------------------------------------------------
Section 2 of the Merger Guidelines describes two general types
of anticompetitive effects that potentially could result from a
merger: (1) Unilateral effects and (2) coordinated interaction. The
Merger Guidelines Commentary describes these anticompetitive effects
as follows:
A horizontal merger is likely to lessen competition
substantially through coordinated interaction if it creates a
likelihood that, after the merger, competitors would coordinate
their pricing or other competitive actions, or would coordinate them
more completely or successfully than before the merger. A merger is
likely to lessen competition substantially through unilateral
effects if it creates a likelihood that the merged firm, without any
coordination with non-merging rivals, would raise its price or
otherwise exercise market power to a greater degree than before the
merger.\43\
---------------------------------------------------------------------------
\43\ See the Merger Guidelines Commentary, p. 22.
---------------------------------------------------------------------------
Paragraph 2 of DOJ's Complaint against Abitibi and Bowater
alleges that
After the merger, the combined firm will have the incentive and
ability to withdraw capacity and raise newsprint prices in the North
American newsprint market.
Paragraph 19 of DOJ's Complaint against Abitibi and Bowater
alleges that
The proposed transaction would combine Defendants' large share
of newsprint capacity, thereby expanding the quantity of newsprint
sales over which the merged firm would benefit from a price
increase. This would provide the merged firm with an incentive to
close capacity sooner than it otherwise would to raise prices and
profit from the higher margins on its remaining capacity.
While DOJ has not disclosed the economic models it used in its
investigation of the Abitibi-Bowater merger, these allegations in
Paragraphs 2 and 19 of the Complaint are consistent with a
unilateral effects theory of competitive harm, specifically a
unilateral effects theory of competitive harm based on the
application of a dominant firm model. The Merger Guidelines
Commentary describes the application of the dominant firm model as
follows:
The Agencies' analysis of unilateral competitive effects draws
on many models developed by economists. The simplest is the model of
monopoly, which applies to a merger involving the only two
competitors in the relevant market. One step removed from monopoly
is the dominant firm model. That model posits that all competitors
but one in an industry act as a ``competitive fringe,'' which can
economically satisfy only part of total market demand. The remaining
competitor acts as a monopolist with respect to the portion of total
industry demand that the competitive fringe does not elect to
supply. This model might apply, for example, in a homogeneous
product industry in which the fringe competitors are unable to
expand output significantly.\44\
---------------------------------------------------------------------------
\44\ See Merger Guidelines Commentary, p. 25.
---------------------------------------------------------------------------
In our opinion, a dominant firm model is the appropriate model
to assess the competitive effects of the Abitibi-Bowater merger. In
our submissions to DOJ, we described our application of the dominant
firm model to this merger.\45\ Our dominant firm model incorporated
the key characteristics of the newsprint industry including the
capacity share of the dominant firm (i.e., a combined Abitibi and
Bowater), the variable cost of the dominant firm, the industry price
elasticity of demand, the industry operating rate, the excess
capacity of fringe firms, and prevailing price levels. In our
application of the dominant firm model we took into consideration
multi-period dynamics, a decline in the NA demand for newsprint, and
an increase in the rate of decline in the NA demand for newsprint.
---------------------------------------------------------------------------
\45\ See Section K of the White Paper: Dominant Firm Model
(pages 120-124); Attachment K to the White Paper: Technical Appendix
to Section K Dominant Firm Model (pages 1-8), and Supplement 1 to
the White Paper: Additional Analysis Based on the Dominant Firm
Model (DFM) Including a Revision of the DFM Designed to Consider
Multi-period Dynamics (pages 24-33).
---------------------------------------------------------------------------
Based on our application of the dominant firm model, we
predicted that, under a wide range of dominant firm capacity shares
and other assumptions, the merged firm would have both the incentive
and ability to remove capacity from the market to raise the price of
newsprint. In particular, we were able to show that under a wide
range of assumptions the dominant firm would hypothetically be able
to close newsprint capacity to raise newsprint prices well above
competitive levels at dominant firm capacity shares well below 37
percent.
The results of our application of the dominant firm model are
consistent with the observed joint dominant firm behavior of Abitibi
and Bowater during the period 2002 to 2006 as discussed in Section
B.2. above and with the observed dominant firm behavior of the
newly-merged AbibitiBowater as discussed in Section B.4 above.
[[Page 32854]]
Attachment A--Curricula Vitae of John H. Preston and Kent W. Mikkelsen,
PhD
Curriculum Vitae
John H. Preston
Office
Economists Incorporated, 1200 New Hampshire Avenue, NW., Suite 400,
Washington, DC 20036, (202) 833-5237, preston.j@ei.com
Home
18505 SE Heritage Oaks Lane, Tequesta, FL 33469, (561) 575-2310,
jhp2004@comcast.net
Education
A.B. English, Dartmouth College (1966), M. A. Economics, University
of Michigan (1972), Candidate in Philosophy in Economics, University
of Michigan (1974)
Professional Experience (Consulting)
Senior Vice President, Economists Incorporated (December 1998-
Present), Vice President, Economists Incorporated (December 1995-
December 1998), Senior Economist, Economists Incorporated (April
1985-December 1995)
Selected Matters
Timberlawn v. Tenet Healthcare, et al. Provided affidavit,
deposition testimony, and trial testimony on behalf of defendants in
the alleged monopolization of psychiatric hospitals in the Dallas
area by NME.
Proposed Abitibi-Consolidated/Donohue newsprint merger. On
behalf of NAA, provided analysis to DOJ concerning the likely
anticompetitive effects of the merger.
Proposed MCI/Sprint Merger. Provided affidavits to DOJ, FCC, and
European Commission analyzing the competitive effects of the
proposed merger on behalf of British Telecom and AT&T. Testified
before the European Commission on this matter.
Coated Groundwood Paper Anti-Dumping Investigation. Helped
prepare response to Antidumping investigation of the ITC on behalf
of European groundwood paper manufacturers. Participated in
presentation to ITC.
Proposed SBC/AT&T and Verizon/MCI mergers. On behalf of BT,
analyzed competitive effects of the two telecommunications mergers.
Provided affidavits to DOJ, FCC and European Commission and made
presentations to DOJ and FCC staffs.
British Telecom/AT&T Global Venture. Provided economic analysis
on a wide range of competition issues concerning the global venture,
including presentations to the European Commission and DOJ.
PacifiCare/FHP merger. Analysis of the impact of this proposed
merger on the provision of Medicare HMO services in California. Made
written and oral presentations to the FTC staff and senior
management.
WellPoint/HSI merger. Analysis of the competitive effects of
this proposed merger of two of the largest HMOs in California and
participation in meetings with DOJ.
Sale of General Dynamics' Missile Division to Hughes Aircraft
and General Dynamics' Jet Fighter Division to Lockheed. Helped
prepare antitrust analysis and participated in presentations to DOJ
and FTC on these defense industry mergers.
Professional Experience (Antitrust Division)
Economist (January 1975-April 1985), Economic Policy Office,
Antitrust Division, U.S. Department of Justice
Honors
Special Achievement Award for work on U.S. v. Hospital Affiliates
International, Inc. and American Health Services, Inc. (1980)
Outstanding Performance Rating (1980-1981)
Outstanding Performance Rating (1981-1982)
Outstanding Performance Rating (1982-1983)
Outstanding Performance Rating (1983-1984)
Meritorious Award (1983)
Selected Matters Testimony Affidavit
U.S. v. Hospital Affiliates International, Inc., and American
Health Services, Inc. In 1980, submission of an affidavit to the
U.S. District Court in New Orleans analyzing the competitive effects
of the proposed merger of three psychiatric hospitals in New
Orleans, LA.
Selected Matters Deposition Testimony
U.S. v. British Columbia Forest Products, et al. In 1981,
deposition testimony on the preparation of the trial exhibits for
the challenge of an acquisition of a coated groundwood paper plant
by a firm partially owned by two other manufacturers of coated
groundwood paper.
U.S. v. State Board of Certified Public Accountants of
Louisiana. In 1984, deposition testimony on product and geographic
market definition and competitive effects of restrictions on
advertising and solicitation by the Louisiana board of accountants.
Grand Jury Testimony
U.S. v. Gary L. McAliley et al. In 1980, testimony before a
grand jury in Alabama on the effects of an alleged agreement between
attorneys in Coffee County, Alabama to raise fees for real estate
closings.
Other Filed Cases
U.S. v. National Medical Enterprises, et al. Hospital merger
case.
U.S. v. American Consulting Engineers Council. Prohibitions on
free designs and on participation in design competitions.
U.S. v. Alaska Board of Registration for Architects, Engineers
and Land Surveyors. Competitive bidding ban.
U.S. v. First Multiple Listing Service. Alleged exclusion of
competitors by owners of an essential facility.
Investigations
Georgia-PacifIc Acquisition of Hudson Pulp & Paper. This merger
was investigated by DOJ for antitrust implications in a number of
paper and paperboard product lines.
Acquisition of Hospital Affiliates International by Hospital
Corporation of America (1981). Merger of two major hospital
management companies.
South Florida Physicians' Boycott (1983). Boycott by physicians
to place pressure on the legislature to enact malpractice insurance
legislation favorable to physicians.
Stanislaus Preferred Provider Organization (SPPO) (1984).
Agreement by physician members of SPPO not to contract with any
other PPOs allegedly in order to forestall the development of PPO
competition in Stanislaus County.
Policy Matters
The Division's position on the Health Care Cost Containment Act
of 1983 (1984). This position was delivered in testimony by Charles
F. Rule, Deputy Assistant Attorney General, to a Senate
Subcommittee.
Letter to the Health Care Financing Administration (HCFA)
(1984). This letter expressed the Division's views on certain
proposals which would restrict the dissemination of information
collected by Professional Review Organizations.
The Division's policy toward the health care sector in general
and preferred provider organizations (PPOs) in particular (1985).
This policy was expressed in a paper presented by J. Paul McGrath,
Assistant Attorney General, to the National Health Lawyers
Association and the ABA.
Business Review commenting on plans by the Southwest Michigan
Health Systems Agency (HSA) (1982). The HSA wanted to publish rates
charged by hospitals within the HSA.
Business Review commenting on a proposal by the Maryland Health
Care Coalition (1982). The Coalition wanted to collect and
disseminate information concerning the incentive effects of
different types of insurance policies.
Letters to the ABA and State Supreme Courts (1982-1984). These
letters expressed the Division's views on restrictions on
advertising and solicitation contained in the ABA's Model Rules.
Publications
``An antitrust analysis of the Alliant decision and defense
industry mergers,'' International Merger Law, April 1993 (w/Philip
B. Nelson) [Note: a shorter version appeared in Economists Ink
(Winter 1993), a newsletter published by Economists Incorporated.]
``Coated Groundwood Paper Anti-Dumping Investigation,''
Economists Ink (Winter 1993).
Curriculum Vitae
Kent W. Mikkelsen
Office
Economists Incorporated, 1200 New Hampshire Ave., NW., Suite 400,
Washington, DC 20036, (202) 833-5240, mikkelsen.k@ei.com
Home
3012 Fayette Road, Kensington, MD 20895, (301) 946-8901
Background
Born: September 20, 1954, married, 3 children
Education
Ph.D., Economics, Yale University, 1984
M.Phil., Economics, Yale University, 1981
M.A., Economics, Yale University, 1980
B.A., Economics, Brigham Young University, 1978, summa cum laude
[[Page 32855]]
Fellowships, Honors and Awards
College Valedictorian, Brigham Young University, 1978
H. B. Earhart Fellow, 1978-1979
University Fellow, Yale University, 1978-1980
Richard Bernhard Fellow, 1980-1981
Research Scholar, International Rice Research Institute, 1981
Fields of Concentration
Industrial Organization, Economic Development
Professional Experience
1986-present: Senior Vice President, Economists Incorporated
1984-1986: Economist, Economic Analysis Group, Antitrust Division,
U.S. Department of Justice
1983-1984: Visiting Assistant Professor, University of Michigan
1982: Acting Instructor, Yale University
1981-1982: Teaching Fellow, Yale University
1979-1983: Research Fellow, Yale University
Testimony
Expert witness for Government in United States v. Calmar Inc.
and Realex Corp., United States District Court, District of New
Jersey, Civil Action No. 84-5271.
Expert witness for Defendant in Sunbelt Television, Inc. v.
Jones Intercable, Inc., United States District Court, Central
District of California, Case No. CV-91-3506 WDK (Kx).
Expert witness for Defendant in Stag-Parkway, Inc. v. The
Dometic Corporation, United States District Court, Northern District
of Georgia, Case No. 1-91-CV-2579-JOF.
Expert witness for Plaintiff in Thomas L. Hopkins (Commonwealth
of Virginia) v. Smithfield Foods, Inc., Virginia Circuit Court, Isle
of Wight County, No. 96-125.
Expert witness for Defendant in Elpizo Limited Partnership v.
Marriott International, Inc. and Host Marriott Corporation v.
Maryland Hospitality, Inc., Court of Common Pleas for Philadelphia
County, Pennsylvania, October Term, 1994, No. 607.
Expert witness for Plaintiff in Thomas L. Hopkins (Commonwealth
of Virginia) v. Smithfield Foods, Inc., Virginia Circuit Court, Isle
of Wight County, No. 97-80.
Expert witness for Defendant in Consumer Health Foundation v.
Humana Group Health Plan, Inc., et al., United States District
Court, District of Columbia, Case No. 1:98CV02920 (GK).
Expert witness for Defendant in United States v. Broadcast
Music, Inc. United States District Court, Southern District, New
York, 64 Cir. 3787 (LLS).
Expert witness for Defendants Advance Stores Company, Inc. and
Discount Auto Parts, Inc. in Coalition for a Level Playing Field LLC
et al. v. AutoZone, Inc., et al., United States District Court,
Eastern District of New York, No. CV 00 0953 (LDW) (ETB).
Expert witness for Defendant in United States v. Broadcast
Music, Inc. United States District Court, Southern District, New
York, 64 Cir. 3787 (LLS), remand proceeding.
Expert witness for Defendants in Ramallo Bros. Printing, Inc. v.
El Dia, Inc. et al., United States District Court, District of
Puerto Rico, Civil No. 02-2400 (JAF).
Expert witness for Defendant in Marco Island Cable, Inc. v.
Comcast Cablevision of the South, Inc., United States District
Court, Middle District of Florida, Case No. 2:04cv-26-FtM-29-DNF.
Testimony, Federal Communications Commission En Banc Hearing
Regarding Local Television Ownership Rules, February 12, 1999.
Testimony, United States Senate Committee on Commerce, Science,
and Transportation, Hearing on Media Ownership, May 22, 2003.
Selected Consulting Matters
Detroit Free Press and Detroit News Joint Operating Agreement
(JOA)--Prepared economic and business analysis used in hearing
before Administrative Law Judge.
Soft Drink Price Fixing--Analysis of evidence of price fixing
and estimation of damages in Department of Justice investigations
and private damage suits against various CocaCola bottlers.
Federal Communications Commission Inquiry into Cable
Television--Supervised and wrote up research projects regarding
cable rates and vertical market structure submitted with briefs
filed by TCI.
GenCorp acquisition from Goodyear, Department of Justice
review--Analyzed demand and supply-side substitution for vinyl
laminates.
York acquisition of Hyster, Federal Trade Commission review--
Analyzed geographic market definition in forklift trucks.
Stag-Parkway v. Dometic--For defendant, testified regarding lack
of injury and damages due to price discrimination in sales to
distributors.
Sunbelt Television v. Jones Intercable--For defendant, testified
regarding market definition and monopoly power in local advertising
and critiqued plaintiff's damage study.
Kiwifruit antidumping investigation by International Trade
Commission--Coordinated preparation of economic analysis for New
Zealand respondents.
State of Virginia v. Smithfield Foods--For plaintiff, evaluated
the economic gain defendant received through non-compliance with
environmental laws.
Federal Communications Commission Inquiries into Broadcast
Television--For three broadcast networks, prepared comments on the
economic effects of prime-time access rules and station ownership
rules.
Elpizo Ltd Partnership v. Marriott--For defendant, analyzed
plaintiff's damages model and testified regarding inappropriateness
of plaintiff's damages model.
Media Ownership Rules--Researched and submitted three separate
papers to FCC on behalf of ABC, CBS and Newspaper Association of
America.
Cable & Wireless Optus acquisition of AAPT--Presented analysis
of multiple telecommunications markets to Australian Competition and
Consumer Commission.
TeleCell Cellular, Inc. et al. v. GTE Mobilnet of South Texas
Limited Partnership--For defendant, analyzed damages claims of
plaintiffs for compensation allegedly less favorable than another
cellular agent.
Kesmai Corp. et al. v. America Online--For defendant, analyzed
plaintiff's claim of damages to Internet games business.
Federal Communications Commission En Banc Hearing--Presented
testimony on FCC local television ownership rules.
Consumer Health Foundation v. Humana--For defendant, evaluated
damages from alleged delay in releasing payment.
API v. Granite--Advisor to court-appointed special master making
findings on below-cost pricing in road construction.
Hearst Acquisition of San Francisco Chronicle--For Hearst,
prepared analysis showing prospects for competition by San Francisco
Examiner outside the JOA and incremental contribution of Examiner to
JOA profits.
Denver Post-Denver Rocky Mountain News JOA--For applicants,
analyzed probable failure and incremental unprofitability of the
News.
United States v. BMI--For defendants, testified about a
reasonable royalty rate for a music performing right blanket
license.
Vitamin Price Fixing Case--Submitted expert reports finding no
incentive for two vitamin producers to participate in conspiracies
involving vitamins they did not manufacture.
Newspaper-Broadcast Cross-Ownership Rule--For the Newspaper
Association of America, submitted a paper to the FCC on structural
change since 1975 and potential benefits of joint ownership.
Advance-Discount Robinson-Patman Case--For defendants, testified
about drawing cost inferences from pricing data.
U.S. Senate Commerce Committee Hearing--Presented testimony
supporting elimination of three FCC rules governing ownership of
broadcast stations.
IPSCO v. Mannesmann Steel Mill Case--For defendant, analyzed
damages from deficiencies of a steel mill.
TRICO v. NKK et al. Steel Mill Case--For plaintiff, analyzed
damages from deficiencies of a steel mill.
FCC ``Omnibus'' Broadcast Ownership Proceeding--For CBS, Fox and
NBC, analyzed station ownership, news broadcast and diversity
issues.
FCC Cable Bundling and Retransmission--For Disney, submitted
analysis of proposals to mandate a la carte cable programming and
value of cable retransmission rights for ABC stations.
Heavy-Duty Trucks--For defendant Mack Trucks, submitted expert
report discussing market definition, market power, alleged
anticompetitive practices and damages.
Printing Monopolization--For defendant El Dia, testified on
market definition, dangerous probability, and alleged
anticompetitive practices including predatory pricing.
Dissolution of Birmingham JOA--For Birmingham News Post-Herald
and Birmingham Post-Herald, presented to DOJ an analysis of the
incremental unprofitability of the Post Herald.
Cable Monopolization--For defendant Comcast, testified on
alleged monopolization
[[Page 32856]]
and anticompetitive practices including exclusive contracts and on
damages.
Regulatory Impact--For a consortium of telecommunications firms
in Bermuda, analyzed the impact of proposed regulatory changes.
Attachment B--White Paper by Economists Incorporated, Submitted on
Behalf of the NAA to DOJ on April 11, 2007
Economists Incorporated
An Economic Analysis of the Competitive Effects of the Proposed
Abitibi-Bowater Merger
Submitted to DOJ on Behalf of NAA
John H. Preston, Kent W. Mikkelsen, PhD, Economists Incorporated,
Washington, DC, April 11, 2007.
Table of Contents
Section A. Overview of the White Paper
1. Introduction.
2. Summary of Our Analysis and Our Main Conclusions.
3. A Note on Our Sources.
Section B. Product and Geographic Market Definition
1. Introduction.
2. A Description of Newsprint and Uncoated Groundwood Specialty
Grades.
3. Product Market Definition.
4. Geographic Market Definition.
Section C. Analysis of the Increase in Concentration That Would Result
From the Proposed Merger
1. Analysis of the Increase in Concentration in the NA Newsprint
Market Based on Estimated 2006 Capacity.
2. Analysis of the Increase in Concentration in the East of the
Rockies Newsprint Market Based on Estimated 2006 Capacity.
Section D. Analysis of the Increase in Concentration and Decrease in
Capacity in the NA Newsprint Market 1995-2006
1. The Increase in Concentration in the NA Newsprint Market
1995-2005 as Described by Abitibi and Bowater.
2. Concentration in the NA Newsprint Market in 1995.
3. Acquisitions and Exits of NA Newsprint Manufacturers since
1995.
4. Analysis of the Reduction of Newsprint Capacity in North
America 1995 to 2006.
Section E. NA Newsprint Demand and Supply
1. Introduction.
2. NA Demand (Quantity Purchased) 1999-2006.
3. Causes of the Decline in NA Newsprint Demand 1999-2006.
4. Projected NA Newsprint Demand 2006-2008.
5. Production, Shipments and Operating Rates of NA Newsprint
Mills 1999-2006.
6. The Price of Newsprint per Metric Tonne (Eastern U.S., 30
lb.) 1999 to 2006 by Quarter.
Section F. Evidence From Presentations to Investment Analysts and Other
Public Information That Abitibi and Bowater Have Used Their Control
Over Newsprint Capacity and the Newsprint Industry Operating Rate To
Significantly Raise the Price of Newsprint 2002 to 2006
1. Introduction.
2. Presentation by John Weaver, President and CEO of Abitibi, at
the Citigroup Conference in December 2006.
3. Presentation by David Paterson, President and CEO of Bowater,
at the Citigroup Conference in December 2006.
4. Presentation by John Weaver, President and CEO of Abitibi, at
the Credit Suisse First Boston Investment Analysts Conference in
March 2004.
5. Interview of John Weaver Titled ``Tighter supply/demand
balance boosts newsprint hike prospects says Abitibi's Weaver.''
Section G. An Analysis of Permanent Newsprint Capacity Reductions
Between 2002 and 2006
1. Introduction.
2. Chart G1: Shares of NA Newsprint Capacity by Manufacturer
2002 arid 2006.
3. Chart G2: Permanent Reduction of NA Newsprint Capacity by
Manufacturer During the Period 2002-2006.
4. Chart G3: Percentage of Total NA Permanent Newsprint Capacity
Reduction by Manufacturer During the Period 2002-2006.
5. Chart G4. Permanent Reduction of Newsprint Capacity Over the
Period 2002-2006 as a Percentage of Own 2002 NA Capacity by
Manufacturer.
Section H. Four Articles by Two Newsprint Industry Experts Describing
the Abitibi-Bowater Strategy To Raise Price by Closing Capacity
1. Introduction.
2. Article by Harold M. Cody Titled ``New Paradigm: Newsprint
Demand Falls, Prices Soar.''
3. Three Articles by RISI Senior Economist Andrew Battista
Analyzing the Strategy of Abitibi and Bowater to Shut Down Capacity
to Maintain High Operating Rates and Increasing Prices.
Section I. Abitibi's Newsprint Capacity Closures 1999 to 2001
Section J. A Comparison of Newsprint Prices With the Prices of Uncoated
Groundwood Specialty Grades 3Q 1999 to 4Q 2006
1. Introduction.
2. The Adverse Impact of the Increases in Input Prices and the
Appreciation of the Canadian Dollar Has Fallen More Heavily on
Producers of Uncoated Groundwood Specialty Grades Than on Producers
of Newsprint.
3. Comparing Quarterly Prices tbr Newsprint and Uncoated
Groundwood Grades from 3Q 1999 Though 4Q 2006.
4. Abitibi's Variable Costs to Produce Newsprint and Uncoated
Groundwood Specialty Grades Have Been Relatively Constant for the
Period 2001-2005.
5. Applying the Percentage Price Changes for the Uncoated
Groundwood Specialty Grades to the 3Q 1999 Price of Newsprint to
Determine the Effect on Newsprint Revenues from Sales to NA
Customers.
Section K. Dominant Firm Model
Section L. Conclusions
Attachments
Attachment A Links to Newsprint-Related Web Sites
Attachment B Additional Analysis of Uncoated Groundwood Specialty
Grades and Tables B1 to B7 for Section B
Attachment C Tables C1 to C3 for Section C
Attachment D Tables D1 to D4 for Section D
Attachment K Technical Appendix to Section K Dominant Firm Model
Section A. Overview of the White Paper
1. Introduction
Economists Incorporated has been asked by the Newspaper
Association of America (``NAA''), an association of U.S. daily
newspapers, to prepare an economic analysis of the likely
competitive effects of the proposed Abitibi-Bowater merger in the
North American (``NA'') newsprint market (``White Paper'') with the
intent to provide that analysis to the U.S. Department of Justice to
assist the department in its investigation of the proposed merger.
The objective of this White Paper is to analyze the economic
effects of the proposed merger of Abitibi and Bowater and to
identify any anticompetitive consequences of the proposed merger.
In Section A below, we summarize our analysis and main
conclusions of each section. Sections A, B, C, D, and K have
attachments containing data and analysis related to the analysis in
the section.
The last subsection of Section A contains a table of contents to
the White Paper. Attachment A to Section A provides a list of the
Internet addresses of newsprint manufacturers and other Web sites
most frequently cited in the White Paper.
2. Summary of Our Analysis and Our Main Conclusions
a. Introduction
This section provides a summary of our analysis and our main
conclusions reached in Sections B through L.
b. Section B. Market Definition
In Section B, we conclude that the relevant product market is
newsprint and that the relevant geographic market is NA. We also
provide some evidence that East of the Rockies may be a relevant
geographic market.
Our analysis shows that new Chinese capacity is likely to be
largely if not entirely absorbed in Asia over the next couple of
years and will not have a significant impact on the NA market. That
is also the expectation of Abitibi, Bowater, and the PPPC. If there
is an effect on the NA newsprint market from the new Chinese
capacity, it is likely to be indirect. Some NA mills may be
displaced from some Asian accounts by the new Chinese capacity.
However, the effect on the NA newsprint market of any displacement
is not likely to be significant. Abitibi and Bowater, who
[[Page 32857]]
combined account for about 70% of exports from North America, expect
strong export growth in 2007. Abitibi expects its exports from NA to
grow by 10% in 2007 and Bowater expects its exports from NA to grow
by 5% to 6% in 2007.
Attachment B to Section B provides additional analysis of the
relation between uncoated groundwood specialty grades and newsprint.
One analysis compares prices of four uncoated groundwood grades with
the price of newsprint. A second analysis shows the estimated
combined Bowater and Abitibi capacity share of several uncoated
groundwood specialty segments.
c. Section C. Analysis of the Increase in Concentration That Would
Result From the Proposed Merger
In Section C, we identify all of the suppliers to the NA
newsprint market and estimate their 2006 newsprint capacity by mill.
Based on estimated 2006 capacity, we show that the combined Abitibi-
Bowater would have a capacity share of 45.0%. The premerger HHI is
1,380, the change in the HHI that would result from the merger is
962 and the post-merger HHI is 2,342. According to Sec. 1.51(c) of
the Merger Guidelines, markets with post-merger HHIs above 1,800 are
highly concentrated and that HHIs of this magnitude create the
presumption that the merger would be ``likely to create or enhance
market power or facilitate its exercise.''
Based on estimated 2006 capacity, we also calculate capacity
shares and HHIs for a possible East of the Rockies relevant
newsprint market. We show that the combined Abitibi-Bowater would
have a capacity share of 54.3%. The pre-merger HHI is 1,876, the
change in the HHI that would result from the merger is 1,445 and the
post-merger HHI is 3,321.
Attachment C to Section C contains tables showing 2006 estimated
capacity by NA mill and the capacity share and HHI calculations for
the NA newsprint market and the East of the Rockies market.
d. Section D. Analysis of the Increase in Concentration and Decrease in
Capacity in the NA Newsprint Market 1995-2006
Due primarily to acquisitions by Abitibi and Bowater between
1995 and 2001, the NA newsprint market was transformed from an
unconcentrated market in 1995 to a highly concentrated market in
2000 with Abitibi's acquisition of Donohue in April 2000. Bowater's
acquisition of Alliance in 2001 and Norske Skog's acquisition of
Pacifica, also in 2001, further increased concentration in an
already highly concentrated market. This section analyzes the
mergers and increase in concentration that occurred between 1995 and
2006.
Both John Weaver, President and CEO of Abitibi, and David
Paterson, President and CEO of Bowater, have made presentations to
investment analyst conferences describing the significant
consolidation in the NA newsprint market and the roles of Abitibi
and Bowater in achieving that consolidation. Slides from their
presentations illustrating the increase in consolidation are
included in this section.
This section also shows that there has been a 20.7% reduction in
NA newsprint capacity between 1995 and 2006. Most of that reduction
has occurred since 2002. Of the 16 firms that remain in the NA
newsprint market today, Abitibi and Bowater combined account for
83.6% of that capacity reduction and Catalyst accounts for 15.9%.
The other 13 firms account for 0.5%.
Attachment D to Section D contains tables showing how Abitibi
and Bowater significantly increased their shares of newsprint
capacity between 1995 and 2001 and tables showing that Abitibi and
Bowater account for most of the reduction in NA newsprint capacity
between 1995 and 2006, which primarily occurred after 2002.
e. Section E. NA Newsprint Demand and Supply
This section provides charts showing annual and quarterly data
concerning NA newsprint demand and supply from 1999 through 2006.
Almost all of the data are from standard industry sources PPPC and
RISI. Chart E7 in Section E shows a steady decline in quarterly NA
newsprint demand (quantity purchased) between 1999 and 2006. The
chart also shows quarterly newsprint prices (30 lb., Eastern U.S.)
over that same period. Chart E7 shows that while NA newsprint demand
(quantity purchased) fell 18.0% between the third quarter of 2002
and the third quarter of 2006, the price of news print increased an
aggregate of 49.0% over that same period.
Section E also analyzes the causes of the decline in newsprint
consumption over time by U.S. daily newspapers. We conclude that the
primary causes are declining newspaper circulation, declining
advertising lineage, and newspaper efforts to reduce the consumption
of newsprint by reducing the width of newspaper pages, by switching
to lower basis weight paper, and by moving some content to the
newspaper Web site. Declining circulation and advertising lineage
should be regarded as exogenously shifting the newspapers' demand
curve for newsprint downward. These declines are unrelated to the
price of newsprint. While newspaper efforts to conserve on newsprint
are largely in reaction to increasing newsprint prices, they should
be regarded as efforts to permanently shift the demand curve for
newsprint downward. If the price of newsprint drops significantly,
it is improbable that newspapers will respond by increasing the
width of newspaper pages or return content to the newspaper that was
placed on Web sites. As long as the relative prices for higher and
lower basis weight paper remain approximately the same, as seems
likely, newspapers will have no incentive to switch back to higher
basis weight paper.
f. Section F. Evidence From Presentations to Investment Analysts and
Other Public Information That Abitibi and Bowater Have Used Their
Control Over Newsprint Capacity and the Newsprint Industry Operating
Rate To Significantly Raise the Price of Newsprint 2002 to 2006
As noted above, the price of newsprint increased 49.0% from the
third quarter of 2002 to the third quarter of 2006 while the demand
for newsprint (quantity demanded) declined 18.0%. Since the
reductions in newsprint demand were largely caused by exogenous
factors, the price of newsprint would be expected to decline holding
the supply curve constant. However, the supply was not held constant
during this period. Abitibi and Bowater responded to continual
downward shifts in the demand curve by indefinitely idling and
permanently closing their own capacity. Each downward shift in the
demand curve was met with an upward shift in the supply curve
sufficient to maintain maximum practical NA newsprint industry
operating rates. At maximum practical operating rates price
increases can successfully be imposed as they were throughout this
four-year period by Abitibi and Bowater. The remaining firms in the
industry generally followed the announced price increases of Abitibi
and Bowater within a month or two.
John Weaver, President and CEO of Abitibi, has been describing
this strategy in slide show presentations at investment analyst
conferences since 2003. David Paterson, who became President and CEO
of Bowater in April 2006, discussed this strategy at an investment
analysts' conference in December 2006. Section F documents the
AbitibiBowater strategy to use their control of capacity to raise
the price of newsprint through an analysis of relevant slides
presented and described by Weaver and Paterson at investment analyst
conferences. This section also contains excerpts from an interview
of Weaver that relate to this strategy.
g. Section G. An Analysis of Permanent Newsprint Capacity Reductions
Between 2002 and 2006
Section G contains an analysis of permanent newsprint capacity
reduction in NA between 2002 and 2006. The analysis shows that 18.0%
of NA newsprint capacity was removed from the market between the end
of 2000 and the end of 2006. Abitibi and Bowater combined were
responsible for 80.0% of the permanent capacity removals and
Catalyst was responsible for 7.3%. Of manufacturers that remain in
the market today, Abitibi and Bowater combined account for 89.4% of
the total capacity removals and Catalyst accounts for 8.1%. The
other 13 remaining firms account for 2.5% of the capacity removals.
Through these permanent capacity removals, Abitibi reduced its
own capacity by 30.7% and Bowater reduced its own capacity by 24.0%.
Catalyst also reduced its newsprint capacity by a significant
proportion--22.7%. The other 13 newsprint manufacturers that remain
in the market today reduced their capacity by a combined 1.0%.
h. Section H. Four Articles by Two Newsprint Industry Experts
Describing the Abitibi-Bowater Strategy to Raise Price by Closing
Capacity
The Abitibi-Bowater strategy to use their control of capacity to
raise newsprint prices is well known within the newsprint industry.
Every newspaper newsprint buyer that we talked to described the
Abitibi-Bowater strategy. This section analyzes four articles by two
newsprint experts. These articles accurately describe the Abitibi-
Bowater
[[Page 32858]]
strategy. The titles of the articles are ``(1) New Paradigm:
Newsprint Demand Falls, Prices Soar,'' (2) ``Will operating rates
climb high enough in 2003 to support rising newsprint prices in the
U.S.?,'' (3) ``Is rising newsprint demand necessary to support
higher prices in 2004?,'' and (4) ``Newsprint producers must rely on
supply reductions to support rising prices.''
i. Section I. Abitibi's Newsprint Capacity Closures 1999 to 2001
Between the third quarter of 1999 and the second quarter of
2001, newsprint prices increased 30.2% as shown in Section E6.
Abitibi's permanent capacity removals immediately before and during
that period were a significant cause of the price increases. Abitibi
removed 450,000 metric tonnes of newsprint capacity from the market
in 1999 or almost 3% of NA capacity. In conjunction with its
acquisition of Donohue in April 2000, Abitibi announced that it
would remove an additional 400,000 metric tonnes of newsprint
capacity from the market during 2000 and 2001. Section I documents
Abitibi's permanent newsprint capacity removals between 1999 and
2001.
j. Section J. A Comparison of Newsprint Prices With the Prices of
Uncoated Groundwood Specialty Grades 3Q 1999 to 4Q 2006
Over the last four years there have been significant increases
in energy, fiber, and transportation costs faced by NA newsprint
manufacturers. Newsprint mills in Eastern Canada have been
especially hard hit. In addition, the appreciation of the Canadian
dollar relative to the U.S. dollar has effectively raised the cost
of Canadian newsprint mills while lowering the cost of U.S.
newsprint mills.
In this section, we compare the price of newsprint from the
third quarter of 1999 to the second quarter of 2006 with the prices
of four uncoated groundwood specialty grades. We find that the
quarterly prices for newsprint as a percentage of its price in 3Q
1999 were significantly higher than the quarterly prices for three
of the four uncoated groundwood specialty grades over the period 4Q
1999 to 2Q 2006. Based on these results, it is implausible that the
increases in newsprint prices were caused by the increases in input
prices. We find that the price trend of one uncoated groundwood
specialty grade was similar to that of newsprint. It appears that
Abitibi and Bowater are the dominant providers of that grade as
well.
Section J presents a slide from an Abitibi presentation showing
that Abitibi's variable cost of newsprint production has been
virtually flat between 2001 and 2005. Since all or nearly all of the
newsprint price increases over the period 2002 to 2006 were led by
Abitibi, it seems unlikely that increases in Abitibi's input costs
are a plausible justification for the price increases.
In Section J, we also calculate quarterly newsprint revenues
over the period 3Q 1999 to 2Q 2006 based on actual NA newsprint
consumption and actual newsprint prices. We then apply the quarter
to quarter percentage price changes for each of the four uncoated
groundwood specialty grades to the 3Q 1999 newsprint price and
multiply the resulting adjusted newsprint prices by actual NA
demand. For the three grades with percentage changes in prices
significantly below the percentage changes in newsprint prices,
total revenues over the period are reduced by $4.7 billion to $7.4
billion.
k. Section K. Dominant Firm Model
Based on our analysis in Sections B through J, we conclude that
Abitibi and Bowater have acted as a joint dominant firm since at
least the end of 2002 and perhaps since 2000. Abitibi and Bowater
have jointly used capacity closures to raise the price of newsprint
well above competitive levels. By removing capacity from the
newsprint market in a timely way Abitibi and Bowater have been able
to maintain maximum practical operating rates for the newsprint
industry which has directly led to the price increases.
Section K presents a theoretical dominant firm model which
formally explains the joint behavior of Abitibi and Bowater. Using
current data, we use the dominant firm model to predict whether it
would be profitable for a merged Abitibi-Bowater to further increase
the price of newsprint through additional capacity closures. We show
that it would be profitable for the merged firm to close additional
capacity to achieve a 5% price increase.
Attachment K of Section K is a technical appendix in which the
equations of the dominant firm model are formally derived.
l. Section L. Conclusions
Based on our analysis in Sections B through J, we conclude that
the joint strategy of Abitibi and Bowater to close NA newsprint
capacity to raise the price of newsprint is anticompetitive and has
caused significant economic harm to U.S. daily newspapers and other
NA purchasers of newsprint.
We predict that if the proposed merger is allowed to proceed,
the ability of the merged entity to pursue the Abitibi-Bowater
strategy of closing capacity to raise the price of newsprint will be
strengthened. The market power of the merged firm will be more
effectively employed than Abitibi and Bowater were able to do as
separate but coordinating firms. The possibility that one of the two
firms would stop coordinating, resulting in a price decrease, will
be eliminated once the two firms are merged. The merged entity will
have an increased incentive and ability to use its control over
capacity to raise the price of newsprint significantly above
competitive levels. Newsprint consumers, whom the antitrust laws are
designed to protect, will suffer additional significant competitive
harm.
3. A Note on Our Sources
In conducting our analysis of the competitive effects of the
proposed AbitibiBowater merger, we relied on a wide variety of
newsprint industry sources. Amongst the most important sources for
data and other information concerning the NA newsprint industry are
RISI and the Pulp and Paper Products Council (``PPPC''). RISI is the
leading source of information about the pulp, paper, and forest
products industries in NA and worldwide. The PPPC is a private
organization that compiles demand and supply data and conducts
forecasts for North American producers of pulp and paper products,
including manufacturers of newsprint. We also relied on information
that Abitibi and Bowater make publicly available on their Web sites
as well as similar information on the Web sites of other newsprint
manufacturers. We interviewed a number of U.S. newspaper newsprint
buyers who gave us their perspective on the NA newsprint market and
the likely competitive effects of the proposed merger. The NAA also
provided us with data.
B. Product and Geographic Market Definition
1. Introduction
The principles of product and geographic market definition are
set out in the U.S. Department of Justice (``DOJ'') and Federal
Trade Commission (``FTC'') Horizontal Merger Guidelines (``Merger
Guidelines'').\1\ Following the Merger Guidelines methodology,
product and geographic markets are defined from the perspective of
consumers of the products of the merging firms. With respect to the
proposed merger of Abitibi and Bowater, the two companies
manufacture newsprint, uncoated groundwood specialty grades, and
other pulp, paper, and forest products which they sell in NA and, in
some cases, in other regions of the world. We have been asked to
determine likely competitive effects of an Abitibi-Bowater merger on
the sale of newsprint in NA. Our provisional product market is
newsprint and our provisional geographic market is NA.\2\
---------------------------------------------------------------------------
\1\ The Merger Guidelines were issued on April 2, 1992 and
revised on April 8, 1997.
\2\ In Section B2 in Attachment B and Section J below, we
provide some evidence that suggests that the proposed merger of
Abitibi and Bowater may have an adverse competitive effect
concerning at least one uncoated groundwood specialty grade.
---------------------------------------------------------------------------
2. A Description of Newsprint and Uncoated Groundwood Specialty
Grades
a. Introduction
The main focus of this analysis is on the production and sale of
newsprint in NA, but in applying the methodology of the Merger
Guidelines to a provisional newsprint product market, it is
necessary to consider demand and supply substitution possibilities
regarding closely related uncoated groundwood specialty grades.
While we do not reach any firm conclusions on relevant product
markets within the uncoated groundwood specialty grade segment, we
provide considerable information about the grade structure of that
segment and the manufacturers of uncoated groundwood specialty
grades in this section.
b. Newsprint
Newsprint is used to print newspapers, inserts, flyers and other
advertising materials. In the U.S., the main purchasers of newsprint
are newspaper publishers (both daily and non-daily) and commercial
printers. In 2006, U.S. daily newspapers accounted for 80.0% of the
U.S. consumption of newsprint.\3\ U.S. demand for newsprint
accounted for 88.9%
[[Page 32859]]
of total NA demand for newsprint in 2006.\4\ The PPPC estimated 2006
NA newsprint capacity at 12,625,000 metric tonnes.\5\
---------------------------------------------------------------------------
\3\ Source: December 2006 PPPC Flash Report.
\4\ Source: December 2006 PPPC Flash Report. The PPPC Flash
Report does not provide data on Canadian consumption of newsprint
nor does it provide data on purchases of newsprint by Canadian daily
newspapers. The difference between annual demand and annual
consumption is the change in inventories from the prior December.
\5\ See PPPC's March 29, 2006 NA Mechanical Printing Papers
Forecast.
---------------------------------------------------------------------------
Newsprint is the lowest quality and least expensive uncoated
groundwood paper. The main ingredient of newsprint is groundwood
pulp, also known as mechanical pulp, recycled fiber (old newspapers
(ONP) and old magazines (OMG)), or a combination of groundwood pulp
and recycled fiber. Chemical pulp is usually added to the pulp
furnish to improve runnability on printing presses.
Although newsprint must meet the exacting standards of modem
printing presses, it is a commodity grade. About half the newsprint
sold in NA today has a basis weight of 30 lb. (48.8 grams per square
inch) and about half has a basis weight of 27.7 lb. (45.0 grams per
square inch). Over the last four or five years newspapers have been
gradually switching from the heavier basis weight newsprint to the
lighter basis weight newsprint.\6\
---------------------------------------------------------------------------
\6\ There is a financial gain for a newspaper from switching to
the lower basis weight paper, but it is not a large gain. The switch
to the lower basis weight reduces a newspaper's consumption of
newsprint by 8.5% holding the square footage of newsprint purchased
constant, but the gain to the newspaper from the reduced consumption
is mostly offset by the higher price that the newspaper must pay for
the lower basis weight paper. Based on February newsprint prices,
the net gain to a newspaper from switching would be a cost saving
per metric tonne of about 2.7% or $16.94.
---------------------------------------------------------------------------
c. Uncoated Groundwood Specialty Grades
(1). The Similarities and Differences Between Newsprint and Uncoated
Groundwood Specialty Grades
To evaluate demand and supply substitution possibilities in a
provisional newsprint market, it is necessary to describe in some
detail the similarities and differences between newsprint and higher
quality and higher value uncoated groundwood specialty grades. See
Attachment B for (a) a comparison of the price of newsprint with the
prices of four uncoated groundwood specialty grades and (b) Abitibi-
Bowater HHIs based on estimated 2006 capacity and capacity shares by
manufacturer for uncoated groundwood specialty grade segments.
Newsprint is a type of uncoated groundwood paper, but to
distinguish newsprint from other uncoated groundwood grades, the
paper industry refers to the other uncoated groundwood grades as
uncoated groundwood specialty grades or higher value uncoated
groundwood grades. Uncoated groundwood paper is also referred to as
uncoated mechanical paper.
RISI estimated the 2006 NA capacity of uncoated groundwood
specialty grades as 6,915,000 metric tonnes or somewhat more than
half of 2006 NA newsprint capacity.\7\ Some uncoated groundwood
specialty grades are produced on machines that never produced
newsprint, some uncoated groundwood specialty grades are produced on
machines that have been converted from newsprint production, and
some uncoated groundwood grades are produced on machines that also
produce newsprint. Machines that produce both newsprint and uncoated
groundwood specialty grades are called ``swing'' machines.
---------------------------------------------------------------------------
\7\ See the 2006 RISI Fact and Price Book, p. 163. The PPPC
forecast 2006 NA capacity for uncoated groundwood specialty grades
as 6,360,000 metric tonnes. See the PPPC March 2006 forecast. We can
account for some but not all of the difference between the RISI
estimate and the PPPC forecast.
---------------------------------------------------------------------------
There are significant similarities in the production process for
newsprint and the production processes for uncoated groundwood
specialty grades. Indeed, many machines that currently produce
uncoated groundwood specialty grades were formerly newsprint
machines. The main ingredient of newsprint and uncoated groundwood
specialty grades is pulp produced from some combination of
groundwood pulp, recycled fiber, and chemical pulp. The grades vary
by brightness, gloss, basis weight, opacity, and strength.
Generally, newsprint has the lowest value combination of these
characteristics. Higher value uncoated groundwood specialty grades
are glossier and brighter than newsprint.
Slide 13 below, which is from Abitibi's presentation of
financial results for Q1 2006 in June 2006, shows uncoated and
coated printing paper grades.\8\
---------------------------------------------------------------------------
\8\ This presentation is available on Abitibi's Web site under
Investor Relations/Presentations & Webcasts.
---------------------------------------------------------------------------
A graph appearing in this comment is not able to be reprinted
here. Copies of the comment with the graph are available at the
Department of Justice Antitrust Division Web site, http://
www.usdoj.gov/atr, at the Antitrust Documents Group of the
Department of Justice Antitrust Division, 450 Fifth Street, NW.,
Suite 1010, Washington, DC 20530, (202) 514-2481, and at the Office
of the Clerk of the United States District Court for the District of
Columbia, 333 Constitution Avenue, NW., Washington, DC 20001.
The slide is titled ``Paper Spectrum'' and states that ``Two key
properties, brightness and gloss, define paper grade groups.'' The
blue ovals in Slide 13 represent printing paper products produced by
Abitibi. The blue ovals also identify the main uncoated groundwood
specialty grades. The white ovals identify coated groundwood grades
(all of coated 5 and some of coated 4) and coated
free sheet grades (some of coated 4 and all of coated
3).\9\ These coated grades are not produced by Abitibi.
However, Bowater does produce 3, 4, and 5
coated paper.
---------------------------------------------------------------------------
\9\ There are two additional coated free sheet grades not shown
in Slide 13, coated 1 and coated 2.
---------------------------------------------------------------------------
The uncoated groundwood specialty grades can be divided into two
categories: Glossy and non-glossy. The glossy grades are
distinguished primarily by their degree of glossiness. The non-
glossy grades are distinguished primarily by their degree of
brightness. These distinctions are apparent in Slide 13.
Newsprint is a non-glossy grade. Newsprint is the least bright
and, with the exception of the bulky book grade (ABIbook), the least
smooth of the non-glossy grades. In terms of smoothness and
brightness, the bulky book grade is closest to newsprint followed by
the directory grade and the Hi-Brite grade (ABIbrite). Bulky book
paper is typically used for paperback books and coloring books.
Directory paper is somewhat brighter and smoother than newsprint and
is also lighter (basis weight typically 22.1 lb. vs. 30 lb. or 27.1
lb. for newsprint) and is used primarily for the printing of
telephone directories. The typical brightness of newsprint is 58.
The brightness of Hi-Brite grades ranges from 65 to 75. Hi-Brite
grades are used for printing inserts and flyers and in other similar
commercial printing applications. The brightness of Super Hi-Brite
grades (Abitibi grades EO, IO, and AO) ranges from 75 to 85.\10\
Abitibi's Super Hi-Brite Grades compete with uncoated free sheet for
the printing of books and may also be used in commercial printing
applications.
---------------------------------------------------------------------------
\10\ We follow two practices of Abitibi in terminology and
brightness ranges. Abitibi calls the Highbright and Super-bright
grades Hi-Brites and Super Hi-Brites, respectively. Abitibi sells
Hi-Brites in the brightness range 65-75 and Super Hi-Brites in the
brightness range 75 and over. The PPPC categorization limits High-
brights to the brightness range >= 65 to less than 75. Super-
brights, according to the PPPC have a brightness level >= 75.
---------------------------------------------------------------------------
The glossy uncoated groundwood specialty grades are
supercalendered (SC) and soft nip calendered (SNC) grades. The gloss
in SC and SNC grades is produced by adding clay fillers to the pulp
furnish. After the SC paper roll comes off of the paper machine,
gloss and smoothness are imparted to the paper by running the paper
through a series of rolls called supercalenders. The gloss of SNC
paper is typically achieved by an on-machine soft-nip calender.
Slide 13 shows several SC grades (SCA, SCB+, SCB) which vary
primarily by the degree of glossiness. SC and SCN grades are used in
printing inserts, flyers, and catalogs. SC grades are also used in
printing magazines.\11\ The New York
[[Page 32860]]
Times Sunday Magazine is printed on SC paper.
---------------------------------------------------------------------------
\11\ The PPPC classifies uncoated groundwood specialty grades
into three categories: High-Gloss, Standard, and Lightweight. The
High-Gloss category includes all grades with a gloss >= 26, a
smoothness <= 2.5 (the lower the smoothness measure, the smoother
the surface of the paper), and a brightness >= 65. The grades
included in this category ranked from highest to lowest gloss,
highest to lowest smoothness, and highest to lowest brightness are
SCA+, SCA, SCB, and SNC+. The low gloss SNC and SCC grades have a
gloss >= 20 but less than 26, a smoothness measure greater than 2.5
and a brightness measure >= 60 ISO. The PPPC places these latter two
grades in the Standard Category even though the other Standard
Category grades are non-glossy. The Standard Category is defined
primarily in terms of brightness and is not defined in terms of
smoothness or gloss except for the SNC and SCC grades. The other
grades in the Standard Category are Superbright (brightness >= 75
ISO), High-bright (brightness >= 65 but less than 75 ISO), Bulky
book (brightness <= 60 ISO), and Other (no brightness requirement).
All High-Gloss and Standard grades have a basis weight >= 40 grams
per square meter. The Lightweight category contains one grade:
Directory paper. Directory paper has a basis weight of less than 40
grams per square meter.
---------------------------------------------------------------------------
Slide 23 below is from the presentation of John Weaver,
President and CEO of Abitibi, to a Credit Suisse First Boston Global
Basics investment analysts conference on March 4, 2004.\12\ This
slide provides additional information on the relation between
uncoated groundwood grade categories, the brand names of Abitibi
products in each category, and the end uses served by each
category.\13\
---------------------------------------------------------------------------
\12\ This presentation is available on Abitibi's Web site under
Investor Relations/Presentations & Webcasts.
\13\ MFS means machine-finished surface. Hi-Brite and Bulky Book
grades are MFS grades. All finishing to the surface of the paper is
accomplished on the paper machine. In contrast, the surface
finishing to SC grades is usually accomplished on off-machine
supercalenders. The Abitibi Alternative Offset and Equal Offset
grades are primarily sold as a substitute for uncoated free sheet
(UFS) grades for the printing of books. The Alternative Offset and
Equal Offset grades as well as the Innovative Offset grade (not
shown in Slide 23) are also MFS grades.
[GRAPHIC] [TIFF OMITTED] TN10JN08.001
Slide 12 below is from the presentation at the announcement of
the AbitibiBowater merger.\14\ Compared to Slide 13 discussed above,
it provides a somewhat different perspective on the relation between
the quality and value of uncoated and coated printing paper grades.
It shows that newsprint is the lowest valued and lowest quality
grade. The qualities indicated in the slide are brightness, opacity,
paper gloss, print gloss, basis weight, and strength. Slide 12 shows
that the two closest grades to newsprint in terms of value and
quality are the Bulky Book and Hi-Brite grades.\15\
---------------------------------------------------------------------------
\14\ See the AbitibiBowater merger announcement presentation,
``Creating a Global Leader in Paper and Forest Products,'' January
29, 2007. This presentation is available on Abitibi's Web site under
Investor Relations/Presentations & Webcasts.
\15\ Directory paper is not shown in Slide 12. If it were
included in Slide 12, it would probably be placed between the Bulky
Book and Hi-Brite grades as is indicated in Slide 13.
---------------------------------------------------------------------------
A graph appearing in this comment is not able to be reprinted
here. Copies of the comment with the graph are available at the
Department of Justice Antitrust Division Web site, http://
www.usdoj.gov/atr, at the Antitrust Documents Group of the
Department of Justice Antitrust Division, 450 Fifth Street, NW.,
Suite 1010, Washington, DC 20530, (202) 514-2481, and at the Office
of the Clerk of the United States District Court for the District of
Columbia, 333 Constitution Avenue, NW., Washington, DC 20001.
3. Product Market Definition
a. Likely Demand Substitution Responses by NA Newsprint Customers
Assuming a hypothetical monopolist of newsprint imposed ``a
`small but significant and nontransitory' increase in price'', would
current newsprint customers switch in sufficient numbers to other
paper grades to defeat the attempted price increase? \16\ There are
four pieces of evidence that suggest current newsprint customers are
unlikely to switch to other grades of paper in sufficient numbers to
defeat such an attempted price increase.
---------------------------------------------------------------------------
\16\ See the Merger Guidelines, Sec. 1.1 Product Market
Definition.
---------------------------------------------------------------------------
(1) Newsprint is the lowest quality and least expensive uncoated
groundwood grade. Newsprint is designed to run on the printing
presses of daily newspapers. We are unaware of any daily newspaper
that has responded to increases in the price of newsprint in the
past by switching to a higher quality and higher priced uncoated
groundwood specialty grade. We believe it is implausible that in the
future newspapers will switch to any higher quality and higher
priced uncoated groundwood specialty grade if there is a relative
increase in the price of newsprint. Every newspaper newsprint buyer
we talked to said that if the price of newsprint rose 5% to 10%
following the proposed merger they would have no alternative but to
pay the increased price. They said they could not switch to other
types of paper nor could they turn to suppliers outside of NA for
any significant quantity of newsprint.
(2) Estimates of the elasticity of demand for newsprint have
consistently been quite low (i.e., consistently quite inelastic). A
2004 study estimated that the U.S. demand
[[Page 32861]]
elasticity for newsprint was 0.36.\17\ A hypothetical monopolist
could profitably raise the price 5% to 10% and considerably more in
a market with a demand elasticity of 0.36.
---------------------------------------------------------------------------
\17\ While we have not attempted to estimate the demand
elasticity for the NA newsprint market, we note that an article in
2004 reported on an analysis that estimated the elasticity of the
U.S. demand for newsprint at 0.36 taking into account structural
changes in U.S. demand. See Jari Kuuluvainen, ``Structural Change in
U.S. Newsprint Demand: GDP and Price Elasticities,'' University of
Helsinki, Department of Forest Economics, Reports 34, 2004,
p. 8. A demand elasticity of 0.36 is in the same range as demand
elasticities reported in earlier articles. An article in 1997
reported the demand elasticity in NA at 0.22. Other estimates cited
in this article have been about twice as large. Estimates of demand
elasticity vary from 0.22 to 0.44. These estimates all indicate a
fairly inelastic demand curve for newsprint. See Ylbing Zhang and
Joseph Buongiorno, ``Communication Media and Demand for Printing and
Publishing Papers in the United States,'' Forest Science 43(3)
(August) 1997, p. 372. The results of our analysis of the proposed
Abitibi-Bowater merger are consistent with an inelastic demand
curve.
---------------------------------------------------------------------------
(3) Over the period 2002 to 2006, average annual newsprint
prices rose a total of 42.6%. Over that same period the consumption
by U.S. daily newspapers declined by 13.6%. The ratio of the
percentage decline in newsprint consumption by U.S. daily newspapers
to the percentage increase in the price of newsprint was 0.32.\18\
Daily newspapers account for 80% of U.S. newsprint consumption and
non-daily newspapers and commercial printers account for the
remaining 20%. Over the four-year period 2002-2006, newsprint
consumption for this latter category of newsprint customers declined
15.2%. The absolute ratio of the percentage decline in newsprint
consumption by U.S. non-daily newspapers and commercial printers to
the percentage increase in the price of newsprint was 0.36. When
total U.S. newsprint consumption is considered, the percentage
decline over the four-year period was 13.9% and the absolute ratio
of the percentage decline in newsprint consumption to the percentage
increase in the price of newsprint was 0.33. These results are
consistent with the estimated U.S. newsprint demand elasticity of
0.36 discussed immediately above.
---------------------------------------------------------------------------
\18\ The ratio is expressed as an absolute number. Sources: PPPC
NA Monthly Newsprint Bulletins and PPPC Flash Reports. As discussed
in Section E below, much of the decline in the demand of U.S. daily
newspapers has not been caused by the rise in newsprint prices.
---------------------------------------------------------------------------
(4) As shown in Table B1 in Attachment B, the prices of three
major uncoated groundwood specialty grades are significantly above
the price of newsprint even before the reduction in printing surface
due to the switch to a heavier basis weight is taken into account.
Taking the reduction in printing surface into account, as a
newsprint customer rationally would, a buyer of 30.0 lb. newsprint
who switched to 35 lb. SCB, Hi-Brite 65, or SCA, would face an
equivalent price increase per metric tonne of 30.0 lb. newsprint
ranging from 34.0% to 47.0% based on February 2007 prices.\19\ Table
B1 in Attachment B does show a financial gain to a newsprint buyer
of 27.7 lb. from switching to 22.1 lb. directory paper. However, the
information provided to us by newsprint buyers leads us to conclude
that the lower basis weight and thinner directory paper would not be
suitable for use in a newspaper or for running on newspaper printing
presses. The lowest basis weight newsprint that we are aware of that
is being used to print newspapers is 26.4 lb. (43.0 g/m\2\)
newsprint. Our understanding is that 26.4 lb. newsprint is used
primarily if not entirely on flexographic printing presses and not
on the predominant offset printing presses.\20\
---------------------------------------------------------------------------
\19\ RISI Pulp & Paper Week does not publish prices for the
lowest quality uncoated groundwood specialty glossy grades, SNC and
SCC. Abitibi-Bowater Slide 12 above, which plots quality against the
value of uncoated groundwood grades, placed SNC and SCC in between
SCB and Hi-Brite in terms of value and quality. In Table B1 in
Attachment B, 35 lb. SCB is priced 8.8% below the price of 35 lb.
SCA. If 35. lb. SNC and SCC grades were priced 8.8% below the price
of 35 lb. SCB, their prices would still be 22.1% above the price of
30 lb. newsprint.
\20\ According to RISI 2006 Fact and Price Book, p. 145, offset
presses account for 85% of the presses at U.S. daily newspapers,
letterpress presses account for 10%, and flexographic presses
account for 5%.
---------------------------------------------------------------------------
b. Identifying Participants in the NA Newsprint Market
(1) Introduction
In Section C below we identify NA capacity to produce newsprint
by manufacturer mill. This capacity participates in the NA newsprint
market. According to the Merger Guidelines, it is also necessary to
identify those firms that could participate in the NA newsprint
market through a supply response.
The antitrust agencies' methodology for determining whether such
capacity should be included is described in ``Sec. 1.32 Firms That
Participate Through Supply Response'' of the Merger Guidelines.
Sec. 1.32 notes that the agencies ``will identify other firms [or
capacity] not currently producing or selling the relevant product in
the relevant area as participating in the relevant market if their
inclusion would more accurately reflect probable supply responses.
These firms are termed `uncommitted entrants.' These supply
responses must be likely to occur within one year and without the
expenditure of significant sunk costs of entry and exit, in response
to a `small but significant and nontransitory' price increase.''
Sec. 1.32 further notes that ``[i]f a firm [or capacity] has the
technological capability to achieve such an uncommitted supply
response, but likely would not (e.g., production would render such a
response unprofitable), that firm [or capacity] will not be
considered to be a market participant.''
The most likely type capacity for inclusion as a participant in
the newsprint market would be uncoated groundwood specialty grades
produced on so-called ``swing'' machines. The next most likely type
of capacity would be newsprint machines that have been converted to
the exclusive production of groundwood specialty grades without the
expenditure of capital funds to rebuild the machine, to add or
reconfigure pulping capability, or add off-machine finishing
equipment. In cases where the conversion of a newsprint machine to
an uncoated groundwood specialty grade has required a significant
expenditure of capital funds, it is the least likely that that
capacity should be included as a participant in the newsprint
market. Similar analytical considerations apply to uncoated
groundwood specialty machines that have never produced newsprint.
(2) Swing Machines
A certain amount of newsprint is produced on so-called ``swing''
machines. That is, the same machine is used to produce both
newsprint and one or more higher quality and higher priced uncoated
groundwood specialty grades. For example, some manufacturers may be
able to produce Hi-Brite grades and Directory paper on the same
machine as newsprint. It is likely that Bulky Book paper can also be
produced on the same machine as newsprint. The Catalyst 2006 annual
report, p. 9, states that ``Capacities in the above table can vary
as the Company is able to switch production between products,
particularly newsprint, directory, and machine-finished [i.e., Hi-
Brite] uncoated grades.''
Bowater's 2005 Annual Report states on p. 4 that it has
newsprint and uncoated groundwood swing machines at the following
mills: Calhoun, TN, Thunder Bay, ON, Gatineau, QC, and Dalhousie,
NB. Abitibi's 2005 Annual Report, p. 10, indicates that Abitibi may
have swing newsprint machines at its Belgo, QC, Iroquois Falls, ON,
and Grand Falls, NL mills. Since the annual report does not provide
a capacity breakdown by machine, it cannot be determined from the
table on p. 10 which, if any, of the machines at these mills are
producing both newsprint and uncoated groundwood paper. The annual
report also indicates that Abitibi's Fort William mill in Thunder
Bay, ON has a newsprint and uncoated groundwood swing machine. The
mill's only paper machine is shown with a capacity of 107,000 metric
tonnes for newsprint and 38,000 metric tonnes for uncoated
groundwood grades.
The PPPC 2003 NA Newsprint Capacity Survey (March 3, 2003)
states on p. 2 that at the time of the survey there were 17 machines
in NA that were classified as ``swing'' machines. The PPPC noted
that the
[[Page 32862]]
number of swing machines had been declining due to increased machine
specialization and the conversion of newsprint machines to other
grades. We do not know the total current capacity of NA ``swing''
machines by NA mill, but believe that the newsprint capacity of each
swing machine is reported by manufacturers to the PPPC in proportion
to the actual or anticipated production of newsprint on the machine.
As the Merger Guidelines suggest, it would be necessary to
determine if it would be profitable to switch the capacity on swing
machines used to make uncoated groundwood specialty grades to
newsprint production in the event of an increase in the price of
newsprint. If it would be profitable, then the capacity of the swing
machine used to make uncoated groundwood specialty grades should be
included as participating in the newsprint market through a supply
response. If it would not be profitable, then that capacity would
not be included. We are aware of no publicly-available information
that could be used to address this issue.
(3) Machines That Have Been Converted From Newsprint Production
In contrast to the use of swing machines to produce both
newsprint and uncoated groundwood specialty grades, some newsprint
manufacturers have converted newsprint machines to the production of
higher quality and higher priced uncoated groundwood grades (e.g.,
SC grades). That is, these machines are no longer used to
manufacture newsprint. In some cases, these machine conversions
required significant investment expenditures and non-trivial down
times. To the extent that it would not be profitable to produce
newsprint on a converted newsprint machine ``in response to a `small
but significant and nontransitory' price increase,'' the capacity of
that machine should not be regarded as participating in the market
(supply response within one year) or as en entrant (entry within two
years). See Sec. 1.32 as discussed above and ``Sec. 3 Entry
Analysis'' of the Merger Guidelines. In some cases, however, it may
be profitable to produce newsprint on a converted newsprint machine
``in response to a `small but significant and nontransitory' price
increase.'' This analysis can also be applied to machines that have
never produced newsprint but are used to produce closely related
uncoated groundwood specialty grades.
Below we provide two examples of recent conversions by Abitibi
from newsprint to uncoated groundwood specialty grades. In 2005,
Abitibi removed about 118,000 metric tonnes of newsprint capacity by
converting a newsprint machine at its Shawinigan (Belgo). QC mill to
Hi-Brite production. The conversion consisted of an increase in
bleaching capacity at the Belgo mill. The cost was about C$15
million.\21\ It seems likely that the Belgo machine could still
technically produce newsprint. Hi-Brites are essentially brighter
newsprint, once called improved newsprint. If the Belgo mill were
owned by a firm that could not influence the price of newsprint
through the removal of capacity from the market, that firm
potentially might have the incentive to switch some of the capacity
of the converted machine back to newsprint in response to a relative
increase in the price of newsprint. To determine whether that
incentive exists requires knowledge of alternative profitability
scenarios involving different mixes of Hi-Brite and newsprint
production. Abitibi is quite unlikely to use the Belgo machine to
produce newsprint in the event of an increase in the price of
newsprint since part of Abitibi's objective in converting the
machine to Hi-Brites was likely to remove newsprint capacity from
the newsprint market in order to raise the price of newsprint.
---------------------------------------------------------------------------
\21\ Sources: Abitibi 2005 Annual Report p. 28, Abitibi 2004
Annual Report, p. 42, and Abitibi-Bowater Merger Announcement
Presentation, p. 17.
---------------------------------------------------------------------------
In 2003 and 2004, Abitibi converted about 170,000 metric tonnes
of newsprint capacity at its Alma, QC mill to the production of
Super Hi-Brites. The total cost of the conversion exceeded C$200
million. The conversion likely included an expansion of bleaching
capacity and a rebuild of the paper machine. It seems unlikely that
this machine would be used to produce newsprint under any
foreseeable circumstances.
In 2002, Great Northern Paper rebuilt its No. 11 uncoated
groundwood specialty machine at its Millinocket, ME at a cost of
$103 million. After the mill was sold to Katahdin in 2003, the new
owners made additional improvements to the machine to enable it to
produce high quality SCA and SCA+ paper for magazines and catalogs.
The machine was down 17 months before being restarted in 2004. Part
of this downtime was due to the bankruptcy of Great Northern.\22\ We
do not know if the investment and lost downtime required to convert
the Katahdin machine to SC paper is representative nor do we know if
SC machines are technically capable of producing newsprint. Assuming
SC machines are technically capable of producing newsprint, it seems
unlikely to us that owners of SC capacity would find it profitable
to divert part of their SC capacity to the production of newsprint
in the absence of a substantial increase in the relative price of
newsprint.\23\
---------------------------------------------------------------------------
\22\ Source: ``Katahdin Paper, The Maine Chance,'' Manufacturing
in Action, September 2004. While, strictly speaking, this machine
was not converted from newsprint to SC grades, it seems likely that
prior to the conversion the machine was producing paper close to the
quality of newsprint.
\23\ SNC grades are typically made on former newsprint machines
with an on-machine soft-nip calendar added to the paper machine. SNC
grades are comparable to SCB and SCC grades as discussed above.
While technically they likely could produce newsprint, it seems
unlikely that an SNC machine that has switched away from newsprint
production would switch back to newsprint production in the event of
a 5% to 10% increase in the price of newsprint relative to the price
of SNC grades. Both Abitibi and Bowater manufacture SNC grades.
---------------------------------------------------------------------------
Referring to Slide 13 in Section B.2.b.(1) above, the most
likely capacity to be converted to the production of newsprint in
the event of a relative increase in the price of newsprint would be
Directory, Bulky Book, and Hi-Brite. The machines used to produce
these grades are technically closest to the machines used to produce
newsprint. As discussed above, the machines used to produce SC
grades and Super Hi-Brite grades have been significantly upgraded
from newsprint machines or from lower quality uncoated groundwood
grades. It seems unlikely that it would be profitable to use these
machines to produce newsprint even if the price of newsprint were
increased significantly.
Directory paper is sold under one- to three-year contracts that
specify both price and volume. About 80% to 90% of directory paper
is sold under contract. The other 10% to 20% is sold on the spot
market. The main buyers of Directory paper are RBOCs and independent
publishers of telephone directories. The demand for Directory paper
has shown strong growth since 2004 and contract price increases of
10% are expected in 2007.\24\ It seems unlikely to us that owners of
Directory capacity could divert Directory capacity that is being
sold under contract. To the extent that some owners of Directory
capacity have excess capacity, they might use that capacity to
produce newsprint in the event of a relative increase in the price
of newsprint. However, with a growing demand for Directory paper,
the use of that capacity to produce newsprint is likely to be short-
lived.
---------------------------------------------------------------------------
\24\ RISI Fact & Price Book, pp. 168-169.
---------------------------------------------------------------------------
As shown in Table B-7 in Attachment B, Abitibi and Bowater
control 76.5% of the NA Hi-Brite capacity and 100% of the Hi-Brite
capacity East of the Rockies. Abitibi and Bowater also appear to
control most of the Bulky Book capacity although we were not able to
obtain a Bulky Book capacity figure for Bowater.
(4) Machines Producing Uncoated Groundwood Specialty Grades That Have
Never Produced Newsprint
There are at least three machines producing uncoated groundwood
specialty grades that have been designed specifically to produce
those grades. These are high-speed, high-capacity machines use to
produce high-quality SC paper. These machines are owned by Stora
Enso and Madison paper. It is highly unlikely that these machines
would ever be used to produce newsprint under any conceivable
circumstances.
c. Conclusions Regarding Product Market Definition
(1) Newsprint Market
Based on our analysis in Section B.3.a. above of the likelihood
of demand substitution in the event of a relative increase in the
price of newsprint, we conclude that the relevant product market is
no larger than newsprint.
(2) Participating Manufacturers in the NA Newsprint Market
Current newsprint suppliers are participants in the NA newsprint
market.\25\ Based on our analysis in Section B.3.b., we considered
whether it was likely that capacity used to manufacture uncoated
groundwood grades could be considered likely participants through a
supply response following the Merger Guidelines
[[Page 32863]]
methodology. Our conclusion is that there is undoubtedly some swing
capacity that should be included as likely participants in the NA
newsprint market. There are no public data available to quantify the
amount of swing capacity that should be included but a significant
portion of that swing capacity is likely controlled by Abitibi and
Bowater.
---------------------------------------------------------------------------
\25\ In Section B.4 below, we consider whether the geographic
market is narrower or broader than NA.
---------------------------------------------------------------------------
Abitibi and Bowater also control a very large portion of Bulky
Book and Hi-Brite, the next most likely capacity to participate in
the NA newsprint market, and control virtually all of that capacity
East of the Rockies.\26\
---------------------------------------------------------------------------
\26\ As explained in Section B.3.b.(3) above, it is unlikely
that manufacturers of Directory paper would divert more than a small
amount of Directory capacity to the production of newsprint and that
diversion is likely to be short-lived.
---------------------------------------------------------------------------
Several of the newspaper newsprint buyers we interviewed said
that they were unaware of any newsprint machine that had been
converted to production of uncoated groundwood specialty grades that
had subsequently been converted back to the production of newsprint.
Given the steady decline in the NA demand for newsprint since at
least 1999 this is not a surprising result. As shown in Section E2
below, NA demand (quantity purchased) for newsprint has declined
every year from 1999 to 2006 for a total decline of 25.5% over that
period.\27\
---------------------------------------------------------------------------
\27\ The last de novo entry into the NA American newsprint
market was in 1990 by Atlantic Newsprint and Alberta Newsprint.
Inland Empire installed a small newsprint machine in 2001 to replace
an old newsprint machine. That machine is the only new newsprint
machine installed in NA since 1991. Source: ``Newsprint: A Pulp &
Paper Market Focus Book (1999), pp. 19-20. These facts and the 25.5%
decline in NA consumption since 1999 indicate that de novo entry
into this high capital cost industry is unlikely for the foreseeable
future.
---------------------------------------------------------------------------
We conclude that the participants in the NA newsprint market are
the current NA newsprint producers. These 16 NA newsprint producers
are identified in Tables C1 and C2 attached to Section C1.
4. Geographic Market Definition
a. Introduction
The methodology for geographic market definition is described in
Sec. 1.2 of the Merger Guidelines. The methodology is similar to
the methodology used to define relevant product markets.
Absent price discrimination, the Agency will delineate the
geographic market to be a region such that a hypothetical monopolist
that was the only present or future producer of the relevant product
at locations in that region would profitably impose at least a
``small but significant and nontransitory'' increase in price,
holding constant the terms of sale for all products produced
elsewhere. That is, assuming that buyers likely would respond to a
price increase on products produced within the tentatively
identified region only by shifting to products produced at locations
of production outside the region, what would happen? \28\
---------------------------------------------------------------------------
\28\ See the Merger Guidelines, Sec. 1.21 General Standards.
---------------------------------------------------------------------------
In this section we consider whether the relevant geographic
market is narrower or broader than our provisional geographic market
of NA.
b. Is the Relevant Geographic Market Narrower Than NA?
There is evidence that the relevant market may be narrower than
NA. Based on interviews with buyers of newsprint for newspaper
publishers, newsprint mills located West of the Rockies rarely ship
to customers located East of the Rockies and vice versa.\29\
According to these buyers, the high cost to transport newsprint from
West Coast newsprint mill locations to customers located East of the
Rockies makes newsprint produced in West Coast mills non-competitive
with newsprint manufactured at mills located East of the Rockies.
Even if there were a relative 5% to 10% increase in the price of
newsprint sold East of the Rockies, these buyers believe that it
would not be profitable for West Coast mills to begin shipping
newsprint in significant quantities to customers located East of the
Rockies.
---------------------------------------------------------------------------
\29\ This evidence also implies that newsprint sold to customers
West of the Rockies may also be a relevant market. Since Bowater is
not a majority owner of any mill on the West Coast the merger would
not have a competitive effect in a West of the Rockies newsprint
market. Bowater does have a 40% minority interest in the Ponderay
Newsprint mill, which is located in Usk, WA. Abitibi does own two
newsprint mills West of the Rockies. These mills are located in
Snowflake, AZ and Mackenzie, BC.
---------------------------------------------------------------------------
We do not have the information necessary to determine if
newsprint sold to customers located East of the Rockies is a
relevant geographic market for the purposes of assessing the
competitive effects of the merger. Primary sources of information on
whether such a geographic market can be properly defined would
include West Coast newsprint mills and customers located East of the
Rockies. An analysis of comparative freight rates from West Coast
mills and mills located East of the Rockies to East of the Rockies
newsprint customers would also be useful in determining whether
there is a relevant East of the Rockies market. For the purposes of
calculating capacity shares and HHIs in Section C below, it is
assumed that East of the Rockies is a relevant geographic market.
c. Is the Relevant Geographic Market Broader Than NA?
(1) Introduction
There has been considerable speculation in the trade press
concerning the likely impact of new Chinese newsprint capacity on NA
purchasers of newsprint and NA newsprint mills. While some buyers of
newsprint have shown an interest in newsprint from China, it appears
from press reports that the only newsprint that they have bought
from Chinese mills is for test runs. There is no current indication
that they intend to buy significant amounts of newsprint from China
within the next one to two years. To the extent that there are
imports of newsprint from China in the near-term, it is likely that
the phenomena will be short-lived.
If there is an effect of the new Chinese capacity on NA
newsprint mills, it will likely be on the displacement of export
sales from NA mills to current customers located in Asia. It is
likely that the new Chinese newsprint capacity will be largely
absorbed in Asia over the next several years.
(2) Current and Past NA Import Levels
Imports of newsprint into NA have not been a significant source
of supply for NA newspaper publishers and other NA purchasers of
newsprint. In 1999, imports accounted for only 3.3% of NA newsprint
purchases.\30\ Since 1999, imports have accounted for 2.0% or less
of NA purchases. See Section E2 below. Imports have been falling
since 2004 both in absolute quantities and as a percentage of NA
demand. In 2006, imports accounted for just 1.5% of NA newsprint
purchases. For the first two months of 2007, imports have fallen
56.1% compared to the first two months of 2006. Imports accounted
for 0.7% of NA newsprint purchases for the first two months of
2007.\31\
---------------------------------------------------------------------------
\30\ Sources: December 2006 and December 2005 PPPC NA Newsprint
Statistics-Flash Report (``Flash Report'') and December 2001-2004
PPPC NA Newsprint Statistics Monthly Bulletin (``PPPC Monthly
Bulletin'').
\31\ Source: February 2007 Flash Report.
---------------------------------------------------------------------------
In the latter part of the 1990s, there was an increase in NA
imports to about 555,000 metric tonnes in 1998 (about 4.3% of NA
consumption).\32\ Almost all of the increase was due to imports from
South Korea and Russia.
---------------------------------------------------------------------------
\32\ Sources: RISI 2006 Fact and Price Book, p. 142, and Pulp &
Paper 2000 NA Factbook, p. 190.
---------------------------------------------------------------------------
There were a number of unique circumstances that accounted for
the increase in imports from South Korea to NA. These include (1)
significant new efficient capacity coming on line in South Korea;
(2) a very steep devaluation of the South Korean won relative to the
U.S. dollar; (3) a significant recession in South Korea and Asia
which reduced Asian demand for newsprint; and (4) strikes at
newsprint mills in British Columbia which removed about 1 million
metric tonnes of annual newsprint capacity from the NA market.\33\
As the South Korean and Asian economies began to recover, as the
South Korean won began to appreciate against the U.S. dollar, and as
the strikes at the British Columbia mills were settled, the new
South Korean capacity was largely absorbed in Asia. NA publishers,
however, have continued to import some newsprint from South Korea,
although at significantly reduced amounts from the 1998 peak. NA
imports from all sources, including South Korea and Russia, declined
from the 1998 peak of 555,000 metric tonnes to about 221,000 metric
tonnes in 2000. NA imports have remained at the 2000 level or
slightly below until declining to 142,000 metric tonnes in 2006.
---------------------------------------------------------------------------
\33\ See Economists Incorporated's submission to DOJ concerning
the proposed acquisition of Alliance by Bowater, dated May 7, 2001,
pp. 15-18.
---------------------------------------------------------------------------
Imports from Russia also increased during the latter part of the
1990's though not as significantly as imports from South Korea.
Newspaper publishers found that newsprint from Russian mills was
unreliable both in terms of quality and delivery. As a consequence,
imports from Russia declined to a low level by 2000.
[[Page 32864]]
(3) The Likelihood of Imports From China
(a) Projected Growth in Global Newsprint Demand
Martine Hamel, head of market research for the PPPC, estimates
growth in newsprint demand for all regions of the world over the
period 2006 to 2008.\34\ See Slide 39 below. The slide shows
negative growth for NA for all three years. Western Europe is
expected to have positive growth in 2006 and 2007 before
experiencing negative growth in 2008. All other regions are shown
with positive growth for all three years.
---------------------------------------------------------------------------
\34\ Source: At the November 2, 2006 joint NPA/NAA Newsprint
Conference, Martine Hamel, VP, COO and head of market research for
the PPPC, presented a report titled ``Review and Forecast of
Newsprint Demand and Supply'' (``PPPC 2006 NPA/NAA Presentation'').
The Presentation reviews global demand and supply of newsprint for
the first nine months of 2006 and earlier years and forecasts global
demand and supply of newsprint for the period 2006-2008.
[GRAPHIC] [TIFF OMITTED] TN10JN08.002
---------------------------------------------------------------------------
(b) Projected Growth in Chinese and Other Asian Newsprint Demand
Slide 36 below from Martine Hamel's presentation shows the
forecast growth of Chinese demand for newsprint. Chinese newsprint
demand is projected to increase by 3.1% in 2006, 8.7% in 2007, and
14.0% in 2008.
[[Page 32865]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.003
Slide 37 shows growing demand in the rest of Asia (excludes
Japan, South Korea, and China). The projected demand growth in China
and the rest of Asia \35\ was likely the primary reason for the
installation of the new newsprint capacity in China.
---------------------------------------------------------------------------
\35\ We assume that the growth projections in Slides 36, 37, and
39 above correspond to similar projections that were available to
Chinese officials responsible for investments in new newsprint
capacity.
---------------------------------------------------------------------------
[[Page 32866]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.004
(c) Projected Growth in Global Newsprint Supply
Slide 42 below from Martine Hamel's presentation shows that
virtually all of the growth in global newsprint capacity over the
period 2005-2008, is expected to come from the installation of new
Chinese capacity. This growth in Chinese newsprint capacity is
partially offset by reductions in NA newsprint capacity.
[[Page 32867]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.005
(d) Evidence From the PPPC 2006 NPA/NAA Presentation That the New
Chinese Newsprint Capacity Is Expected To Be Mostly Absorbed in Asia
Over the Next Several Years
Martine Hamel of the PPPC also estimates that exports from Asia
to other regions of the world will total 60,000 metric tonnes per
year over the period 2005 to 2008.\36\ See Slide 49 below. The slide
shows that despite the significant increase in Chinese newsprint
capacity, exports from Asia to other regions of the world are not
expected to be significant.
---------------------------------------------------------------------------
\36\ While Slide 49 does not specify whether the 60,000 metric
tonnes of exports from Asia is per year or for the entire four-year
period, we conservatively assume that the figure is an annual
average estimate.
---------------------------------------------------------------------------
[[Page 32868]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.006
We expect that many of these exports from Asia would be to
regions other than NA since, as shown in Slide 39 above, demand in
those regions is growing while demand in NA is decreasing
significantly. To the extent there were exports from Asia in 2005
and 2006, these exports did not have a significant impact on the NA
newsprint market since imports into NA in 2005 and 2006 actually
declined each year from the prior year.
Slide 49 also shows exports from Asia during the period 1996 to
1999 at a rate of 360,000 metric tonnes per year. For five of the
six years 1995 to 2000, Asian newsprint capacity increased by a
greater percentage than is projected for the three years 2006 to
2008. As was discussed above, this capacity came on line at the same
time that the Asian region was undergoing a steep economic decline
and steep decline in the demand for newsprint. The new Chinese
capacity is coming on line at a time of significant growth in demand
for newsprint in China and in the rest of Asia. See Slides 36 and 37
above. This significant projected growth in newsprint demand
increases the likelihood that the new Chinese capacity will be
absorbed in Asia over the next several years. This projected growth
was undoubtedly a major factor in the PPPC's forecast of 60,000
metric tonnes of exports per year from Asia for the period 2005-
2008, compared to the much higher export total of 360,000 metric
tonnes per year from Asia that occurred over the period 1996-1999.
(e) Evidence From the Heads of Abitibi and Bowater That the New Chinese
Newsprint Capacity Is Expected To Be Mostly Absorbed in Asia Over the
Next Several Years
The heads of Abitibi and Bowater also expect that the new
Chinese capacity will be absorbed in Asia over the next several
years.
John Weaver, President and CEO of Abitibi, gave a presentation
to Citigroup's 11th Annual Global Pulp & Forest Products Conference
on December 7, 2006 (``Citigroup Conference''). During the Q&A that
followed his slide show presentation, Weaver was asked about the
impact of the new Chinese newsprint capacity on the global and NA
newsprint markets.\37\
---------------------------------------------------------------------------
\37\ We have provided DOJ with a copy of the audio recording of
Weaver's remarks at the Citigroup Conference. The copy is a .wma
file and can be played on Windows Media Player (``WMP''). If the
copy is played on WMP, the time expressed as minutes and seconds is
shown as the recording proceeds. Weaver's discussion of the
possibility of imports from the new Chinese capacity begins at 24:04
into the recording. We have also provided DOJ with a copy of the
slide show that Weaver presented to the Citigroup Conference. The
slide show of Weaver's presentation is available under Investor
Relations/Presentations & Webcasts on Abitibi's Web site. According
to Abitibi's Web site, the audio recording of Weaver's remarks at
the Citigroup Conference is no longer available on the Web site.
---------------------------------------------------------------------------
Weaver begins his response by saying that ``There will be a
trend in the international market in [2007] but it won't be China.''
He said that he does not know of any deal that a publisher has
signed that is not a trial. He said that there had been only 242
tonnes of imports from China so far in 2006.
``So I don't really expect to see any significant imports of
Chinese paper to North America [in 2007],'' he said. He also said
that based on most of the calculations he has seen, including those
by Abitibi, ``it's hard to see the economic benefit of the Chinese
coming.'' \38\
---------------------------------------------------------------------------
\38\ While he does not elaborate further on this statement, he
appears to be saying that it would be more profitable for the
Chinese mills to sell their newsprint closer to home rather than to
incur the additional freight costs to ship newsprint to NA.
---------------------------------------------------------------------------
He said that he does expect there will be some Chinese exports.
He specifically mentions that Abitibi has seen Chinese exports in
India. He said, ``I really feel that the phenomena of Chinese
oversupply may be short-lived.'' He gives several reasons. He
mentions 1.7% growth in global newsprint demand. He also says that
the Chinese government recently announced that they would close
their smaller polluting newsprint mills in 2007 and 2008, which
would reduce the amount of Chinese newsprint capacity.
David Paterson, President and CEO of Bowater, also gave a
presentation at the 2006 Citigroup Conference. Paterson addressed
the issue of new Chinese capacity during his slide show presentation
(Slides 14 and 15).\39\
[[Page 32869]]
He notes the strong growth in demand globally for newsprint except
in the U.S. He also notes the strong growth in the demand for
newsprint in China.
---------------------------------------------------------------------------
\39\ We have provided DOJ with a copy of the audio recording of
Paterson's remarks at the Citigroup Conference. The copy is a .wma
file and can be played on Windows Media Player (``WMP''). If the
copy is played on WMP, the time expressed as minutes and seconds is
shown as the recording proceeds. Paterson's discussion of the
possibility of imports resulting from the new Chinese capacity
begins at 11:59 into the recording. We have also provided DOJ with a
copy of the slide show that Paterson presented to the Citigroup
Conference. The slide show of Paterson's presentation is available
under Investor Relations/Presentations on Bowater's Web site.
According to Bowater's Web site, the audio recording of Paterson's
remarks at the Citigroup Conference is no longer available on the
Web site.
---------------------------------------------------------------------------
He asks, ``Where will those Chinese tonnes go as they start up
and come into the market?'' Paterson said Bowater believes they will
flow into Asia and that there will be some coming into NA. He said
that U.S. newspapers were talking openly about importing newsprint
from China into the east coast and the west coast of the U.S. But,
he said, ``Having said that, I think most of the tonnes will show up
in places like Singapore, Malaysia, India, Brazil. These are all
high growth markets.'' He said that newsprint consumption in India
was up 17% so far this year. He said that Bowater sees the Chinese
in India and that Chinese newsprint sales are growing.
He said, ``There is room for those tonnes to go. It will be a
difficult 12 to 18 months as they find a home.'' He said there were
also other forces affecting Chinese tonnage, primarily Chinese
demand as well as the change in their tariff system. He said if the
government does what it said it is going to do and eliminates tariff
protection for exports, then high-cost Chinese capacity will start
shutting down.\40\
---------------------------------------------------------------------------
\40\ Paterson appears to referring to a 13% rebate to Chinese
newsprint exporters on a 17% import tax that newsprint mills must
pay on imported raw materials. If this rebate has been eliminated,
the cost of newsprint exports has been increased, especially exports
made from recycled paper (ONP). Chinese newsprint mills are major
importers of recycled paper. The two newest Chinese newsprint
machines are recycled paper machines. The price of ONP has nearly
doubled since last fall to $180 per tonne. This will make the new
Chinese newsprint capacity and other Chinese capacity that relies on
ONP less competitive against Abitibi and Bowater who rely primarily
on wood fiber for their pulp needs. See ``Paper Chase,'' by Andrew
Bary, Barron's On-Line, April 5, 2007.
---------------------------------------------------------------------------
In their audio remarks, both Weaver and Paterson, emphasized the
export opportunities for NA newsprint manufacturers created by the
global growth in the demand for newsprint. Abitibi and Bowater
foresee a healthy increase in overseas shipments in 2007 due to the
projected growth in newsprint demand in other regions. Abitibi and
Bowater account for about 70% of total exports from NA to overseas
locations. In a news report, Weaver said he expected Abitibi to
increase its offshore shipments by 10% in 2007 and Paterson
anticipated a 5% to 6% increase in offshore shipments from NA.\41\
---------------------------------------------------------------------------
\41\ See ``Abitibi, Bowater turning to export markets to counter
declines in NA,'' RISI, February 12, 2007.
---------------------------------------------------------------------------
(f) Evidence That Buyers of Newsprint for U.S. Daily Newspapers
Generally Do Not Have Plans To Buy Newsprint From China Within the Next
Several Years
Several of the newspaper newsprint buyers we talked to indicated
that they had tested Chinese newsprint but that they had no
immediate plans to purchase newsprint from Chinese mills. Factors
that they cited were an unknown track record, the lack of a
relationship, the need to assure reliability of delivery and
quality, and the need to assure service. While price is an extremely
important factor to a newsprint buyer, another important factor is
the need to assure an adequate and reliable supply of newsprint at
all times since newspapers print on a daily basis.\42\
---------------------------------------------------------------------------
\42\ Two newspaper publishers, Gannett and the Tribune Co., have
been publicly identified as conducting test runs using Chinese
newsprint. See ``Tribune's Second Test of Chinese Newsprint a
Success,'' by Jim Rosenberg, Editor & Publisher, December 11, 2006.
According to the article, Gannett and the Tribune Co. said the
results of the tests were successful. According to the article, a
Gannett executive said last year that Gannett expects to buy Chinese
newsprint but would not specify the quantity it planned to purchase
or when purchases might commence. The Tribune Co. continues to run
tests on Chinese newsprint for its Los Angeles Times printing
operation. After the Tribune's first successful test run in November
2006 at its Orlando (FL) Sentinel printing plant. John Cannizzo,
Tribune's senior manager of group operations, is quoted as saying
```If it turns out we can get, say, 1,000 tons shipped in a
reasonable time and on a consistent basis, (buying Chinese
newsprint) might be a viable option in 2007'. [* * *] We're not in a
great hurry. We just want to see if this might work.'' ``Tribune
marks `successful' test of Chinese mill's newsprint,'' by Chuck
Moozakis, Newspapers & Technology, December 2006. The newsprint
tested in Orlando was originally intended for a test at the Los
Angeles Times plant. However, the paper's cores and chucks were not
compatible with the Chinese rolls, according to the article. That
problem has since been solved. It seems unlikely that it would be
profitable to ship newsprint from China through the Panama Canal to
an east coast location, given the much greater shipping costs.
---------------------------------------------------------------------------
The buyers emphasized the need to develop a very close
relationship with their suppliers. Buyers emphasized that it would
take several years of low-volume purchases to establish the trust
and track record needed to increase their level of purchases.
Several buyers believed that if Chinese newsprint were shipped
to the U.S., it would only be economically feasible to ship the
paper to west coast ports to supply newspaper printing plants
located close to the docks.
d. Conclusions Regarding Geographic Market Definition
(1) Relevant Geographic Market
We conclude that the geographic market is no larger than NA. It
is possible that the relevant geographic market may be narrower than
NA. Some evidence suggests that there may be a relevant East of the
Rockies geographic market. To conclude that there is a relevant East
of the Rockies market it would be necessary to determine if West of
the Rockies newsprint mills could profitably ship newsprint to East
of the Rockies customer locations in response to a ``small but
significant and nontransitory'' increase in price in sufficient
quantities to make the price increase unprofitable.
(2) The Likely Effect of New Chinese Newsprint Capacity on the NA
Newsprint Market
While there is new Chinese newsprint capacity that has come on
line recently, it appears that that capacity will be largely
absorbed in Asia over the next couple of years. There may be some
limited sales to U.S. publishers by Chinese mills over the next
couple of years. Most publishers we talked to showed little interest
in buying newsprint from Chinese mills. They placed great emphasis
on trust, reliability and a close relationship with their newsprint
suppliers. Currently they have no relationship with any of the
Chinese mills and believe that establishing the trust and
reliability necessary to buy more than nominal amounts of newsprint
would take at least a couple of years if not longer.
If there is to be an effect on the NA newsprint market from the
new Chinese newsprint capacity, it would likely be an indirect one.
It is possible that some NA suppliers who currently export to Asia
will be displaced from some of their customers by the new Chinese
capacity. If so, that would create excess capacity at their NA mills
used to supply the Asian market. As discussed above, however,
Abitibi and Bowater expect newsprint exports from NA to increase,
not decrease. The export growth opportunities that Abitibi and
Bowater expect to be able to take advantage of should be available
to other NA mills that export newsprint, including those that may be
displaced from Asian customers by the new Chinese capacity.
C. Analysis of the Increase in Concentration That Would Result From the
Proposed Merger
1. Analysis of the Increase in Concentration in the NA Newsprint
Market Based on Estimated 2006 Capacity
According to Sec. 1.51(b) of the DOJ/FTC Horizontal Merger
Guidelines (``merger guidelines'') the NA newsprint market is
currently moderately concentrated. Based on estimated 2006 NA
newsprint capacity, the pre-merger HHI is 1,380. If the merger is
consummated, the change in the HHI would be 962 and the post-merger
HHI would be 2,342.\43\ See Chart CI below.
---------------------------------------------------------------------------
\43\ Table C1 in Attachment C identifies the owner, location and
capacity for each NA newsprint mill. Table C1 also provides detailed
information on the methods and sources relied upon for the estimate
of the market shares. Table C2 in Attachment C shows the calculation
of the capacity shares and HHIs by manufacturer based on the mill-
level data contained in Table C1. Table C2 is the source for both
Charts C1 and C2.
---------------------------------------------------------------------------
[[Page 32870]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.007
According to Sec. 1.51(c) of the merger guidelines, markets
with post-merger HHIs above 1,800 are highly concentrated and HHIs
of the magnitude shown in Chart C1 create the presumption that the
merger would be ``likely to create or enhance market power or
facilitate its exercise.'' This section of the merger guidelines
states in part that:
Where the post-merger HHI exceeds 1800, it will be presumed that
mergers producing an increase in the HHI of more than 100 points are
likely to create or enhance market power or facilitate its exercise.
The presumption may be overcome by a showing that factors set forth
in Sections 25 of the Guidelines make it unlikely that the merger
will create or enhance market power or facilitate its exercise, in
light of market concentration and market shares.
Pre-merger, Abitibi has a 27.4% market share based on estimated
2006 capacity and Bowater has a 17.5% share. Following the merger
Abitibi-Bowater would have a combined share of 45.0%. The next
largest newsprint manufacturer, White Birch would have a 9.0% share.
See Chart C2 below.
[GRAPHIC] [TIFF OMITTED] TN10JN08.008
[[Page 32871]]
2. Analysis of the Increase in Concentration in the East of the
Rockies Newsprint Market Based on Estimated 2006 Capacity
Based on estimated 2006 east of the Rockies newsprint capacity,
the pre-merger HHI is 1,876. If the merger is consummated, the
change in the HHI would be 1,445 and the post-merger HHI would be
3,321. See Chart C3 below.\44\ In terms of pre-merger and post-
merger HHIs, an east of the Rockies newsprint market would be more
concentrated than a NA newsprint market.
---------------------------------------------------------------------------
\44\ The source for Charts 3 and 4 is Table C3 in Attachment C.
[GRAPHIC] [TIFF OMITTED] TN10JN08.009
Abitibi has only two west of the Rockies mills (Mackenzie, BC
and Snowflake, AZ) and, as noted above, Bowater does not own a
majority interest in any west of the Rockies newsprint mill.
Virtually all of their combined capacity is located east of the
Rockies. Pre-merger, Abitibi has a 30.8% market share based on
estimated 2006 east of the Rockies capacity and Bowater has a 23.4%
share. Following the merger, AbitibiBowater would have a combined
share of 54.3%. The next largest newsprint manufacturer, White
Birch, would have a 12.1% share. See Chart C4 below.\45\
---------------------------------------------------------------------------
\45\ The following North America newsprint manufacturers have
all of their newsprint capacity in mills located west of the
Rockies: Catalyst, North Pacific, Blue Heron, Ponderay, Howe Sound,
and Inland Empire. In addition, the following NA newsprint
manufacturers have some but not all of their newsprint capacity in
mills located west of the Rockies: Abitibi (561,000 metric tonnes)
and SP Newsprint (395,000 metric tonnes).
---------------------------------------------------------------------------
[[Page 32872]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.010
D. Analysis of the Increase in Concentration and Decrease in Capacity
in the NA Newsprint Market 1995-2006
1. The Increase in Concentration in the NA Newsprint Market 1995-
2005 as Described by Abitibi and Bowater
a. Description of the Increase in Concentration by John Weaver,
President and CEO of Abitibi
John Weaver, the President and CEO of Abitibi, has discussed the
increase in consolidation in the NA newsprint market in a number of
presentations to investment analysts. Slide 5 below is from a
presentation that Weaver made at the UBS Global Paper and Forest
Products Conference on September 18, 2003. The presentation was
titled ``Is the Industry Positioned to Reap the Benefits of Its
Restructuring?'' and is available on the Abitibi Web site. Slide 5
shows Abitibi with a 32% capacity share and Bowater with a 19%
capacity share in NA.
[[Page 32873]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.011
Slide 3 below from Abitibi's UBS presentation states that the
capacity share of the top 5 NA newsprint producers more than doubled
from 35% in 1995 to 73% in 2002. Slide 3 also identifies the
acquisitions and mergers that occurred over the period 1995 to 2002
that enabled the share of the top 5 newsprint producers in North
America to rise from 35% to 73%.\46\ Some of these mergers involved
companies that Abitibi and Bowater eventually acquired.
---------------------------------------------------------------------------
\46\ Seven mergers identified in the lower right hand corner of
the slide involve overseas transactions. In 2003, Abitibi was a 50%
owner of PanAsia, a large Asian newsprint producer. In 2005, Abitibi
sold its interest in PanAsia to the other 50% owner, Norske Skog, in
order to reduce its debt, part of which was incurred in the Donohue
acquisition in 2000. See Abitibi presentation ``Divesting PanAsia: A
Good Price at the Right Time,'' September 2005, pp. 5-6. This
presentation is available on the Abitibi Web site.
---------------------------------------------------------------------------
[[Page 32874]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.012
Slide 4 below from Abitibi's UBS presentation shows the
acquisitions that enabled Abitibi to increase its NA newsprint
capacity share from 11.2% in 1995 \47\ to 32% in 2003. All of these
acquisitions occurred between 1995 and 2000.
---------------------------------------------------------------------------
\47\ Source: ``Newsprint: A Pulp & Paper Market Focus Book,'' p.
113, 1999.
---------------------------------------------------------------------------
[[Page 32875]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.013
b. Description of the Increase in Concentration by David Paterson,
President and CEO of Bowater
In a slide show presentation at the Annual Citigroup Paper and
Forest Products Conference on December 7, 2006, David Paterson,
President and CEO of Bowater, spoke to investment analysts about
Bowater's product lines, efforts to reduce costs, and financial
results. Referring to Slide 12,\48\ Paterson noted that there had
been significant consolidation in the newsprint industry and that he
expected that consolidation would continue. See Slide 12 below.
Slide 12 shows that in 1995 the top 5 producers had a combined share
of 49%.\49\ If the Abitibi-Bowater merger is allowed to be
completed, the chart shows that the merged entity will have a share
equal to the 49% share of the top 5 firms in 1995. The chart also
shows the pre-merger share of the top 5 firms increased from 49% in
1995 to 75% in 2006.
---------------------------------------------------------------------------
\48\ The slide show is titled ``Bowater: Citigroup Global Paper
and Forest Products Conference, December 2006'' (``Paterson 2006
Citigroup slide show''). Both the slide show and an audio recording
of Paterson's remarks are available on the Bowater Web site.
\49\ Slide 3 in the Weaver UBS presentation discussed above
shows the top 5 NA newsprint producers with a 35% capacity share in
1995, 14% lower than the capacity share shown in the Paterson
presentation. Page 113 of ``Newsprint: A Pulp & Paper Market Focus
Book'' (1999) shows the top 5 newsprint producers with a 42.5%
capacity share.
---------------------------------------------------------------------------
[[Page 32876]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.014
2. Concentration in the NA Newsprint Market in 1995
In 1995, Abitibi Price was the largest newsprint manufacturer
with a capacity share of 11.2% and Bowater was the third largest
firm with a capacity share of 8.1%. Their combined share was 19.4%.
If Abitibi-Price and Bowater had merged in 1995, the pre-merger HHI
would have been 545, the change in the HHI would have been 183 and
the post-merger HHI would have been 728. According to Sec. 1.51(a)
of the Merger Guidelines, markets with post-merger HHIs below 1,000
are unconcentrated. See Chart D1 below which includes both the HHIs
for the 1995 hypothetical AbitibiBowater merger and the HHIs for the
proposed 2007 AbitibiBowater merger.\50\
---------------------------------------------------------------------------
\50\ The sources for Chart D1 are Table D1 (1995 capacity
shares) in Attachment D and Table C2 (2006 capacity shares) in
Attachment C.
[GRAPHIC] [TIFF OMITTED] TN10JN08.015
[[Page 32877]]
3. Acquisitions and Exits of NA Newsprint Manufacturers Since 1995
See Table D1 in Attachment D for capacity shares and HHIs for
all 33 NA manufacturers of newsprint in 1995. Based on Table D1,
Table D2 in Attachment D identifies all acquisitions and exits in
the NA newsprint market since 1995.
See Chart D2 below, which shows capacity shares for the top 20
newsprint manufacturers in 1995.\51\
---------------------------------------------------------------------------
\51\ The source for Chart 2 is Table D1 in Attachment D. The
source for Table D1 is ``Newsprint: A Pulp & Paper Market Focus
Book,'' (1999), p. 113. In 1995 Avenor was a 40% minority owner of
Ponderay Newsprint. For the purposes of this analysis, Ponderay is
listed as a separate firm. Bowater acquired its current 40% interest
in Ponderay when it acquired Avenor in 1998. In 1998, Ponderay had a
capacity of 240,000 metric tonnes (1998 Bowater Annual Report, p.
3).
[GRAPHIC] [TIFF OMITTED] TN10JN08.016
The chart also shows which of the top 20 newsprint manufacturers
in 1995 were acquired directly and indirectly by Abitibi and Bowater
after 1995.\52\ As Table D2 shows, Abitibi also indirectly acquired
Finley Forest Industries, the 27th largest newsprint manufacturer in
1995 with a capacity share of 1.2%. Bowater also directly acquired
Alliance, the 24th largest manufacturer in 1995 with a capacity
share of 1.3%.
---------------------------------------------------------------------------
\52\ An example of a direct acquisition is Abitibi's acquisition
of Donahue in 2000. Donahue had acquired QUNO in 1996. When Abitibi
acquired Donahue in 2000, it also indirectly acquired QUNO.
---------------------------------------------------------------------------
4. Analysis of the Reduction of Newsprint Capacity in North America
1995 to 2006
In 1995, there were 16,093,000 metric tonnes of NA newsprint
capacity. In 2006, there were an estimated 12,760,000 metric tonnes
of NA newsprint capacity, a reduction of 20.7%, most of it occurring
since 2002.\53\ Utilizing the data and other information in Table C2
in Attachment C and Tables D1 and D2 in Attachment D, it is possible
to identify the sources for the reduction of newsprint capacity in
North America since 1995. This is a two-step process. The first step
is to adjust the 1995 capacities and shares shown in Table D1 to
account for subsequent acquisitions while eliminating the acquired
firms from the list of manufacturers, See Table D3 in Attachment D.
As shown in Table D2, there were 34 manufacturers of newsprint in
North America. After all acquisitions since 1995 are accounted for,
21 manufacturers remain. There has been a reduction of 14 newsprint
manufacturers through acquisition since 1995.\54\ Through direct and
indirect acquisitions, Abitibi accounted for five of those newsprint
manufacturer reductions and Bowater four. Table D3 shows that
Abitibi also accounted for 46.8% of the acquired capacity and
Bowater 21.2% for a combined total of 68.0%. Through these
acquisitions, Abitibi increased its capacity share by 22.7% from
11.2% to 34.0% and Bowater increased its capacity share by 10.3%
from 8.1% to 18.4%. Abitibi and Bowater increased their combined
capacity share by 33.0% from 19.4% to 52.4%. The second step is to
subtract estimated 2006 newsprint capacity from adjusted 2005
newsprint capacity. See Table D4 in Attachment D. Table D4 shows
that the total net reduction in capacity between 1995 and 2006 was
3,333,000 metric tonnes. Table D5 below summarizes the results in
Table D4.
---------------------------------------------------------------------------
\53\ Sources: See Table DI in Attachment D and Table C2 in
Attachment C.
\54\ The net reduction in firms is 13 because a new firm was
added to the NA newsprint market in 1999 when Bowater sold its East
Millinocket, ME newsprint mill to Great Northern Paper. Following
Great Northern's subsequent bankruptcy, Katahdin acquired the East
Millinocket mill in 2003 and produced newsprint until it converted
its newsprint capacity to uncoated groundwood specialty grades in
2005-2006. In 1998, the newsprint capacity of the East Millinocket
mill was $168,000 metric tonnes (1998 Bowater Annual Report, p. 4).
[[Page 32878]]
Table D5.--Summary of the Net Capacity Reduction in NA Newsprint Capacity 1995-2006
----------------------------------------------------------------------------------------------------------------
Percent of net
Percent of capacity
Net capacity total net reductions
changes 1995- capacity 1995-2006 for
2006 changes 1995- 5 firms that
2006 remain in the
market
----------------------------------------------------------------------------------------------------------------
Abitibi......................................................... (1,964) 58.9 60.9
Bowater......................................................... (731) 21.9 22.7
Catalyst........................................................ (514) 15.4 15.9
Tembec.......................................................... (15) 0.5 0.5
North Pacific................................................... (2) 0.1 0.1
Net Capacity Reductions for 5 Firms That Remain in the NA (3,226) 96.8 100.0
Newsprint Market Today.........................................
Net Capacity Additions or No Capacity Change for 11 Firms That 630 (18.9) ..............
Remain in the NA Newsprint Market Today........................
Net Capacity Reduction of the 16 Firms That Remain in the NA (2,596) 77.9 ..............
Newsprint Market Today.........................................
5 Firms That Exited from the NA Newsprint Market Between 1995 (737) 22.1 ..............
and 2006.......................................................
-----------------------------------------------
Total Net Capacity Reduction 1995-2006...................... (3,333) 100.0 ..............
----------------------------------------------------------------------------------------------------------------
The firms in Table D5 can be divided into three categories: (1)
Firms remaining today in the NA newsprint market that had a net
reduction in capacity over the period 1995 to 2006; (2) firms
remaining today in the NA newsprint market that had a net addition
in capacity over the period 1995 to 2006; and (3) firms who exited
from the NA newsprint market between 1995 and 2006.\55\ As Table D5
shows, there are 5 firms in the first category, 11 firms in the
second category, and 5 firms in the third category.
---------------------------------------------------------------------------
\55\ Of the five firms that exited from the NA newsprint market,
four of those firms converted their newsprint capacity to other
groundwood grades. Only Garden State exited by permanently closing
its newsprint mill.
---------------------------------------------------------------------------
The first and third categories total 3,963,000 metric tonnes in
net capacity reductions. These net capacity reductions are partially
offset by 630,000 metric tonnes in net capacity additions by 10 of
the 16 firms that remain in the market today.\56\ After this offset
is taken into account, the total net reduction in NA newsprint
capacity is 3,333,000 metric tonnes.
---------------------------------------------------------------------------
\56\ One of the remaining firms had no change in capacity. There
could be several reasons for the net increases in capacity. These
may include speed-ups and other improvements to existing newsprint
capacity and switching capacity from the production of uncoated
groundwood grades to newsprint. The increase for Inland Empire is
due to the installation of a new newsprint machine in 2001 and the
permanent closure of the machine it replaced. There were no other
installations of new newsprint machines in North America between
1995 and 2006. Some of the additions may not be real (e.g., they may
result from methodological differences in reporting or estimating
capacity in 1995 and 2006 or they may result from errors).
---------------------------------------------------------------------------
The first category in Table D5 shows that the reductions by
Abitibi, Bowater, and Catalyst account for 99.5% of NA capacity
reductions by firms that (a) had net capacity reductions between
1995 and 2006 and (b) remain in the market today.\57\ Abitibi
accounts for 60.9% of the net capacity reduction, Bowater for 22.7%
of the net capacity reduction, and Catalyst for 15.9% of the net
capacity reduction.\58\ Combined, Abitibi and Bowater account for
83.6% of the net reduction in NA newsprint capacity since 1995 shown
in the first category. See Chart D3 below.
---------------------------------------------------------------------------
\57\ The net capacity reduction shown for North Pacific is not
meaningful. In 2004, Tembec permanently closed one newsprint machine
at its mill in Kapuskasing, ON.
\58\ It should be noted that some of the net capacity reduction
for Abitibi, Bowater, and Catalyst occurred in acquired firms after
1995 but prior to their acquisitions by Abitibi, Bowater, or
Catalyst. The most significant such capacity reduction is the
closure of a 184,000 metric tonne capacity newsprint machine by
MacMillan Bloedel in 1996. MacMillan Bloedel was subsequently
acquired by Pacifica which was subsequently acquired by Norske
Canada (later renamed Catalyst). This machine closure accounts for
35.8% of Catalyst's total net capacity reduction shown in Table D5
and Chart D3. Capacity reductions after 1995 by firms before they
were acquired by Abitibi or Bowater make up a much smaller
percentage of their respective net capacity reductions. Taking into
account these prior capacity reductions for the three acquiring
firms, Abitibi's share of the net capacity reduction of firms that
remain in the market would increase to 66.2%, Bowater's share would
increase to 21.8% and Catalyst's share would decrease to 11.4%.
Source: ``Newsprint: A Pulp & Paper Market Focus Book,'' p. 20
(1999).
---------------------------------------------------------------------------
[[Page 32879]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.017
As Table D4 indicates, Abitibi lost 6.5% in capacity share.
Bowater lost 0.9% in capacity share, and Catalyst lost 2.1% in
capacity share between 1995 (adjusted 1995 capacity) and 2006 due to
their net capacity reductions. Combined, the three firms lost 9.5%
in newsprint capacity share. The 5 firms that exited the NA
newsprint market lost a combined 4.6% in newsprint capacity share.
E. NA Newsprint Demand and Supply
1. Introduction
The demand for newsprint by daily newspapers is derived from the
demand for newspapers by readers and advertisers. Demand, as used in
this sense, means the demand curve for newsprint and the demand
curve for newspapers. If the demand for newspapers declines
independent of the price of newsprint, the demand curve for
newspapers will shift downward causing the newspaper's derived
demand curve for newsprint to also shift downward.
Chart E2 below shows the total NA average quarterly demand for
newsprint 1999 to 2006.\59\ Demand, as used in this sense, means the
quantity of newsprint purchased during a quarter. The same is true
with respect to Chart E1 below, except that the period over which
quantity is purchased is a year. Chart E2 shows that while there
were quarters where demand increased from the prior quarter the
overall trend is declining in demand. Demand in Q4 2006 was 24.5%
lower than demand in Q1 1999. Chart E2 cannot explain the causes of
this decline in demand (i.e., quantity purchased); it can only show
that demand (i.e., quantity purchased) did generally decline over
the 32 quarters.
---------------------------------------------------------------------------
\59\ Annual demand equals annual consumption plus the change in
inventories held by customers from the prior December.
---------------------------------------------------------------------------
Chart E6 shows quarterly prices for newsprint. Prices declined
from the Q1 1999 to Q3 1999, generally increased from Q3 1999 to Q2
2001, declined significantly from Q2 2001 before bottoming out in Q2
and Q3 2002, and generally increasing from Q3 2002 to Q3 2006 before
declining somewhat in Q4 2006. Just considering the period from Q3
2005 to Q3 2006 the price of newsprint increased by an aggregate of
$222 or 49.0% while demand (quantity purchased) declined by an
aggregate of 521,000 metric tonnes or 18.0%.
This section, as well as Sections D and F, explores the likely
causes of the significant and sustained increase in newsprint prices
over the two periods described above while newsprint demand
(quantity purchased) was either flat or steadily declining. See
Chart 7 below, which combines Chart 2 and Chart 6. In seeking the
explanation for the likely causes, we make three main observations:
(1) The decline in demand (quantity purchased) over the period
1999 to 2006 was due primarily by downward shifts in the demand
curve for newspapers caused by declining circulation and advertising
lineage independent of increases in the price of newsprint. The
downward shifts in the demand curve for newspapers caused downward
shifts in the derived newsprint demand curve.
(2) Holding the newsprint supply curve constant, downward shifts
in the newsprint demand curve would be expected to lead to lower
newsprint prices. That has not happened. The steady rise in
newsprint prices over the two periods was primarily caused by the
strategic and coordinated removals of newsprint capacity from the
market by Abitibi and Bowater in response to the downward shifts in
the newsprint demand curve, These upward shifts of the supply curve
maintained maximum operating rates and increased newsprint prices.
Both Abitibi and Bowater pursued the approach of reducing capacity,
which was highly successful in achieving a steady increase in the
price of newsprint.
(3) It is not plausible that increases in the price of inputs
used to manufacture newsprint or the appreciating Canadian dollar
are a significant cause of the price increases.
The reduction in newsprint capacity by Abitibi and Bowater and
its relationship to the maintenance of high operating rates and
rising prices was recognized as a strategic move by newsprint
producers, newsprint buyers, and newsprint industry analysts, as
this passage from The Global Pulp & Paper Fact Book 2006 \60\ on p.
152 indicates.
---------------------------------------------------------------------------
\60\ The Global Pulp & Paper Fact Book 2006 is published by
RISI.
Even though demand continued to decline during the 2003-2006
period, newsprint producers have steadily raised prices during the
past several years. Through a policy of closing mills and either
shutting newsprint machines or converting them to added-value
grades, newsprint producers have kept supply and demand relatively
balanced, and operating rates high enough to support the progression
of supply-driven price increases. By third quarter of 2006 the
market average stood at $675/tonne with another $20/tonne increase
proposed by some producers for August 1 and by others for September
---------------------------------------------------------------------------
1.
The Global Pulp & Paper Fact Book 2006 does not identify any
newsprint manufacturers but noted that unnamed newsprint
manufacturers had a ``policy of closing mills and either shutting
newsprint machines or converting them to added-value grades'' in
order to keep ``supply and demand relatively balanced, and operating
rates high enough to support the progression of supply-driven price
increases.'' (Emphasis
[[Page 32880]]
added) The identification of those newsprint manufacturers will be
the subject of Section F.
2. NA Demand (Quantity Purchased) 1999-2006
NA annual newsprint demand (quantity purchased) has fallen 25.5%
on an annual basis between 1999 and 2006. See Chart E1 below.\61\ In
1999, imports accounted for 3.3% of NA demand. Since 1999, imports
have accounted for 2.0% or less of NA purchases. As Chart E1 shows,
imports of newsprint into North America have not been a significant
source of supply for NA newspaper publishers and other NA purchasers
of newsprint. In 2006, imports supplied just 1.5% of NA newsprint
consumption. For the first two months of 2007, imports have fallen
56.1% compared to the first two months of 2006.\62\
---------------------------------------------------------------------------
\61\ Sources: December 2006 and December 2005 PPPC NA Newsprint
Statistics-Flash Report (``Flash Report''). and December 2001-2004
PPPC NA Newsprint Statistics Monthly Bulletin (``PPPC Monthly
Bulletin'').
\62\ Source: February 2007 Flash Report.
[GRAPHIC] [TIFF OMITTED] TN10JN08.018
Chart E2 below shows NA demand (quantity purchased) by quarter
from Q1 1999 to Q4 2006. Quarterly NA demand (quantity purchased)
has decreased from Q4 1999 to Q4 2006 by 28.8%
[[Page 32881]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.019
3. Causes of the Decline in NA Newsprint Demand 1999-2006
a. Estimates of the Causes of the Decline in NA Newsprint Demand by the
PPPC
There are three main causes of the decline in NA newsprint
demand over the period 1999-2006: (a) Declining newspaper
circulation; (b) declining newspaper ad linage; and (c) newspaper
efforts to conserve on the consumption of newsprint.\63\ These
conservation efforts include reducing the width of newspapers,
switching to lighter basis weight paper (i.e., thinner paper), and
eliminating certain sections of the newspaper and placing them on
the newspaper's Web site (e.g., stock tables and TV listings).
---------------------------------------------------------------------------
\63\ In 2006, U.S. daily newspapers accounted for 71.3% ofNA
newsprint demand and 80.2% of US. newsprint demand and U.S.
newsprint demand accounted for 88.7% of NA newsprint demand. Source:
December 2006 Flash Report.
---------------------------------------------------------------------------
In the March 2007 edition of Pulp & Paper Magazine, Bill Moore
of Moore & Associates, a recycled paper consulting firm, states that
``[t]he decline in newsprint consumption in North America is
structural and very little can be done at this point to change the
situation.'' \64\
---------------------------------------------------------------------------
\64\ ``Another side of the decline of newspapers.'' Mr. Moore
believes that local governments should put more effort into
encouraging citizens in their communities to recycle old newspapers.
---------------------------------------------------------------------------
Mr. Moore described how the decline in newspapers has led to the
decline in the production of newsprint.
The reasons for this decline in NA newsprint production have
been well documented and are related to a series of factors in the
decline of newspapers:
Newspaper readership in the U.S. has been steadily
declining for a number of years and the downward trend has
accelerated in the last few years.
Many newspapers have moved to smaller formats, tighter
margins, and also the use of a lower basis weight sheet.
More advertising and classifieds have moved to the web.
Stock pages, and even the classical in-depth reporting
that newspapers were known for, have been eliminated from many
papers. The recent Wall Street Journal changes resulted in a 15%
reduction in the use of newsprint [by that newspaper]!
Martine Hamel, head of market research for the PPPC, has
estimated the relative size of each of these effects \65\ on the
consumption of newsprint by U.S. daily newspapers.\66\ Slide 17 of
the 2005 PPPC Presentation below shows that for the first nine
months of 2005 compared to the first nine months of 2004,
consumption by U.S. daily newspapers declined 4.9%. Declines in ad
linage and circulation accounted for about 63% of the consumption
decline and switching to lower basis weight paper (i.e., grammage
reduction) accounted for about 31% of the consumption decline. Other
(presumably other conservation methods including width reductions)
accounted for 6%.
---------------------------------------------------------------------------
\65\ See the presentations to the November 2005 and 2006 Joint
NPA/NAA Newsprint Conference titled ``Review and Forecast of
Newsprint Demand and Supply'' (``PPPC 2005 and 2006 NPA/NAA
Presentations''). NPA is the Newsprint Producers Association.
\66\ Annual NA demand equals shipments to North America by NA
mills plus imports from overseas. Annual newsprint consumption by NA
customers equals NA demand minus the change in newsprint inventories
from the prior December. In 2006, the change in inventories at U.S.
daily newspapers was a decline of 58,000 metric tonnes or 0.8%
(absolute) of NA consumption and demand. The PPPC publishes
inventory data for U.S. newsprint customers but not Canadian
newsprint customers.
---------------------------------------------------------------------------
[[Page 32882]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.020
Slide 8 of the PPPC 2006 NPA/NAA Presentation shows a 7.8%
decline in U.S. daily newsprint consumption for the first nine
months of 2006 compared to the first nine months of 2005.\67\ The
decline in ad linage and circulation account for about 55% to 60% of
the decline and grammage reduction and other conservation methods
such as width reductions account for 40% to 45% of the decline.
---------------------------------------------------------------------------
\67\ U.S. daily newspapers accounted for 83.7% of the decline in
NA demand between 2005 and 2006. Other US. newsprint customers
accounted for 14.1% of the decline and Canadian customers accounted
for 23% of the decline.
[GRAPHIC] [TIFF OMITTED] TN10JN08.021
b. Distinguishing Between Shifts in the Newsprint Demand Curve and
Movements Along the Newsprint Demand Curve
If the newsprint supply curve shifts upward and to the left due,
say, to the permanent closure of newsprint capacity, a new
equilibrium price and quantity will be established. The new price
will be higher than the old price and the new quantity purchased
will be lower than the old quantity purchased. This can be described
as a movement along the demand curve caused by the shift of the
supply curve upward and to the left. The effect of the supply curve
shift on equilibrium price and quantity will depend upon the price
elasticity of demand. If the demand curve is highly inelastic in the
region of the supply curve shift,\68\ then price
[[Continued on page 32883]]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
]
[[pp. 32883-32932]] United States v. Abitibi-Consolidated Inc. et al.; Response to
Public Comment on the Proposed Final Judgment
[[Continued from page 32882]]
[[Page 32883]]
would likely rise significantly and quantity of newsprint purchased
would be little reduced from the previous level. If demand were
elastic in the region of the supply curve shift, then, compared to
an inelastic demand curve, the resulting equilibrium price would be
lower and the resulting quantity reduction would be greater.
---------------------------------------------------------------------------
\68\ While we have not attempted to estimate the demand
elasticity for the NA newsprint market, we note that an article in
2004 reported on an analysis that estimated the elasticity of the
U.S. demand for newsprint at 0.36 taking into account structural
changes in U.S. demand. See Jari Kuuluvainen, ``Structural Change in
U.S. Newsprint Demand: GDP and Price Elasticities,'' University of
Helsinki, Department of Forest Economics, Reports 34, 2004,
p. 8. A demand elasticity of 0.36 is in the same range as demand
elasticities reported in earlier articles. An article in 1997
reported the demand elasticity in North America at 0.22. Other
estimates cited in this article have been about twice as large.
Estimates of demand elasticity vary from 0.22 to 044. These
estimates all indicate a fairly inelastic demand curve for
newsprint. See Ylbing Zhang and Joseph Buongiorno, ``Communication
Media and Demand for Printing and Publishing Papers in the United
States,'' Forest Science 43(3) (August) 1997, p. 372. The results of
our analysis of the proposed Abitibi-Bowater merger are consistent
with an inelastic demand curve.
---------------------------------------------------------------------------
Newspapers, of course, buy newsprint to help meet the demands of
their customers, the readers and advertisers. Their demands for
newspapers are exogenous to the newspapers' demand for newsprint.
That is, their demand for newspapers is shaped by factors completely
independent of the market for newsprint.\69\ If the demand for
newspapers declines because, say, readers and advertisers are moving
from newspapers to the Internet, this movement will result in the
newspaper demand curve for newsprint shifting downward and to the
left. As a result of the shift of the demand curve down the supply
curve, both price and quantity purchased will decline.
---------------------------------------------------------------------------
\69\ While it is certainly possible that some newspapers have
been able to pass some portion of the last four years' of newsprint
price increases on to newspaper customers, we are unaware of any
such examples. To the extent there are such examples, they are
likely to be insignificant in comparison to the aggregate magnitude
of the newsprint price increases.
---------------------------------------------------------------------------
When newspapers narrow the width of the page or buy lower basis
weight newsprint or move stock tables from the newspaper to their
web sites, they are permanently removing newsprint demand from the
market. In so doing, they are shifting the demand curve downward and
to the left. While the conservation efforts are no doubt largely in
response to the four years of newsprint price increases, they do not
indicate movements along the demand curve. They indicate shifts in
the demand curve. If newsprint prices declined by 10 percent, it is
implausible that newspapers would go back to wider webs or start
running stock tables in the newspaper again. As long as the relative
prices for higher and lower basis weight paper remain approximately
the same, as seems likely, newspapers will have no incentive to
switch back to higher basis weight paper.
The demand removal through conservation efforts is directly
analogous to the capacity removal that has been taking place in the
NA newsprint market, particularly since 2002. The capacity removals
shift the supply curve upward and to the right. The demand removals
shift the demand curve downward and to the left. The major
difference between the two is that the capacity removals occur more
quickly and have a much greater impact on price than the demand
removals. The narrowing of the width of newspapers from 50 inches to
48 inches would be the equivalent of a 4 percent reduction in price.
The move from 30 lb. newsprint to 27.7 lb. newsprint \70\ will only
save a newspaper an equivalent of a 2.7% reduction in the price of
30 lb. newsprint.\71\ If the price of 30 lb. newsprint were $630 per
metric tonne (as it was in February 2007), a 2.7% net savings in
newsprint purchases would be equivalent to a $16.94 reduction per
metric tonne in the price of 30 lb. newsprint.
---------------------------------------------------------------------------
\70\ Basis weight correlates with the thickness of the newsprint
sheet. The higher the basis weight, the thicker the newsprint sheet
and vice-versa. Most newsprint in North America is sold in two basis
weights 30 lb. and 27.7 lb. Many of the largest newspapers and
newspaper chains in the U.S. have switched from 30 lb. basis weight
to 27.7 lb. basis weight newsprint in the last several years.
\71\ Holding constant the square footage of printing surface
purchased, the move to 27.7 lb. newsprint by the customer will
reduce the tonnage needed by 8.5%. However, the newspaper will be
paying more per metric tonne for the reduced amount of newsprint.
According to Pulp & Paper Week, the February 2007 price of 30 lb.
newsprint delivered in the eastern U.S. was $630 per metric tonne
and the price of 27.7 lb. newsprint was $670 per metric tonne. At
these prices, the cost per tonne purchased will increase by 6.3%.
When these two effects are combined, the newspaper will save 2.7% or
$16.94. per metric tonne. Whether the newsprint manufacturer will
financially benefit from the switch depends on the relationship
between the manufacturer's variable costs to produce the lower basis
weight paper and the higher basis weight paper. If the
manufacturer's variable cost to produce the lower basis weight paper
is not too far above the variable cost to product the higher basis
weight paper, the profits of the manufacturer could actually
increase as a result of the switch.
---------------------------------------------------------------------------
Slide 5 of the 2006 PPPC presentation shows that in 2006, about
half of the newsprint shipped by NA mills to NA customers was 27.7
lb. newsprint. That implies that only half of the 2.7% or $16.94
cost savings potentially available to newsprint customers had been
realized even though prices had steadily risen over the prior four
years. Slide 6 in the same presentation also shows that conservation
efforts on the part of newsprint customers take years to accomplish
in the aggregate and even then, some and perhaps many customers will
never convert. The same general comments can be made with respect to
the reduction of page widths to 48 inches from 50 inches. Finally,
newsprint buyers have said that the low-hanging fruit has been
picked and that the opportunities for cost savings from future
efforts to conserve on newsprint are reaching the point of
diminishing returns.
4. Projected NA Newsprint Demand 2006-2008
The PPPC forecasts a 5.9% decline in NA newsprint demand in 2007
and an additional 3.3% decline in NA newsprint demand in 2008.\72\
See Slide 10 below.\73\
---------------------------------------------------------------------------
\72\ While Slide 10 forecasts a 4.9% decline in NA demand for
2006, the actual decline was 6.0%. Source: December 2006 PPPC Flash
Report.
\73\ Source: 2006 NPA/NAA Presentation.
[GRAPHIC] [TIFF OMITTED] TN10JN08.022
Assuming the PPPC forecast is reasonably accurate, NA demand
will fall by a total of 879,000 metric tonnes over the two-year
period. Assuming no change in overseas shipments from NA mills or in
imports by NA customers from 2006 levels, NA manufacturers would
have to temporarily idle or permanently shut down 1,055,000 metric
tonnes of capacity during 2007 and
[[Page 32884]]
2008 in order to maintain a 95% industry operating rate.\74\ That
amount of capacity reduction would represent 8.4% of current NA
capacity and 19.1% of the current combined Abitibi-Bowater capacity.
---------------------------------------------------------------------------
\74\ The industry operating rate for 1996 was 94% down 2% from a
96% operating rate in 2005 and 2004. Source: December 2005 and 2006
PPPC Flash Reports.
---------------------------------------------------------------------------
5. Production, Shipments, and Operating Rates of NA Newsprint Mills
1999-2006
Shipments by NA mills to NA customers and overseas customers
declined significantly over the period 1999 to 2006. See Chart E3
below.\75\ Shipments to NA customers declined by 24.1% and shipments
to overseas customers declined by 25.8%.
---------------------------------------------------------------------------
\75\ Sources: December 2005 and 2006 PPPC Flash Reports and
December 2001-2004 PPPC NA Newsprint Statistics Monthly Bulletin
(``PPPC Monthly Bulletin'').
[GRAPHIC] [TIFF OMITTED] TN10JN08.023
As a result of the decline in shipments to NA and overseas
customers, NA newsprint production declined by 24.5% between 1999
and 2006. Due to newsprint mill closures, newsprint machine shut
downs, and newsprint machine conversions to other grades, NA
newsprint capacity has declined by 23.7% during the same period.
Chart E4 below shows capacity and production by quarter over the
period 1999 to 2006. The chart shows that both capacity and
production have declined steadily from the beginning of 2001 through
the end of 2006.
[[Page 32885]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.024
Chart E5 below shows the quarterly operating rates \76\ of NA
newsprint mills for the period 1999-2006. After the operating rate
reached 97.3% in the third and fourth quarters of 2000, the
operating rate dropped slightly to 96.0% in the first quarter of
2001 and then plunged sharply for the rest of 2001 reaching a low of
86.0% in the third quarter of 2001. This plunge corresponds to the
widening gap between capacity and production shown in Chart E4 over
the same period. The sharp decline in the operating rate was caused
by the 18.7% decline in the NA demand for newsprint that occurred
between the third quarter of 1999 and the first quarter of 2002. The
decline in newsprint demand followed the significant slowing of the
U.S. economy that began in the first quarter of 2001 and which was
exacerbated by the economic disruption caused by the attacks of
September 11, 2001. \77\ After the third quarter of 2001, the
operating rate increased fairly steadily reaching 96.3% in the first
quarter of 2004 and remaining at about 96% for the next two years.
The operating rate then mostly declined throughout 2006 falling to
93.0% in the fourth quarter of 2006.
---------------------------------------------------------------------------
\76\ The operating rate is production as a percentage of
capacity.
\77\ From the fourth quarter of 2000, the U.S. Real Gross
Domestic Product declined for three consecutive quarters before
increasing in the fourth quarter of 2001. Source: Economic Report of
the President, February 2003, Table B.2--Real Gross Domestic Product
1959-2002, p. 278.
[GRAPHIC] [TIFF OMITTED] TN10JN08.025
[[Page 32886]]
6. The Price of Newsprint per Metric Tonne (Eastern U.S., 30 lb.)
1999 to 2006 by Quarter
Chart E6 below shows the price of newsprint per metric tonne by
quarter for the period 1999 to 2006.\78\ The price is the delivered
price per metric tonne in the eastern United States for 30 lb. basis
weight newsprint.
---------------------------------------------------------------------------
\78\ The source for the quarterly prices is RISI. RISI
calculates quarterly prices based on monthly prices that appear in
the RISI publication Pulp & Paper Week.
[GRAPHIC] [TIFF OMITTED] TN10JN08.026
The price of newsprint increased $145 or 30.2% from the third
quarter of 1999 to the second quarter of 2001 before falling by $172
or 27.5% through the second quarter of 2002. As Chart E6 shows,
price increased in 5 of the 7 quarters during the period of the
price rise. In the other two quarters, price was unchanged.
After the bottom was reached in the second and third quarters of
2002, the price of newsprint steadily increased over the next four
years from $453 to $675 in the third quarter of 2006. This was an
increase of $222 or 49.0%. As Chart E6 shows, price increased in 14
of the 16 quarters over this four-year period. In one quarter, the
price was unchanged and in one quarter the price declined by $5. In
the fourth quarter of 2006, price decreased slightly to $660.
Combining Chart E2 and Chart E6, shows the two sustained price
increases from the end of 1999 to the beginning of 2001 and from the
end of 2002 to the end of 2006. During the first period demand was
more or less flat and during the second period demand was steadily
trending downward. See Chart E7 below.
[[Page 32887]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.027
F. Evidence From Presentations to Investment Analysts and Other Public
Information That Abitibi and Bowater Have Used Their Control Over
Newsprint Capacity and the Newsprint Industry Operating Rate To
Significantly Raise the Price of Newsprint 2002 to 2006
1. Introduction
In Section D above, the significant increase in concentration in
the NA newsprint industry between 1995 and 2006 and the significant
decrease in newsprint capacity over that same period were analyzed.
Due primarily to acquisitions by Abitibi and Bowater between 1995
and 2001, the NA newsprint market was transformed from an
unconcentrated market in 1995 to a highly concentrated market in
2000 with Abitibi's acquisition of Donohue in April 2000. Bowater's
acquisition of Alliance in 2001 and Norske Skog's acquisition of
Pacifica, also in 2001, further increased concentration in an
already highly concentrated market.
The key to increasing newsprint prices is maintaining high
newsprint industry operating rates. Before 1995 no newsprint
producer had a market share large enough to cause an increase in the
market price. Without the acquisitions of newsprint capacity that
they made between 1995 and 2001 (described in Section D above),
Abitibi and Bowater could not have profitably pursued a strategy to
increase the market price even through coordinated interaction. With
the increased capacity under their control, Abitibi and Bowater
gained that power and have jointly used it to play the role of a
dominant firm. Publicly available information shows that Abitibi and
Bowater have acted in a coordinated manner to strategically idle and
shut down newsprint capacity sufficient to maintain high industry
operating rates and increase the price of newsprint. With the
possible exception of Catalyst,\79\ the remaining firms in the
market have played the role of fringe firms. As fringe firms, they
have been generally allowed to operate at full capacity while
Abitibi and Bowater determine the amount of their own capacity to
idle and shut down as needed to maintain high operating rates for
the NA newsprint industry.
---------------------------------------------------------------------------
\79\ If there is a West of the Rockies relevant market (as well
as an East of the Rockies relevant market), it seems possible that
Catalyst has played the role of a dominant finn in that market in
much the same way that Abitibi and Bowater have played that role in
the NA newsprint market or in an East of the Rockies relevant market
should such a market exist. Catalyst's newsprint mills are located
entirely within British Columbia. Evidence relating to the
possibility of Catalyst acting as a dominant firm in a West of the
Rockies market is discussed at the end of Section G.5.
---------------------------------------------------------------------------
Since the end of 2002, Abitibi and Bowater have used their
dominant control over NA newsprint capacity to raise operating rates
and the price of NA newsprint significantly above competitive
levels. Between the third quarter of 2002 and the third quarter of
2006, the price of newsprint has increased by an aggregate of 49.0
percent even though the demand for newsprint declined 16.5 percent
over that same period.\80\
---------------------------------------------------------------------------
\80\ As analyzed in Section E above, the decline in newspaper
demand for newsprint was due mostly to downward shifts of the demand
curve for NA newsprint and does not indicate a movement up the
demand curve in response to upward shifts of the supply curve.
---------------------------------------------------------------------------
John Weaver, the President and CEO of Abitibi, and David
Paterson, the President and CEO of Bowater, made separate
presentations at the 11th Annual Citigroup Global Paper and Forest
Products Conference on December 7, 2006 (``Citigroup Conference'').
These presentations are discussed in more detail in Sections F.2.
and F.3. below. Weaver emphasized the importance of maintaining a
``balance'' in the demand and supply of newsprint. Weaver introduced
a slide which shows the positive relationship between the level of
the newsprint industry operating rate and the percentage change in
the list price of newsprint.\81\ He said that industry demand and
supply had been in ``balance'' since 2003 and that manufacturers had
been able to improve pricing significantly since 2003. He also said
that the industry was currently operating at full capacity.
---------------------------------------------------------------------------
\81\ The percentage change shown in the slide is the percentage
change of a price in a given month from the June 2000 price of
newsprint.
---------------------------------------------------------------------------
Paterson of Bowater stated that the ``industry'' had ``responded
fairly aggressively'' to declines in demand and that Bowater was
``taking action'' to remove capacity from the market. He described
the removal of more than 10% of Bowater's newsprint capacity from
the market during 2006. During the Q&A, he said that to maintain
cash flow and dividend payments, Bowater needed to stay ahead of the
demand curve to maintain an operating rate that would give Bowater
``pricing leverage''. He said ``I can do that'' by shutting down
Bowater's high cost assets hopefully before price erosion has set in
with any significance. From these remarks, it is clear that that the
control of capacity is used by Abitibi and Bowater not only to raise
newsprint prices but to prevent prices from falling from current
levels.
This section discusses information primarily from Abitibi and
Bowater presentations to investment analysts. This evidence is
consistent with and supportive of our hypothesis that Abitibi and
Bowater acted as a joint dominant firm to raise the price of
newsprint significantly above competitive levels from the end of
2002 through 2006. Section I below discusses
[[Page 32888]]
Abitibi's closures of newsprint capacity over the period 1999 to
2001 and the relation of those closures to increases in the
operating rate and increases in newsprint prices.
This section provides evidence of Abitibi's and Bowater's
anticompetitive conduct for the period 2002-2006, based on (a) John
Weaver's presentation at the December 2006 Citigroup Conference
(Section F.2.); (b) David Paterson's presentation at the same
Citigroup Conference (Section F.3.); (c) John Weaver's presentation
to the Credit Suisse First Boston investment analysts conference in
March 2004 (Section F.4.); and (d) an interview of John Weaver by
paperloop.com in February 2004 (Section F.5.).
2. Presentation by John Weaver, President and CEO of Abitibi, at
the Citigroup Conference in December 2006
John Weaver, president and CEO of Abitibi, spoke for about 30
minutes at the December 2006 Citigroup Conference. His presentation
consisted of commentary on slides prepared by Abitibi \82\ and a
follow-up Q&A session with investment analysts.\83\
---------------------------------------------------------------------------
\82\ The 27 page slide show is titled ``Our Story on Paper.''
\83\ The slide show is available on Abitibi's Web site under
Investor Relations/Presentations and Web casts. According to
Abitibi's Web site, the audio recording of Weaver's comments at the
Citigroup Conference is no longer available on the Web site.
---------------------------------------------------------------------------
Slide 9 of Weaver's presentation shows the relation between the
level of the newsprint operating rate and percentage change in the
list price of newsprint between July 2000 and September 2006.\84\
The list price is expressed as a percentage of the June 2000 list
price. List prices are based on RISI data and operating rates are
based on PPPC data. See Slide 9 below. This slide with some
variations has been presented by Abitibi to investment analyst
groups since June 5, 2003. These presentations are archived on the
Abitibi Web site.
---------------------------------------------------------------------------
\84\ John Weaver's presentation to the June 5, 2003 Scotia
Capital Materials Conference appears to be the first presentation
where Abitibi provided a slide (Slide 15) showing the relation
between the newsprint operating rate and the price of newsprint. See
the investment analyst presentations on the Abitibi Web site. As
discussed in Section H.3.a. below, Slides 9 and 15 may have been
inspired by a similar figure published in an article by a RISI
senior economist in paperloop.com on February 20, 2003. While there
are obvious differences between Slides 9 and 15 and the figure that
appeared in the RISI economist's article, the differences are
superficial. The fundamental economic relationships that are
illustrated in Slide 9 and in the figure in RISI economist's article
are identical.
[GRAPHIC] [TIFF OMITTED] TN10JN08.028
Slide 9 and Slide 10, which follow are titled ``Industry Supply/
Demand Balance.'' Slide 9 is sub-titled ``Newsprint List Price and
Operating Rate.'' Slide 9 shows that beginning in September of 2000,
price rose about 12% above the June 2000 price by April 2001. As the
U.S. economy went into negative growth in 2001, price plunged by 33%
(from 12% above the June 2000 price to 21% below the June 2000
price) reaching the bottom in July 2002. Price then rose in a fairly
uninterrupted path from 21% below the June 2000 price to 20% above
the June 2000 price by September 2006.
The operating rate bottomed out at the end of 2001, about 6
months before the bottoming out of price. The operating rate then
rose in fits and starts to above 95% by early 2004. Price rose
accordingly, lagging the increase in the operating rate by several
months. As will be discussed below, Weaver describes a 95% operating
rate as a full capacity rate for the industry.
Weaver said that demand and supply have more or less been in
balance since 2003.\85\ He said that manufacturers have been able to
improve pricing significantly over this period [as is clearly
depicted in Slide 9].
---------------------------------------------------------------------------
\85\ Weaver's remarks on Slides 9 and 10 begin at about 5:31
into the copy of the audio recording that we have provided to DOJ.
---------------------------------------------------------------------------
Weaver said that the industry had been at a 95%+ operating rate
for past 2 years and since mill inventories were declining, a 95%
operating rate is ``for all intents and purposes the full operating
rate.\86\ We can't really make
[[Page 32889]]
any more tonnes than we are making now. I am talking about the
industry there.''
---------------------------------------------------------------------------
\86\ Full operating capacity is usually considered to be 98% of
theoretical full capacity. How can 95% be full operating capacity as
Weaver stated? Newsprint operating rates are calculated by the PPPC.
If Abitibi indefinitely idles a machine in order to maintain the
maximum practical industry operating rate, that machine is still
counted as available capacity by the PPPC even though the machine
has been strategically idled. If the Abitibi newsprint machine
remains idled for a long enough period of time the PPPC will
eventually remove that capacity from its capacity forecasts and
Flash Reports. At the time Weaver spoke to the Citigroup Conference
in December 2006, Abitibi and Bowater had each indefinitely idled
one newsprint machine. In addition, Stora Enso's newsprint machine
had been shut down for almost a year due to labor and energy
problems. If the capacity of these three machines were not included
in the calculation of industry operating rates, the industry would
be operating at 98% of total capacity. The Stora Enso machine was
re-started at about the time Weaver was giving his presentation at
the December 2006 Citigroup Conference.
There is also a distinction between market-related downtime and
the strategic idling of capacity. If a relatively small newsprint
producer takes market-related downtime, it is because the producer
does not have enough orders to keep operating. It is likely that the
producer intends to restart the machine as soon as it can book
enough orders, perhaps through offers of discounts. With Abitibi and
Bowater, the motivation is generally, though not always, different.
[Both Abitibi and Bowater have taken market-related downtime since
2002.] Their goal is maximum operating rates. They are using the
indefinite idling of capacity as a lever to raise prices.
We are unaware of any Abitibi or Bowater indefinitely idled
newsprint capacity that has been restarted. The capacity has either
been shut down or has remained indefinitely idled. The subject of
determining the ``real'' operating rate as opposed to the PPPC
official operating rate is discussed further in Section H.3.c.
below.
---------------------------------------------------------------------------
Slide 10 below shows the newsprint industry supply/demand
balance from January 2004 through September 2006.
[GRAPHIC] [TIFF OMITTED] TN10JN08.029
Demand (quantity purchased) is defined as NA consumption plus
net exports. Referring to Slide 10, Weaver said ``month after month
production is equal to consumption'' and since mill inventories are
flat or trending down, ``there is no excess capacity in the
marketplace today. It [i.e., production] is all being consumed.''
3. Presentation by David Paterson, President and CEO of Bowater, at
the Citigroup Conference in December 2006
David Paterson of Bowater, also made a presentation at the
Citigroup Conference on December 7, 2006. The format was similar to
Weaver's presentation.\87\
---------------------------------------------------------------------------
\87\ The slide show is available on Bowater's Web site under
Investor Relations/Presentations. According to Bowater's Web site,
the audio recording of Paterson's comments at the Citigroup
Conference is no longer available on the Web site.
---------------------------------------------------------------------------
The note at the bottom of Slide 13 of Paterson's presentation
says ``Balanced newsprint capacity & demand.'' See Slide 13 below.
The slide plots the quantity of NA demand and supply over the period
2000 to 2006. The slide shows similar downward slopes over time for
both demand and supply. Paterson said ``North American demand.
That's not the slope you want clearly but the industry has responded
fairly aggressively.'' \88\ He said that ``I think that the real
challenge is that if that slope continues at the rate it is in the
fourth quarter, clearly actions will need to be taken.'' He said
that Bowater has removed 300,000 metric tonnes of newsprint capacity
(or more than 10% of Bowater's total capacity) in 2006 from the NA
market. The capacity removals were accomplished by a machine
conversion at Bowater's Calhoun, TN mill to uncoated groundwood
specialty grades (150,000 metric tonnes) and by a shut down of PM
4 at Bowater's Thunder Bay, ON mill (150,000 metric
tonnes). He said that Bowater also took significant downtime on PM
5 at Thunder Bay in the fall.\89\ ``We are taking action,''
he said.
---------------------------------------------------------------------------
\88\ Paterson's remarks on Slide 13 begin at about 10:44 of the
copy of the audio recording we have provided to DOJ.
\89\ PM 5 at Thunder Bay was only temporarily idled and
has been restarted. PM 4 at Thunder Bay has been
indefinitely idle. If it is restarted it is unlikely that it will be
producing newsprint according to news reports.
---------------------------------------------------------------------------
[[Page 32890]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.030
During the Q&A that followed the slide show, Paterson was asked
about maintaining cash flow and dividend payments. Paterson said
that in the near term, newsprint pricing is stable but that any
significant decline in prices would cause another round of closures,
primarily Canadian assets.\90\ Paterson said that Bowater's U.S.
mills are more efficient than Bowater's Canadian mills. He said if
Bowater just had U.S. mills, the newsprint business would be pretty
good at today's prices. But in Canada, due to energy and currency
issues, age of equipment and other reasons, ``there is not a lot of
margin left in the Canadian assets.''
---------------------------------------------------------------------------
\90\ Paterson's response to the question on how Bowater will
sustain its cash flow begins at about 27:35 of the copy of the audio
recording we have provided to DOJ.
---------------------------------------------------------------------------
Paterson said he thinks about near-term cash management as using
two tools to sustain cash flow.--``One is newsprint pricing and the
ability to manage that and that's critical. I've got two and a half
million tonnes [of capacity], so the math is pretty compelling.
Every $10 bucks, with a company our size, that's $25 million in
revenue that I've got to protect. So that's number one.''
Paterson then elaborated on the second tool that Bowater uses to
sustain near-term cash flow:
``Number two is we have to stay ahead of that curve, that demand
curve that you mentioned to sustain cash flow. So my belief [* * *]
is that we have to move faster to stay ahead of that [demand] curve
to maintain an operating rate that gives us some pricing leverage in
the market and I can do that. We know which our high cost assets are
and we will shut them down hopefully before rather than after price
erosion with any significance. So that's the second tool. Now what
does that do? My spread between best and worst assets is quite
significant. So without doing anything else, I can lower my total
manufacturing costs pretty significantly. I've got to balance that
against--you know these assets are generating cash and we need to
pay down debt and do other things.''
He said that the Bowater Board of Directors is committed to
paying dividends and that the board challenges management to
generate operating cash flow on a sustainable basis to pay dividends
and interest payments.
4. Presentation by John Weaver, President and CEO of Abitibi, at
the Credit Suisse First Boston Investment Analysts Conference in
March 2004
John Weaver gave a presentation at the Credit Suisse First
Boston Credit Global Basics Conference on March 3, 2004 (``Credit
Suisse Conference''). Three consecutive slides presented by Weaver
relate to the closure of Abitibi's capacity in order to raise
industry operating rates and prices.
Slide 13 below is an earlier version of Slide 9 that Weaver
presented at the December 2006 Citigroup Conference. Slide 13 shows
that the price of newsprint lags the NA operating rate by about a
quarter. When the operating rate begins to fall, the newsprint price
will begin to fall several months later. Similarly, when the
operating rate begins to rise, the newsprint price will begin to
rise several months later.
[[Page 32891]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.031
Slide 14 below shows NA monthly newsprint production, capacity
and operating rate from mid-1996 through January 2004.
[[Page 32892]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.032
Note that capacity hit a monthly high of 1,378 metric tonnes in
1998. Between 1998 and 2001, capacity declined by about 5%. Abitibi
began removing newsprint capacity from the newsprint market in 1999
and announced additional newsprint capacity removals in conjunction
with its acquisition of Donohue in April 2000. These capacity
closures are discussed in Section I below. Between the end of 2001
and the end of 2003, an additional 7% of capacity, compared to the
1998 peak, was removed from the market. Some of this capacity
removal was due to the closure of the Garden State mill at the end
of 2001. In addition, several other manufacturers converted small
newsprint machines to other groundwood grades as is discussed in
Sections D.3. and D.4. above. Slide 14 projects additional capacity
reduction in 2004 to bring the total reduction as a percentage of
the 1998 peak to 12.8%. Between the 1998 peak through projected
2004, Abitibi and Bowater accounted for almost 80% of the total
reduction.
Slide 15 below shows that Abitibi removed 977,000 metric tonnes
of capacity from the NA newsprint market in 2003. About 43% of the
removal was due to temporary rotating downtime (i.e., market related
downtime). The remaining 57% of the 2003 capacity removal was due to
the indefinite idling of capacity. Abitibi calculated the 2003
industry operating rate at 87%. This calculation excludes Abitibi's
indefinitely idled capacity from total NA newsprint capacity (i.e.,
the denominator of the operating rate calculation). The exclusion of
Abitibi's indefinitely idled capacity from total NA capacity
indicates that the capacity was withheld from the market for the
strategic purpose of raising the industry operating rate and
increasing the price of newsprint.
[[Page 32893]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.033
Slide 15 also shows Abitibi's projected 2004 capacity removals.
In 2004, Abitibi was projected to remove 1,075,000 metric tonnes of
newsprint capacity from the NA market. Rotating downtime was not
expected to account for any of the capacity removal in 2004. Abitibi
projected that it would achieve its capacity removal in 2004 by
increasing indefinitely idled capacity by 202,000 metric tonnes, by
permanently closing 230,000 metric tonnes of capacity, and by
converting 85,000 metric tonnes of capacity to uncoated groundwood
specialty grades. Slide 15 also shows projected 2004 NA capacity
(excluding indefinitely idled capacity) declining by 498,000 metric
tonnes from 2003. Abitibi's projected increase in capacity removal
in 2004 accounts for all of the projected reduction in total NA
newsprint capacity from 2003.\91\ The 2004 industry operating rate
was projected to rise from 87% in 2003 to 99% in 2004. This
calculation does not include Abitibi's indefinitely idled capacity
in NA capacity. Slide 15 illustrates numerically the key role that
Abitibi's indefinitely idled capacity played in achieving the
projected maximum industry newsprint operating rate in 2004.
---------------------------------------------------------------------------
\91\ In fact, Abitibi's projected increase in capacity removals
between 2003 and 2004 exceeds the projected decline in industry
capacity by 19,000 metric tonnes.
---------------------------------------------------------------------------
5. Interview of John Weaver Titled ``Tighter Supply/Demand Balance
Boosts Newsprint Hike Prospects Says Abitibi's Weaver''
John Weaver, President and CEO of Abitibi, was interviewed by
Will Mies, Editorial Director, Paperloop Information Products. The
interview was published on paperloop.com on February 11, 2004.
The article describes Abitibi's aggressive ``focused downtime''
strategy. While the term ``focused downtime'' strategy is not
explicitly defined in the article, it clearly means that the
newsprint machine or newsprint mill has been indefinitely idled. It
should be noted that none of the mills mentioned in the article
subject to Abitibi's ``focused downtime'' strategy in December 2003
have re-opened. The Port-Alfred, QC and Sheldon, TX mills have been
permanently closed. The Lufkin, TX mill remains indefinitely idled.
Abitibi-Consolidated has been aggressively pursuing a ``focused
downtime'' strategy. On Dec. 14 the company indefinitely idled its
Lufkin, Texas, and Port-Alfred, Que., newsprint mills, extended
downtime at its Sheldon, Texas, mill and permanently shut two
machines at the latter two mills with 230,000 tonnes/yr. of
capacity. As a result, the company began the year with one million
tonnes of newsprint capacity removed from the market--and this
excludes the conversion of the company's Alma, Que., to Equal Offset
paper production later this year. Last year the company took 977,000
tonnes of newsprint downtime and 887,000 tonnes in 2002.
As used by Abitibi, ``focused downtime'' or the indefinite
idling of capacity means that this capacity has been removed from
the market to maintain high newsprint industry operating rates. The
capacity would not be restarted if the effect would be to lower the
operating rate from its current and, presumably, high level.
However, it seems plausible that indefinitely idled capacity would
be restarted if there were sufficient increases in newsprint demand
that the restart would not adversely affect the industry operating
rate. Since demand has been consistently declining in recent years,
none of Abitibi's indefinitely idled machines has been restarted. As
noted above, most have been permanently closed. ``Focused downtime''
or the indefinite idling of capacity should not he confused with
market related downtime. As discussed in Section F.4 above market
related downtime, called ``rotating downtime'' in Slide 15, was a
temporary idling of capacity that would be brought back on line as
demand rebounds to expected levels.
When asked about Abitibi's pricing goal, ``Weaver said that
AbitibiConsolidated's goal is to `return newsprint prices back to
their trend line level' which would eventually bring prices on
standard newsprint up to around $585-595/tonne level.''
Weaver was asked if consolidation is working (i.e., Abitibi's
acquisitions of Stone-Consolidated and Donohue that occurred in 1997
and 2000). His reply was included in the quote below.
The acquisition of Donohue followed the 1997 merger with Stone-
Consolidated; both events were followed by significant capacity
shutdowns, downtime and rationalization. Has all of the money spent
on the vision of consolidation begun to pay off for shareholders?
``There have been a number of signs that consolidation is working,
such as the inventory control we have seen over the past several
years and several supply-driven price increases over the last two
years,'' Weaver said.
[[Page 32894]]
``All of the consolidators have taken out significant cost by
closing their high cost capacity and reconfiguring their
companies,'' \92\ he said. But none of the acquiring companies could
foresee at the time of their acquisitions that they would have to
carry the debt through a three-year economic downcycle, he added.
``When the economy recovers, we will see the real returns from
consolidation.'' (Emphasis added)
---------------------------------------------------------------------------
\92\ This statement can only apply to Abitibi, Bowater and
Catalyst.
---------------------------------------------------------------------------
G. An Analysis of Permanent Newsprint Capacity Reductions Between 2002
and 2006
1. Introduction
Section D.4. above analyzed the permanent capacity reductions
that occurred in the NA newsprint industry between 1995 and 2006.
The analysis showed that of the firms that (a) had net capacity
reductions between 1995 and 2006 and (b) remain in the market today,
Abitibi and Bowater combined accounted for 83.6% of those permanent
capacity reductions. Catalyst accounted for most of the remaining
permanent capacity reductions. The analysis in this section focuses
on permanent newsprint capacity reductions in North America between
2002 and 2006. As documented in Section E.6., newsprint prices rose
an aggregate of 49.0% between the third quarter of 2002 and the
third quarter of 2006. Of the newsprint manufacturers that remain in
the market today, Abitibi and Bowater combined accounted for 89.4%
of the permanent reductions of NA newsprint capacity between the end
of 2002 and the end of 2006. Charts G1 to G4 provide an analysis of
the NA permanent capacity reductions during this period.
2. Chart G1: Shares of NA Newsprint Capacity by Manufacturer 2002
and 2006
Chart G1 below shows the shares of NA newsprint capacity by
manufacturer for 2002 and 2006.\93\ At the end of 2002, NA newsprint
capacity was 15,555,000 metric tonnes and at the end of 2006,
estimated NA newsprint capacity was 12,760,000 metric tonnes.
---------------------------------------------------------------------------
\93\ The Sources for Charts G1 to G4 are as follows: (I) For
estimated 2006 NA newsprint capacity, see Tables C1 and C2 in
Attachment C. (2) The sources for 2002 newsprint capacity are as
follows: (a) Abitibi 2002 Annual Report, p. 28; (b) Bowater 2002
Annual Report, p. 6; (c) for Catalyst, Katahdin Paper, and Irving
Paper, see 2003 capacity shown in PPPC's July 9, 2004 ``Update of
North American Mechanical Printing Papers Capacity Forecast''; (d)
for total 2002 NA newsprint capacity, see ``North American Newsprint
Capacity: Results of PPPC's 2003 Capacity Survey,'' March 3, 2003.
The Abitibi and Bowater annual reports are available on their
respective Web sites. The two PPPC capacity surveys are available on
the PPPC Web site under Press Releases.
[GRAPHIC] [TIFF OMITTED] TN10JN08.034
Chart G1 shows that the combined Abitibi and Bowater NA capacity
share declined from 51.4% to 45.0% between the end of 2002 and the
end of 2006 and that Catalyst's share declined by 0.3%. Including
Katahdin and Irving, the shares of all other NA newsprint
manufacturers increased from 42.8% to 49.6%.\94\ Katahdin and Irving
converted their newsprint capacity to the production of uncoated
groundwood specialty grades in 2005-2006. Excluding Katahdin and
Irving, the shares of all other NA newsprint manufacturers increased
from 41.0% to 49.6% from the end of 2002 to 2006.
---------------------------------------------------------------------------
\94\ At the end of 2006 there were 16 newsprint manufacturers
operating in North America. This total includes the Ponderay
newsprint mill in which Bowater has a 40% ownership-interest. The
category ``All Other NA Manufacturers 2006'' includes 13 firms. See
Tables C1 and C2 in Attachment C for more details.
---------------------------------------------------------------------------
3. Chart G2: Permanent Reduction of NA Newsprint Capacity by
Manufacturer During the Period 2002-2006
Chart G2 below shows the permanent reduction of NA newsprint
capacity by manufacturer during the period 2002 to 2006.
[[Page 32895]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.035
There were 2,795,000 metric tonnes of capacity permanently
removed from the NA newsprint market from the end of 2002 to the end
of 2006. Abitibi and Bowater combined accounted for 2,258,000 metric
tonnes that were permanently removed \95\ and Catalyst accounted for
205,000 metric tonnes. The conversion of the Katahdin and Irving
newsprint capacity to uncoated groundwood specialty grades accounted
for 270,000 metric tonnes of capacity removal. All other NA
newsprint manufacturers accounted for 62,000 metric tonnes of
capacity removal.
---------------------------------------------------------------------------
\95\ The capacity reduction totals for Abitibi and Bowater do
not include the capacity of their newsprint machines that are
currently indefinitely idled. Abitibi has two indefinitely idled
newsprint machines. One machine (PM 2) is at its indefinitely idled
Lufkin, TX mill. It has a capacity of 150,000 metric tonnes and has
been idled since December 2003. The other machine (PM 7) is at
Abitibi's Grand Falls, NL mill. It has a capacity of 60,000 metric
tonnes and has been indefinitely idled since the end of 2005.
Bowater's 4 paper machine at its Thunder Bay, ON mill has
been indefinitely idled since September 2006. It has a capacity of
146,000 metric tonnes.
---------------------------------------------------------------------------
Tembec's closure of a 35,000 metric tonne capacity newsprint
machine at its Kapuskasing, ON mill accounted for more than half of
this total.
4. Chart G3: Percentage of Total NA Permanent Newsprint Capacity
Reduction by Manufacturer During the Period 2002-2006
Chart G3 below shows the percentage of total NA permanent
newsprint capacity reduction by manufacturer during the period 2002
to 2006.
[[Page 32896]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.036
The percentage calculations are based on the capacity reduction
figures shown above in Chart G3. Combined, Abitibi and Bowater
accounted for 80.8% of the permanent capacity removals over this
period and Catalyst accounted for 7.3%. Of manufacturers that remain
in the market today, Abitibi and Bowater combined account for 89.4%
of the total capacity removals and Catalyst accounts for 8.1%. The
two manufacturers who converted their newsprint capacity to uncoated
groundwood specialty grades accounted for 9.7% of the total
permanent capacity reduction. All other NA newsprint manufacturers
accounted for 2.2% of the total capacity removals and 2.5% of the
capacity removals by the manufacturers that remain in the market
today.
5. Chart G4. Permanent Reduction of Newsprint Capacity Over the
Period 2002-2006 as a Percentage of Own 2002 NA Capacity by
Manufacturer
Chart G4 below shows the permanent reduction of NA newsprint
capacity over the period 2002 to 2006 as a percentage of each
manufacturer's own capacity at the end of 2002.
[GRAPHIC] [TIFF OMITTED] TN10JN08.037
[[Page 32897]]
Between the end of 2002 and the end of 2006, NA newsprint
capacity was reduced by 18.0%. Through permanent capacity removals,
Abitibi reduced its own capacity by 30.7% and Bowater reduced its
own capacity by 24.0%. Catalyst also reduced its newsprint capacity
by a significant proportion--22.7%. The other 13 newsprint
manufacturers that remain in the market today reduced their capacity
by a combined 1.0%.
Catalyst is the largest newsprint manufacturer West of the
Rockies. Catalyst's removal of a significant amount of its own
newsprint capacity from the market suggests the possibility of a
relevant West of the Rockies newsprint market and a relevant East of
the Rockies newsprint market. Norske Skog's acquisition of Pacifica
in 2001 may have given it the incentive and ability to shut down
capacity to raise the industry operating rate and increase prices in
a West of the Rockies market.\96\ If there is a West of the Rockies
relevant newsprint market, Catalyst may have been playing the same
role in a West of the Rockies market as Abitibi and Bowater were
playing in an East of the Rockies market (i.e., shut down capacity
to raise the industry operating rate and increase prices). All of
Abitibi's and Bowater's capacity reductions have occurred in mills
located East of the Rockies. Bowater has no mills West of the
Rockies and Abitibi has only a limited newsprint manufacturing
presence West of the Rockies.
---------------------------------------------------------------------------
\96\ See Section D.1. above for more details. The Norske Skog
and Pacifica newsprint mills were all located in British Columbia.
Norske Skog's Canadian newsprint assets were renamed Norske Canada
after the Pacifica acquisition and then renamed Catalyst in 2005.
Norske Skog sold its interest in Catalyst in 2006.
---------------------------------------------------------------------------
H. Four Articles by Two Newsprint Industry Experts Describing the
AbitibiBowater Strategy to Raise Price by Closing Capacity
1. Introduction
Four articles by two newsprint industry experts are cited in
this section describing the strategy of Abitibi and Bowater to raise
the price of newsprint through the closure of capacity. The first
article does not specifically identify Abitibi and Bowater, but the
events described can only apply to Abitibi, Bowater and, possibly,
Catalyst. The four articles are evidence that the Abitibi-Bowater
strategy is well understood throughout the newsprint industry by
buyers and sellers alike. The four articles also provide
confirmation of our analysis in this White Paper.
2. Article by Harold M. Cody Titled ``New Paradigm: Newsprint
Demand Falls, Prices Soar.''
Harold M. Cody, Contributing Editor to Paper Age, published an
article in the May/June 2006 edition of Paper Age titled ``New
Paradigm: Newsprint Demand Falls, Prices Soar.'' The following
passage confirms how newsprint industry consolidation has permitted
unnamed manufacturers to strategically shut down capacity to raise
newsprint prices despite a ``steady five year decline in demand.''
North American newsprint consumption continued its steady five-
year decline last year and newspaper publishers faced similar
difficulties. In early 2006, demand continued to drop at an
accelerating rate. But producers continue to fight the fight as
evidenced by the almost hard-to-believe fact that prices are now
reaching the highest levels in five years in spite of all this.
Continuing the boxing parallel, these prolonged tribulations
clearly illustrate just how adept U.S. and Canadian newsprint
producers really are at fighting. They have been able to quickly and
decisively cut supply in response to these challenging conditions,
masterfully reducing capacity via either shutdowns or conversions to
other grades.
The closure of 3.5 million metric tpy of newsprint capacity
since 2001 has kept operating rates for the most part above 95%,
fueling the steady increase in prices from a bottom of about $475/
mton in 2002 to more than $650/mton or higher on lightweight grades
by early 2006. Consolidation has also had an impact, as the top five
newsprint producers control nearly 75% of capacity, and maybe even
more importantly, the top three hold more than 50%. (Emphasis
added.)
Cody notes that, despite the continual decline in newsprint
demand, ``[t]hey have been able to quickly and decisively cut supply
in response to these challenging conditions, masterfully reducing
capacity via either shutdowns or conversions to other grades.'' Cody
does not identify who ``they'' are, but his description of events
can only apply to Abitibi, Bowater, and, possibly, Catalyst. He says
that the capacity reductions have ``kept operating rates for the
most part above 95%, fueling the steady increase in prices.''
3. Three Articles by RISI Senior Economist Andrew Battista
Analyzing the Strategy of Abitibi and Bowater to Shut Down Capacity
to Maintain High Operating Rates and Increasing Prices
a. ``Will operating rates climb high enough in 2003 to support rising
newsprint prices in the U.S.?'' (February 20, 2003)
Andrew Battista, senior economist at RISI, published an article
\97\ in February 2003 titled ``Will operating rates climb high
enough in 2003 to support rising newsprint prices in the U.S.'' This
was the first of three articles Battista wrote over a two year
period analyzing the unfolding AbitibiBowater strategy to use their
control over capacity to raise the price of newsprint.
---------------------------------------------------------------------------
\97\ Source: paperloop.com, February 20, 2003. RISI is the major
NA and global source of data, information, news, and analysis on the
pulp, paper, and forest products industries.
---------------------------------------------------------------------------
At the time Battista wrote this article, newsprint prices were
just starting to increase after the 28% decline in newsprint prices
between the second quarter of 2001 and the second and third quarters
of 2002, caused primarily by the U.S. recession that began in late
2000/early 2001 and the economic aftermath of 9/11.
Producers finally got the ball moving in the other direction
with a $35/tonne rise (of the proposed $50/tonne) last autumn.
Newsprint manufacturers hope to capitalize on this momentum and push
hard for the next $50/tonne increase announced for March 1.
Battista describes the economic relationships between production
costs, operating rates and the price of newsprint.
There are two predominant drivers of product prices: Production
costs and operating rates. Both are highly and positively correlated
with newsprint prices through mechanisms that are well understood.
When production costs inflate, newsprint profit margins fall. Buyers
may balk at paying more for newsprint when ONP [recycled old
newspapers] gets more expensive, but as cost pressure mounts, the
least competitive mills edge closer to shutdown unless newsprint
prices also rise.
The closure of a mill will result in higher operating rates.
Likewise, a rise in demand usually leads to a tighter market (higher
operating rates) in which paper becomes increasingly scarce, and
hence, more valuable.
* * * * *
Rising costs support higher prices, but do not guarantee them in
the short term. We still need to forecast the supply/demand balance
in order to get a handle on pricing.
Battista provides an analysis of the relationship between
operating rates and changes in newsprint prices.
When we plot operating rates against the (quarter-to-quarter)
percent change in prices (as in Figure 1), we clearly see a high
degree of correlation between the two series.\98\
---------------------------------------------------------------------------
\98\ Note that Slide 9 contained in John Weaver's presentation
to the Citigroup Conference in December 7, 2006 is a close variation
of Battista's Figure 1. In presentations to investment analyst
conferences by John Weaver and Pierre Rougeau, a close variation of
the Battista figure is included in all or almost all such
presentations beginning with Weaver's presentation to the June 5,
2003 Scotia Capital Materials Conference. The Scotia investment
analysts conference was held a little bit more than three months
after the Battista article was published. A similar slide is
included in the most recent Abitibi presentation on March 20, 2007,
which was by Rougeau, who is Abitibi's Senior Vice-President for
Corporate Development and CFO.
---------------------------------------------------------------------------
[[Page 32898]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.038
Furthermore, we observe that the goodness of fit in this
relationship is best with a one-quarter lag on operating rates. This
fact reinforces the hypothesis of a causal relationship; higher
operating rates lead to higher prices. In other words, a tight
market in the summer tends to yield higher prices in the autumn. But
how tight is ``tight''?
Closer examination of Figure 1 shows us that sustained operating
rates in excess of 95% are typically required to lift newsprint
prices.
Battista then analyzes the newsprint price increase that had
occurred since the market hit bottom in mid-2002 and the prospects
for further price increases in 2003 and 2004.
Last autumn's increase stands as an exception to that rule [that
sustained operating rates in excess of 95% are required to lift
newsprint prices]. The oddly timed price hike led publishers to
complain that the market fundamentals did not justify an increase
and forced producers to argue that they needed a rise just to stay
alive.
But the massive market downtime taken by producers held
inventory levels in check and led to a compromise increase (buyers
accepted $35/tonne of the proposed $50/tonne). And although market
recovery seems to be on hold during the winter months, with
operating rates hovering between 92% and 93%, signs point to a
tighter market in 2003. Abitibi-Consolidated Inc. and Bowater Inc.
recently announced plans to withdraw 270,000 tonnes of combined
capacity at Alma, Que., and Calhoun, Tenn.
In addition, ad lineage will likely continue along a gradual
growth path and support a steady rise in newsprint demand. These
factors should push operating rates above 94% this spring and summer
before cresting [at] 95% toward the end of 2003.
Therefore, rising ONP costs and the threat of additional mill
shutdowns may spur some positive pricing momentum this spring and
once again, a portion of the $50/tonne sought on March 1 may be
accepted. Continued market discipline through downtime will support
prices a bit by keeping mill inventories low, but downtime does not
affect the market as powerfully as the permanent removal of
capacity.
North American newsprint producers will struggle to get prices
to crest [at] $500/tonne by the fourth quarter of this year because
operating rates will struggle to get above the 95% threshold in time
to have much impact.
No new capacity will come online in North America in 2004, and
we forecast newspaper advertising lineage growth to accelerate.
Operating rates will likely top 97% for the year next year, and cost
pressure probably will not subside. [Emphasis added]
b. ``Is rising newsprint demand necessary to support higher prices in
2004?'' (December 11, 2003)
Battista followed up his February 2003 article with an article
\99\ published in December 2003 titled ``Is rising newsprint demand
necessary to support higher prices in 2004?'' His answer is that
capacity closures will be sufficient to cause rising prices. He
describes the removal of significant amounts of newsprint capacity
from the market. The only capacity closures and conversions he
describes are by Abitibi and Bowater. Like Weaver and Paterson in
Section F above, Battista describes industry efforts to restore
``balance'' between supply and demand and forecasts the likelihood
of a price increase, as the following excerpt indicates.
---------------------------------------------------------------------------
\99\ Source: paperloop.com, December 11, 2003.
---------------------------------------------------------------------------
Just yesterday, Abitibi-Consolidated announced its intention to
idle, or keep idle, its mills at Sheldon, Lufkin, and Port-Alfred.
Over 750,000 metric tonnes per year (mtpy) will be indefinitely
removed from the market. Perhaps more importantly, though, the
company will permanently shut down two machines, one in Port-Alfred
and one in Sheldon. This latter action will remove 230,000 mtpy from
the North American newsprint market, permanently. Furthermore,
closures and conversions at Abitibi-Consolidated's mill at Alma and
Bowater's mills at Calhoun and Catawba in addition to any market-
related downtime taken next year by anyone will further exacerbate
the 7-year downward trend in North American newsprint supply. The
point is that producers' efforts to reconcile supply with demand
have come a long way toward restoring balance in the market. A
strong rebound in demand next year would undoubtedly spark a sharp
rise in newsprint prices, but as capacity continues to fall, prices
could jump even without a recovery in newsprint consumption.
(Emphasis added)
The extremely tight market for newsprint in 2000 pushed the
average transaction price over $600/tonne by the end of the year.
Several successive years of approximately 2% annual gains in demand
against virtually flat supply led to extraordinarily high operating
rates (near 100%) in the autumn of 2000. However, the turnaround in
2001 proved to be bitterly sharp for newspapers and newsprint
manufacturers, alike. In the three years since, flailing newspaper
advertising lineage pulled North American newsprint demand down by
over 12% or approximately 1.4 million tonnes on an annual basis.
Mills struggled and eventually succeeded in matching the
declines in demand with permanent closures and downtime. True
operating rates (which count temporarily idled capacity as if it
were available capacity) stayed below 90% throughout 2003, and we
further know that production corresponded with demand during 2002-
2003 because producer inventories remained low. This producer
discipline had its first impact last summer when it effectively
stopped the year-and-a-half long slide in prices, and has since
permitted three partially successful increases (thanks also to
rising production costs and the Canadian dollar).
If we now include Abitibi-Consolidated's latest permanent cuts
to the announced list of newsprint capacity withdrawals, we see that
the drop in North American newsprint
[[Page 32899]]
supply over the last three years amounts to nearly 1.3 million mtpy.
This reduction nearly matches the aforementioned (1.4 million tonne)
drop in domestic demand over the same period. If domestic shipments
or exports improve at all next year over the four levels endured
during the second half of 2003, the industry operating rate will
move to between 93% and 95% for the year. We predict that a moderate
rise in both demand and exports will cause the gap between shipments
and practical capacity (98% of theoretical capacity) to vanish, just
as it did during the tight market of 2000 (see Figure 1). Thus,
operating rates could top 97% in late 2004 not adjusting for any
ongoing downtime.
[GRAPHIC] [TIFF OMITTED] TN10JN08.039
What then will happen to newsprint prices in 2004? Given that,
in all likelihood, the North American operating rate in newsprint
will climb above 95% sometime in 2004 perhaps as early as the
spring--prices will surely rise. When we plot operating rates
against the (quarter-to-quarter) percent change in prices (as in
Figure 2), we clearly see a high degree of correlation between the
two series. Furthermore, we observe that the goodness-of-fit in this
relationship is best with a one-quarter lag on operating rates. This
fact reinforces the hypothesis of a causal relationship; higher
operating rates lead to higher prices. In other words, a tight
market in the summer tends to yield higher prices in the autumn. But
how tight is ``tight''?
[GRAPHIC] [TIFF OMITTED] TN10JN08.040
Closer examination of Figure 2 shows us that sustained operating
rates in excess of 95% are typically required to lift newsprint
prices. The half-successful increases since last summer provide a
very noteworthy exception, but are attributable to the massive
[[Page 32900]]
downtime and rising production costs borne by North American
newsprint mills over the period. Therefore, should downtime continue
to be taken through 2004 as the true industry operating rate crests
95%, paper will be extremely scarce even though demand may be not
much higher than during 2003. The average transaction price for
newsprint might not get above $600/tonne next year, but this latest
move by Abitibi-Consolidated brings the supply-and-demand balance
much closer to where it stood 3 years ago, when newsprint last
topped $600/tonne. (Emphasis added)
c. ``Newsprint producers must rely on supply reductions to support
rising prices'' (October 14, 2004)
In October 2004, Battista wrote a third article on the use of
reductions and downtime of newsprint capacity to raise the price of
newsprint.\100\ The article was titled ``Newsprint producers must
rely on supply reductions to support rising prices.'' By this time
it had become clear to Battista that increases in demand were likely
to be anemic at best, and that higher newsprint prices would come
about as a result of the manufacturers' ``zeal'' in further reducing
capacity.
---------------------------------------------------------------------------
\100\ Source: paperloop.com, October 14, 2004.
---------------------------------------------------------------------------
Last year, in the RISI Viewpoint, I wrote that rising newsprint
demand would not be necessary to support higher North American
newsprint prices in 2004. Over the first eight months of the year,
U.S. demand is off 0.8%, and Canadian demand is down 2.0% from 2003.
And yet, average prices climbed $30/tonne higher this spring and are
in the midst of another bitterly fought $50/tonne hike that could
take them above $575/tonne before the end of the year.
After three consecutive years of declines in newsprint demand,
seasonally adjusted U.S. consumption among all users is finally
showing marginal improvement on a quarterly basis. The year-over-
year figures will probably show some growth in the current quarter
if only because the market during 4Q03 was so weak. And even though
we expect to see solid, 3%, expansion in North American GDP in 2005,
print advertising and newspaper circulation will likely continue to
underperform and, at best, yield a meager 0.8% gain in domestic
newsprint consumption. Nevertheless, we foresee U.S. newsprint
prices climbing above $600/tonne in 2005 owing to producers' ongoing
zeal to match the declining market with supply reductions.
Battista then discusses the removal of idled Abitibi and Bowater
newsprint capacity from the official PPPC total. His discussion
illustrates why it is misleading to rely on official PPPC capacity
numbers to calculate operating rates. Based on these misleading
capacity numbers, the official PPPC newsprint operating rate was
92%. In reality, the ``real'' operating rates were 98% to 99% which
explains the sustained rise in newsprint prices from the end of 2002
through the time the article was written. According to Battista, the
capacity the PPPC had removed from its official total a few weeks
before his article was published raised the official operating rate
to over 95% but still below the ``real'' operating rate of 98% to
99%. Battista anticipated that the PPPC would remove additional
capacity from the official total in the first quarter of 2005, which
would then align the official operating rate with the ``true''
operating rate. Note that with one minor exception,\101\ Abitibi and
Bowater account for all of the capacity removals in 2004 and 2005
that are discussed by Battista.
---------------------------------------------------------------------------
\101\ Tembec closed paper machine 1 at its Kapuskasing,
ON mill. The machine had a newsprint capacity of 35,000 metric
tonnes.
---------------------------------------------------------------------------
Several weeks ago, the PPPC officially removed some idled
capacity that had been inoperative for more than one year: Bowater's
PM3 at Thunder Bay, and Abitibi's PM5 and PM7 at Sheldon. The move
suddenly took 480,000 tpy from the North American capacity base and
lifted operating rates by more than 3% to over 95%. Furthermore,
over the next two to three months, several more idled machines will
have to come out of the official numbers. Abitibi's remaining
machines at La Baie (Port Alfred) and PM2 at Lufkin were officially
idled last December and account for approximately 430,000 tonnes of
annual capacity. Also, accounting for Tembec's idled PM1 at
Kapuskasing will pull an additional 35,000 tpy in early 2005.
The supply reductions in 2005 could run deeper still. Abitibi
may soon announce the conversion of yet another newsprint machine to
Alternative Offset/Equal Offset. The company has high expectations
for this growing market. Such a conversion would probably be in
addition to possible permanent closures at Sheldon and La Baie. (The
PPPC reporting change temporarily removes those machines from the
books, but Abitibi is rumored to be considering permanent shutdowns
at these sites.) Bowater is also expected to make aggressive moves
out of newsprint in the year ahead, although no details have yet
been made public.
The forthcoming PPPC cuts will effectively boost the North
American newsprint operating rate to 98%-99% in the first quarter of
2005. If another machine or two were to stop manufacturing
newsprint, the market would be as tight as the white-hot market in
2000 and paper would be extremely hard to find. Prices next year
will almost certainly rise even if demand fails to show any
improvement at all.
Battista next discusses, as he did in his two previous articles,
the relation between operating rates and price changes and he
forecasts high operating rates for 2005. Also, as he did before, he
includes a figure plotting NA newsprint operating rates against
changes in price with one adjustment. In the figure below, Battista
adjusts the operating rate for ``downtime,'' presumably to reflect
the ``true'' operating rate rather than the PPPC official operating
rate. The comparable figure that was included in his December 2003
article above reflects the PPPC official operating rate. The figure
shows the PPPC official operating rate bottoming out at 84% at the
end of 2001 and then rising to about 90% by the third quarter of
2002 before leveling out at or slightly below 90% through the third
quarter of 2003. The figure below, which adjusts for downtime, shows
the ``real'' operating rate bottoming out at perhaps 89% at the end
of 2001 and then rising very quickly to above 95% by mid-2002 and
generally remaining at that level or above through the third quarter
of 2004.\102\
---------------------------------------------------------------------------
\102\ The figure shows a dip in the ``real'' operating rate
below 95% to 94% in the second quarter of 2004 before returning
above 95% in the third quarter of 2004. Slide 15 discussed in
Section F.4. above shows that Abitibi believed that the ``true''
operating rate was 87% in 2003 but that it would rise to 99% in 2004
due almost entirely to additional capacity removals by Abitibi.
---------------------------------------------------------------------------
Historically speaking, when the North American operating rate
climbs above 95% for two or more consecutive quarters, prices rise.
This relationship exhibits a very tight correlation and makes good
intuitive sense as well. Newsprint prices inflate when either demand
jumps or supply falls such that the market is tighter than average.
As noted above, the current operating rate is slightly higher than
95%, which means--in conjunction with rising ONP costs and a strong
Canadian dollar--the current price increase ought to be moderately
successful. Indeed, despite the fact that some suppliers have opted
to delay implementation to October 1, other mills tell us that their
order books are full through the balance of 2004.
Looking ahead, to 2005, it seems highly unlikely that operating
rates will dip below 95%. The tiny projected gains in demand may
fail to materialize, but falling capacity will lift the newsprint
industry's utilization rate. Moreover, ongoing ONP inflation and
persistent appreciation of the Canadian dollar will further induce
producers to push for higher newsprint prices next year. The rise of
the loonie, since the end of 2002, effectively wiped out all of the
newsprint pricing gains for Canadian mills, and we expect the
Canadian dollar to appreciate further over the next several months.
Because of all of these factors, average pricing will consequently
crest the $600/tonne threshold by next spring, and could get a
second boost in the autumn. The size of a second increase in 2005
and the ease of its acceptance, of course, will depend on: (1)
Whether leading producers shutter more capacity, and (2) demand not
evaporating as it did in 2001. (Emphasis added)
[[Page 32901]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.041
I. Abitibi's Newsprint Capacity Closures 1999 to 2001
This section briefly reviews Abitibi's newsprint capacity
closures between 1999 and 2001 and their likely impact on newsprint
operating rates and prices.\103\
---------------------------------------------------------------------------
\103\ On behalf of the NAA and U.S. daily newspaper publishers,
Economists Incorporated submitted to DOJ analyses of the likely
competitive effects of the proposed acquisition of Donohue by
Abitibi in 2000 and the proposed acquisition of Alliance by Bowater
in 2001. Those analyses are still relevant to an understanding of
the competitive conditions in the newsprint industry at that time as
well as an understanding of the likely competitive effects of the
currently proposed Abitibi-Bowater merger. There were two
submissions to DOJ concerning the proposed acquisition of Donohue by
Bowater. They are dated March 1, 2000 and March 31, 2000. The
submission to DOJ concerning the proposed acquisition of Alliance by
Bowater is dated May 7, 2001.
---------------------------------------------------------------------------
According to the Abitibi 1999 Annual Report (p. 6), Abitibi
removed 450,000 metric tonnes of newsprint capacity from the market
in 1999 almost 3% of NA capacity.
``We want to fully implement our capacity rationalization
program in 1999, and together with the planned newsprint conversion
next year, you'll see us close or convert 350,000 tonnes ''--John
Weaver, 1998 Annual Report
In fact, Abitibi-Consolidated permanently removed 450,000 tonnes
of excess newsprint capacity in 1999, or nearly 3% of NorthAmerican
capacity. We will continue to be a results-driven Company that
benchmarks objectives and accomplishes them.
According to the Abitibi 2000 Annual Report, (p. 23), Abitibi
announced in conjunction with its acquisition of Donohue in April
2000 that Abitibi would remove an additional 400,000 metric tonnes
of newsprint capacity from the market during 2000 and 2001.
High-cost newsprint capacity rationalization program. In
conjunction with the acquisition of Donohue, the Company announced
its intention to permanently remove 400,000 tonnes of high-cost
newsprint capacity. As part of this program, the Company shut down
its 130,000 tornne West Tacoma newsprint mill, located in
Steilacoom, Washington, in December 2000.
One paper machine with an annual capacity of 70,000 tonnes was
shut down at the Lufkin, Texas mill, on November 1st, 2000 as part
of the modernization program of the mill. At the end of December
2000, the Company shut down a value-added paper machine with an
annual capacity of 45,000 tonnes at the Kenogami, Quebec mill. The
value-added groundwood paper grades produced on these machines will
replace newsprint production at other mills.
Abitibi closed 200,000 metric tonnes of newsprint capacity in
2000 and 200,000 metric tonnes in 2001 for a total removal of
850,000 metric tonnes of newsprint capacity over the three year
period or about 5% of NA newsprint capacity that existed at the
beginning of 1999.
Abitibi's removal of 450,000 metric tonnes of newsprint capacity
in 1999 raised the industry operating rate by almost 3%. In Section
E.6., we noted that newsprint prices increased $145 or 30.2% between
the third quarter of 1999 and the second quarter of 2001. As Chart
E5 in Section E shows, the operating rate increased from 93.0% in
the second quarter of 1999 to 97.7% in the fourth quarter of 1999,
and, except for one quarter, remained above 97% through the end of
2000. Without the newsprint capacity removals of Abitibi during
1999, the industry operating rate would have been at 95% or somewhat
below during the period 4Q 1999 to 2Q 2001. While prices may still
have increased at these lower operating rates, the magnitude of the
price increases would likely have been significantly lower than what
actually occurred.
The ``Pulp & Paper North American 2000 Factbook,'' p. 194,
summarizes the effect of Abitibi's capacity closures on the three
$50 per metric tonne price increases that occurred between September
1999 and September 2000. The Factbook does not identify any other
manufacturers that closed capacity from the market during this
period.
Adding to market tightness and lending support to the price
increases was Abitibi-Consolidated's vow to remove 400,000 mtons of
newsprint from the North American market by 2001. In July 2000,
Abitibi announced the closure of its 130,000 mtpy West Tacoma,
Wash., newsprint mill at year-end. The company had already idled the
No. 2 paper machine at the mill in 1999. Also in 1999, Abitibi idled
the No. 7 paper machine at Iroquois Falls, Ont. (24,000 mtpy of
newsprint). In addition, Abitibi idled and then subsequently sold
its 125,000 mtpy Chandler, Que., mill with the condition that the
new owners not produce newsprint.
The Factbook excerpt above notes that Abitibi's Chandler, QC
newsprint mill was sold with the condition that the new owners not
make newsprint. Abitibi closed the Chandler mill in 1999 and sold it
in 2000. The condition that the Chandler mill not be used by the new
owners to produce newsprint suggests that the mill's variable costs
for producing newsprint were below prevailing newsprint prices at
the time and that it would have been profitable for the new owners
to use the mill to produce newsprint.
J. A Comparison of Newsprint Prices With the Prices of Uncoated
Groundwood Specialty Grades 3Q 1999 to 4Q 2006
1. Introduction
In Section B above we described the similarities and differences
between newsprint and uncoated groundwood specialty grades. The
higher value uncoated groundwood grades generally are brighter than
newsprint (i.e., the fibers in the pulp furnish have been subjected
to more bleach) or glossier (i.e., clay is added to the pulp
furnish). While newsprint is the lowest-quality and lowest value
groundwood grade, the main inputs used to produce newsprint and
uncoated groundwood specialty grades, in particular energy and
fiber, are the same. Rises in common input costs should have a very
similar impact on both NA newsprint mills and NA mills that produce
uncoated groundwood specialty grades, other things being equal.
In Section J.2. below we explain why the impact of the increase
in input prices over
[[Page 32902]]
the past several years has been greater on Canadian mills than U.S.
mills. In addition, the appreciation of the Canadian dollar to the
U.S. dollar has also adversely affected Canadian mills compared to
U.S. mills. We explain why these twin effects fall more heavily on
NA manufacturers of uncoated groundwood paper in the aggregate than
on NA manufacturers of newsprint in the aggregate.
In Section J.3. we compare the quarterly price of newsprint from
the third quarter of 1999 to the second quarter of 2006 with the
quarterly prices of four uncoated groundwood specialty grades. We
find that the quarterly prices for newsprint as a percentage of its
quarterly price in 3Q 1999 were significantly higher than the
quarterly prices for three of the four uncoated groundwood specialty
grades over the period 4Q 1999 to 2Q 2006. Based on these results,
it is implausible that the increases in newsprint prices were caused
by the increases in input prices. We find that the price trend of
one uncoated groundwood specialty grade was similar to that of
newsprint. It appears that Abitibi and Bowater are the dominant
providers of that grade as well.
Section J.4. presents evidence that Abitibi's variable costs
have been relatively constant since 2001. Since nearly all of the
newsprint price increases over the period 2002 to 2006 were led by
Abitibi, it seems unlikely that increases in Abitibi's input costs
are a plausible justification for the price increases.
In Section J.5. we calculate quarterly newsprint revenues over
the period 3Q 1999 to 2Q 2006 based on actual NA newsprint demand
and actual newsprint prices. We then apply the quarter to quarter
percentage price changes for each of the four uncoated groundwood
specialty grades to the 3Q 1999 newsprint price and multiply the
resulting adjusted newsprint prices by actual NA demand. For the
three grades with percentage changes in prices significantly below
the percentage changes in newsprint prices, total revenues over the
period are reduced by $4.7 billion to $7.4 billion.
2. The Adverse Impact of the Increases in Input Prices and the
Appreciation of the Canadian Dollar Has Fallen More Heavily on
Producers of Uncoated Groundwood Specialty Grades Than on Producers
of Newsprint
Both newsprint producers and producers of uncoated groundwood
specialty grades have been subjected to increasing costs of inputs
in recent years. The inputs that have increased in cost include
fiber (both wood and recycled), energy and transportation.
Advantages that Canadian mills once enjoyed in lower energy and
fiber costs have been reversed.\104\ Canadian mills are now at a
cost disadvantage.
---------------------------------------------------------------------------
\104\ Source: ``Global Pulp & Paper Fact & Price Book 2006,''
pp. 149-152, which is published by RISI. (``RISI Fact & Price
Book''). While the discussion of the increasing costs and declining
fortunes faced by Canadian newsprint mills is in the newsprint
section of the RISI publication, the discussion clearly would also
apply to Canadian mills that produce uncoated groundwood specialty
grades.
---------------------------------------------------------------------------
At the December 2006 Citigroup Conference, David Paterson of
Bowater stated that Bowater's U.S. mills (which are all in the
southeastern U.S. with the exception of the Ponderay mill in
Washington) were more efficient than Bowater's Canadian mills (which
are all in Eastern Canada). He said that due to energy and currency
issues (discussed immediately below), the age of the equipment and
other reasons, there is not much margin left at Bowater's Canadian
newsprint mills. He also said that if Bowater had only U.S. mills,
the Bowater's newsprint business would be pretty good at ``today's''
prices.\105\
---------------------------------------------------------------------------
\105\ Source: Audio recording of Paterson's comments at the
December 2006 Citigroup Conference, starting at about 27:35. We have
provided a copy of this audio recording to DOJ. The recording is no
longer available on the Abitibi Web site.
---------------------------------------------------------------------------
In addition to increases in the cost of inputs, Canadian mills
have been adversely affected by a significant increase in the value
of the Canadian dollar relative to the U.S. dollar.\106\ For a given
price increase, U.S. mills will benefit more than Canadian mills if
the value of the Canadian dollar is rising relative to the U.S.
dollar. The combined effects of the input cost increases and the
increasing value of the Canadian dollar have reduced the
profitability of Canadian mills relative to U.S. mills.
---------------------------------------------------------------------------
\106\ Newsprint is priced in U.S. dollars per metric tonne but
the costs to the Canadian mill of producing a metric tonne of
newsprint are denominated in Canadian dollars. If the value of the
Canadian dollar increases relative to the U.S. dollar, the Canadian
mill will receive fewer Canadian dollars from the sale of a metric
tonne of newsprint to a U.S. customer when the U.S. dollars from the
sale are converted to Canadian dollars.
---------------------------------------------------------------------------
A greater percentage of NA uncoated groundwood capacity is in
Canada compared to the percentage of NA newsprint capacity in
Canada.\107\ In addition, Canadian uncoated groundwood specialty
mills ship a greater percentage of their output to U.S. customers
than the percentage of output that Canadian newsprint mills ship to
U.S. customers.\108\ As a result, the impact of increases in input
costs and the appreciating Canadian dollar should fall more heavily
on NA uncoated groundwood specialty manufacturers in the aggregate
than on NA newsprint manufacturers in the aggregate.\109\
---------------------------------------------------------------------------
\107\ In 2005, 71.9% of uncoated groundwood specialty grade
capacity was in Canada and 28.1% was in the U.S. By comparison,
61.4% of NA newsprint capacity was in Canada and 38.6% was in the
U.S. Source: RISI Fact & Price Book, pp. 147, 148, and 164.
\108\ In 2005, Canadian manufacturers of uncoated groundwood
specialty grades shipped 76.6% of their output to U.S. customers. In
contrast, Canadian newsprint mills shipped 61.2% of their output to
U.S. customers. Source: RISI Fact & Price Book, pp. 142, 149, and
164.
\109\ About 65.3% of Abitibi's NA newsprint capacity is in
Canada and about 57.1% of Bowater's NA newsprint capacity is in
Canada. For Abitibi and Bowater combined, 62.1% of their NA
newsprint capacity is in Canada. See Table C1 in Attachment C. The
increase in costs at their Canadian newsprint mills implied by the
appreciation of the Canadian dollar is partially offset by the
implied corresponding decrease in costs at the U.S. newsprint mills
of Abitibi and Bowater. After Abitibi and Bowater, the next two
largest newsprint manufacturers in NA in terms of capacity are White
Birch and Kruger. See Table C2 in Attachment C. As can be determined
from Table C1, 79.6% of White Birch's capacity is in Canada and
100.0% of Kruger's capacity is in Canada. The appreciation of the
Canadian dollar has adversely affected White Birch's and Kruger's
manufacturing costs more than it has Abitibi's or Bowater's
manufacturing costs.
---------------------------------------------------------------------------
3. Comparing Quarterly Prices for Newsprint and Uncoated Groundwood
Grades From 3Q 1999 Though 4Q 2006
There are two reasons to assume that price increases over the
period should be greater for uncoated groundwood specialty grades
than for newsprint over the period 3Q 1999 to 4Q 2006. First, the
growth rate in consumption over this period has been positive for
uncoated groundwood specialty grades in the aggregate, while the
growth rate in consumption has been negative for newsprint. Between
1999 and 2006, total NA uncoated groundwood specialty grade
consumption grew at a compound average growth rate of 3.1% per year.
Over that same period, the compound average growth rate of NA
newsprint consumption was a negative 4.0%.\110\ Positive growth
rates in consumption are usually associated with rising prices and
negative growth rates in consumption are usually associated with
falling prices.\111\
---------------------------------------------------------------------------
\110\ Source: RISI Fact & Price Book, p. 142 and p. 169. The
RISI Fact & Price Book does not provide annual consumption data by
uncoated groundwood specialty grade.
\111\ The 3Q 1999 price per metric tonne for each grade was as
follows: newsprint = $480; Directory (22.1 lb.) = $733; Hi-Brite 65
(35 lb.) = $621; SCA (35 lb.) = $717; SCB (35 lb.) = $623. Source:
RISI Fact and Price Book, p. 150 and p. 167.
---------------------------------------------------------------------------
Second, as described in Section J.2. above, the rise in input
costs and the appreciation in the Canadian dollar relative to the
U.S. dollar have fallen more heavily on NA producers of uncoated
groundwood specialty grades in the aggregate than on NA newsprint
manufacturers in the aggregate.
Chart J1 below reflects the quarterly average price of newsprint
and four uncoated specialty grades over the period 3Q 1999 to 2Q
2006.\112\ The prices for each grade are expressed as a percentage
of that grade's price for 3Q 1999. Three $50 per metric tonne price
increases were implemented from September 1999 to September
2001.\113\ 3Q 1999 was selected for the initial date of the analysis
shown in Chart J1, because that was the quarter when the initial $50
price increase was announced. As was described in Section I above,
Abitibi began closing capacity in 1999. The ``Pulp & Paper North
American 2000 Factbook,'' p. 194. cited
[[Page 32903]]
Abitibi's past closures and announced future closures as ``[a]dding
to market tightness and lending support to the price increases.''.
---------------------------------------------------------------------------
\112\ Source: RISI Fact and Price Book, p. 150 and p. 167 and
Pulp & Paper Week. Except for newsprint, the prices are the average
of the high and low prices for each quarter. The uncoated groundwood
specialty grades were priced in short tons. These prices were
converted to price per metric ton by multiplying by the ratio of the
number of pounds in a metric tonne to the number of pounds in a
short ton (2205/2000). The price for Directory paper is a spot
price. About 80% to 90% of Directory paper is sold under one- to
three-year contracts to RBOCs and independent directory publishers.
\113\ Source: ``Pulp & Paper North American 2000 Factbook,'' p.
194.
[GRAPHIC] [TIFF OMITTED] TN10JN08.042
Chart J1 shows in a broad sense similar price movements for
newsprint and the four uncoated specialty grades. For each of these
grades, price rose from 3Q 1999 to 2001, followed by a rapid decline
as the U.S. recession set in. Prices bottomed out in 2002 or so and
began to climb until Q2 2006. However, the magnitudes and rates of
the price movements are quite different for the five grades. The
prices of both newsprint and Hi-Brites (brightness level = 65) rose
significantly more than the other three grades between 3Q 1999 and
2001 and between bottoming out in 2002 and 4Q 2006. Chart J1 shows
prices increasing within a quarter or two of bottoming out for these
two grades. The price of both grades rose steadily from the bottom.
The newsprint price rose to 39% above its 3Q 1999 price by 4Q 2006
and the Hi-Brite price rose to 36% above its 3Q 1999 price by 3Q
2005 before declining somewhat to 32% by 4Q 2006. In terms of
dollars per metric tonne, the newsprint price in 4Q 2006 was $185
above its 3Q 1999 price and the Hi-Brite price was $196 above its 3Q
1999 price.
The Directory, SCA and SCB grades had much smaller price
increases in the run-up to 2001 and, after the decline to 2002, the
recovery in prices took much longer to occur than for newsprint and
the Hi-Brite grade. The bottoms for the SCA and SCB prices were much
deeper as a percentage of their 3Q 1999 prices than was the case for
the bottoms for newsprint and Hi-Brite prices. The prices for the
SCA and SCB grades also stayed at their bottoms for a much longer
period of time than was the case for the prices for newsprint and
the Hi-Brite grade. By 4Q 2006, the SCA price was 1.3% below its 3Q
1999 price and the SCB price was 3.1% above its 3Q 1999 price. In
terms of dollars per metric tonne, the SCA price in 4Q 2006 was $11
below its 3Q 1999 price and the SCB price was $24 above its 3Q 1999
price. The price of Directory paper as a percentage of its 3Q 1999
price did not fall nearly as deeply as did the SCA and SCB prices
and it recovered more quickly. By 4Q 2006, the Directory paper price
was 7.8% or $61 above its 3Q 1999 price.
Why should 4Q 2006 prices for newsprint and Hi-Brites be so much
higher than their 3Q 1999 prices both in percentage terms and as an
absolute change in price compared to SCA, SCB, and Directory paper
prices? One possible answer is that not only are Abitibi and Bowater
dominant in newsprint, they are also dominant in Hi-Brites. During
our interviews with newspaper newsprint buyers, we learned that
there was also concern that the proposed Abitibi-Bowater merger
could lead to higher Hi-Brite prices and Super Hi-Brite prices.\114\
In addition to newsprint, these buyers also purchase these two
uncoated groundwood specialty grades. We were told that Abitibi and
Bowater are the only suppliers of Hi-Brite and Super Hi-Brite grades
East of the Rockies. We were also told by the buyers that they were
unaware of any European suppliers of Hi-Brites or Super Hi-Brites.
---------------------------------------------------------------------------
\114\ The RISI Fact & Price Book does not provide a price series
for Super-Brites.
---------------------------------------------------------------------------
Our analysis of uncoated groundwood specialty grades in
Attachment B confirms the statements of the newspaper newsprint
buyers cited above regarding the availability of Hi-Brite and Super
Hi-Brite suppliers. See Tables B5 and B6 in Attachment B. Besides
Abitibi and Bowater, the only suppliers of Hi-Brites and Super Hi-
Brites in NA that we were able to identify \115\ were Catalyst,
North Pacific, and Blue Heron, all of whose mills are located West
of the Rockies.\116\ In an NA relevant geographic market, Abitibi
and Bowater would have a combined share of 76.5% of capacity based
on our analysis. In an East of the Rockies relevant geographic
market, Abitibi and Bowater would have a combined share of 100.0% of
capacity.
---------------------------------------------------------------------------
\115\ Because information on producers of specific uncoated
groundwood specialty grades is often sketchy, our analysis should be
regarded as a first approximation.
\116\ Neither Abitibi nor Bowater produce Hi-Brite and Super Hi-
Brite grades at mills located West of the Rockies.
---------------------------------------------------------------------------
The price comparisons shown in Chart J1 are not consistent with
a hypothesis that newsprint price increases observed over the past
four years are due to the rising costs of inputs. If the newsprint
price increases were caused by input cost increases, we should at a
minimum see similar price increases for newsprint and the four
uncoated groundwood specialty grades. As argued above, the price
increases should, in fact, be greater for uncoated groundwood
specialty grades than for newsprint since the impact of the cost
increases falls more heavily on uncoated groundwood specialty
producers in the aggregate than it does on newsprint producers in
the aggregate. In addition, the price increases should be greater
for the uncoated groundwood specialty grades because of the steady
demand growth for the specialty grades in contrast to the steady
demand decline for newsprint.
The price comparisons shown in Chart J1 are consistent with the
hypothesis that Abitibi and Bowater have jointly exercised
[[Page 32904]]
significant market power in the NA newsprint market. The price
comparisons shown in Chart J1, the observations of newspaper
newsprint buyers cited above, and our own confirming analysis
strongly suggest that Abitibi and Bowater have also jointly
exercised significant market power in the sale of Hi-Brite paper to
NA customers. The newspaper newsprint buyers we talked to also noted
that the price increase of newsprint and the price increases of Hi-
Brites and Super Hi-Brites tend to track each other. As Chart J1
shows, that is certainly the case with respect to price increases of
Hi-Brites and newsprint and, as argued above, it is likely due to
Abitibi's and Bowater's joint exercise of market power in the
newsprint market and in the sale of Hi-Brites.
4. Abitibi's Variable Costs To Produce Newsprint and Uncoated
Groundwood Specialty Grades Have Been Relatively Constant for the
Period 2001-2005
While there have been cost increases in inputs used to make
newsprint and uncoated groundwood specialty grades in recent years,
Abitibi has been able to implement cost-saving measures to maintain
relatively constant variable costs of producing these grades over
the period 2001 to 2005.
See Slide 25 below from the December 2006 Citigroup Conference
presentation of Abitibi's John Weaver. The slide shows the cost of
goods sold (or variable costs) for uncoated groundwood specialty
grades (called commercial printing papers or CPP by Abitibi),
newsprint and wood products. The slide shows variable costs (in
Canadian $) actually declining slightly for both newsprint and
uncoated groundwood paper specialty grades from 2001 to 2005.\117\
---------------------------------------------------------------------------
\117\ Slide 25 shows that Abitibi's variable cost to produce
newsprint in 2005 was C$523. It was also C$523 in 2006. Source:
presentation by Pierre Rogeau, Abitibi Senior VP for Corporate
Development and CFO, at the Goldman Sachs Conference, 3/20/07, Slide
24.
---------------------------------------------------------------------------
In the audio recording of Weaver's comments on Slide 25, he said
that despite the Canadian dollar and all the increase in input costs
such as energy and fiber, ``You can see for the last 5 years Abitibi
has basically managed to keep our costs relatively flat through all
these escalating input costs. So I think this shows the focus of the
company on cost reduction.'' \118\
---------------------------------------------------------------------------
\118\ These comments begin at about 17:30 of the audio recording
of Weaver's presentation, a copy of which we have provided to DOJ.
The audio recording of Weaver's presentation and the 2006 Citigroup
conference is no longer available on Abitibi's Web site. The slide
show, however, is still available.
[GRAPHIC] [TIFF OMITTED] TN10JN08.043
Since all or nearly all of the newsprint price increases over
the period 2002 to 2006 were led by Abitibi, it seems unlikely that
increases in Abitibi's input costs are a plausible justification for
the price increases.
5. Applying the Percentage Price Changes for the Uncoated
Groundwood Specialty Grades to the 3Q 1999 Price of Newsprint to
Determine the Effect on Newsprint Revenues from Sales to NA
Customers
We applied the percentage price changes calculated for the four
uncoated groundwood specialty grades shown in Table J1 to the 3Q
1999 newsprint price ($480 per metric tonne) to generate four series
of adjusted newsprint prices. Next we multiplied the actual
newsprint price series and the four adjusted newsprint price series
by quarterly NA demand (quantity purchased) shown in Chart E2.
Finally, we summed over the 30 quarters to derive total revenues
based on the five newsprint price series. The results are shown
below.
[[Page 32905]]
Table J1.--Total Newsprint Revenues Over the Period 3Q 1999 to 2Q 2006 Based on Quarterly Demand and Five
Quarterly Newsprint Price Series
[In billions of dollars]
----------------------------------------------------------------------------------------------------------------
Actual Actual
newsprint newsprint Actual Actual
Actual price price newsprint newsprint
newsprint adjusted by adjusted by price price
price (30 lb) directory hi-brite 65 adjusted by adjusted by
(22.1 lb) (35 lb) price SCA (35 lb) SCB (35 lb)
price % change % change price % change price % change
----------------------------------------------------------------------------------------------------------------
Total Revenues Based on Actual $44.1 $39.4 $44.4 $36.6 $37.5
and Adjusted Newsprint Prices..
Total Revenues Based on Actual 0.0 4.7 (0.3) 7.5 6.6
Newsprint Prices Minus Total
Revenues Based on Adjusted
Newsprint Prices...............
----------------------------------------------------------------------------------------------------------------
Table J1 is broadly suggestive of the scope of overcharges to NA
newsprint customers due to the behavior of Abitibi and Bowater over
the period 3Q 1999 to 4Q 2006. In this context, it must be noted
that we have done no analysis of the demand and supply conditions
for the Directory, SCA, and SCB grades to ensure they are good ``but
for'' world candidates. Nor have we done any analysis to determine
the appropriate methodology to determine overcharges to NA newsprint
customers. With this caveat and assuming that the price changes for
Directory, SCA, and SCB paper over the period 3Q 1999 to 4Q 2006
represent a range of appropriate ``but for'' worlds and the
methodology used to calculate the results in Table J1 is
appropriate, overcharges to NA newsprint customers over the period
3Q 1999 to 4Q 2006 totaled in the range of $4.7 billion to $7.5
billion due to the anticompetitive behavior of Abitibi and Bowater.
K. Dominant Firm Model
The preceding sections, especially Sections F through J, have
provided evidence that Abitibi and Bowater have acted to decrease
newsprint output and increase the price of newsprint over the past
four years. Their behavior can be interpreted as two firms acting
together like a dominant firm. This section discusses a simple model
of dominant firm behavior adapted to the newsprint industry. A more
detailed description of this model can be found in Attachment 4.
The model allows us to address two questions:
In theory, how could Abitibi and Bowater, acting
together or as a merged entity, profitably raise price?
Do the current conditions in the newsprint industry
suggest that Abitibi and Bowater actually have the ability
profitably to raise price further?
The model assumes that the industry is composed of a dominant
firm (or firms) with a significant market share. The rest of the
industry is made up of a large number of smaller firms, none of
which is large enough to affect significantly the market price on
its own. All firms produce the same undifferentiated product. Each
firm is assumed to have a well-defined ``full capacity'' output
level which cannot be exceeded at reasonable cost within the
relevant time frame. It is further assumed that imports are unlikely
to increase significantly from current low levels. These assumptions
provide a reasonably accurate, if somewhat simplified,
representation of the North American newsprint industry today.\119\
---------------------------------------------------------------------------
\119\ The model makes the simplifying assumptions that a firm
cannot expand its capacity and that imports do not increase. It may
be more accurate to say that the supply response of capacity-
constrained fringe firms and foreign producers is believed to be
very small for small to moderate price increases. Relaxing the
model's strict assumptions slightly does not change the general
conclusions of the discussion. The effects of relaxing assumptions
are discussed in Attachment 4.
---------------------------------------------------------------------------
Dominant Firm Strategy
Under the conditions outlined above, the strategy available to
the dominant firm is to remove fringe firms as competitive
constraints by allowing them to fill up their plants. Once the
fringe firms are operating at full capacity, they no longer can
compete to draw sales away from the dominant firm. The dominant firm
can then effectively behave as a monopolist with respect to the
``residual demand''--i.e., that portion of industry demand that is
not satisfied by the fringe firms operating at full capacity. In
this monopoly position, the dominant firm can raise price above the
initial, competitive level.
Conceptually, one can think of the dominant firm's strategy as
involving two steps. In Step 1, the dominant firm allows the fringe
firms to reach full capacity. One way to do this is for the dominant
firm to remove some of its productive capacity from the market,
either temporarily or permanently. Customers that previously
purchased from the dominant firm must then increase their purchases
from fringe firms. Total industry output is unchanged, but a portion
of industry output shifts from the dominant firm to the fringe
firms. Once the fringe firms have reached full capacity, the
dominant firm can take Step 2 and raise price without fear of being
undercut by the fringe firms. The fringe firms will tend to raise
their price along with the dominant firm, since they cannot produce
any more product. Failure to raise price to the level of the
dominant firm's price would unnecessarily sacrifice profit.
The same two conceptual steps can be achieved if the dominant
firm simply announces a significant price increase. Initially,
fringe firms behaving competitively do not follow the price
increase. To the extent possible, customers divert their purchases
from the higher-priced dominant firm to the lower-priced fringe
firms. Once the fringe firms reach their capacity constraint,
however, remaining purchases must be made from the dominant firm at
its higher price. The dominant firm is the only available supplier
capable of satisfying the ``residual demand.''
Applying the Model to the Newsprint Industry
In Attachment 4, the model is expressed formally using equations
and various parameters. Whether the dominant firm will adopt this
strategy depends on the associated gains and losses. The gains and
losses depend on various factors, including initial capacity
utilization of the fringe firms, the current market price, the
dominant firm's variable contribution margin, the percentage price
increase and the elasticity of demand. These factors are set forth
in Table K1 below. Public sources provide at least a rough estimate
of the values of these parameters for the North American newsprint
industry, as shown in Table KI. Using these estimated values, the
model predicts that it would be profitable under current conditions
for a dominant firm with the combined shares of Abitibi and Bowater
to exercise market power through the dominant firm strategy.
Table K1.--Estimated Parameter Values for Dominant Firm Model
------------------------------------------------------------------------
Factor Name Symbol Current value
------------------------------------------------------------------------
1............ Initial capacity Uc............. \120\ 95%
utilization of fringe.
[[Page 32906]]
1a........... Maximum cap. Um............ \121\ 98%
utilization of fringe.
2............ Initial industry unit P1............. \122\ $625
price.
3............ Dominant firm's unit C.............. \123\ $531
variable cost.
4............ Hypothetical price R.............. 5%
increase.
5............ Industry elasticity of E............. \124\ 0.36
demand.
6............ Initial share of S............. \125\ 41.5%
dominant firm.
------------------------------------------------------------------------
Using the parameter values in Table K1, the model predicts that
the price increase yielding the greatest profit for a dominant firm
under these conditions would be approximately 48 percent. If price
were to increase by such a large percentage, it is quite possible
that some of the assumptions of the model would have to be modified.
In particular, if extremely high prices were sustained for a period
of years, fringe firms may invest to expand their capacity, and
imports may become a more significant factor than they are at
current price levels. To avoid triggering these responses, the price
increase a dominant firm would take might be lower than the
estimated 48 percent above current levels.\126\ Even allowing for
such adjustments, the simple model presented here points to the
profitability of a significant price increase. Changing various
estimated parameters within a reasonable range does not alter this
finding.
---------------------------------------------------------------------------
\120\ The PPPC February 2007 Flash Report shows the operating
rate for North American newsprint mills for the first two months of
1997 at 95%.
\121\ According to Andrew Battista, senior RISI economist,
``practical [maximum] capacity'' is ``98% of theoretical capacity.''
See. ``Is rising newsprint demand necessary to support higher prices
in 2004?'' (paperloop.com, December 11, 2003).
\122\ Pulp & Paper Week, February 19, 2007 and RISI news report,
March 19, 2007.
\123\ Abitibi reported its average cost of newsprint production
in 2006 as C$523 (US$461). Abitibi Senior VP for Corporate
Development and CFO Pierre Rougeau presentation to 2007 Goldman
Sachs Paper & Forest Products Investor Day, 3/20/07, Slide 24.
Abitibi's firm-wide cost of distribution is 15.2 percent of its
firm-wide cost of production, averaged over 2002-2005. Abitibi 2005
Annual Report, p. 42. Using Abitibi's average delivered cost is
conservative. In reality, Abitibi and Bowater pursuing a dominant
firm strategy would tend to idle their highest cost plants first,
chiefly those located in Eastern Canada.
\124\ Jari Kuuluvainen, ``Structural Change in U.S. Newsprint
Demand: GDP and Price Elasticities,'' University of Helsinki,
Department of Forest Economics, Reports 34, 2004, p. 8.
\125\ Sum of Abitibi and Bowater current shares adjusted for
partial ownership of certain machines and mills by Abitibi and
Bowater. See Tables Cl and C2 in Attachment 2. Since Abitibi has
announced its intention to buy the minority owner's share of Augusta
newsprint, 100% of that capacity is assigned to Abitibi for the
purposes of this analysis.
\126\ But note that the price of newsprint increased by 49%
between the third quarter of 2002 and the third quarter of 2006
without triggering expansion by fringe firms or an increase in
imports.
---------------------------------------------------------------------------
The model assumes Abitibi's average cost of production as the
unit variable cost.
See Table K1 above. It is quite likely that the capacity that
Abitibi and Bowater would idle when pursuing a dominant firm
strategy would be their highest cost capacity. In his December 2006
presentation to the Citigroup Conference, Abitibi Bowater's David
Paterson was asked how Bowater would be able to maintain sufficient
cash flow to pay for dividends and interest payments if newsprint
prices declined from current levels. As quoted in Section F.3 above
from an audio recording of his remarks, Paterson responded,
So my belief[. . .]is that we have to move faster to stay ahead
of that [demand] curve to maintain an operating rate that gives us
some pricing leverage in the market and I can do that. We know which
our high cost assets are and we will shut them down hopefully before
rather than after price erosion with any significance.
Earlier in his presentation, Paterson had stated that Bowater's
high-cost newsprint assets were located in Eastern Canada and that
``there is not a lot of margin left in the Canadian assets.''
Section L. Conclusions
Based on our economic analysis of the likely competitive effects
of the proposed Abitibi-Bowater merger contained in Sections B
through K above, we conclude that the merger, if it is permitted to
proceed, will have very significant adverse competitive and economic
effects on U.S. newspaper publishers and other NA consumers of
newsprint.
Through their joint behavior over the past four years, Abitibi
and Bowater have demonstrated that their combined share of NA
newsprint capacity was large enough to enable them to consistently
raise the price of newsprint in the face of steadily declining NA
newsprint demand. Abitibi and Bowater matched declining consumption
year after year with the amount of capacity removal needed to
maintain high operating rates and increasing newsprint prices. This
strategy has been remarkably successful as this White Paper
documents. The title of one of the articles cited in Section H,
``New Paradigm: Newsprint Demand Falls, Prices Soar,'' captures this
paradox of ``soaring'' prices in the face of declining consumption.
The fact that Abitibi and Bowater have been able to profitably
reduce their own capacity to raise the price of newsprint is direct
evidence that they have jointly possessed and exercised market power
over a sustained period of time. A small firm would have no
incentive unilaterally to close capacity to raise the price of
newsprint because the loss of net margin from the closed capacity
would outweigh the gain in margin from the price increase on the
capacity that it would still operate.
As we have documented in this White Paper, the NA newsprint
market was unconcentrated in 1995 but became highly concentrated by
2000 primarily due to mergers by Abitibi, Bowater, and the newsprint
firms they acquired. Without these mergers, Abitibi and Bowater
would have been unable to pursue their highly effective and highly
anticompetitive joint strategy.
The newspaper newsprint buyers whom we talked to believe that it
is certain that a combined Abitibi and Bowater will continue to
pursue this anticompetitive strategy, but the merged firm will be
able to do so more effectively. Coordination difficulties, costs,
and uncertainties that Abitibi and Bowater faced as separate firms
in their exercise of joint dominance would be removed by a merger.
Future capacity closures to raise the price of newsprint will be
more optimal and timely from the viewpoint of the merged firm and
more harmful to NA consumers of newsprint. Without a merger,
imperfect coordination between Abitibi and Bowater may break down in
the coming months or years. With a merger, perfect coordination is
certain.
Attachment A--Links to Newsprint-Related Web Sites
Two tables appearing in this comment are not able to be
reprinted here. Copies of the comment with the tables are available
at the Department of Justice Antitrust Division Web site, http://
www.usdoj.gov/atr, at the Antitrust Documents Group of the
Department of Justice Antitrust Division, 450 Fifth Street, N.W.,
Suite 1010, Washington, D.C. 20530, (202) 514-2481, and at the
Office of the Clerk of the United States District Court for the
District of Columbia, 333 Constitution Avenue, N.W., Washington,
D.C. 20001.
Attachment B--Additional Analysis of Uncoated Groundwood Specialty
Grades and Tables B1 to B7 for Section B
A. Comparing the Price of Newsprint With the Prices of Four
Uncoated Groundwood Specialty Grades
Since the quality and value of newsprint is lower than the
quality and value of all uncoated groundwood specialty grades, we
would expect that newsprint would have a lower price. Table B1 below
compares the February 2007 price (Eastern U.S.) of 30 lb. newsprint
with the price of 35 lb. Hi-Brites (65 brightness level), the price
of 35 lb. SCA,
[[Page 32907]]
and the price of 35 lb. SCB. Table B1 also compares the price of
27.7 lb. newsprint with the price of 22.1 lb. directory paper.\1\
---------------------------------------------------------------------------
\1\ Source: RISI Pulp & Paper Week, February 19, 2007. Except
for newsprint, the prices are the average of the high and low prices
for February 2007. The price for directory paper is a spot price.
About 80% to 90% of directory paper is sold under one to three year
contracts to RBOCs and independent directory publishers. The prices
in Pulp & Paper Week for the four uncoated groundwood specialty
grades were per short ton. These prices were converted to price per
metric tonne by multiplying the short ton prices by the ratio of the
weight in pounds of a metric tonne (2,205 lbs.) to the weight in
pounds of a short ton (2,000 lbs.). RISI notes that for the two
newsprint grades and the four uncoated groundwood specialty grades
that there had been some discounting below transaction prices.
---------------------------------------------------------------------------
The SCB, Hi-Brite 65, and SCA 35 lb. February 2007 prices were
17.3% to 23.4% higher than the price of 30 lb. newsprint. If a
newsprint buyer switched from 30 lb. newsprint to one of these
higher basis weight grades, the buyer would incur a 14.3% reduction
in printing surface. Taking the reduction in printing surface into
account, a buyer of 30.0 lb. newsprint who switched to 35.0 lb. SCB,
Hi-Brite 65, or SCA, would face an equivalent price increase per
metric tonne of 30.0 lb. newsprint ranging from 34.0% to 47.0% based
on February 2007 prices.
Table B1.--Comparing February 2007 Newsprint Prices With the Prices of Four Uncoated Groundwood Specialty Grades
----------------------------------------------------------------------------------------------------------------
Percent Percent
Price Percent price increase increase
February 2007 difference difference (decrease) in (decrease) in
price per over the over the square footage the effective
metric tonne newsprint newsprint per metric price per
price price tonne metric tonne
----------------------------------------------------------------------------------------------------------------
Newsprint (30.0 lb.)............ $630.00
Hi-Brite 65 (35 lb.)............ 777.26 $147.26 23.4 (14.3) 41.0
SCA (35 lb.).................... 810.34 180.34 28.6 (14.3) 47.0
SCB (35 lb.).................... 738.68 108.68 17.3 (14.3) 34.0
Newsprint (27.7 lb.)............ 670.00
Directory (22.1 lb.)............ 810.34 140.34 20.9 27.2 (12.0)
----------------------------------------------------------------------------------------------------------------
Source: RISI Pulp & Paper Week, February 19, 2007, p. 3.
Table B1 shows that the February 2007 price of 22.1 lb.
directory paper was 20.9% higher than the price of 27.7 lb.
newsprint. If a buyer of 27.7 lb. newsprint switched to the lower
basis weight paper, the buyer would gain 27.2% in printing surface
per metric tonne. Taking this increase in printing surface into
account, a buyer of 27.7 lb. newsprint who switched to 22.1 lb.
directory paper would receive an equivalent price reduction of 12.0%
per metric tonne of 27.7 lb. newsprint based on February 2007
prices. However, as discussed in Section B.3.a.(4), the information
provided to us by newsprint buyers leads us to conclude that the
lower basis weight and thinner directory paper would not be suitable
for use in a newspaper or for running on newspaper printing presses.
B. An Analysis of Estimated 2006 Abitibi and Bowater Shares of
Uncoated Groundwood Specialty Grade Segments
1. Introduction
It is beyond the scope of this White Paper to delineate product
markets composed of one or more uncoated groundwood specialty
grades. Nonetheless, each of these grades is in some relevant
product market. Both Abitibi and Bowater are significant producers
of uncoated groundwood specialty grades.
We have estimated capacities and capacity shares for the
following uncoated groundwood specialty grade segments: (1) All
uncoated groundwood specialty grades; (2) directory paper; (3) SC/
SNC glossy grades; (4) Hi-Brites/Super Hi-Brites; and (5) Bulky Book
and Other. Attachment 1 contains tables showing capacity and
capacity shares for NA mills for each of the first four segments
shown above.\2\ We also prepared a fifth table which shows East of
the Rockies capacity for mills producing Hi-Brites and Super Hi-
Brites. These five tables are discussed below.
---------------------------------------------------------------------------
\2\ We could only identify Bulky Book capacity for Abitibi and
Tembec. It appears that Bowater produces paper for the Bulky Book
segment but Bowater does not specifically identify the amount of its
capacity used to produce bulky book paper. Other firms may also
produce Bulky Book paper, but we have not been able to identify
them.
---------------------------------------------------------------------------
Our primary source for the estimated capacity and capacity
shares was the Uncoated Mechanical Papers chapter from the RISI 2006
Fact and Price Book (pp. 161-173). RISI provides capacity by
manufacturer for total uncoated groundwood specialty grades,
directory paper, and SC/SNC grades. Because most of the remaining
capacity is for Hi-Brites and Super Hi-Brites, RISI implicitly
provides capacity estimates for those two grades combined.
We supplemented the RISI uncoated groundwood specialty grade
capacity data with the following sources: (1) Reported capacity for
Abitibi and Bowater shown on p. 17 of their merger announcement
presentation; \3\ (2) Web sites of manufacturers; (3) annual reports
and other public documents produced by manufacturers; and (4) online
searches for additional information about manufacturers and their
uncoated groundwood specialty capacity. While the results of our
data search are preliminary and were subject to some exercise of
judgment, we believe these results provide a good first
approximation of manufacturer shares in each of the five segments
described above. Additional data search would likely further refine
the data.\4\
---------------------------------------------------------------------------
\3\ This capacity is reported as uncoated mechanical by Abitibi
or Bowater mill. The capacity is not further broken down by specific
grades.
\4\ The availability and accuracy of capacity data for
manufacturers of uncoated groundwood specialty grades appears to be
lower than for newsprint manufacturers.
---------------------------------------------------------------------------
2. Abitibi-Bowater HHIs Based on Estimated 2006 Capacity and
Capacity Shares by Manufacturer for Uncoated Groundwood Specialty
Grade Segments
Tables B2 through B6 at the end of Attachment B show Abitibi-
Bowater HHIs based on estimated 2006 capacity and capacity shares by
manufacturer for the following uncoated groundwood specialty grade
segments: (a) All uncoated groundwood specialty grade capacity in
NA; (b) all directory paper in NA; (c) all SC/SNC glossy paper
capacity in NA; (d) all Hi-Brite & Super Hi-Brite non-glossy paper
capacity in NA; and (e) all HiBrite & Super Hi-Brite non-glossy
paper capacity East of the Rockies. The results from Tables B2
through B6 plus Bulky Book and Other are summarized in Table B7
below.
[[Page 32908]]
Table B7.--Abitibi-Bowater HHIs Based on Estimated 2006 NA Capacity and Capacity Shares by Manufacturer for Uncoated Groundwood Specialty Grade Segments
--------------------------------------------------------------------------------------------------------------------------------------------------------
Total NA East of the
uncoated NA directory NA hi-brites & rockies hi- NA bulky book
groundwood lightweight NA SC/SNC super hi- brites & super and other
specialty paper paper brites non- hi-brites paper
grades glossy paper paper
--------------------------------------------------------------------------------------------------------------------------------------------------------
Total Segment Capacity (1,000 Metric Tonnes)............ 6,997 1,291 3,360 2,122 1,624 224
Abitibi Capacity Share.................................. 30.2% 10.7% 26.0% 44.8% 58.5% 66.5%
Bowater Capacity Share.................................. 14.3% 0.0% 9.8% 31.8% 41.5% 0.0%
Combined Abitibi-Bowater Capacity Share................. 44.5% 10.7% 35.8% 76.5% 100.0% 66.5%
Pre-Merger HHI.......................................... 1,516 2,319 1,454 3,286 5,144 0
Change in the HHI....................................... 749 0 511 1,392 4,856 0
Post-Merger HHI......................................... 2,265 2,319 1,965 4,679 10,000 0
--------------------------------------------------------------------------------------------------------------------------------------------------------
Sources: RISI 2006 Global Pulp & Paper Fact & Price Book, pp. 163, 165, and 166, Abitibi-Bowater merger announcement presentation, p. 17, manufacturer
Web sites, manufacturer annual reports, and other publicly available information.
Assuming the uncoated groundwood specialty segments shown in
Table B7 above were relevant product and geographic markets, four of
the segments (total NA uncoated groundwood specialty grades, NA SC/
SNC glossy paper, NA Hi-Brite/Super Hi-Brite non-glossy paper, and
East of the Rockies Hi-Brite/Super Hi-Brite non-glossy paper) show
an increase in the HHI significantly greater than 100 resulting from
an Abitibi-Bowater merger and these same four segments show a post-
merger HHI greater than 1,800. In the case of NA Hi-Brite/Super Hi-
Brite capacity, the post-merger HHI is 4,679. In the case of East of
the Rockies Hi-Brite/Super Hi-Brite capacity, the post-merger HHI is
10,000. Two of the segments (Directory Paper and Bulky Book and
Other) show no change in the HHI resulting from an Abitibi-Bowater
merger. According to Sec. 1.51(c) of the Merger Guidelines:
Where the post-merger HHI exceeds 1800, it will be presumed that
mergers producing an increase in the HHI of more than 100 points are
likely to create or enhance market power or facilitate its exercise.
Imports into NA vary by segment: (a) 2005 imports of SC paper
into NA were 13.5% of 2006 NA SC/SCA capacity; (b) 2005 imports of
Lightweight (Directory) paper into NA were 6.5% of 2006 NA Directory
paper capacity; (b) all other 2005 imports were 1.9% of all other
2006 NA Uncoated Groundwood Specialty Grade capacity (i.e., Hi-
Brite/Super Hi-Brite, Bulky Book, and Other).\5\
---------------------------------------------------------------------------
\5\ The source for the 2005 import data is the RISI 2006 Fact &
Price Book, pp. 164 and 169. Canadian imports were not broken by SC,
lightweight, and other. We assumed that the Canadian percentage
breakdown was the same as the U.S. percentage breakdown for these
three categories. The sources for the 2006 NA capacities by category
are shown in Tables B2-B5.
---------------------------------------------------------------------------
BILLING CODE 4410-11-M
[[Page 32909]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.044
[[Page 32910]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.045
[[Page 32911]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.046
[[Page 32912]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.047
[[Page 32913]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.048
BILLING CODE 4410-11-C
Attachment C--Tables C1 to C3 for Section C
Table C1.--Estimate of 2006 U.S. and Canadian Newsprint Capacity by Mill
------------------------------------------------------------------------
U.S. Newsprint Mills
-------------------------------------------------------------------------
Est. 2006
capacity
State/city Company name and notes metric
tonnes
------------------------------------------------------------------------
Alabama
Claiborne.................... Alabama River Newsprint 264,000
Company.
(Abitibi owns 100% of
Alabama Newsprint.)
Coosa Pines.................. Bowater Incorporated.... 328,000
Arizona
Snowflake.................... Abitibi-Consolidated Inc 375,000
California
Pomona....................... Blue Heron Paper Company 150,000
(The company is owned by
employees. The mill was
acquired from Smurfit
in 2005. Blue Heron
recently announced the
Pomona mill would be
indefinitely idled
beginning May 6, 2007.)
Georgia
Augusta...................... Augusta Newsprint 426,000
Company.
(Abitibi owns 52.5% of
Augusta Newsprint.
Woodbridge Co. owns the
other 47.5%. Abitibi
has announced its
intention to buy
Woodbridge's 47.5%
share in Augusta
Newsprint.)
Dublin....................... SP Newsprint Company.... 565,000
(The company is owned by
3 newspaper
publishers.)
Louisiana
DeRidder..................... Boise Cascade 405,000
Corporation.
(Abitibi is the
exclusive marketing and
sales agent for the
newsprint produced at
the DeRidder mill.)
Mississippi
Grenada...................... Bowater Incorporated.... 249,000
Oregon
Newberg...................... SP Newsprint Company.... 395,000
(The company is owned by
3 newspaper publishers.
The mill was acquired
from Smurfit in 1999.)
Oregon City.................. Blue Heron Paper Company 140,000
(The company is owned by
employees. The mill was
acquired from Smurfit
in 2000.)
[[Page 32914]]
Tennessee
Calhoun...................... Bowater Incorporated 382,000
(Southern Division).
(Bowater owns 51% of one
newsprint machine at
the Calhoun mill with
approx. 205,000 metric
tonnes of capacity. The
Herald Company, Inc.
owns the other 49%.
Bowater owns 100% of
remaining Calhoun
newsprint capacity.)
Texas
Lufkin....................... Abitibi-Consolidated Inc 150,000
(The mill has been idled
indefinitely since
December 2003.)
Virginia
Ashland...................... Bear Island Paper 235,000
Company.
(The mill is owned by
White Birch, a
privately-held
company.)
Washington
Longview..................... North Pacific Paper 675,000
Company (NORPAC).
(JV between Weyerhauser
and Nippon Paper
(Japan)).
Milwood...................... Inland Empire Paper 135,000
Company.
(The mill is owned by 2
newspaper publishers.)
Usk.......................... Ponderay Newsprint 249,00
Company.
(Bowater owns 40% of
Ponderay and is
managing partner. The
remaining 60% of
Ponderary is owned by 5
newspaper publishers.)
------------------------------------------------------------------------
------------------------------------------------------------------------
Canadian Newsprint Mills
-------------------------------------------------------------------------
Est. 2006
capacity
Province/city Company name and notes metric
tonnes
------------------------------------------------------------------------
Alberta
Whitecourt Alberta Newsprint 269,000
Company Ltd.
(JV between the Stern
Group and West Fraser
Timber.).
British Columbia
Campbell River............... Catalyst................ 321,000
(Norske Canada was re-
named Catalyst in 2005.
Norske Skog sold its
minority interest in
Norske Canada in 2005.
Catalyst is publicly
traded.).
Crofton...................... Catalyst................ 198,000
(Norske Canada was re-
named Catalyst in 2005.
Norske Skog sold its
minority interest in
Norske Canada in 2005.
Catalyst is publicly
traded.).
Mackenzie.................... Abitibi-Consolidated Inc 186,000
Port Mellon.................. Howe Sound Pulp & Paper 215,000
Ltd.
(JV between Canfor (BC)
and Oji Paper (Japan)).
Powell River................. Catalyst................ 181,000
(Norske Canada was re-
named Catalyst in 2005.
Norske Skog sold its
minority interest in
Norske Canada in 2005.
Catalyst is publicly
traded.).
Manitoba
Pine Falls................... Pine Falls Paper Company 185,000
Ltd.
(Mill is owned by
Tembec, a publicly-
traded company.).
New Brunswick
Dalhousie.................... Bowater Maritimes Inc... 213,000
Bowater now owns 100% of
Bowater-Maritimes. It
recently acquired
minority interests from
two Japanese paper
companies..
Newfoundland
Corner Brook................. Kruger Inc. (Corner 440,000
Brook Pulp and Paper
Ltd.).
(Kruger is a privately-
held company.).
Grand Falls.................. Abitibi-Consolidated Inc 191,000
(Includes capacity of PM
7 (capacity = 60,000
metric tonnes), which
has been indefinitely
idled since the end of
2005.).
Nova Scotia
Liverpool.................... Bowater Mersey Paper 253,000
Company Ltd.
(Bowater owns 51% of
Bowater Mersey. The
Washington Post owns
the other 49%.).
Nova Scotia
Port Hawkesbury.............. Stora Enso North 190,000
American Corp..
(Newsprint machine
restarted at end of
November 2006 after
being idled for almost
a year due to labor
contract problems and
high energy costs.).
Ontario
Iroquois Falls............... Abitibi-Consolidated Inc 240,000
Kapuskasing.................. Spruce Falls Inc........ 330,000
(Mill is owned by
Tembec, a publicly-
traded company.).
[[Page 32915]]
Thorold...................... Abitibi-Consolidated Inc 414,000
Thunder Bay.................. Bowater Canadian Forest 380,000
Products Inc.
(Includes capacity of PM
4 (capacity = 146,000
metric tonnes), which
has been indefinitely
idled since September
2005.).
Whitby....................... Atlantic Newsprint Co... 150,000
(Atlantic Newsprint is a
business unit within
the Atlantic Group, a
privately-held
company.).
Quebec
Amos......................... Abitibi-Consolidated Inc 207,000
Baie Comeau.................. Abitibi-Consolidated Inc 577,000
Bromptonville................ Kruger Inc.............. 310,000
(Kruger is a privately-
held company.).
Clermont..................... Abitibi-Consolidated Inc 354,000
(Abitibi owns 51% of one
newsprint machine at
the Clermont mill with
approx. 219,000 metric
tonnes of capacity. The
New York Times Co. owns
the other 49%. Abitibi
owns 100% of the
remaining Clermont
newsprint capacity.).
Gatineau..................... Bowater Canadian Forest 432,000
Products, Inc.
Masson....................... Papier Masson Ltd....... 240,000
(Acquired by White Birch
in 2006.).
Quebec....................... Stadacona, Inc.......... 410,000
(Acquired by White Birch
in 2004.).
Riviere-du-Loup.............. F.F. Soucy Inc.......... 265,000
(Owned by White Birch, a
privately-held company).
Shawinigan (Belgo)........... Abitibi-Consolidated Inc 116,000
Trois-Rivieres............... Kruger Inc.............. 370,000
(Kruger is a privately-
held company.).
------------------------------------------------------------------------
Sources and Notes
1. The capacity estimates for Abitibi, Bowater and Ponderay
Newsprint mills are from the Abitibi-Bowater merger announcement
presentation, ``Creating a Global Leader in Paper and Forest
Products,'' January 29, 2007, p. 17. http://
www.abitibiconsolidated.com/aciwebsitev3.nsf/site/en/images/pdf/
Final_Investor_Presentation.pdf/$file/Final_Investor_
Presentation.pdf.
2. The capacity estimates for the White Birch Paper newsprint
mills are from the White Birch Paper Web site: http://
www.whitebirchpaper.com/en/p2.html.
3. The capacity estimates for the Kruger newsprint mills are
from the Kruger Web site: http://www.kruger.com/english/D_
Newsprint/Newsprint_INTRO_A.html.
4. The capacity estimates for the Catalyst newsprint mills are
from the Catalyst Web site: http://www.catalystpaper.com/aboutus/
aboutus_ourdivisions.xml.
5. The capacity estimates for the Tembec newsprint mills are
from the Tembec 2006 Annual Report, p. 29. http://www.tembec.com/
public/Investisseurs/Rapports-financiers.html.
6. The capacity estimate for the Alberta Newsprint mill is from
the Alberta Newsprint Web site: http://www.albertanewsprint.com/
profile/information.htm.
7. The capacity estimate for the Stora Enso's Port Hawkesbury,
NS newsprint mill is from the Stora Enso Web site: http://
www.storaenso.com/CDAvgn/main/0,,1_-3429-4370-,00.html. The Port
Hawkesbury mill, including its newsprint machine, was idled on
December 2005 due to labor contract and energy cost problems. The
newsprint machine was restarted at the end of November 2006
following the resolution of these problems. See http://
www.paperage.com/2006news/11_27_2006stora.html.
8. Annual capacity estimates for the SP Newsprint, North
Pacific, Boise Cascade, Blue Heron, Howe Sound, Atlantic Newsprint,
and Inland Empire mills in Table C1 are from the July 2004
preliminary forecast shown in the Pulp and Paper Products Council
(PPPC) July 9, 2004 update titled ``Update of North American
Mechanical Printing Papers Capacity Forecast.'' This update can be
found on the PPPC Web site under press releases: http://
www.pppc.org/en/1_0/index.html. The Web sites for these seven
manufacturers did not clearly and unambiguously identify their
respective annual newsprint mill capacities.
9. Capacity at Abitibi's Kenora ON, La Baie (Port-Alfred) QC,
and Stephenville NF newsprint mills are included in the PPPC July
2004 preliminary forecast. Those mills have been permanently closed.
See the Abitibi 2005 Annual Report, p. 18 and the Abitibi 2004
Annual Report, p. 50. The PPPC July 2004 preliminary forecast also
shows Abitibi's Alma, QC mill with newsprint capacity. The Alma
mill's newsprint capacity has been converted to the production of
higher value uncoated groundwood specialty grades. See the Abitibi
2004 Annual Report, p. 50.
10. The PPPC July 2004 update also notes on p. 1 that the
capacities of three newsprint machines at Abitibi's Sheldon, TX mill
and the 3 newsprint machine at Bowater's Thunder Bay ON
mill that had been idled for over a year were no longer included in
the forecast. The Abitibi Sheldon, TX mill has been permanently
closed. See Abitibi 2004 Annual Report, p. 50. The Bowater Thunder
Bay 3 newsprint machine will not be restarted according to
Bowater. See Bowater February 6, 2007 news release ``Bowater
Announces Fourth Quarter and Full Year 2006 Financial Results,''
Note 1. ``Based on the continued decline of North American newsprint
consumption through the third quarter of 2006, Bowater now has no
plans to restart the machine.''
11. The PPPC March 2006 forecast of 2006 NA newsprint capacity
is 12,625,000 metric tonnes. Compared to the total in Table C2
above, this is a difference of 135,000 metric tonnes or 1.4%. In its
forecast, the PPPC does not provide a breakdown by manufacturer or
by mill so the reasons for the difference cannot be ascertained with
certainty. The PPPC does not include the 150,000 metric tonne
capacity of Abitibi's Lufkin, TX mill in its 2006 forecast because
the mill has been indefinitely idled since December 2003. The
capacity of the Lufkin, TX mill is included in Tables C1-C3,
however, because Abitibi continues to count the Lufkin capacity in
its public documents, including the Abitibi-Bowater merger
announcement presentation. See the Abitibi-Bowater merger
announcement presentation, ``Creating a Global Leader in Paper and
Forest Products,'' January 29, 2007, p. 17. From an antitrust
perspective, it is appropriate to include the Lufkin, TX capacity in
Abitibi's total newsprint capacity if the mill could be re-started
within a year. See the product market discussion in Section B
regarding ``Firms That Participate Through Supply Response.'' If the
Lufkin, TX mill's capacity is added to the PPPC 2006 forecast, the
difference between the Table C2 total and the PPPC forecast for 2006
is reduced to 15,000 metric tonnes or 0.1%.
[[Page 32916]]
12. Two other mills included in the PPPC July 2004 preliminary
forecast, Katahdin Paper and Irving Paper, no longer manufacture
newsprint. Their newsprint capacity has been converted to the
production of higher value uncoated groundwood specialty grades. In
addition, the PPPC July 2004 preliminary forecast shows a small
amount of newsprint capacity at Kruger's Manistique, MI mill. The
mill no longer produces newsprint and Kruger no longer owns the
mill.
13. According to an article in Editor & Publisher by Debra
Garcia, dated March 28, 2007,''Blue Heron Paper Co. [recently]
announced it would indefinitely idle its 140,000 tonnes/year 100%
recycled newsprint mill in Pomona, Calif., due to high wastepaper
and energy costs and declining newsprint consumption. The shutdown
is slated to begin about May 6.''
14. The information on which the notes in Table C1 are based can
generally be found on manufacturer web sites, including annual
reports and 10K reports available as pdf files on the web sites of
publicly-traded newsprint manufacturers. The source for Abitibi's
plans to purchase the remaining 47.5% interest in Augusta Newsprint
is the Abitibi presentation ``Our Story on Paper'' by President and
CEO John Weaver at the Citigroup 11th Annual Global Paper and Forest
Products Conference, December 7, 2006, p. 26, which is available on
the Abitibi Web site.
BILLING CODE 4410-11-M
[[Page 32917]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.049
[[Page 32918]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.050
Attachment D--Tables D1 to D4 for Section D
[[Page 32919]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.051
[[Page 32920]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.052
[[Page 32921]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.053
[[Page 32922]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.054
BILLING CODE 4410-11-C
Attachment K--Technical Appendix to Section K Dominant Firm Model
Dominant Firm Model
This section provides a formal model of a dominant firm in an
industry with fixed capacity constraints producing a homogeneous
product. Fringe firms are assumed to be price-takers. Imports are
assumed to be fixed. Under these conditions, a dominant firm may
find it profitable to remove fringe firms as competitive constraints
by allowing them to fill up their plants (Step 1). Once the fringe
firms are operating at full capacity, they no longer can compete to
draw sales away from the dominant firm. The dominant firm can then
effectively behave as a monopolist with respect to the ``residual
demand''--i.e., that portion of industry demand that is not
satisfied by the fringe firms operating at full capacity. In this
monopoly position, the dominant firm can raise price above the
initial, competitive level (Step 2).
Whether the dominant firm will adopt this strategy of reducing
output to bring the fringe to capacity and then raising its price
depends on the associated gains and losses from doing so. The gains
and losses, in turn, depend on various factors discussed below. The
losses can be thought of in two parts, L1 and L2, corresponding to
Step 1 and Step 2.
L1: In Step 1, the dominant firm gives up some of its sales to
the fringe firms. The cost of doing this is the variable profit that
the dominant firm would have earned on those sales. This variable
profit can be calculated as the forgone quantity times the unit
variable margin on those sales. The forgone quantity is the quantity
needed to move the fringe firms from their initial capacity
utilization to full capacity utilization. The unit variable margin
is the difference between the initial industry price and the
dominant firm's unit variable cost for the capacity that it idles.
The greater is L1, the less likely it is that the benefits of
the dominant firm strategy will outweigh the costs. Three factors
are particularly important in determining the magnitude of L1. The
first factor is the capacity utilization of the fringe firms, and
the second and third factors pertain to the variable profit margins
on the lost sales.
Factor 1. Initial capacity utilization of the fringe
firms. if the fringe firms are operating at a high level of capacity
utilization, the quantity that the dominant firm must give up to
move them to full capacity is relatively small, and L1 is
proportionately small. On the other hand, if initial capacity
utilization is low, the dominant firm will have to give up a larger
quantity to bring the fringe firms to full capacity, and L1 will
tend to be large.
Factor 2. Initial price level. Suppose the initial
industry price level is low relative to the variable cost of the
capacity to be idled. This means that the dominant firm's variable
margin is low for the idled capacity, and the
[[Page 32923]]
profits it loses by giving up quantity to the fringe firms, L1, is
correspondingly low. By contrast, if the initial price level is high
relative to the variable cost of the capacity to be idled, the
profits lost on each unit of quantity given up to the fringe firms
are relatively high, making L1 large.
Factor 3. Dominant firm's variable cost of production.
The variable cost of production operates as the flip side of the
initial price level. The higher the variable cost (relative to
price), the smaller is L1, and the lower is variable cost (relative
to price), the greater is L1. (Note that the relevant variable
margin is the margin in those plants that the dominant firm would
remove from production. Rationally, the dominant firm would first
remove its capacity with the highest costs. For this reason, using
the firm-wide average variable cost margin overstates the loss of
margin in L1. The same point applies to L2 below.)
L2: In Step 2, when the dominant firm raises price above the
initial level, industry customers will tend to respond by reducing
their total purchases. This relationship between price and quantity
demanded follows the basic ``law of demand.'' In order to keep the
competitive fringe at full capacity, the dominant firm absorbs this
entire decrease in quantity. As with L1, the reduction in profits in
L2 is the reduction in quantity times the variable margin on those
sales. We have already noted how the initial price and variable cost
of production, Factor 2 and Factor 3, are important in determining
the variable margin. Two additional factors also affect L2.
Factor 4. Percentage price increase. Obviously, the
greater the percentage price increase, the larger will be the
associated loss of quantity along the demand curve.
Factor 5. Elasticity of demand. Elasticity of demand is
defined as the percentage change in quantity demanded that occurs in
response to a one-percent change in price. The greater the
elasticity of demand, the larger is the loss of quantity resulting
from the price increase, and the larger is L2. Since the dominant
firm is absorbing the quantity reduction for the entire industry,
the appropriate demand elasticity to use is the industry demand
elasticity.
The dominant firm strategy will be adopted only if the benefit
or gain (G) exceeds the sum of L1 and L2. The gain the dominant firm
receives from the strategy is that it receives a higher price on all
of its remaining output. The relevance of the initial price level
(Factor 2) and the percentage increase in price (Factor 4) is quite
apparent. The other factor determining the dominant firm's profit
gain is its initial sales and market share.
Factor 6. Initial sales and share of the dominant firm.
To determine the quantity of sales on which the dominant firm will
enjoy the price increase, one takes the dominant firm's initial
quantity and subtracts the quantity reductions associated with L1
and L2.
To see the role of initial share, take the rather extreme case
in which the dominant firm has low initial sales due to a low share,
and that its sales are approximately equal to the quantity losses
associated with L1 and L2. In that position, the dominant firm could
absorb the quantity needed to move the fringe to full capacity and
absorb the decrease in quantity resulting from the increased price.
However, the dominant firm would have little or no remaining sales
to make at the higher price, and hence little or no benefit or gain
from the strategy. In this situation, the dominant firm will not
adopt the price increase strategy. By contrast, if the dominant firm
has large initial sales due to a large initial share, it is more
likely to still have a large quantity to sell after absorbing the
losses (L1 and L2). In this situation, the dominant firm would
realize a large gain from the price increase, and the price increase
strategy is more likely to be adopted by the dominant firm than if
it had a low initial share.
Note that the share of the dominant firm also affects L1. A
large initial share for the dominant firm indicates that there is a
smaller competitive fringe. This would reduce the amount of quantity
that must be absorbed in Step 1 to bring the fringe to full capacity
(i.e., reduces L1) for any given level of fringe capacity
utilization.
Though mentioned last, the share of the dominant firm may have
the greatest relevance because it is the only factor directly
affected by merger enforcement policy. The initial share affects the
likelihood of a significant price increase and the potential
magnitude of a price increase, both of which are central antitrust
concerns.
Mathematical model
Table K1 shows the six factors discussed above, the symbol used
in this attachment to represent each factor, and an estimate of the
current value of each factor. Factor 1a, the maximum potential
capacity utilization rate for the fringe firms, is added to assist
in calibrating the model to current industry conditions.
Table K1.--Estimated Parameter Values for Dominant Firm Model
------------------------------------------------------------------------
Factor Name Symbol Current value
------------------------------------------------------------------------
1............ Initial capacity Uc............. \1\ 95%
utilization of fringe.
1a........... Maximum cap. Um............. \2\ 98%
utilization of fringe.
2............ Initial industry unit P1............. \3\ $625
price.
3............ Dominant firm's unit C.............. \4\ $531
variable cost.
4............ Hypothetical price R.............. 5%
increase.
5............ Industry elasticity of E.............. \5\ 0.36
demand.
6............ Initial share of S.............. \6\ 41.5%
dominant firm.
------------------------------------------------------------------------
Under the strategy modeled here, the dominant firm first reduces
its output through removal of capacity from the market to the point
that the fringe firms reach their maximum capacity. The reduction in
dominant firm profits in this first step is L1. The dominant firm
then raises price. This price increase further reduces the dominant
firm's profits through a further reduction in quantity. This profit
reduction is L2. The firm increases its profits through an increase
in the price at which it sells its remaining units. This profit
increase is G. The dominant firm strategy is likely to be adopted if
G - L1 - L2>O.
---------------------------------------------------------------------------
\1\ The PPPC February 2007 Flash Report shows the operating rate
for North American newsprint mills for the first two months of 1997
at 95%.
\2\ According to Andrew Battista, senior RISI economist,
``practical [maximum] capacity'' is ``98% of theoretical capacity.''
See. ``Is rising newsprint demand necessary to support higher prices
in 2004?'' (paperloop.com, December 11, 2003)
\3\ Pulp & Paper Week, February 19, 2007 and RISI news report,
March 19, 2007.
\4\ Abitibi reported its average cost of newsprint production in
2006 as C$523 (U.S.$461). Abitibi Senior VP for Corporate
Development and CFO Pierre Rougeau presentation to 2007 Goldman
Sachs Paper & Forest Products Investor Day, 3/20/07, Slide 24.
Abitibi's firm-wide cost of distribution is 15.2 percent of its
firm-wide cost of production, averaged over 2002-2005. Abitibi 2005
Annual Report, p. 42. Using Abitibi's average delivered cost is
conservative. In reality, Abitibi and Bowater pursuing a dominant
firm strategy would tend to idle their highest cost plants first,
chiefly those located in Eastern Canada.
\5\ Jan Kuuluvainen, ``Structural Change in U.S. Newsprint
Demand: GDP and Price Elasticities,'' University of Helsinki,
Department of Forest Economics, Reports 34, 2004, p. 8.
\6\ Sum of Abitibi and Bowater current shares adjusted for
partial ownership of certain machines and mills by Abitibi and
Bowater. See Tables C1 and C2 in Attachment 2. Since Abitibi has
announced its intention to buy the minority owners share of Augusta
newsprint, 100% of that capacity is assigned to Abitibi for the
purposes of this analysis.
---------------------------------------------------------------------------
LI is the product of the dominant firm's per-unit variable
margin and the quantity reduction needed to bring the fringe firms
to their maximum capacity. Per-unit variable margin is represented
as P1-C. For convenience, and in the absence of more exact
information about the actual shape of the cost curve, it is assumed
that the
[[Page 32924]]
dominant firm's unit variable costs are constant in the relevant
range.\7\
---------------------------------------------------------------------------
\7\ The dominant firm may be able to reduce its losses in L1 and
L2 if, instead of idling capacity, it can ``dump'' some of its
production in overseas markets from which they will not be re-
imported.
---------------------------------------------------------------------------
As a further convenience, quantity units will be chosen such
that the industry's total nominal capacity is one unit. Under this
assumption, the total capacity of the dominant firm is S and the
total capacity of the fringe firms is 1-S. Maximum practical
capacity, Um, is permitted to he below maximum nominal capacity. To
change fringe firms' capacity utilization from the initial level,
Uc, to Um requires that the fringe's quantity be increased, and the
dominant firm's quantity be decreased, by (1-S) (Um-Uc). Thus
[1] LI = (P1-C) (1-S) (Um-Uc)
Once fringe firms are operating at maximum capacity, the
dominant firm raises price by some percentage R. The dominant firm
absorbs the entire reduction in industry quantity demanded resulting
from the price increase. The quantity reduction is given by the
product of R (the percentage price increase), E (the industry
elasticity of demand), and Uc (initial industry quantity demanded).
As before, unit variable margin for the dominant firm is given by
P1-C. The profit reduction due to the loss of quantity resulting
from the price increase is given by
[2] L2 = (P1-C) (R E Uc)
The profit increase the dominant firm gains from raising price
is the price increase multiplied by the quantity the dominant firm
will sell after the price increase. The change in price is R
multiplied by P1. The quantity sold is the dominant firm's initial
quantity, S Uc, less the quantity reductions associated with L1 and
L2, which are (1-S) (Um-Uc) and (R E Uc), respectively. The profit
increase can be written
[3] G = (RPI) [(SUc)-(1-S) (Um-Uc)-(R E Uc)]
The entire profit consequences of the dominant firm strategy can
be expressed as
[4] G-L1-L2 = (R P 1) [(S Uc)-(1-S) (Um-Uc)-(R E Uc)]
-[(P1-C) (1-S) (Um-Uc)]-[(P1-C) (R E Uc)]
From Equation [4] one can find the profit-maximizing price
increase, R*, by taking the first derivative with respect to R,
setting the derivative equal to zero, and solving for R*. The
resulting expression is
[GRAPHIC] [TIFF OMITTED] TN10JN08.058
Results and Sensitivities
Equation [5] can be solved using the parameter values in [Table
3.1]. The model predicts that the profit-maximizing price increase
for a dominant firm under these circumstances would be approximately
48 percent above current levels.
This result should not be viewed as a prediction that price will
necessarily increase by 48 percent above current levels. If price
were to increase by such a large percentage, it is quite possible
that some fringe firms would make investments that would increase
capacity. It is also possible that imported newsprint would become a
significant factor. It also is possible that newsprint purchasers
would consider additional alternatives if price were to increase by
such a large percentage. Conceptually, reactions could be
accommodated in the model by reflecting additional loss of quantity
experienced by the dominant firm.\8\
---------------------------------------------------------------------------
\8\ For instance, suppose that a 5 percent increase in price
would result in a 1 percent loss of sales to imports or expanded
fringe firms. The profit-maximizing price increase for a dominant
firm with a 41.5 percent share would then be 27 percent rather than
48 percent.
---------------------------------------------------------------------------
Several of the parameters in Table K1 are estimated; hence,
their true value could be higher or lower than shown. Significant
further price increases are predicted by the model even if some of
the parameters are altered. As explained above, production cost in
the plants that Abitibi and Bowater would idle when pursuing a
dominant firm strategy would likely be higher than the average cost
used in the model.
However, suppose that the level of variable cost were 20 percent
lower than shown in Table K1. Suppose further that the elasticity of
demand were 20 percent larger than shown in Table K1. With these
changed parameters, the profit-maximizing price increase would still
be 30 percent.
Attachment C--Supplement 1 to the White Paper by Economists
Incorporated, Submitted on Behalf of the NAA to DOJ on July 9, 2007
Economists Incorporated
An Economic Analysis of the Competitive Effects of the Proposed
Abitibi-Bowater Merger
Response to Issues Raised at Our Meeting With the DOJ Staff on
April 20, 2007
Submitted to DOJ on Behalf of NAA
John H. Preston, Kent W. Mikkelsen, Ph.D., Economists Incorporated,
Washington, DC, July 9, 2007.
A. Introduction
On April 11, 2007, Economists Incorporated presented an economic
analysis of the likely competitive effects of the proposed Abitibi-
Bowater merger in the North American (``NA'') newsprint market
(``White Paper'') to the U.S. Department of Justice (``DOJ'') to
assist the Department in its investigation of the proposed merger.
This economic analysis was prepared on behalf of the Newspaper
Association of America (``NAA''), an association of U.S. daily
newspapers.
The evidence we presented to DOJ in the White Paper demonstrates
that Abitibi and Bowater jointly exercised market power to raise
newsprint prices significantly above competitive levels during the
period 2002 to 2006. We do not believe that any alternative
explanation of the aggregate 49% increase in newsprint prices from
the third quarter of 2002 through the third quarter of 2006 is
remotely plausible.\1\ We label our hypothesis that Abitibi and
Bowater jointly exercised market power over the period 2002 to 2006
the ``Dominant Firm Hypothesis.'' \2\ We label the principal
competing hypothesis the ``Competitive Response Hypothesis.''
---------------------------------------------------------------------------
\1\ See especially Section J of the White Paper, which shows
that it is implausible that the newsprint price increases were
primarily due to input cost increases or the appreciation of the
Canadian dollar relative to the U.S. dollar. The analysis in Section
J is based on a comparison of price increases for newsprint and
price increases for several closely related uncoated groundwood
specialty grades over the period 3Q 1999 though 4Q 2006. [Note: When
the Canadian dollar appreciates relative to the U.S. dollar, the
cost of producing newsprint in Canadian mills increases in terms of
U.S. dollars relative to the cost of producing newsprint in U.S.
mills and vice versa. Newsprint is priced in U.S. dollars.J The
implications of the divergence of NA operating rates between the
production of newsprint and the production of uncoated groundwood
specialty grades from 2002 to 2006 are discussed in Section C.3.
below. This divergence in operating rates provides additional
support for the conclusions in Section J of the White Paper.
\2\ Our analysis and evidence for the Dominant Firm Hypothesis
were presented in Sections F through K of the White Paper.
---------------------------------------------------------------------------
On April 20, 2007, we met with the DOJ staff investigating the
proposed merger to discuss our White Paper. In our discussion with
DOJ, several questions were raised concerning our analysis and
evidence regarding the joint exercise of market power by Abitibi and
Bowater. One staff member suggested that the rise in the price of
newsprint might be explained as a competitive response by newsprint
producers to the appreciation of the Canadian dollar relative to the
U.S. dollar. Another staff member asked whether the maximum
practical operating rate for the production of uncoated groundwood
specialty grades might be lower than the maximum practical operating
rate for the production of newsprint.\3\ The staff also asked us if
the
[[Page 32925]]
acceleration in the rate of decline of NA newsprint consumption \4\
might eliminate the ability of a merged Abitibi-Bowater to engage in
the type of anticompetitive behavior that we had alleged.
---------------------------------------------------------------------------
\3\ During our meeting with DOJ, we had pointed out that the
significantly lower price increases for uncoated groundwood
specialty grades compared to the price increases for newsprint over
the period 2002 to 2006 could be largely explained by the
significantly lower operating rates for uncoated groundwood
specialty grades. See Section J of the White Paper and Section C.3.
below.
\4\ See the NA newsprint consumption and production statistics
for the first five months of 2007 presented in Section B.1.a below.
---------------------------------------------------------------------------
We divide our response to issues raised by the DOJ staff into
the following five sections:
Section A. Introduction
Section B. Events Since the Merger Was Announced in January 2007
Confirm the Dominant Firm Hypothesis
1. In 2007, NA Newsprint Demand and Prices Have Declined
Significantly While the Value of the Canadian Dollar Relative to the
U.S. Dollar Has Increased Significantly
2. Abitibi and Bowater Have Not Taken Significant Actions To
Remove Newsprint Capacity From the Market Since They Announced Their
Merger in January 2007
3. Newsprint Industry Analysts and Competitors of Abitibi and
Bowater Do Not Expect Abitibi and Bowater To Take Any Significant
Action To Remove Newsprint Capacity From the Market Until After They
Have Merged
Section C. Additional Evidence That Abitibi and Bowater Exercised
Market Power Over the Period 2002 to 2006
1. Based on Publicly Available Information, the Cash Costs of NA
Newsprint Mills Were Below the Price of Newsprint in 2003 and 2005.
2. Based on Publicly Available Information, the Cash Costs of NA
Newsprint Mills Were Below the Price of Newsprint in 4Q 2006
3. A Comparison of Operating Rates for Newsprint and Uncoated
Groundwood Specialty Grades 1999 to 2006
Section D. Additional Analysis Based on the Dominant Firm Model (DFM)
Including a Revision of the DFM Designed To Consider Multi-Period
Dynamics
1. Introduction
2. The Relevance of a Paper by Matthew Gentzhow to Our
Conclusions Regarding the DFM
3. Would the Dominant Firm Strategy Be Profitable for Abitibi or
Bowater Acting Independently?
4. What Are the Effects on Dominant Firm Behavior of a Decline
in Demand?
5. A Description of a Revision of the DFM Designed to Consider
Multi-period Dynamics
Section E. Conclusion
B. Events Since the Merger Was Announced in January 2007 Confirm the
Dominant Firm Hypothesis
1. In 2007, NA Newsprint Demand and Prices Have Declined
Significantly While the Value of the Canadian Dollar Relative to
the U.S. Dollar Has Increased Significantly
a. NA Newsprint Demand Declined Significantly During the First Five
Months of 2007
Table 1 below shows the percentage change in selected newsprint
statistics for the first five months of 2007 compared to the first
five months of 2006.\5\ Table 1 also shows the percentage change in
selected newsprint statistics for the twelve months of 2006 compared
to the twelve months of 2005.
---------------------------------------------------------------------------
\5\ See the Pulp and Paper Products Council (``PPPC'') Newsprint
Flash Reports for May 2007, issued June 21, 2007, and December 2007,
issued January 25, 2007. As apparently calculated by the PPPC, NA
demand equals shipments from NA mills to NA customers plus imports
from overseas mills to NA customers.
Table 1.--Percentage Change From Prior Year for Selected PPPC Newsprint
Statistics--May 2007 YTD vs. May 2006 YTD and December 2006 YTD vs.
December 2005 YTD
------------------------------------------------------------------------
Percent change
Percent change December 2006
May 2007 year- year-to-date
PPPC newsprint flash report category to-date vs. vs. December
May 2006 year- 2005 year-to-
to-date date
------------------------------------------------------------------------
Total NA Demand......................... -10.8 -6.0
Consumption by U.S. Dailies............. -9.1 -7.1
Imports from Overseas Mills............. -51.3 -25.2
Shipments from NA Mills to NA Customers. -10.1 -5.6
Shipments by NA Mills to Overseas 5.6 -9.8
Customers..............................
Total Shipments by NA Mills............. -7.3 -6.4
------------------------------------------------------------------------
During the period January 2007 to May 2007 NA demand declined by
10.8% compared to the first five months of 2006 and consumption by
U.S. daily newspapers declined by 9.1%. Imports of newsprint from
overseas mills to NA customers declined by 51 .3% to an annual rate
of 79,000 metric tonnes. At this rate, imports will account for 0.8%
of NA demand in 2007.\6\
---------------------------------------------------------------------------
\6\ In the White Paper, we concluded that significant imports by
NA customers from new Chinese newsprint capacity were unlikely. See
Section BA. of the White Paper for our analysis. The import
statistics for the first five months of 2007 support that
conclusion.
---------------------------------------------------------------------------
Table 1 also shows that shipments by NA newsprint mills to NA
customers declined by 10.1% over the first five months of 2007.
Partially offsetting the decline in shipments to NA customers,
exports from NA mills to overseas customers increased by 5.6%. Total
shipments by NA mills to both NA customers and overseas customers
were down 7.3% for the five-month period.
Since March 2007, there has been a gradual improvement in NA
demand and total shipments from NA mills to NA customers and
overseas customers.\7\ See Table 2 below.
---------------------------------------------------------------------------
\7\ See PPPC Flash Reports for March and April 2007. This is a
``gradual improvement'' in the sense that the decline in NA demand
and total shipments from NA mills was lower in April and May 2007
compared to the first three months of 2007.
Table 2.--Percentage Change from Prior Year for Selected PPPC Newsprint Statistics--January 2007, February 2007,
March 2007, April 2007 and May 2007
----------------------------------------------------------------------------------------------------------------
Percent change Percent change
PPPC newsprint flash report January 2007 February 2007 Percent change Percent change Percent change
category vs. January vs. February March 2007 vs. April 2007 vs. May 2007 vs.
2006 2006 March 2006 April 2006 May 2006
----------------------------------------------------------------------------------------------------------------
Total NA Demand................. -10.5 -12.7 -13.4 -9.7 -8.7
Consumption by U.S. Dailies..... -9.1 -9.4 -8.7 -9.8 -9.2
Imports from Overseas Mills..... -58.1 -47.3 -62.6 -38.4 -68.7
Shipments from NA Mills to NA -9.6 -12.3 -12.6 -9.4 -7.2
Customers......................
Shipments by NA Mills to -17.2 10.1 7.0 -0.5 29.0
Overseas Customers.............
[[Page 32926]]
Total Shipments by NA Mills..... -10.8 -9.9 -8.6 -7.7 -0.7
----------------------------------------------------------------------------------------------------------------
Between January 2007 and March 2007, the rate of decline in
total NA newsprint demand was higher than previously. In March 2007,
NA demand was down 13.4% compared to March 2006. However, in April
and May 2007, the rate of decline slowed. By May 2007, the decline
in NA demand dropped to 8.7%.\8\ The decline in shipments from NA
mills to NA customers was almost cut in half: a decline of 12.6% in
March vs. a decline of 7.2% in May.\9\ In May 2007, total shipments
from NA mills were down only 0.7% compared to May 2006 due to both
the improvement in shipments to NA customers and strong export
growth. After falling to 93% in March and April 2007, the operating
rate for NA mills increased to 94% in May 2007. In 2006, the
operating rate was 95% for all three months.
---------------------------------------------------------------------------
\8\ The decline in consumption by U.S. daily newspapers did not
change significantly over the three months: -8.7% in March, -9.8% in
April, and -9.2% in May.
\9\ The decline in imports was reduced from -62.6% in March to -
38.4% in April before increasing to -68.7% in May.
---------------------------------------------------------------------------
A comparison of the two columns in Table I reflects the gradual
improvement in newsprint operating results over the period March
2007 to May 2007. The decline in consumption by U.S. daily
newspapers increased from 7.1% for the twelve months of 2006 to 9.1%
for the first five months of 2007, an increase of 2.0%. The decline
in total shipments from NA newsprint mills increased from 6.4% for
the twelve months of 2006 to 7.3% for the twelve months of 2006, and
increase of 0.9%. Operating rates at NA newsprint mills for both the
first five months of 2007 and the twelve months of 2006 were 94%.
We conclude that while there has been a modest increase in the
rate of decline in newsprint consumption by U.S. daily newspapers
for the first five months of 2007 compared to the twelve months of
2006, the overall operating results for NA newsprint mills over the
two periods are not significantly different. As Table 2 shows, the
operating results between 2006 and 2007 have been narrowing over the
period March to May, not widening.
b. NA Newsprint Prices Declined Significantly During the First Five
Months of 2007 While the Value of the Canadian Dollar Increased
Significantly
The price of newsprint (30 lb, Eastern U.S.) reached a peak of
$675 per metric tonne in May 2006 and stayed at $675 through
September 2006 before declining gradually to $660 in December 2006.
From December 2006 to June 2007, the NA newsprint price fell $75 to
$575, a decline of 11.4%.\10\
---------------------------------------------------------------------------
\10\ The source for the monthly newsprint prices is the RISI
publication Pulp & Paper Week.
---------------------------------------------------------------------------
While the price of newsprint was declining by 11.4% between
December 2006 and June 2007, the value of the Canadian dollar was
increasing 8.2% from $0.868 per U.S. dollar in December 2006 to
$0.939 per U.S. dollar in June 2007.'' \11\
---------------------------------------------------------------------------
\11\ The source for the average monthly exchange rates is
FXHistory: Historical currency exchange rates, Oanda.com.
---------------------------------------------------------------------------
The RISI Pulp & Paper Week edition of May 21, 2007 shows a chart
on page 11 comparing the price of newsprint on one vertical axis
with the value of the Canadian dollar per U.S. dollar on the other
vertical axis from May 2005 to April 2007. The chart shows both
values tracking each other fairly closely in the 20 months from May
2005 through December 2006. From January 2007 through April 2007 the
two values continuously diverge with the value of the Canadian
dollar steadily increasing and the price of newsprint steadily
decreasing.
Chart I below is an adaptation of the Pulp & Paper Week chart.
It shows the percentage change from the respective May 2005 values
for both the price of newsprint \12\ and the exchange rate for the
Canadian dollar in terms of U.S. dollars.\13\ Between May 2005 and
December 2006, the maximum difference between the two series in any
month was 3.3%. In December 2006 the percentage changes from their
respective May 2005 values were almost identical (a 9.1% increase
for the price of newsprint and a 9.0% increase for the value of the
Canadian dollar). In January 2007, the two series began to diverge.
As Chart 1 shows, the divergence reached 21.2% in June 2007 as the
value of the Canadian dollar increased to 17.9% above its May 2005
value and the price of newsprint declined to 3.3% below the May 2005
price.
---------------------------------------------------------------------------
\12\ Source: RISI publication Pulp & Paper Week.
\13\ Source: FXHistory: Historical currency exchange rates,
Oanda.com.
---------------------------------------------------------------------------
[[Page 32927]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.055
c. Implications of the Recent Decline in NA Newsprint Demand and Price
and the Appreciation of the Canadian Dollar
Between the third quarter of 2002 and the third quarter of 2006,
the price of NA newsprint rose an aggregate of 49% despite a steady
decline in NA newsprint consumption.\14\ As we argued in the White
Paper, the strategic closure of newsprint capacity by Abitibi and
Bowater was a joint exercise of market power responsible for the
price increases. We believe these actions and their effects are well
documented in the White Paper. As an alternative to the Dominant
Firm Hypothesis, the Competitive Response Hypothesis asserts that
the price increases are due to competitive responses to the
appreciation of the Canadian dollar and increases in the prices of
inputs.
---------------------------------------------------------------------------
\14\ See Chart E6 on p. 71 of the White Paper which shows
steadily rising newsprint prices in the face of steadily declining
newsprint demand.
---------------------------------------------------------------------------
As discussed below, since the merger announcement in early
January and likely several months earlier, Abitibi and Bowater have
stopped strategically closing capacity to raise the price of
newsprint. In our view and the view of newsprint industry analysts
and newsprint competitors of Abitibi and Bowater,\15\ the reason
that Abitibi and Bowater have stopped strategically closing capacity
is the concern that it could very well lead to the rejection of the
merger by U.S. and/or Canadian antitrust authorities. It is also our
view and the view of newsprint industry analysts and newsprint
competitors of Abitibi and Bowater\16\ that if the merger is
approved in the U.S. and Canada, a merged AbitibiBowater will take
the actions necessary to restore the ``balance'' between newsprint
demand and supply to again raise the price of newsprint above
competitive levels.
---------------------------------------------------------------------------
\15\ See Section B.3.a. below.
\16\ See Section B.3a. below.
---------------------------------------------------------------------------
The current decline in newsprint prices is the true competitive
response to the decline in NA newsprint demand. In our view, the
decline in newsprint prices is occurring because Abitibi and Bowater
perceive it would be imprudent to close significant capacity during
the merger review period. The current decline in newsprint prices is
indicative of the declines that would have occurred over the period
2002 to 2006 had Abitibi and Bowater not intervened with their
strategic removal of capacity.
The widening divergence between the percentage change in the
appreciation of the Canadian dollar and the percentage change in NA
newsprint prices from December 2006 to June 2007 as shown in Chart I
is further evidence that the correlation between the appreciation of
the Canadian dollar and the rise in the price of newsprint in prior
years was due to the strategic behavior of Abitibi and Bowater and
was not a competitive response to the appreciation.
Of course, higher newsprint costs must be reflected in newsprint
prices and, as newsprint demand declines, the highest cost capacity
will be forced to exit from the market. In 2007, we observe
newsprint prices approaching or dropping below the cash costs of the
highest cost mills. One mill (the Blue Heron Pomona, CA mill) has
been indefinitely idled because it apparently can no longer cover
its cash costs. In our view, the operation of the NA newsprint
market in the face of declining demand in 2007 is reflective of a
competitive market due to the temporary absence of the exercise of
market power by Abitibi and Bowater.\17\
---------------------------------------------------------------------------
\17\ According to a RISI news note dated June 29, 2007, Kruger
announced a $25 per metric tonne price increase for 30 lb. newsprint
effective September 1, 2007, According to RISI, ``Kruger is North
America's fourth-largest newsprint producer in terms of capacity
with 1.15 million tonnes/yr of production, all of it located in
tEasterni Canada. Contacts said it was the first time they could
remember that the company had sought to initiate a price increase
round.'' We view Kruger's announced price increase as a competitive
response primarily to the appreciation of the Canadian dollar, an
action taken in the absence of the exercise of market power by
Abitibi and Bowater since their merger announcement in January 2007.
It is plausible that NA newsprint prices have fallen close to the
cash costs of one or more Kruger newsprint mills, necessitating the
price increase announcement. See Section C.2. below for a discussion
of 4Q 2006 cash costs of NA newsprint mills. Whether the price
increase announced by Kruger will be successfully implemented or not
will depend mainly on the amount of excess capacity at NA newsprint
mills in September and succeeding months.
---------------------------------------------------------------------------
2. Abitibi and Bowater Have Not Taken Significant Actions To Remove
Newsprint Capacity from the Market Since the Merger Was Announced
in January 2007
Abitibi and Bowater began their merger discussions in June 2006
and concluded them with their merger announcement on January 29,
2007. As antitrust economists, we would expect that during the
merger review by regulatory authorities neither Abitibi nor Bowater
would take any actions that could be construed by antitrust
regulators as anticompetitive, including the significant removal of
capacity from the market to raise the price of newsprint.\18\ It is
likely that even before January 29, 2007, Abitibi and Bowater felt
constrained from taking actions to
[[Page 32928]]
aggressively remove capacity from the market.
---------------------------------------------------------------------------
\18\ See ``Background of the Combination,'' in AbitibiBowater
Amendment 3 to the Form S-4 Registration Statement (``Form S-4'')
filed with the SEC. June 4, 2007 pp. 70-78.
---------------------------------------------------------------------------
We are not aware of any actions by Abitibi since June 2006 to
indefinitely idle or permanently shut down newsprint capacity. No
such actions are identified in the Abitibi 2006 Annual Report or in
Abitibi's report on its 2007 first quarter results,\19\ nor are we
aware of any such actions identified in the trade press. In March
2007, Bowater indefinitely idled the No. 3 newsprint machine at its
Gatineau, QC mill due to weak demand and increasing costs of
recycled fiber and took downtime at other unidentified newsprint
mills.\20\
---------------------------------------------------------------------------
\19\ See Abitibi 2006 Annual Report, pp-23-34 and Abitibi First
Quarter 2007 Report to Shareholders, pp. 6-7. In June 2007, Abitibi
shut down its Grand Falls, NL mill for three weeks to repair the
damage from a fire at the mill. See RISI news note, June 21, 2007.
\20\ See the Bowater 10-Q Report for 1Q 2007, p. 19. According
to RISI economist Kevin Conley, ``Bowater is also responding to the
sharp decline in demand and rapid rise in fiber prices, curtailing
newsprint production at their Gatineau mill in Quebec. The company
also stated they have selected other machines for downtime that are
heavily dependent on recycled fiber.'' See ``Surviving the downturn
in North American newsprint'', by Kevin Conley, RISI Economist, RISI
News Service, April 19, 2007. The newsprint capacity of the No. 3
machine at the Gatineau mill is approximately 115,000 metric tonnes
per year. Bowater also indefinitely idled its No. 4 newsprint
machine at its Thunder Bay, ON mill in September 2006. See the
Bowater 10-Q Report for 1Q 2007, p. 19 and p. 23. Bowater
subsequently stated that it would restart this paper machine in May
2007 producing specialty grades rather than newsprint. The newsprint
capacity of the Thunder Bay machine was 146,000 metric tonnes.
---------------------------------------------------------------------------
These actions by Bowater, however, fall far short of the
capacity removals needed to restore the ``balance'' between NA
newsprint supply and demand. According to RISI economist Kevin
Conley, ``At this point, the announced reduction in North American
supply [i.e., the closure of Blue Heron's Pomona, CA mill and
Bowater's curtailment of production at its Gatineau, QC mill] could
not possibly keep pace with the continued decline in North American
demand.'' \21\
---------------------------------------------------------------------------
\21\ ``Surviving the downturn in North American newsprint'', by
Kevin Conley, RISI Economist, RISI News Service, April 19, 2007. In
our view, the current idling of newsprint capacity at Bowater's
Gatineau, QC mill, is a competitive response and not a strategic
capacity closure in pursuit of a joint dominant firm strategy.
---------------------------------------------------------------------------
3. Newsprint Industry Analysts and Competitors of Abitibi and
Bowater Do Not Expect Abitibi and Bowater to Take Any Significant
Action to Remove Newsprint Capacity from the Market Until After
They have Merged
a. Comments in the Trade Press
(1) ``We would expect that Abitibi and Bowater will be focused
primarily on closing the merger, and therefore, unlikely in our
opinion to rationalize any newsprint capacity in IH 2007,'' Goldman
Sachs analyst Richard Skidmore told investors.\22\
---------------------------------------------------------------------------
\22\ ``Market abuzz over merger: concerns center on pricing and
customer relationships,'' Pulp & Paper Week, February 5, 2007, p.
11.
---------------------------------------------------------------------------
(2) ``No one will close any capacity because they figure
AbitibiBowater will do it for them. And Abitibi and Bowater will
figure they can't be too aggressive on pricing or close capacity
until their deal closes,'' said one contact.\23\
---------------------------------------------------------------------------
\23\ ``Market abuzz over merger: concerns center on pricing and
customer relationships,'' Pulp & Paper Week, February 5, 2007, p.
11.
---------------------------------------------------------------------------
(3) North American newsprint capacity now exceeds orders,
resulting in a declining market. Salman Partners indicated that the
majority of newsprint producers are waiting to see what will happen
after the merger of Abitibi-Consotidated Inc. with Bowater Inc.
later this year before making any decisions on shutdowns.\24\
---------------------------------------------------------------------------
\24\ ``Steeper Decline in Newsprint Data Reported in February,''
Debra Garcia, Editor & Publisher, March 28, 2007.
---------------------------------------------------------------------------
(4) At this point, the announced reduction in North American
supply could not possibly keep pace with the continued decline in
North American demand. It appears producers are waiting for the
Abitibi/Bowater merger to be finalized in the hope that the new
company will close necessary capacity to balance the market and
bring an end to falling newsprint prices. However, this merger of
North America's two largest newsprint producers will not be
completed until the third quarter of 2007, at the earliest.\25\
---------------------------------------------------------------------------
\25\ ``Surviving the downturn in North American newsprint'', by
Kevin Conley, RISI Economist, RISI News Service, April 19, 2007.
---------------------------------------------------------------------------
(5) Other suppliers are hoping the union of the two companies
will go through smoothly in anticipation that AbitibiBowater will
quickly make the industry's capacity cuts. They see it as a silver
bullet for the whole industry, allowing them to reap the benefits of
a tighter North American paper market without the necessity of
cutting production themselves.\26\
---------------------------------------------------------------------------
\26\ ``The making of a merger: secret talks that could have
derailed AbitibiBowater deal set tantalizing questions for
analysts,'' Pulp & Paper Week, May 7, 2007, p.8. The title of the
Pulp & Paper Week article refers to other strategic options Bowater
was considering as alternatives to a merger with Abitibi. According
to the AbitibiBowater Form S-4 filing: ``Throughout the period from
July 2006 through December 2006, Bowater continued to consider a
wide range of strategic alternatives with third parties, including
acquisitions of assets or businesses and sales or distributions of
certain of its businesses, and members of senior management had
informal discussions with their counterparts at other paper
companies. Bowater's Board of Directors was regularly updated on the
status of these discussions. These discussions did not advance
beyond intermediate stages in respect of transactions that would
have precluded a combination with Abitibi. In August 2006, Bowater
commenced discussions with a paper producer regarding a possible
transaction in which Bowater would acquire the paper producer and
possibly either sell or spin-off its newsprint assets. However, due
to significant tax and structuring issues that would have made
execution difficult and potentially adversely impact shareholder
value, as well as significantly differing views as to the parties'
respective valuations, the parties determined not to proceed with
discussions regarding a possible transaction. During this period,
Bowater also explored the potential sale of certain of its newsprint
assets to another newsprint manufacturer. These discussions were
terminated in January 2007.'' See AbitibiBowater Amendment 3 to the
Form S-4 Registration Statement (``Form S-4'') filed with the SEC.
June 4, 2007, p. 71.
---------------------------------------------------------------------------
(6) Dillon expected a further newsprint price hike attempt later
this year, despite the sluggish market. To be successful, the two
biggest producers, Abitibi and Bowater, would have to support it,
and that is not likely to occur until after the merger is completed
``due to concerns that such a move might he misread by regulators,''
said Dillon.\27\
---------------------------------------------------------------------------
\27\ ``Newsprint Prices Continue to Sink,'' Debra Garcia, Editor
& Publisher, July 5, 2007. Chip Dillon is a newsprint industry
analyst with Citigroup Global Markets.
---------------------------------------------------------------------------
b. Implications of Comments in the Trade Press
From the trade press commentary above, it is apparent that
newsprint industry analysts and newsprint competitors of Abitibi and
Bowater are waiting for the merger to be completed in anticipation
that a merged Abitibi-Bowater will increase NA newsprint prices by
shutting down enough newsprint capacity to create a tight market. It
is also apparent that these same analysts and competitors believe
that Abitibi and Bowater will not take any significant actions to
remove capacity from the market until after their merger review is
completed ``due to concerns that such a move might be misread by
regulators.'' \28\
---------------------------------------------------------------------------
\28\ ``Newsprint Prices Continue to Sink,'' Debra Garcia, Editor
& Publisher, July 5, 2007.
---------------------------------------------------------------------------
C. Additional Evidence That Abitibi and Bowater Exercised Market Power
Over the Period 2002 to 2006
1. Based on Publicly Available Information, the Cash Costs of NA
Newsprint Mills Were Below the Price of Newsprint in 2003 and 2005
a. Description of RISI Newsprint Cash Cost Benchmarking Studies 2003
and 2005
RISI conducts periodic cost benchmarking studies analyzing the
cash cost of producing newsprint for each NA newsprint mill.\29\ The
supply curve for NA newsprint can be shown by arraying the cash
costs by NA mill in ascending order.
---------------------------------------------------------------------------
\29\ See the RISI Web site for more mformation on these
benchmarking studies. RISI publishes these studies every two years.
RISI also provides quarterly updates by CD. In addition, RISI
provides cash cost benchmarking studies by newsprint machine. While
NAA has not acquired any of the newsprint cost benchmarking studies
($12,500 for the 2006 NA newsprint mill study), we expect that the
studies are available to DOJ from Abitibi, Bowater and other
newsprint manufacturers through the discovery process.
---------------------------------------------------------------------------
Chart 2 below compares the cash costs for NA mills in 2003 and
2005. Chart 2 has been adapted from a report by a Canadian
securities analyst for CIBC World Markets (``CIBC report'') \30\ The
vertical axis shows the
[[Page 32929]]
cash costs per metric tonne of newsprint in U.S. dollars for each NA
mill in 2003 and 2005. The horizontal axis of Chart 2 shows the
capacity per NA newsprint mill in 2003 and 2005 arrayed from lowest
cost mill to highest cost mill. Each vertical bar represents one
mill. The paler vertical bars in the foreground of the chart
represent the capacities and cash costs of NA newsprint mills in
2003. The vertical darker bars in the background of the chart
represent the capacities and cash costs of NA newsprint mills in
2005. As the chart shows, the mill locations in 2003 and 2005 are
identified by region: Canada West, Canada East, U.S. Northeast, U.S.
South, and U.S. West. The mills were not further identified in Slide
35 of the CIBC Report, but the mill owners and specific mill
locations (as opposed to regional locations) are identified in the
underlying paperloop.com cost benchmarking study available from
RISI.
---------------------------------------------------------------------------
\30\ See ``World Newsprint Market: Winners and Losers,'' by Don
Roberts, Managing Director, CIBC World Markets, April 24, 2006,
Slide 35. CIBC World Markets was retained by Abitibi in June 2006 as
its financial advisor with respect to the proposed merger with
Bowater. See Form S-4, p. 70. The CIBC report states that the source
for the cost curve comparison is ``Paperloop Benchmarking Service,''
a predecessor to RISI. We have added the four text boxes to the left
of the chart and the two text boxes to the right. In addition, we
have added the two horizontal green lines and the two horizontal red
lines at the top of the chart.
---------------------------------------------------------------------------
A chart appearing in this comment is not able to be reprinted
here. Copies of the comment with the chart are available at the
Department of Justice Antitrust Division web site, http://
www.usdoj.gov/atr, at the Antitrust Documents Group of the
Department of Justice Antitrust Division, 450 Fifth Street, NW.,
Suite 1010, Washington, DC 20530, (202) 514-2481, and at the Office
of the Clerk of the United States District Court for the District of
Columbia, 333 Constitution Avenue, NW., Washington, DC 20001.
Chart 2 shows a reduction in NA newsprint capacity of about 1.4
million metric tonnes between 2003 and 2005. The aggregate NA
capacity shown for 2003 is about 13.5 million metric tonnes and the
aggregate NA capacity shown for 2005 is about 12.1 million metric
tonnes. In Chart 2, the number of NA newsprint mills declined from
48 in 2003 to 44 in 2005.
In 2003 and 2005, Chart 2 shows that most of the highest cost
mills in NA were located in Eastern Canada. In 2005, the top half of
the cost curve is dominated by Eastern Canadian mills with the
exception of one U.S. Northeast mill, three U.S. West mills, and one
Western Canadian mill. The bottom half of the cost curve in 2005 is
dominated by mills located in the U.S. South and in Western Canada.
Between 2003 and 2005, the cost disadvantage of mills in Eastern
Canada increased relative to other NA mills, particularly those
mills located in the U.S. South. CIBC attributes this increased cost
disadvantage ``largely to the strong C$,'' stating that the ``15%
appreciation of the C$ made the cost curve steeper--up another 5%
since then.'' \31\
---------------------------------------------------------------------------
\31\ See CIBC Report, Slide 35.
---------------------------------------------------------------------------
CIBC Slide 35 does not identify the quarter in which the NA mill
cash costs were estimated for either the 2003 or 2005 newsprint cost
benchmarking studies. In Chart 2, the two horizontal green lines
that we have drawn show the NA newsprint price (30 lb., Eastern
U.S.) for 1Q 2003 ($475 per metric tonne) and 4Q 2003 ($527 per
metric tonne).\32\ As indicated by the lower text box on the right
hand side of the chart, the highest mill cash cost in 2003 was about
$430 per metric tonne, which was $45 per metric tonne lower than the
1Q 2003 newsprint price and $97 per metric tonne lower than the 4Q
2003 newsprint price.
---------------------------------------------------------------------------
\32\ The source of the quarterly newsprint prices is the RISI
2006 Fact & Price Book, p. 150. The price of newsprint increased in
each quarter of 2003. See Chart E6 on p. 71 of the White Paper.
---------------------------------------------------------------------------
In Chart 2, the two horizontal red lines that we have drawn show
the NA newsprint price (30 lb., Eastern U.S.) for 1Q 2005 ($580 per
metric tonne) and 4Q 2005 ($637 per metric tonne).\33\ As indicated
by the upper text box on the right hand side of the chart, the
highest mill cash cost in 2005 was about $510 per metric tonne which
was $70 per metric tonne lower than the 1Q 2005 newsprint price and
$127 per metric tonne lower than the 4Q 2005 newsprint price.
---------------------------------------------------------------------------
\33\ The source of the quarterly newsprint prices is the RISI
2006 Fact & Price Book, p. 150. The price of newsprint increased in
each quarter of 2005. See Chart E6 on p. 71 of the White Paper.
---------------------------------------------------------------------------
b. Implications of the RISI 2003 and 2005 Cash Cost Studies
The newsprint capacity removals by Abitibi and Bowater during
the period 2002 to 2006 are analyzed in Sections F through H of the
White Paper. During that time Abitibi and Bowater combined capacity
removals accounted for 80.8% of total NA capacity removals. Catalyst
accounted for 7.3% of the capacity removals and two firms that
exited from the newsprint market to produce uncoated groundwood
specialty grades accounted for 9.7%. The other thirteen newsprint
manufacturers that remain in the NA newsprint market today accounted
for just 2.2% of the capacity removals.\34\
---------------------------------------------------------------------------
\34\ See Chart G3 on p.91 of the White Paper.
---------------------------------------------------------------------------
If the variable cost of the newsprint capacity that Abitibi and
Bowater removed from the market during the period 2002 to 2006 was
less than the price of newsprint, that capacity removal would be
consistent with the hypothesis that Abitibi and Bowater were jointly
exercising market power. Firms in competitive markets do not
generally remove capacity from the market if that capacity is
generating positive profit margin (i.e., when price exceeds variable
cost).
Chart 2 above shows that the price of newsprint exceeded the
2003 cash cost of all NA newsprint mills in 1Q 2003 and 4Q 2003.\35\
Similarly, Chart 2 shows that the price of newsprint exceeded the
2005 cash cost of all NA newsprint mills in 1Q 2005 and 4Q 2005.\36\
Due to the limitations of Chart 2 discussed above, these results
strongly suggest but do not prove that the cash cost of the
newsprint capacity Abitibi and Bowater removed from the market
during this period was less than the price of newsprint at the time
of the capacity removal. However, as we pointed out at our meeting
with the DOJ staff, DOJ should be able to determine if the cash cost
of the capacity removed by Abitibi and Bowater was less than the
price of newsprint at the time of the capacity removal with
information available to DOJ through the discovery process,
including the RISI NA newsprint mill cash cost benchmarking studies.
Such a determination would provide additional evidence that the
capacity removals were an exercise in market power in pursuit of
their dominant firm strategy.
---------------------------------------------------------------------------
\35\ Since the price of newsprint increased in each quarter of
2003, the price exceeded the 2005 cash cost of each mill in the
second and third quarters of 2003 as well.
\36\ Since the price of newsprint increased in each quarter of
2005, price exceeded the 2005 cash cost in the second and third
quarters of 2005 as well.
---------------------------------------------------------------------------
2. Based on Publicly Available Information, the Cash Costs of NA
Newsprint Mills Were Below the Price of Newsprint in 4Q 2006
a. Description of RISI Newsprint Cash Cost Benchmarking Study 4Q 2006
Chart 3 below shows cash costs of NA mills in 4Q 2006. The chart
is adapted from a chart that appeared in a RISI article in April
2007.\37\
---------------------------------------------------------------------------
\37\ See ``Surviving the downturn in North American newsprint''
by Kevin Conley, senior economist, RISI, April 19, 2007.
---------------------------------------------------------------------------
[[Page 32930]]
[GRAPHIC] [TIFF OMITTED] TN10JN08.056
The interpretation of Chart 3 is similar to the interpretation
of Chart 2 except that Chart 3 doesn't provide a color code to
identify mills by region. As in Chart 2, the mill owners and
specific mill locations are not identified. In Chart 3, 43 newsprint
mills are shown and the aggregate NA total capacity is 11.9 million
metric tonnes. The highest cost mill has a cash cost of about $630
per metric tonne which is $30 lower than the December 2006 newsprint
price of $660 (30 lb., Eastern U.S.) as reported by RISI Pulp &
Paper Week. The December 2006 newsprint price is indicated by the
horizontal red line at the top of Chart 3.
b. Implications of the RISI 4Q 2006 Cash Cost Study
Since 4Q 2006, the price of newsprint has dropped from $660 per
metric tonne to $585 in June 2007. In March 2007, Blue Heron
announced that it would be indefinitely idling its Pomona, CA mill
due primarily to significant increases in the cost of recycled fiber
over the past year.\38\ It seems likely that the high cost mill in
Chart 3 at about $630 per metric tonne is the Blue Heron Pomona
mill. If so, when the price of newsprint dropped below $630 to $625
in March 2007, the variable cost of production at the Pomona plant
exceeded the price of newsprint.
---------------------------------------------------------------------------
\38\ The Blue Heron Pomona plant is a 100% recycled fiber plant.
In March 2007, Bowater announced that it was indefinitely idling a
newsprint machine at its Gatineau, QC mill due to high recycled
fiber costs. See ``Surviving the downturn in North American
newsprint'' by Kevin Conley, senior economist, RISI, April 19, 2007.
SP newsprint, which also relies heavily on recycled fiber at its two
mills, recently announced that it was evaluating its strategic
options, including a possible sale of the two mills. One mill is
located in Oregon and the other is located in Georgia. See RISI news
note, May 17, 2007.
---------------------------------------------------------------------------
3. A Comparison of Operating Rates for Newsprint and Uncoated
Groundwood Specialty Grades 1999 to 2006
Section J of the White Paper compared newsprint prices with the
prices of uncoated groundwood specialty grades 3Q 1999 to 4Q 2006.
We showed that price increases for newsprint between 2002 and 2006
greatly exceeded price increases for three of four uncoated
groundwood specialty grades for which data were available.\39\ Since
these three uncoated groundwood specialty grades were more adversely
affected by the increase in input prices and the appreciation of the
Canadian dollar than newsprint was over the period 2002 to 2006,\40\
we would expect to see greater price increases for these uncoated
groundwood specialty grades than for newsprint if the price
increases for newsprint were competitively determined. The fact that
the price increases for these uncoated groundwood specialty grades
were considerably lower than the price increases for newsprint over
this period contradicts the hypothesis that the newsprint price
increases were a competitive response to input price increases and
the appreciation of the Canadian dollar and confirms the Dominant
Firm Hypothesis that the newsprint price increases were due to the
joint exercise of market power by Abitibi and Bowater.
---------------------------------------------------------------------------
\39\ The price changes were measured as a percentage of their
respective 3Q 1999 prices. There was one exception to the
significant divergence between newsprint prices and the prices of
uncoated groundwood specialty grades over the period 2002 to 2006.
The price of Hi-Brite 65 showed a similar increase to that of
newsprint. The explanation for this similarity appears to be that
Abitibi and Bowater are also dominant in the production of Hi-Brite
grades. See p. 115 of the White Paper and Table B7 in Attachment B
of the White Paper.
\40\ See the discussion on pages 110-112 of the White Paper. In
addition, demand for uncoated groundwood specialty grades was
growing over the period 2002-2006 whereas the demand for newsprint
was declining. Other things equal, these divergent growth rates
should have led to higher price increases for uncoated groundwood
specialty grades than for newsprint.
---------------------------------------------------------------------------
During our meeting with DOJ, we pointed out that the
significantly lower price increases for uncoated groundwood
specialty grades compared to the price increases for newsprint over
the period 2002 to 2006 could be largely explained by the
significantly lower operating rates for uncoated groundwood
specialty grades. We were asked by the DOJ staff if the maximum
practical operating rate for the production of uncoated groundwood
specialty grades might be lower than the maximum practical operating
rate for the production of newsprint.
Chart 4 below shows that the operating rates for both newsprint
and uncoated groundwood specialty grades were nearly identical from
1999 to 2001 before diverging in 2002.\41\ In 1999 and 2000, the
operating
[[Page 32931]]
rates for both newsprint and uncoated groundwood specialty grades
were 95% and 97% before falling to 90% in 2001. In 2002, the
operating rate for newsprint exceeded the operating rate for
uncoated groundwood specialty grades by 1%. This gap widened to 3%
in 2003 and 6% in 2004 before narrowing to 2% in 2005 and 1% in
2006. These results show that high maximum practical operating rates
are similarly attainable for uncoated groundwood specialty grades
and provide further support for the hypothesis that the
significantly greater increase in newsprint prices over the period
2002 to 2006 was due to the joint exercise of market power by
Abitibi and Bowater.
---------------------------------------------------------------------------
\41\ Sources for Chart 4: (a) Newsprint operating rates 1999 to
2003 from PPPC North American Newsprint Statistics Monthly Bulletin,
December 2001 to December 2004, and PPPC Newsprint Flash Reports,
December 2005 and December 2006; (b) Uncoated groundwood specialty
grade statistics from RISI Fact and Price Book, p. 164. The relevant
statistics for the U.S. and Canada have been combined to calculate
an NA operating rate for uncoated groundwood specialty grades for
the period 1999 to 2006. The source for the uncoated groundwood
specialty grades has been previously provided by NAA to DOJ.
[GRAPHIC] [TIFF OMITTED] TN10JN08.057
D. Additional Analysis Based on the Dominant Firm Model (DFM) Including
a Revision of the DFM Designed to Consider Multi-period Dynamics
1. Introduction
In Section K and Attachment K of the White Paper, we presented a
model of dominant firm behavior adapted to the newsprint industry.
The model allowed us to address two questions:
In theory, how could Abitibi and Bowater, acting
together or as a merged entity, profitably raise price?
Do the current conditions in the newsprint industry
suggest that Abitibi and Bowater actually have the ability
profitably to raise price further? \42\
---------------------------------------------------------------------------
\42\ As discussed in Section B.1.a. above, operating results at
NA newsprint mills have gradually improved over the period March
2007 to May 2007 after declining over the first three months of
2007. In May 2007, total shipments from NA mills were down only 0.7%
compared to May 2006. See Table 2. One of the questions asked by DOJ
concerned the applicability of the DFM in the context of a
significant accelerating decline in operating results for NA
newsprint mills. Given that the gap in operating results between the
first five months of 2007 and the twelve months of 2006 has been
narrowing over the past three months, this question may be obviated.
---------------------------------------------------------------------------
Using estimated values for the model's parameters, we showed
that the model predicted that it would be profitable under current
conditions \43\ for a dominant firm with the combined shares of
Abitibi and Bowater to exercise market power through the dominant
firm strategy. We concluded that even allowing for adjustments to
the parameter values, the model pointed to the profitability of a
significant price increase. Changing various estimated parameters
within a reasonable range did not alter this finding.
---------------------------------------------------------------------------
\43\ To estimate the parameter values, we used the most current
data publicly available at the time we prepared the White Paper.
---------------------------------------------------------------------------
In this section, we address the following issues:
1. Introduction
2. The Relevance of a Paper by Matthew Gentzhow to Our
Conclusions Regarding the DFM.
3. Would the Dominant Firm Strategy be Profitable for Abitibi or
Bowater Acting Independently?
4. What Are the Effects on Dominant Firm Behavior of a Decline
in Demand?
5. A Description of a Revision of the DFM Designed to Consider
Multiperiod Dynamics.
2. The Relevance of a Paper by Matthew Gentzhow to Our Conclusions
Regarding the DFM
In our April 20 meeting, the DOJ staff mentioned a paper by
Matthew Gentzhow which analyzed how a newspaper's online activities
affect the demand for its print edition.\44\ Using information
concerning the Washington Post, the author concluded that the Post's
online edition reduced readership of the paid newspaper by a
significant but very small amount: eliminating the online edition
entirely would increase readership by only about 1.5% (p. 5).
---------------------------------------------------------------------------
\44\ We believe the article staff referred to is Matthew
Gentzhow, ``Valuing New Goods in a Model with Complementarity:
Online Newspapers'' National Bureau of Economic Research (NBER)
Working Paper 12562, January 24, 2006.
---------------------------------------------------------------------------
The DOJ staff expressed interest in determining the rate at
which the demand for newsprint will decline in the future.
Extrapolating from Gentzhow's paper to newspapers other than the
Post, demand for printed newspapers has been reduced very slightly
by the introduction of newspaper websites. There is nothing in the
article to suggest that newspaper websites (which are now quite
widespread) will cause significant further reduction in the demand
for printed newspapers (and hence newsprint) in the near future.
Data recently published by the NAA on newspaper print copy and
newspaper online advertising revenues are consistent with this
conclusion. On-line advertising revenues at U.S. daily newspapers
increased from 5.5% of total newspaper advertising revenues in the
first quarter of 2006 to 7.1% of total newspaper advertising
revenues in the first
[[Page 32932]]
quarter of 2007.\45\ While this is a non-trivial increase in on-line
advertising revenues as a percentage of total newspaper advertising
revenues, both the percentage increase and overall percentage of on-
line revenues are still quite small relative to total newspaper
advertising revenues.
---------------------------------------------------------------------------
\45\ See ``Newspaper Online Ad Growth Slows--As Print Revenue
Keeps Skidding,'' by Jennifer Saba, Editor & Publisher, May 29,
2007.
---------------------------------------------------------------------------
3. Would the Dominant Firm Strategy be Profitable for Abitibi or
Bowater Acting Independently?
In the White Paper model, as well as in a revised model designed
to consider multi-period dynamics,\46\ a dominant firm with initial
share of about 25.7% (like Abitibi) or about 15.8% (like Bowater)
can increase its profits by acting as a dominant firm. However, the
optimal percentage price increase that either firm would find is
lower than the price increase that would be preferred by a firm with
their combined share (modeled as 41.5%).\47\
---------------------------------------------------------------------------
\46\ See the description of this revised model in Section D.5.
below.
\47\ For the purposes our analysis of the DFM, the individual
Abitibi and Bowater shares as well as their combined share have been
adjusted to account for Abitibi aM Bowater partial ownership of
certain newsprint mills and machines. See Table C.l. in Attachment C
of the White Paper for information on their partial ownership of
certain newsprint capacity.
White Paper Model
------------------------------------------------------------------------
------------------------------------------------------------------------
Dominant Firm Share......... No DF 41.5% 25.7% 15.8%
Price....................... $625 $922 $781 $692
------------------------------------------------------------------------
Revised Model
------------------------------------------------------------------------
------------------------------------------------------------------------
Dominant Firm Share......... No DF 41.5% 25.7% 15.8%
Price....................... $590 $1,166 $782 $647
------------------------------------------------------------------------
Under the White Paper model, the lowest initial dominant firm
share from which it is profitable to engage in the dominant firm
strategy, given the other assumed parameters, is about 16%. Using
the revised model, the corresponding share is about 14.5%.
Both models indicate that it would be profitable for Abitibi or
Bowater acting on its own to reduce capacity and elevate price. In
both models, the dominant firm assumes that all other firms in the
industry will act as fringe, increasing their output in response to
a capacity reduction by the dominant firm. (In other words, there is
no assumption of a coordinated anticompetitive response by the
fringe.) As pointed out in the White Paper, however, both firms have
been actively reducing capacity since at least 2002. We believe it
unlikely that either of these firms assumes that the other firm will
behave as part of the fringe.
4. What Are the Effects on Dominant Firm Behavior of a Decline in
Demand?
a. A Decline in Demand Resulting in a Lower Newsprint Industry Capacity
Utilization Rate
A decline in demand can be interpreted as affecting the initial
conditions. Reducing demand starts the industry off with lower
industry capacity utilization. Decreasing industry capacity
utilization (i.e., increasing excess capacity in the initial
conditions) reduces the optimal price increase for a dominant firm
of a given size.
This question can be addressed with a simple adjustment to the
White Paper model. We assumed that capacity utilization was 95% and
that a dominant firm could begin to raise newsprint prices by
removing capacity to bring utilization to 98%. A fall in demand
could be thought of as changing the starting position from 95%
capacity utilization to something lower: e.g., 90%. Leaving all the
other parameters in the model the same (see Table K1 of the White
Paper), the profit-maximizing dominant firm price increase at
various levels of initial capacity utilization is as follows:
White Paper Model
----------------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------------
Initial Capacity Utilization............................. 95% 90% 80% 70% 63%
DF's profit-maximizing price increase.................... 48% 43% 32% 18% 5%
----------------------------------------------------------------------------------------------------------------
Even if demand for newsprint fell to such an extent that
capacity utilization was 63%, it would still be profitable for the
dominant firm with a 41.5% initial share to withdraw capacity and
raise price 5%.
Using a revised model, a fall in demand can be modeled as
reducing the initial demand level such that, given the existing
industry capacity and cost structure, the industry equilibrium
output is at a lower level of capacity utilization. If demand were
such that initial capacity utilization were as low as 73%, it would
still be profitable for a dominant firm with a 41.5% initial share
to engage in the dominant firm strategy.
Revised Model
----------------------------------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------------------------------
Initial Capacity Utilization............................. 95% 90% 80% 75% 73%
DF's profit-maximizing price increase.................... 98% 79% 47% 32% 26%
----------------------------------------------------------------------------------------------------------------
b. The Effect of an Increase in the Rate of Decline of Demand
Alternatively, a decline in demand can be interpreted as
affecting the rate of decline of demand in future periods. A revised
dominant firm model was created to consider multiple-period
dynamics. To explore the effect of the rate of decline of demand, we
contrasted the profits from two alternative strategies:
DF: The dominant firm acts as a dominant firm in the first
period by withdrawing capacity and raising price, then it accepts
the equilibrium price (given the reduced capacity) in subsequent
periods.
No DF: The dominant firm accepts the equilibrium price and
quantity in all periods.
The dominant firm prefers the strategy that yields the greatest
discounted profit flow. With an initial share of 41.5%, the DF
strategy is preferred even if demand is declining by as much as 20%
per year.\48\ It appears that no reasonable rate of future decline
in demand would cause a dominant firm with this initial share to
abandon dominant firm behavior entirely. Future decline in demand
does not deter the dominant firm from withdrawing capacity and
elevating price in the first period.
---------------------------------------------------------------------------
\48\ This rate is almost double the rate of decline in recent
months. Higher rates of decline were not explored. During the period
January 2007 to April 2007, total NA demand for newsprint declined
11.2% compared to the first four months of 2006. See Section 2.b.
above.
---------------------------------------------------------------------------
[[Continued on page 32933]]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
]
[[pp. 32933-32935]] United States v. Abitibi-Consolidated Inc. et al.; Response to
Public Comment on the Proposed Final Judgment
[[Continued from page 32932]]
[[Page 32933]]
If the dominant firm's initial share is sufficiently low (e.g.,
15%), the No DF strategy is preferred when there is significant
decline in future demand (e.g., 5% or 10% per year). Thus, it is
possible that a dominant firm with a low initial share would act as
a dominant firm when demand is declining slowly but choose not to
act as a dominant firm when demand is declining rapidly. The
intuition is as follows: with a small initial share, the dominant
firm must close a major portion of its capacity to elevate the price
in the first period. Accepting the competitive solution in
subsequent periods, the dominant firm finds that the profits with a
much-reduced output and slightly higher prices (as would result from
the DF strategy) yields lower profits than taking its initial share
of industry output at somewhat lower prices (as would result from
the No DF strategy). When the two alternative strategies are
considered for the initial period and multiple subsequent periods,
the No DF strategy yields higher discounted profits.
Note that if there is an incentive not to act as a dominant
firm, it comes from the assumption that capacity withdrawn by the
dominant firm is permanently withdrawn and cannot be restarted. If
the dominant firm were simply to ``idle'' capacity but retain the
option of restarting the capacity in the future, then it suffers no
penalty in future periods when the dominant firm behavior is no
longer profitable. If capacity can be withdrawn on a temporary
basis, future decreases in demand would not deter a dominant firm
from behaving as a dominant firm when it is otherwise profitable to
do so.
Using a model based on current industry conditions and plausible
projected declines in North American demand for newsprint, we see no
reason to believe that dominant firm behavior in the newsprint
market will cease due to a more rapid decline in industry demand.
The decline in newsprint demand is not new. With the exception of a
few up-ticks in demand, the NA demand for newsprint has been
steadily declining since the fourth quarter of 1999.\49\ As separate
firms, Abitibi and Bowater have been engaging in dominant firm
behavior since at least the third quarter of 2002 in response to the
decline in NA newsprint demand. Even if future rates of decline are
higher than in previous years, the merger of two firms separately
engaged in dominant firm activity in the past increases the
likelihood that such behavior will be profitable in the future.
---------------------------------------------------------------------------
\49\ See Chart E2 on p. 61 of the White Paper.
---------------------------------------------------------------------------
5. A Description of a Revision of the DFM Designed to Consider
Multiperiod Dynamics
The model presented in the White Paper started with a stylized
representation of current conditions and considered whether it would
be profitable for a dominant firm to withdraw capacity. The revised
model includes an expanded structure that permits calculation of an
equilibrium price under various dominant firm behaviors and under
different levels of industry demand. In particular, the revised
model takes into account multi-period dynamics.
1. Information is available showing the variable cost per
delivered tonne of all the mills in the industry as of 4Q 2006. See
Chart 3 in Section C.2.a. above. Mills are arranged in order of
increasing cost. Based on a slightly stylized version of this cost
profile, it is assumed that the cost per tonne of the most efficient
mill is $400, the cost per tonne of the least efficient mill is
$600, and the cost per tonne of the rest of the capacity in the
industry can be approximated by a straight line between these two
end points. The industry cost of $600 per tonne occurs at full
capacity of approximately 12,000,000 tonnes. This cost profile
becomes the industry cost curve and is the supply curve under
competitive conditions. Thus, if output were 12,000,000 tonnes, the
cost of the least efficient mills would be $600 and, in a
competitive equilibrium, $600 would be the price. C = 400 + Q/
60,000.
2. For simplicity, it is further assumed that the dominant firm
and the fringe have the same cost profile at corresponding degrees
of capacity utilization, or in other words, that they have
(approximately) the same mix of mills with various degrees of
efficiency. The cost curve for the dominant firm runs from $400 at
zero or low levels of output to $600 at full capacity utilization;
likewise for the fringe. Added together, the two cost curves make up
the industry supply curve.
3. There is an explicit industry demand equation: Q = A P
[alpha]. This demand function is calibrated using the market
elasticity of demand cited in the literature and assumed in the
White Paper ([alpha] = -0.36). The parameter A is chosen so that
price is equal to cost in the initial scenario of interest.
Decreases in demand are modeled as reductions in A. Reducing A by
10%, for instance, means that the quantity demanded at any given
price would be 90% of what it previously was.
4. We start by looking at a situation in which the industry is
at competitive equilibrium with capacity utilization of 95%. (For
simplicity, we assume that the maximum achievable capacity
utilization is 100%, rather than a lower level such as 98% in the
White Paper model.) Given the industry capacity assumed, 95%
capacity utilization is achieved at an output level of 12,000,000 *
95% = 11,400,000. Given the industry cost curve assumed, cost at
this output level is $590 per tonne. The demand curve is
parameterized with A = 113,347,403 so demand equals supply at this
price and output. The assumption that the industry is currently at a
competitive equilibrium follows the observation that price has been
falling and capacity has not been withdrawn by either Abitibi or
Bowater in the past few months.
5. At this stage, the dominant firm decides whether it is more
profitable to stay at the competitive equilibrium or behave as a
dominant firm, removing capacity from the market to increase price.
When the industry is at a competitive equilibrium, the profit of the
dominant firm is calculated as the area of a right triangle. The
base of the triangle is the segment from $400 to the current
industry cost level. The height of the triangle is the output of the
dominant firm. In the initial scenario, output of the dominant firm
is 95% times the capacity of the dominant firm.
6. If the dominant firm decides to increase price, its profit
has two components. The first is a triangle as described previously
(but with a reduced quantity for the dominant firm). The second is a
rectangle. The height of the rectangle is the dominant firm's output
and the base of the rectangle is the difference between price and
the dominant firm's cost at the relevant output level. (As the
dominant firm reduces capacity, the capacity with highest cost is
eliminated first. For this reason, the marginal cost of the dominant
firm's output declines as it reduces capacity.)
7. With these initial conditions, it is profitable for a firm
with 41.5% share of capacity to remove capacity and increase price--
the profit-maximizing price is almost double the initial price of
$590. (One reason that such a large price increase is predicted is
the assumption that demand elasticity does not increase as price
increases.) At lower initial capacity levels, the profit-maximizing
price is reduced. At an initial capacity level of about 14.5%, the
profit-maximizing price under a dominant firm strategy yields no
more profit than the competitive equilibrium. Separately and
combined, Abitibi and Bowater currently have shares above 14.5%.
8. Suppose that a firm is at 15% initial capacity share. It is
slightly more profitable for the first period to behave as a
dominant firm. However, if demand declines 10% in each subsequent
period, it is not profitable in these subsequent periods to behave
as a dominant firm. The ``dominant firm'' accepts the market
equilibrium in the second period and thereafter. Because the firm
gave up share in the first period, however, its profits in all
subsequent periods are reduced. For a firm with an initial share of
15%, the multi-period discounted profit flow is greater if the firm
does not engage in the dominant firm strategy even in the first
period.
9. Intuitively, whether it will be profitable to behave as a
dominant firm for some number of periods will depend on the firm's
initial share of capacity, the degree of capacity utilization
initially, the rate of decline in demand, and the relevant discount
rate. As noted above, acting as a dominant firm brings no penalty in
later periods if the dominant firm idles, rather than permanently
removes, capacity. In this case, considerations about reduced
capacity in future periods would no longer deter a firm from
pursuing a dominant firm strategy.
E. Conclusion
We met with the DOJ staff on April 20, 2007 to discuss our White
Paper analyzing the likely competitive effects of the proposed
Abitibi-Bowater merger.\50\ This memorandum responds to several
questions raised by the DOJ staff at our meeting. In our White Paper
we provided considerable evidence that Abitibi and Bowater had used
a dominant firm strategy to successfully exercise market power
through strategic capacity closures over the period 2002 to 2006. We
concluded that Abitibi and Bowater, if allowed to merge, would have
an increased incentive and ability to pursue a dominant firm
strategy
[[Page 32934]]
post-merger. The analysis contained in this response memorandum
confirms our White Paper analysis and strengthens our conclusions.
---------------------------------------------------------------------------
\50\ The White Paper was submitted to DOJ on behalf of the
Newspaper Association of America on April 11, 2007.
---------------------------------------------------------------------------
In this response memorandum, we reach six main conclusions:
(1) Events in the NA newsprint market since the Abitibi-Bowater
merger announcement in January 2007 demonstrate how the NA newsprint
market would have functioned absent the exercise of market power by
Abitibi and Bowater. As NA newsprint demand continued to decline in
2007, NA newsprint prices have declined to the cash costs of the
highest cost NA newsprint mills. One mill (Blue Heron in Pomona, CA)
has been indefinitely idled due to its high cash costs of newsprint
production. In the absence of the exercise of a dominant firm
strategy by Abitibi and Bowater while their proposed merger is under
regulatory review, the NA newsprint market is performing
competitively. See Sections B.1., B.2., B.3., and C.2. above.
(2) We conclude that if the merger is approved, Abitibi-Bowater
will have an enhanced incentive and ability to engage in dominant
firm behavior post-merger. As shown by trade press comments cited in
Section B.3.a. above, it is widely anticipated by competitors of
Abitibi and Bowater and by newsprint industry analysts that, once
the merger is approved, Abitibi-Bowater will remove enough newsprint
capacity from the market post-merger to create a tight market,
thereby increasing newsprint prices above competitive levels.
(3) Prior to the merger announcement, changes in the price of
newsprint were closely correlated with changes in the value of the
Canadian dollar per U.S. dollar. Since the merger announcement in
January, the value of the Canadian dollar has increased
significantly while the price of newsprint has declined
significantly. The divergence between the value of the Canadian
dollar and the price of newsprint since the merger announcement
provides strong support for the Dominant Firm hypothesis and
contradicts the Competitive Response hypothesis. See Section B.1.b.
and Chart 1 above.
(4) RlSI benchmarking cash cost studies for NA newsprint mills
strongly suggest that Abitibi and Bowater closed newsprint capacity
over the period 2002-2006 even though the cash cost of that capacity
was below the price of newsprint at the time of the capacity
closures. Such behavior is consistent with the Dominant Firm
hypothesis and contradicts the Competitive Response hypothesis. See
Section C.1. and Chart 2 above.
(5) Between 1999 and 2001, the aggregate operating rates for NA
newsprint mills and NA mills producing uncoated groundwood specialty
grades were nearly identical. Beginning in 2002, the gap between
newsprint mill operating rates and the operating rates of mills
producing uncoated groundwood specialty grades began to widen. In
2004, the aggregate operating rate for newsprint mills was 6%
greater than the aggregate operating rate for mills producing
uncoated groundwood specialty grades. This divergence in operating
rates is consistent with the Dominant Firm hypothesis and
contradicts the Competitive Response Hypothesis. See Section C.3.
and Chart 4 above.
(6) In Section D above, we revise the Dominant Firm Model to
account for multi-period dynamics and the effect of an increase in
the decline of newsprint demand on dominant firm strategy.\51\ We
also analyze whether Abitibi and bowater, acting independently could
profitably pursue a dominant firm strategy. Our analysis shows that
while it would be profitable for both Abitibi and Bowater to
independently pursue a dominant firm strategy, a merged Abitibi-
Bowater would have the incentive and ability to achieve higher
prices and profits though a dominant firm strategy compared to the
firms acting independently. We also show that a dominant firm
strategy would be profitable even in the face of declines in
newsprint demand considerably greater than currently experienced and
over multiple periods.
---------------------------------------------------------------------------
\51\ As shown in Section B.1.a. and Tables 1 and 2 above, the
increase in the decline in NA newsprint mill operating results in
the first three months of 2007 began to slow in April and May 2007.
In May 2007, total shipments by NA newsprint mills were only 0.7%
below the level for May 2006.
---------------------------------------------------------------------------
Attachment D--Supplement 2 to the White Paper by Economists
Incorporated, Submitted on Behalf of the NAA to DOJ on July 20, 2007
Economists Incorporated
An Economic Analysis of the Competitive Effects of the Proposed
Abitibi-Bowater Merger
Response to Issues Raised at Our Meeting With the DOJ Staff on
April 20, 2007
Revision to the July 9, 2007 Response
Submitted to DOJ on Behalf of NAA
John H. Preston, Kent W. Mikkelsen, Ph.D., Economists Incorporated,
Washington, DC, July 20, 2007.
A. Introduction
On July 9, 2007, Economists Incorporated submitted a response
(``DOJ Response'') to issues raised by the Department of Justice
(``DOJ'') staff concerning the likely competitive effects of the
proposed Abitibi-Bowater merger in the North American (``NA'')
newsprint market.\1\ In this paper, we submit two revisions to our
DOJ Response based on publicly-available information that we have
received since we submitted the DOJ Response. The first revision
concerns the strategy of Abitibi-Bowater competitors in the NA
newsprint market who have recently announced a newsprint price
increase effective in September 2007. The second revision concerns
the plausibility of cost savings that Abitibi and Bowater have
claimed will result from the merger.
---------------------------------------------------------------------------
\1\ Our meeting with the DOJ staff was held on April 20, 2007.
The purpose of the meeting was to discuss our economic analysis
(``White Paper'') regarding the likely competitive effects of the
proposed merger. We had submitted the White Paper on April 11, 2007
on behalf of the Newspaper Association of America (``NAA''), an
association of U.S. daily newspapers.
---------------------------------------------------------------------------
B. The Strategy of NA Newsprint Competitors of Abitibi and Bowater Who
Have Recently Announced a Newsprint Price Increase Effective September
1, 2007
In footnote 17 of our DOJ Response, we stated the following:
According to a RISI news note dated June 29, 2007, Kruger
announced a $25 per metric tonne price increase for 30 lb. newsprint
effective September 1, 2007. According to RISI, ``Kruger is North
America's fourth-largest newsprint producer in terms of capacity
with 1.15 million tonnes/yr of production, all of it located in
[Eastern] Canada. Contacts said it was the first time they could
remember that the company had sought to initiate a price increase
round.'' We view Kruger's announced price increase as a competitive
response primarily to the appreciation of the Canadian dollar, an
action taken in the absence of the exercise of market power by
Abitibi and Bowater since their merger announcement in January 2007.
It is plausible that NA newsprint prices have fallen close to the
cash costs of one or more Kruger newsprint mills, necessitating the
price increase announcement. See Section C.2. below for a discussion
of 4Q 2006 cash costs of NA newsprint mills. Whether the price
increase will be successfully implemented or not will depend mainly
on the amount of excess capacity at NA newsprint mills in September
and succeeding months.
Subsequent trade press reports have made it clear that we were
mistaken in our conclusion that Kruger's announced price increase
should be viewed as a ``competitive response'' to the appreciation
of the Canadian dollar. Instead, these subsequent trade press
reports make it clear that the announced price increase is an
anticompetitive continuation of the Abitibi-Bowater Dominant Firm
strategy supported by coordination between Abitibi-Bowater and some
of its leading NA newsprint competitors. According to an article in
the July 16, 2007 edition of Pulp & Paper Week (p.7): \2\
---------------------------------------------------------------------------
\2\ See ``Newsprint: Price hike gains support; merger vote is
dogged by asset sale uncertainties.'' See also RISI news notes
``$25/tonne US newsprint price hike gains momentum,'' July 12, 2007
and ``More North American newsprint supplies support $25/tonne price
hike,'' July 16, 2007.
---------------------------------------------------------------------------
Several newsprint producers including the largest North American
supplier, Abitibi-Consolidated, began telling customers last week
they planned to increase the price of 30-lb newsprint by $25/tonne
effective Sept 1.
The move to raise prices $25 was kicked off at the end of June
by Canadian supplier Kruger, the fourth largest newsprint maker in
North America based on capacity. Contacts said Catalyst and Blue
Heron were among suppliers also planning the increase, and No. 3
ranked White Birch was considering it.
``If this gets followed by capacity reduction announcements it
would put some teeth into it,'' said one contact last week.
North American suppliers depend on Abitibi-Consolidated and
Bowater, which
[[Page 32935]]
hope to merge in the third quarter, to close sufficient capacity to
move North American newsprint supply in line with demand. Contacts
estimate North American newsprint supply outpaces demand by about
500,000 tonnes this year.
No one expects the two companies to remove any capacity until
after the U.S. Dept of Justice (DOJ) and Canada's Competition Bureau
(CCB) disclose whether the terms of the deal require any asset
divestments.
In our view, the most economically reasonable interpretation of
the comments in the Pulp & Paper Week article above is as follows:
(1) Kruger, Catalyst, and Blue Heron announced a $25/tonne price
increase at the end of June and in early July effective September 1,
2007 timed for the anticipated completion of the Abitibi-Bowater
merger.
(2) The price increase will not succeed unless substantial
capacity is closed.
(3) Abitibi-Bowater's NA newsprint competitors ``depend on
Abitibi-Consolidated and Bowater * * * to close sufficient capacity
to move North American newsprint supply in line with demand.'' \3\
---------------------------------------------------------------------------
\3\ See Section B.3. of the DOJ Response for our similar
comments by newsprint industry analysts and competitors of Abitibi-
Bowater.
---------------------------------------------------------------------------
(4) By also announcing a $25 price increase effective September
1, 2007, Abitibi has signaled to its NA newsprint competitors that
it will close the capacity necessary to support the price increase.
(5) Abitibi-Bowater will not close the capacity necessary to
support the price increase before their merger is approved by DOJ
and the CCB, almost certainly out of concern that such an action
would jeopardize regulatory approval of the merger.\4\
---------------------------------------------------------------------------
\4\ See Section B.2. of the DOJ Response for our analysis of
this issue.
---------------------------------------------------------------------------
(6) Abitibi-Bowater will close the capacity necessary to support
the price increase after the merger review period assuming the
merger is approved by DOJ and CCB.
(7) In initiating the $25 price increase to become effective at
the time of the anticipated completion of the Abitibi-Bowater
merger, Kruger and the other Abitibi-Bowater competitors who have
announced the price increase have engaged in coordinated interaction
in support of the Abitibi-Bowater Dominant Finn strategy.
C. According to Abitibi's Largest Shareholder, the Probability is Low
That the Merger Will Achieve the Efficiencies Claimed by Abitibi and
Bowater
In previous submissions to DOJ, we have not addressed the
synergies and other cost savings that Abitibi and Bowater have
claimed will result from the merger. There are two reasons. First,
as we do not have access to the non-public analyses supporting those
claims, we are not in a good position to analyze those claims.
Second, even assuming for the sake of argument that the magnitude of
the claimed efficiencies were likely to be achieved, it is our
opinion that the cost savings would not come close to offsetting the
likely anticompetitive harm from the merger that we have analyzed in
the White Paper and in the DOJ Response.\5\
---------------------------------------------------------------------------
\5\ Based on estimates on pages 5 and 8 of the Abitibi-Bowater
presentation ``Creating a Global Leader in Paper and Forest
Products,'' January 29, 2007, Abitibi and Bowater were claiming that
the merger would achieve cost savings of 1.6% of combined Abitibi-
Bowater sales over all product lines by the end of year 1 and 3.2%
by the end of year 2 and in subsequent years. (For the purposes of
this discussion, we assume these percentages approximately apply to
the combined NA newsprint operations of the two companies.) These
claimed cost savings are small in comparison to the anticompetitive
price increases that we analyzed in the White Paper (an aggregate
price increase of 49% from 3Q 2002 to 3Q 2006) and the
anticompetitive price increases that are likely to occur in future
years if the merger is approved by DOJ and the CCB. The announced
price increase of $25 discussed in Section B above is a 4.3%
increase over the June 2007 newsprint price of $585 per metric tonne
(30 lb., East) as published in Pulp & Paper Week. Of course, if
successfully implemented, the competitive harm from the price
increase to NA newspaper publishers and other NA newsprint customers
would result not just from an increase in the price of newsprint
sales by a merged Abitibi-Bowater but also from an increase in the
price of newsprint sales by all other NA newsprint suppliers.
---------------------------------------------------------------------------
The U.S. Department of Justice and Federal Trade Commission
Horizontal Merger Guidelines set out stringent standards for
determining if claimed efficiencies would be sufficient to prevent a
merger from being anticompetitive.\6\ In our view, the proposed
merger falls far short of satisfying those stringent standards, even
assuming for the sake of argument that all claimed efficiencies are
cognizable as defined in the Merger Guidelines.
---------------------------------------------------------------------------
\6\ See Sec. 4. Efficiencies (Revised April 7, 1997) of the
U.S. Department of Justice and Federal Trade Commission Horizontal
Merger Guidelines. According to the Merger Guidelines, DOJ will
consider only efficiencies that are merger-specific and cognizable.
Cognizable efficiencies are defined as ``merger-specific
efficiencies that have been verified and do not arise from
anticompetitive reductions in output or service.'' The Merger
Guidelines further state that ``When the potential adverse
competitive effect of a merger is likely to be particularly large,
extraordinarily great cognizable efficiencies would be necessary to
prevent the merger from being anticompetitiVe.''
---------------------------------------------------------------------------
Third Avenue Management LLC (TAM) is Abitibi's largest
shareholder with an ownership share of 12.44%.\7\ TAM is a
professional asset management company. In its press releases, TAM
describes itself as follows:
---------------------------------------------------------------------------
\7\ See RISI news note ``Aitibi-Consolidated's biggest
shareholder opposes merger with Bowater,'' July 16, 2007.
---------------------------------------------------------------------------
Third Avenue Management LLC is a New York-based investment
advisory firm that offers its services to private and institutional
clients. Third Avenue adheres to a disciplined bottom-up value
investment strategy to identify investment opportunities in
undervalued securities of companies with high quality assets,
understandable businesses and strong management teams that have the
potential to create value over the long term. Third Avenue
Management has $30 billion in assets under management and offers
value-oriented strategies through mutual funds, separate accounts
and alternative investment vehicles.
On July 16, 2007, TAM announced its opposition to the Abitibi-
Bowater merger. Among the reasons cited for its opposition was that
TAM has ``low confidence'' that the economic benefits and synergies
claimed for the merger will be achieved.
Mr. Wadhwaney noted that, ``We have low confidence that the
alleged economic benefits and synergies claimed by management will
actually be realized, and urge shareholders to read carefully the
risk factors and disclaimers that the companies have identified in
their combined proxy circular.''\8\
---------------------------------------------------------------------------
\8\ Amit Wadhwaney is Portfolio Manager for TAM. See TAM press
release, ``Third Avenue Management Opposes the Proposed Abitibi-
Consolidated Merger with Bowater Incorporated,'' July 16, 2006. TAM
also submitted a 13D filing to the SEC stating its opposition to the
merger.
---------------------------------------------------------------------------
D. Conclusion
If DOJ and the CCB approve the proposed Abitibi-Bowater merger,
anticompetitive price increases to NA newsprint customers, beginning
with the $25 per metric tonne price increase announced for
September, 1, 2007, are virtually certain. If the Third Avenue
Management analysis is correct, the synergies and other cost
reductions claimed by Abitibi and Bowater are unlikely to be
realized.
[FR Doc. E8-11401 Filed 6-9-08; 8:45 am]
BILLING CODE 4410-11-P