Of Counsel: CRAIG W. CONRATH PHILIP R. MALONE ALLEE A. RAMADHAN ALEXANDER Y. THOMAS BRIGID L. THOMAS    Attorneys    Antitrust Division    U.S. Department of Justice    1401 H Street, N.W.    Washington, D.C. 20530

ANNE K. BINGAMAN    Assistant Attorney General JOEL I. KLEIN    Deputy Assistant Attorney General ROBERT B. NICHOLSON MARK S. POPOFSKY    Attorneys    Antitrust Division    Appellate Section    U.S. Department of Justice    10th & Pennsylvania Ave., N.W.    Washington, D.C. 20530    (202) 514-3764

SUMMARY OF THE CASE AND REQUEST FOR ORAL ARGUMENT

Appellants' summary of the case is substantially accurate. The United States notes, however, that although NAT, L.C. and DR Partners both appeal from the district court's June 30, 1995, Judgment and Order of Rescission, they no longer seek review of the relief imposed.

In light of the nature of the issues involved in this case, and the substantial oral argument time requested by Appellants, the United States requests thirty minutes in which to present oral argument.

SUMMARY OF THE CASE AND REQUEST FOR ORAL ARGUMENT

TABLE OF AUTHORITIES

PRELIMINARY STATEMENT

STATEMENT OF ISSUES

STATEMENT OF THE CASE

SUMMARY OF ARGUMENT

ARGUMENT

1. THE COURT MAY AFFIRM THE FINDING OF A SECTION 7 VIOLATION AS LONG AS THE DISTRICT COURT DID NOT CLEARLY ERR IN DETERMINING THAT THE TIMES AND THE MORNING NEWS COMPETE IN THE SAME LOCAL DAILY NEWSPAPER MARKET
2. THE DISTRICT COURT CORRECTLY DETERMINED THAT THE TIMES AND THE MORNING NEWS COMPETE FOR READERS IN THE SAME LOCAL DAILY NEWSPAPER MARKET
1. The District Court Correctly Found That The Times And The Morning News Compete For Readers
1. The Evidence Demonstrated Vigorous Competition Between The Times And The Morning News
2. Appellants' Objections To The District Court's Analysis Are Without Merit
2. The District Court Correctly Excluded Other Media From the Market
3. THE DISTRICT COURT CORRECTLY DETERMINED THAT THE TIMES AND THE MORNING NEWS COMPETE FOR ADVERTISERS IN THE SAME LOCAL DAILY NEWSPAPER MARKET
1. The Times And The Morning News Compete For Advertisers Through The Feedback Loop
2. The District Court Correctly Excluded Other Advertising Vehicles From The Market
4. THE DISTRICT COURT CORRECTLY CONCLUDED THAT THE ACQUISITION MAY SUBSTANTIALLY LESSEN COMPETITION
1. The District Court Properly Found The Acquisition Presumptively Unlawful And That The Defendants Failed To Rebut The Presumption
2. The District Court Correctly Aggregated The Stephens Family's Shares in Donrey And NAT

CONCLUSION

CASES

Alexander v. National Farmers Organization, 687 F.2d 1173 (8th Cir. 1982)

American Crystal Sugar Co. v. Cuban-American Sugar Co., 152 F. Supp. 387 (S.D.N.Y. 1957), aff'd, 259 F.2d 524 (2d Cir. 1958)

Anderson v. City of Bessemer City, 470 U.S. 564 (1985)

Bathke v. Casey's Gen. Stores, Inc., 64 F.3d 340 (8th Cir. 1995)

Besta v. Beneficial Loan Co., 855 F.2d 532 (8th Cir. 1988)

Buffalo Courier-Express, Inc. v. Buffalo Evening News, Inc., 441 F. Supp. 628 (W.D.N.Y. 1977), vacated, 601 F.2d 48 (2d Cir. 1979)

Century Oil Tool, Inc. v. Production Specialties, Inc., 737 F.2d 1316 (5th Cir. 1984)

City of Mt. Pleasant v. Associated Elec. Co-op, 838 F.2d 268 (8th Cir. 1988)

Consolidated Gold Fields PLC v. Anglo American Corp. of South Africa, Ltd., 698 F. Supp. 487 (S.D.N.Y. 1988), aff'd in part and rev'd in part, 871 F.2d 252 (2d Cir. 1989) 43

Consolidated Gold Fields PLC v. Minorco, S.A., 871 F.2d 252 (2d Cir. 1989)

Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984)

Denver and Rio Grande Western R.R. v. United States, 387 U.S. 485 (1967)

F. & M. Schaefer Corp. v. C. Schmidt & Sons, Inc., 597 F.2d 814 (2d Cir. 1979) (per curiam)

Flegel v. Christian Hosp., 4 F.3d 682 (8th Cir. 1993)

FTC v. Coca-Cola Co., 641 F. Supp. 1128 (D.D.C. 1986), vacated as moot, 829 F.2d 191 (D.C. Cir. 1987) (Table)

FTC v. Freeman Hosp., Nos. 95-1448, -2882, 1995 WL 638654 (8th Cir. Nov. 1, 1995)

FTC v. University Health, Inc., 938 F.2d 1206 (11th Cir. 1991)

Grumman Corp. v. LTV Corp., 665 F.2d 10 (2d Cir. 1981)

Gulf & Western Indus., Inc. v. Great Atlantic & Pac. Tea Co., 476 F.2d 687 (2d Cir. 1973)

Hamilton Watch Co. v. Benrus Watch Co., 114 F. Supp. 307 (D. Conn.), aff'd, 206 F.2d 738 (2d Cir. 1953)

H.J., Inc. v. International Tel. & Tel. Corp., 867 F.2d 1531 (8th Cir. 1989)

Morgenstern v. Wilson, 29 F.3d 1291, 1297 (8th Cir. 1994), cert. denied, 115 S. Ct. 1100 (1995)

Paschall v. Kansas City Star Co., 727 F.2d 692 (8th Cir.), cert. denied, 469 U.S. 872 (1984)

Reich v. Bay, Inc., 23 F.3d 110 (5th Cir. 1994)

Ricks v. Riverwood Int'l Corp., 38 F.3d 1016 (8th Cir. 1994)

Sherron v. Norris, No. 95-1265, 1995 WL 672335 (8th Cir. Nov. 14, 1995)

Sun Newspapers, Inc. v. Omaha World-Herald Co., 1983-2 Trade Cas. (CCH) ¶ 65,522 (D. Neb. June 14, 1983), aff'd in part and modified in part, 713 F.2d 428 (8th Cir. 1983) 15

Times-Picayune Pub. Co. v. United States, 345 U.S. 594 (1953)

United States v. Archer-Daniels-Midland Co., 866 F.2d 242 (8th Cir.), cert. denied, 493 U.S. 809 (1989)

United States v. Continental Can Co., 378 U.S. 441 (1964)

United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956)

United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586 (1957)

United States v. Eastman Kodak Co., 63 F.3d 95 (2d Cir. 1995)

United States v. Empire Gas Co., 537 F.2d 296 (8th Cir. 1976), cert. denied, 429 U.S. 1122 (1977)

United States v. Mrs. Smith's Pie Co., 440 F. Supp. 220 (E.D. Pa. 1976)

United States v. Penn-Olin Chem Co., 378 U.S. 158 (1964)

United States v. Philadelphia Nat'l Bank, 374 U.S. 321 (1963)

United States v. Rockford Mem. Corp., 717 F. Supp. 1251 (N.D. Ill. 1989), aff'd, 898 F.2d 1278 (7th Cir.), cert. denied, 498 U.S. 920 (1990)

United States v. Tidewater Marine Serv., Inc., 284 F. Supp. 324 (E.D. La. 1968)

U.S. Anchor Mfg., Inc. v. Rule Indus., Inc., 7 F.3d 986 (11th Cir. 1993), cert. denied, 114 S. Ct. 2710 (1994)

STATUTES

15 U.S.C. 1

15 U.S.C. 18

15 U.S.C. 25

15 U.S.C. 26

28 U.S.C. 1291

28 U.S.C. 1331

28 U.S.C. 1337

OTHER

5 Phillip Areeda & Donald F. Turner, Antitrust Law (1980)

Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines (Apr. 2, 1992)

PRELIMINARY STATEMENT

NAT, L.C. ("NAT") and DR Partners ("Donrey") appeal from the Judgment and Order of Rescission rendered by the Honorable H. Franklin Waters on June 30, 1995. This ruling, entered both in the case brought by the United States and that brought by Community Publishers, Inc. and Shearin Inc. (collectively the "private plaintiffs"), is reported at 892 F. Supp. 1146. Neither Appellant challenges the propriety of the relief fashioned by the court.

The United States' complaint alleged violations of Clayton Act section 7, 15 U.S.C. 18, and Sherman Act section 1, 15 U.S.C. 1, and sought equitable relief. The district court accordingly had jurisdiction pursuant to 15 U.S.C. 25-26 and 28 U.S.C. 1331, 1337. Appellants filed timely notices of appeal on August 1, 1995 (NAT), and August 24, 1995 (Donrey), from the Judgment and Order of Rescission entered by the district court on June 30, 1995. This Court's jurisdiction is invoked pursuant to 28 U.S.C. 1291.

NAT and Donrey also appeal from the court's Opinion and Amended Order, filed on August 25 and August 28, 1995. The United States did not participate in the proceedings that generated these rulings, which involved questions of costs and attorneys fees. The Clerk of the Court accordingly did not name the United States as a party to these appeals (Nos. 95-3355 & 95-3358).⁽¹⁾

FOOTNOTES

1. In this Brief "GX" denotes government exhibits, "PX" the private plaintiffs' exhibits, and "DX" defendants' exhibits. The trial transcript is referenced by "T." and "T. P.I.H." denotes the preliminary injunction hearing transcript. Cross-references to exhibit and transcript citations refer to Appellants' Joint Appendix ("A"), the private plaintiffs' Appendix ("PA"), and the United States' Appendix ("GA").

STATEMENT OF ISSUES

1. Whether this Court may affirm the judgment that the acquisition violates section 7 of the Clayton Act, 15 U.S.C 18, whether or not the Daily Record is in the relevant market.

United States v. Continental Can Co., 378 U.S. 441 (1964)

United States v. Philadelphia Nat'l Bank, 374 U.S. 321 (1963)

United States v. Empire Gas Co., 537 F.2d 296 (8th Cir. 1976), cert. denied, 429 U.S. 1122 (1977)

2. Whether the district court clearly erred in determining that the Times and the Morning News compete for readers in the same local daily newspaper market.

FTC v. Freeman Hosp., Nos. 95-1448, -2882, 1995 WL 638354 (8th Cir. Nov. 1, 1995)

United States v. Archer-Daniels-Midland Co., 866 F.2d 242 (8th Cir.), cert. denied, 493 U.S. 809 (1989)

United States v. Empire Gas Co., 537 F.2d 296 (8th Cir. 1976), cert. denied, 429 U.S. 1122 (1977)

U.S. Anchor Mfg., Inc. v. Rule Indus. Inc., 7 F.3d 986 (11th Cir. 1993), cert. denied, 114 S. Ct. 2710 (1994)

3. Whether the district court clearly erred in determining that the Times and the Morning News compete for advertisers in the same local daily newspaper market.

4. Whether the district court correctly concluded that the effect of the acquisition may be substantially to lessen competition.

Consolidated Gold Fields PLC v. Minorco, S.A., 871 F.2d 252 (2d Cir. 1989)

F. & M. Schaefer Corp. v. C. Schmidt & Sons, Inc., 597 F.2d 814 (2d Cir. 1979) (per curiam)

American Crystal Sugar Co. v. Cuban-American Sugar Co., 152 F. Supp. 387 (S.D.N.Y. 1957), aff'd, 259 F.2d 524 (2d Cir. 1958)

STATEMENT OF THE CASE

The United States brought this action to challenge NAT's acquisition of the Northwest Arkansas Times ("Times"), a transaction that eliminated competition and left the Stephens family dominant in the market for local daily newspapers in the Fayetteville, Arkansas metropolitan area. The Stephens family owns 95% of NAT, a company formed for the sole purpose of acquiring the Times. The family also owns 100% of Donrey, which owns the Morning News of Northwest Arkansas ("Morning News"). The Times, although based in Fayetteville, has significant circulation in the adjacent city of Springdale, and the Morning News, although based in Springdale, has a very significant presence in Fayetteville.⁽¹⁾ Together, the two papers account for virtually 100% of the local daily newspapers sold in the two communities (T. 882-83, GA 116-17).⁽²⁾

NAT obtained the Times from Thomson Newspapers, Inc. ("Thomson") in early 1995. As a consequence of a strategic reorganization, Thomson decided to sell twenty-five of its over one-hundred papers (T. 2275, GA 195). Although Thomson offered the other twenty-four papers as a package deal, it determined that the Times would be extremely valuable to a local purchaser, and therefore decided to sell it separately (T. 2307, GA 200). Several potential buyers expressed interest in the Times; however, Thomson short-circuited the bidding process, concluding an agreement on January 27, 1995, to sell the paper to NAT. Thomson demanded, as a condition of sale to the Stephens family, a $2 million premium over its previously indicated asking price (T. 2281-82, GA 196-97). Apparently anticipating an antitrust challenge (T. 1897-99, GA 157-59), Thomson also obtained indemnification, inter alia, for "any Loss . . . incurred in connection with or arising from . . . any investigation, suit or proceeding challenging the transactions contemplated hereby on the basis of a violation of any Federal or State anti-trust statutes" (DX 1113 at 7, GA 383).

Antitrust litigation came quickly. On Sunday February 5, 1995, a day prior to the scheduled closing, the private plaintiffs served representatives of Thomson and the Stephens family with a complaint alleging violations, inter alia, of section 7 of the Clayton Act, 15 U.S.C. 18, and informed them that a preliminary injunction hearing would be held in two days (1 T. P.I.H. 3-4, GA 17-18; T. 2286-87, 2344-45, GA 198-99, 201-02). Thomson and the Stephens family nonetheless consummated the transaction as planned (T. 2345, GA 202). Four days later, the district court entered a hold-separate order.

On March 28, 1995, the United States filed its complaint challenging the acquisition, and the court consolidated the two actions. Although the private plaintiffs alleged a relevant market encompassing both Benton and Washington Counties, thus including the Daily Record as well as the Times and the Morning News, the government advanced a smaller market consisting of four Washington County zip codes covering virtually all of Fayetteville and Springdale. Unlike the private plaintiffs' market, the government's market included only the Times and the Morning News.

Following a bench trial conducted in early May, the district court held NAT's acquisition of the Times to violate section 7 of the Clayton Act. Employing well-established principles of market delineation, see Community Publishers, Inc. v. Donrey Corp., 892 F. Supp. 1146, 1153-55 (W.D. Ark. 1995), the court first determined, based on the facts in the record, that the two relevant product markets consisted of advertising in, and readership of, local daily newspapers. See id. at 1155-57. Turning to the geographic dimension of the markets, the court concluded that "the Times and the Morning News strongly compete against each other for readers and advertisers in Washington County." Id. at 1158.

The court thus found the relevant markets to encompass both the Times and the Morning News and to exclude media other than local daily newspapers. Based on further findings including, inter alia, that the Morning News and the Daily Record compete in Benton County, see id. at 1162, and that the Times and the Daily Record compete for regional advertisers, see id. at 1163-64, the court determined that "th[e] acquisition would affect market power over the entire two-county area," id. at 1163; accordingly, "Northwest Arkansas" constituted the relevant geographic market. Id. at 1165. In this market, the court found, the merger would leave "Stephens-owned newspapers in Northwest Arkansas [with] in excess of 84% of circulation and 88% of advertising," id. at 1168. The transaction thus presumptively violated section 7 of the Clayton Act⁽³⁾ -- a presumption that, in light of high barriers to entry exacerbated by the acquisition, the defendants failed to rebut. See id. at 1168-69.

Importantly, and in recognition that broadening the market beyond the area in which the Times and the Morning News compete decreased the Stephens family's post-merger market share, the Court held in the alternative that "[e]ven if it were to be found that the relevant geographic market was the Fayetteville Metropolitan area, as the government defines it, there would still clearly be a violation of Section 7 because the Times and the Morning News compete in that area, without question." Id. at 1165 n.15.

Having found the acquisition unlawful, the court ordered rescission.⁽⁴⁾ Thomson subsequently sought from the district court a stay of judgment pending appeal. Along with the plaintiffs, NAT opposed this request, "recogniz[ing] that it faces a substantial burden on appeal in attempting to completely overturn the decision of th[e] highly respected District Court" (Mem. Br. in Support of NAT, L.C.'s Response in Opp. to Motion for Stay Pending Appeal 1 (July 20, 1995), GA 1). Indeed, although contending that the district court committed clear error in "finding that the relevant market consists of Benton and Washington Counties," NAT observed that "given the Court's finding that it would, in any event, conclude that the market includes at least the Times and the Morning News, and the applicable standard of review . . . it would be even more difficult to obtain a second reversal with respect to the Government's case" (id. at 10-11, GA 3-4).

The district court, and subsequently this Court, denied Thomson's request for a stay, and the Times reverted from NAT to Thomson on September 20, 1995. The same day, and pursuant to an agreement publicly disclosed on September 12, 1995, Thomson transferred ownership of the Times to American Publishing Co. The private plaintiffs subsequently filed a motion to dismiss these appeals as moot. That motion has been briefed and remains pending before this Court.

SUMMARY OF ARGUMENT

The district court found, and the evidence at trial demonstrated, that the challenged acquisition threatened to eliminate the vigorous rivalry between the Times and the neighboring Morning News in violation of Clayton Act section 7. Appellants, seeking reversal of the judgment, principally contend that the evidence fails to support the markets the district court found, and that NAT's acquisition of the Times did not threaten competitive harm within the reach of section 7 despite the Stephens family's complete ownership of the Morning News and 95% stake in the Times. Appellants' arguments are without merit.

The record shows that from 1990 until NAT acquired the Times, the two papers engaged in a "newspaper war" that left each with significant readership in the other's primary town of circulation. Participants in this competitive struggle viewed the other paper, and not other media, as threatening to wrest away significant readership and advertising dollars, and actual marketplace behavior confirmed these perceptions. Based on this evidence, as well as other data, the government's expert testified that the relevant market encompassed the Times and the Morning News but excluded other media, and the district court properly credited this testimony.

Unable to demonstrate that the district court clearly erred in finding that the Times and the Morning News compete in the same local daily newspaper market, Appellants contend that the district court's judgment nonetheless must be reversed because the court erroneously included in the market another paper, the Daily Record. But excluding the Daily Record only could increase the Stephens family's post-acquisition market share, and the district court so recognized in holding that the challenged acquisition would be unlawful even in a narrower market that included only the Times and the Morning News. That alternative holding, supported by the district court's findings, is sufficient to condemn the acquisition.

Appellants' final contention, that the district court was compelled to ignore the Stephens family's market dominance because no single member of the family, or set of family members, possessed more than a minority interest in both the Times and the Morning News, elevates form over substance. The district court found that the Stephens family could be expected to act to maximize the wealth of the family as a whole, an objective inconsistent with vigorous competition between the Times and the Morning News. Having found it likely that the papers would be operated in furtherance of a single economic interest, the court properly aggregated the Stephens family's shares in the entities that control the two papers. Even if, as Appellants erroneously suggest, all the anticompetitive effects engendered by the acquisition might later be condemned when they arise under other provisions of the antitrust laws, that is irrelevant in light of Clayton Act section 7's purpose of arresting anticompetitive restraints in their incipiency.

ARGUMENT

1. THE COURT MAY AFFIRM THE FINDING OF A SECTION 7 VIOLATION AS LONG AS THE DISTRICT COURT DID NOT CLEARLY ERR IN DETERMINING THAT THE TIMES AND THE MORNING NEWS COMPETE IN THE SAME LOCAL DAILY NEWSPAPER MARKET

The district court delineated two relevant markets and found the transaction unlawful in both: (1) a market for readership of local daily newspapers in Northwest Arkansas (which includes the Daily Record's area of circulation); and (2) a market for advertising in local daily newspapers in Northwest Arkansas. However, the court held in the alternative that "[e]ven if were to be found that the relevant geographic market was the Fayetteville Metropolitan Area, as the government defines it [(an area in which the Daily Record lacks appreciable readership)], there would still clearly be a violation of Section 7 because the Times and the Morning News compete in that area, without question." 892 F. Supp. at 1165 n.15.

This alternative holding is sufficient to uphold the district court's judgment that the acquisition violates Clayton Act section 7, and to sustain the judgment on this basis, the Court need not decide whether the district court erred in expanding the geographic market to include the Daily Record's area of circulation. For the alternative holding reflects the court's recognition that any error in expanding the geographic market beyond the area in which in Times and Morning News compete with each other would be harmless because a narrower local daily newspaper market, one that excluded the Daily Record, would yield an even higher post-merger market share for the Stephens family than the 84% and 88% calculated for the "Northwest Arkansas" market.

The district court made all the factual findings essential to the narrower market that underpins this holding, specifically finding that the Times and the Morning News belong in the same local daily newspaper market for either readers or advertisers and that other media do not belong in either market.⁽⁵⁾ Moreover, the findings that underlie the district court's conclusion that the acquisition violates section 7 in the broader markets delineated -- the existence of high barriers to entry exacerbated by the acquisition, the absence of demonstrable efficiencies, and the propriety of aggregating the Stephens family's interests in the two papers -- apply equally to the court's alternative holding. As demonstrated below, these findings, and the court's legal analysis, were sound.

2. THE DISTRICT COURT CORRECTLY DETERMINED THAT THE TIMES AND THE MORNING NEWS COMPETE FOR READERS IN THE SAME LOCAL DAILY NEWSPAPER MARKET

Appellants advance essentially two objections to the district court's core finding that the Times and the Morning News compete for readers in the same local daily newspaper market. First, they argue that court erred in finding that the Times and the Morning News belong in the same market⁽⁶⁾; and second, they maintain that the court erred in excluding other media from that market. The district court's findings, however, are not clearly erroneous.⁽⁷⁾

1. The District Court Correctly Found That The Times And The Morning News Compete For Readers

The district court found that "[t]he record is absolutely replete with evidence that the Morning News and the Times compete in the same product market and that they both serve the same locale. Specifically, they are competing for readers in the Washington County area which is comprised mainly of the towns of Fayetteville and Springdale." 892 F. Supp. at 1158. The district court based this finding on evidence demonstrating (1) that the Times and the Morning News strive to cover local stories that appeal to each others' subscriber base; (2) that the two papers have substantial overlapping circulation; (3) that significant "competitive actions and reactions . . . have [been] undertaken [by both papers] in direct response to each other"; and (4) that the papers had a "consistent obsession with each other as `the competition.'" Id. at 1159. The evidence amply supports the district court's finding.

1. The Evidence Demonstrated Vigorous Competition Between The Times And The Morning News

Examination of only a sample of the evidence considered by the district court demonstrates that, from the early 1990s, the Times and Morning News competed vigorously for readers in Fayetteville and Springdale. In May 1990, the Morning News opened an office in Fayetteville and converted to morning delivery (GX 4 at TC 002648, GA 203). In the two years that followed, the Morning News' daily distribution in Fayetteville increased from 2,733 to 4,461 and its Sunday distribution there increased from 3,058 to 4,933 (A 421).⁽⁸⁾ In the same period, the Times' Fayetteville readership declined markedly (A 420).

The Times, conceding that its reverses stemmed from failure to improve the paper in the wake of the Morning News' initiatives (GX 9 at TC 004542, GA 208), and recognizing that "Fayetteville is a very competitive market" (GX 4 at TC 002648, GA 203; GX 9 at TC 004542, GA 208), launched a counterstrike in 1992 -- one designed both to stop Morning News' penetration in Fayetteville and to increase the Times' circulation in Springdale. The Times, for instance, opened a bureau and added a reporter in Springdale (GX 9 at TC 004542-43, GA 208-09), increased nonadvertising content by over five columns a day (id.), introduced a Saturday edition (GX 5 at DMG01-00017, GA 204), and switched to morning delivery (A 412).

The impact of this competition was not lost on the Morning News (PX 39 at DONR-10706, GA 359). Donrey's President noted that the Times' new Saturday edition gave "a large percentage of news space . . . to Springdale news items" (GX 5 at DMG01-00017, GA 204). Indeed, concluding that the Times sought to use the Saturday edition "to hit hard" at areas in which the Morning News circulates (id.), he ordered an acceleration of the Morning News' own plan to add a Saturday edition (GX 7 at DMG01-00015, GA 205; GX 8 at DMG01-00019, GA 207). Also in direct response to the Times' turnaround, the Morning News decided to delay for six months a planned $1.00 subscription price increase (GX 7 at DMG01-00015, GA 205), a delay that ended up amounting to a year (A 412). When the Morning News finally implemented the increase, the Times responded by planning to keep its own prices constant in order to augment circulation in Springdale, noting that it would "have to take [a] revenue hit to get the numbers that are critical to this paper's overall strategy of market growth" (GX 85 at NAT-00141, A 611). The Times similarly delayed for almost a year a planned $.50 to $.75 price increase for Sunday rack sales because it "[could] not go to that price with [the Morning News] still at 50" (id. at NAT 08-00140, A 610; GX 88 at -00134, PA 847; GX 91 at 00122, GA 239).⁽⁹⁾

The Times also hired a new publisher, George Smith, who carried out a program of aggressive competition with the Morning News throughout 1993 and 1994, which he characterized as "the Second Battle of Northwest Arkansas" (GX 1 at NAT01-00277, PA 779). During this period, each paper rapidly reacted to the other's competitive actions. The Times increased its use of color, and the Morning News responded with additional use of color (GX 19 at TC 004827, GA 212). Smith distributed free copies of the Times to Morning News readers on holidays (GX 99 at NAT08-00101, A 632; GX 44, GA 215). After the Times did so in Springdale, the Morning News switched to 365-days a year publication (GX 45, GA 216; T. 1251-53, GA 146-48). The Morning News followed the Times in introducing a travel page (GX 119 at NAT07-00080, GA 282), and the Times, in turn, improved its weather page in response to similar improvements by the Morning News (GX 231, GA 346). Perhaps most visibly, the Times expanded its sports coverage in order to "rattle the [Morning] News' cage" (GX 26 at TC 001961, PA 829), and the Morning News responded by enhancing its own sports coverage (GX 129, at NAT07-00048, GA 314). Smith also sought to take business away from the Morning News more directly. He reported "picking up subscribers" from the Morning News after distributing free copies of the Times (GX 102 at NAT08-00096, PA 868); indeed, at one point Smith disseminated some 4,200 free copies of the Times to Morning News subscribers as part of his plan "to barricade Fayetteville" (GX 58, GA 218). He also sought to expand coverage of regional and local news of interest to Springdale readers (T. 1664, PA 393). In sum, as a Morning News official put it, competition in the region was "a ferocious dog eat dog situation" (PX 39 at DONR-10708, GA 361).

Kenneth Baseman, the government's expert, whose testimony the district court accepted, concluded that this evidence demonstrated that the two papers belonged in the same market. Applying the Merger Guidelines' methodology (T. 861-69, 881-94, GA 95-103, 115-28), Baseman found particularly relevant the extensive evidence, detailed above, that the papers viewed one another as "the competition" and that the papers based business decisions on each others' competitive actions (T. 871-874, GA 105-08). He also found the defendants' contention that the Times and the Morning News do not belong in the same relevant market inconsistent with profit data showing that the outbreak of "competition between the Times and the [Morning] News in the early '90s resulted in approximately a ten-percent reduction [and a] ten percent transfer from profits to quality related expenses" on the part of the Times (T. 875-77, 914-15, 918, GA 109-11, 135-37). The sum of the evidence, Baseman concluded, demonstrated that the two papers significantly checked each others' ability to attempt to exercise market power (T. 876-77, GA 110-11).

2. Appellants' Objections To The District Court's Analysis Are Without Merit

Despite this evidence, Appellants contend that the district court committed clear error in placing the papers in the same market. Their arguments lack force.

a. Appellants first maintain that "local daily newspapers," such as the Times and the Morning News, inherently appeal only to a "particular locale." NAT Br. at 25. The papers "cannot be competitors in a `local' market unless both papers are serving the same local populace, which is not the case here." Id. at 26 (emphasis added). Put more simply, Appellants maintain that the Times and the Morning News do not, and indeed cannot, constitute substitutes for one another in the eyes of consumers because of a near-irrebuttable presumption that "local daily" papers necessarily compete only in their towns of primary circulation. See NAT Br. at 13-15, 25-26.

Such a presumption, however, has no basis in law or policy. It impermissibly abandons the principle that "the reality of the marketplace must serve as the lodestar" of market definition, e.g., United States v. Empire Gas, 537 F.2d 296, 303 (8th Cir. 1976), cert. denied, 429 U.S. 1122 (1977), as well as Appellants' own position that elasticity tests⁽¹⁰⁾ must be applied to newspaper markets, see NAT Br. 46 & n.47. Furthermore, new residents, residents who live between two communities, at the juncture of two communities, or live in one community but work in another, may have "hometown" loyalty to neither and may find news and information germane to a broader region desirable. Indeed, Appellants' argument is refuted by the November 1994 merger of the Morning News' separate Rogers and Springdale daily papers into the Morning News of Northwest Arkansas. As the new name of the paper suggests, "local" daily papers, contrary to Appellants' argument, see NAT Br. at 25, may seek to differentiate themselves on bases other than identification with a "hometown" or primary area of circulation.⁽¹¹⁾

The record, moreover, amply supports the district court's finding that the Times and Morning News each appeal to readers in both Fayetteville and Springdale. For instance, in the two Fayetteville zip codes identified by the government, the Morning News possesses a daily circulation of 3,986 and the Times 8,007 (A 431-32). Clearly, the Morning News has significant appeal in Fayetteville; indeed, a switch of only a little more than 2,000 readers would result in it displacing the Times as the leading Fayetteville paper.⁽¹²⁾ And, although the Times did not achieve the same degree of penetration in Springdale, its presence in the zip codes identified by the government is substantial, with a daily circulation of 1,856 compared to the Morning News' 10,136 (A 431-32).⁽¹³⁾ The record also includes evidence that the Morning News contains local news and advertising of interest to Fayetteville readers (T. 412-15, 453, GA 29-32, 47; T. 701-02, A 71-72; T. 709, PA 127), and that the Times' content similarly appeals to readers in Springdale (T. 453, GA 47), see also supra p.11 (discussing the Times' devotion of increased space to Springdale news items). Indeed, because of their constraining influence on one another, described in detail above, the papers actively sought not to differentiate themselves on the basis of their appeal to their particular "home town" audience, but instead primarily on the overall quality of their content and regional appeal.⁽¹⁴⁾

b. Appellants next contend that the district court erred in placing the Times and the Morning News in the same market because the analysis of their economic expert, Dr. Overstreet, conclusively demonstrated that they were not. See NAT Br. at 26-27. However, the district court rejected Dr. Overstreet's testimony in favor of that of the plaintiffs' experts. See 892 F. Supp. at 1161. "It is axiomatic that a district court has the discretion to evaluate the credibility of expert witnesses and accept the testimony it finds most plausible," FTC v. Freeman, Nos. 95-1448, -2882, 1995 WL 638354, at *8 n.13 (8th Cir. Nov. 1, 1995), and the district court did not abuse its discretion in rejecting Dr. Overstreet's testimony here.

Dr. Overstreet examined the Times' January 1994 $.50 to $1.00 price increase for Sunday rack sales. Based on current profit data, Dr. Overstreet determined that the "critical" elasticity of demand -- the elasticity above which a 10% price increase would be unprofitable for the Times -- was 3.3 (T. 2074-76, A 287-89). Dr. Overstreet then observed that, following implementation of this 100% price increase, during which time the Morning News did not increase its $.50 price, the Times lost approximately 1,300 single copy sales in the short run, or 34% of such sales (T. 2080-81, A 293-94). This yielded an elasticity associated with the price increase, Dr. Overstreet concluded, of (34/100) .34. Because the elasticity calculated was "substantially below the [elasticity of] 3.0 or so that would have been necessary" to make the price increase unprofitable, Dr. Overstreet reasoned that the two papers did not belong in the same market (T. 2081-86, A 294-99).

Dr. Overstreet's calculation was flawed.⁽¹⁵⁾ As an initial matter, taken at face value Dr. Overstreet's testimony implies that the Times was severely underpricing its product. The disparity between the critical elasticity of 3.3 calculated and the actual elasticity for the 100% price increase found of .34 indicates that the Times should have been profitably pricing the paper substantially above $1.00. But it is plainly more likely that Dr. Overstreet performed a flawed calculation than that the Times possessed market power unknown to its management. Confirmation of the calculation's infirmity is provided by three sources.

First, the calculation, as the district court observed, see 892 F. Supp. at 1161, failed to account for the interaction between readership and advertising. Dr. Overstreet's determination of a critical elasticity for the Times of 3.3 and an actual elasticity of .34 both took into account only profits lost from failure to sell circulation to readers.⁽¹⁶⁾ They did not take into account lost advertising revenue from having fewer readers at the higher circulation price, and advertising dollars are the most important component of newspaper profitability (T. 761-62, PA 149-50). Nor, as Dr. Overstreet acknowledged (T. 2197-98, GA 185-86), did either calculation account for the dynamics of the "feedback" loop.⁽¹⁷⁾ Consequently, Overstreet's calculations are meaningless.

Second, Dr. Overstreet admitted that he made no attempt, through multiple regression analysis or otherwise, to sort out all of the effects on quantity sold other than price (T. 2187-89, GA 172-74).⁽¹⁸⁾ Most significantly, contemporaneously with the increase in price to $1.00, the Times redesigned its Sunday sports section as well as the paper's front page (GX 110 at NAT10-00047, GA 253). According to the Times' publisher, the new sports section in particular "was necessary" to gain acceptance for the new $1.00 rate (id.). Consequently, as the district court noted, see 892 F. Supp. at 1161, the "quality-adjusted" price increase likely was much less than 100%; this, however, Dr. Overstreet ignored in calculating the .34 elasticity.

Finally, Dr. Overstreet's calculation of an elasticity for a past price increase sheds little light on what elasticity the Times would face for a price increase initiated today (at a higher prevailing price). But it is the latter that is the concern in defining markets for the purpose of analyzing mergers. It might well be the case that, at a price of $1.00, a small price increase might yield a much higher elasticity than the .34 calculated. Dr. Overstreet, however, overlooked this problem completely.

c. Appellants similarly contend that district court committed reversible error in failing to accept Dr. Overstreet's testimony that instances in which the Times and Morning News did not immediately respond to each others' price increases evinced a low elasticity. See NAT Br. at 27-28. Although evidence of rapid reactions to price increases is probative of competition, competitive responses need not be instantaneous. See United States v. Continental Can Co., 378 U.S. 441, 455 (1964). Moreover, the very authority cited by Appellants recognizes that refusing to follow a price increase may constitute a competitive response. See Buffalo Courier-Express, Inc. v. Buffalo Evening News, Inc., 441 F. Supp. 628, 637 (W.D.N.Y. 1977), vacated on other grounds, 601 F.2d 48 (2d Cir. 1979).

Nothing in the record required the district court to reject the view that the Morning News' failure to follow the Times' Sunday rack price increase until mid-1995 reflected the constraining force of the Times' improved quality. Indeed, the timing of the price increase -- following the acquisition -- suggests that it resulted from the Times' diminished competitive influence once the Stephens family controlled both papers. Appellants' reliance on the Morning News' failure to raise its home delivery price to $6.00 until April 1993, when the Times effected a $6.00 to $6.50 home delivery price increase in January 1991, is similarly misplaced. Appellants ignore that the Morning News immediately reacted to the Times' price increase by raising its own price from $4.50 to $5.00 (A 412), and that the Morning News' delayed a further $1.00 increase because of the Times (GX 7 at DMG01-00015, GA 205) -- reactions plainly probative of competition.⁽¹⁹⁾

d. Finally, Appellants contend that the district court erred in accepting the analysis of the government's expert, Kenneth Baseman. See NAT Br. at 17-18 n.20. Baseman's testimony was fatally deficient, they argue, because he made no attempt to measure elasticities directly or to conduct "any economic analysis" to support his opinion. NAT Br. at 17 & n.20. These objections are misconceived.

No court ever has required an actual measurement of elasticities.⁽²⁰⁾ Reliable data for performing such calculations often does not exist. See, e.g., U.S. Anchor Mfg., Inc. v. Rule Indus., Inc., 7 F.3d 986, 995 (11th Cir. 1993). If it were required, antitrust markets commonly could not be defined. It is for precisely this reason that the Guidelines, which Appellants concede supply appropriate market definition principles, see Nat Br. at 10-11, 16-17, 45-46; Donrey Br. at 43, provide for consideration of "all relevant evidence," Guidelines §1.11,⁽²¹⁾ from which elasticity may be gauged indirectly. And consistent with the Guidelines, this Court recognizes that a "broad range of evidence . . . may be of value in determining" the relevant market. Freeman, 1995 WL 638354 at *10; see also United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 246 (8th Cir.) (explaining that the tools of market delineation "are not [to be] used to obscure competition but to recognize competition, or the lack of competition, to the extent such exists" (citing United States v. Continental Can Co., 378 U.S. 441, 453 (1964)), cert. denied, 493 U.S. 809 (1989); cf. Empire Gas, 537 F.2d at 303 ("In defining the relevant part of commerce for any product the reality of the marketplace must serve as the lodestar.").

Accordingly, and contrary to Appellants' contention, see NAT Br. at 17 n.20, Baseman followed the Guidelines in relying on evidence of significant competitive interaction between the two papers as well as business documents demonstrating that the papers consistently viewed each other as one another's principal competitor. See Guidelines § 1.11.⁽²²⁾ This evidence, moreover, plainly is probative of the papers' presence in the same relevant market. See Freeman, 1995 WL 638345, at *9 ("We agree that testimony of market participants is relevant to a determination of a proper geographic market . . . ."); Morgenstern v. Wilson, 29 F.3d 1291, 1297 (8th Cir. 1994) (relying on the testimony of market participants to include other suppliers in the market), cert. denied, 115 S. Ct. 1100 (1995); U.S. Anchor, 7 F.3d at 995 (detailing factors that "serve as useful surrogates for cross-elasticity data" including "whether industry firms routinely monitor each other's actions and calculate and adjust their own prices (at least in part) on the basis of other firms' prices" (internal quotations omitted)); FTC v. Coca-Cola Co., 641 F. Supp. 1128, 1132 & n.8 (D.D.C. 1986) ("Analysis of the market is a matter of business reality -- a matter of how market is perceived by those who strive to profit in it." (citing cases)), vacated as moot, 829 F.2d 191 (D.C. Cir. 1987) (Table).

Baseman also properly relied on profit data that, he concluded, was inconsistent with the view that the papers belonged in different markets (T. 934-35, A 109-10). Appellants argue that Baseman "conducted no study of the financial data underlying his hypothesis that the reduction in profits was indicative of competition." NAT Br. at 17 n.20. However the evidence demonstrated that the Times, during a period in which its expenditures rose and its profits declined 21%, experienced the greatest percentage increase in both nonadvertising content and staffing for all Thompson Southern Group newspapers during 1989-1993 (GX 82 at TC 003607, GA 223; GX 82A, GA 224), and Baseman testified that these are "standard measures of quality in the empirical literature on newspapers" (T. 916, A 98), an industry in which Baseman, as the court noted, see 892 F. Supp. at 1156, had much experience (T. 860-61, 870, GA 94-95, 104). During this period, moreover, the Times' revenues actually rose (GX 82 at TC 003607, GA 223) because, unlike the rest of the country, the early 1990s marked a period of growth for Northwest Arkansas (T. 934-35, A 109-10; T. 2174, GA 171). The Times' reduced profits, and a similar decline for the Morning News, thus did not reflect a recession; rather, coupled with quality improvements that far outstripped Thomson's other papers, the data evinces the presence of competition (T. 935, A 110).

2. The District Court Correctly Excluded Other Media From the Market

Appellants next maintain that the district court clearly erred in including in the market only "local daily papers." Their contention should be rejected.

1. Appellants initially argue, see NAT Br. at 29-31 & n.29, that the district court erred in relying on "the contemporaneous, prelitigation records of the various newspaper organizations and personnel involved in [this] case" that "made frequent comparisons between the Times and the Morning News" but "did not make comparisons between the Times and any other media outlet." 892 F. Supp. at 1155. They first contend that the court misread the record as a factual matter, claiming that "the Times' internal documents and business plans directed to its corporate parent, Thomson, are replete with references to non-daily newspapers and other media competitors, as well as the neighboring papers." NAT Br. at 29-30. However, documents noting media other than local daily papers generally do so in the context of a concern for advertising, not readership.⁽²³⁾ On the crucial question of competition for readership, George Smith's monthly Times manager's reports for 1993 and 1994 contain no mention of television or radio competition for readers.⁽²⁴⁾ As for state-wide, out-of-state and national papers, the Times regarded these papers as complements for readers, not substitutes, because they did not contain the same sort of local and regional coverage as the Times and the Morning News,⁽²⁵⁾ a judgment confirmed by the testimony of readers (T. 455-56, GA 49-50). And, with respect to much lower circulation daily, weekly, or biweekly publications in the region, there is no evidence that these publications were viewed as serious competitors for readers in Springdale and Fayetteville.⁽²⁶⁾ The documents, then, support the district court's limitation of the market to "local daily newspapers."

Appellants' second argument is that the repeated references to the Morning News and the Times as each others' principal competitor for readers in these documents, and the absence of significant mention of other media, reflect not market reality but simply an arbitrary decision by Thomson for "each of its 144 papers to discuss as competition all newspapers within a 35-mile radius of it." NAT Br. at 30. However, the Thomson Southern Group business plan cited, in addition to listing newspapers within 35 miles, also listed "Other Serious Competitors" (GX 82 at TC 003532, GA 222). For many of Thomson's papers this category included other media, such as "television stations," "radio," and other papers and printed publications (id.). But for the Fayetteville market, as Baseman noted (T. 874, 937-38, GA 108, 138-39), there were no such entries. This document thus supports the court's exclusion of other media. Moreover, that the Times' documents occasionally mention other media, albeit almost only in the context of advertising, refutes Appellants' contention that the Times' reports merely reflect a policy on the part of Thomson to consider only neighboring local daily papers.

2. Appellants, citing H.J., Inc. v. International Tel. & Tel. Corp., 867 F.2d 1531 (8th Cir. 1989), next maintain that the perceptions of the competitive environment contained in these documents are insufficient to exclude other media from the market. NAT Br. at 30-31; see also id. at 11, 13 n.17, 34-35 (contending that the district court erroneously failed to define the relevant market from the perspective of consumers). However, this circuit has made clear that "testimony of market participants is relevant to a determination of [the] market." Freeman, 1995 WL 638354 at *9.⁽²⁷⁾ When such evidence has been rejected as deficient, it is not because market participant testimony never can suffice to delineate the relevant market, but because, in context, the particular evidence was not sufficiently probative of the crucial question before the court.

For instance, in H.J., the plaintiffs' internal documents consisted of "casual" and "conclusory statements" that "alternatively discount[ed] and recognize[ed]" another product as competition. 867 F.2d at 1540. Such contradictory statements did not constitute "hard evidence" that permitted the conclusion that two products did not compete. Id. Similarly, this Court, examining its recent precedents on geographic market definition, explained in Freeman that "the views of market participants are not always sufficient to establish the relevant market, especially when their testimony fails to specifically address the practicable choices available to consumers." Freeman, 1995 WL 638354 at *9 (discussing Bathke v. Casey's Gen. Stores, Inc., 64 F.3d 340, 345-46 (8th Cir. 1995) (emphasis added)). The point of these cases is that the evidentiary force of competitors' perceptions depends on context, which is generally for the district court to evaluate. See Freeman, 1995 WL 638354 at *9-*10. They announce no requirement, as Appellants insist, that the perceptions of competitors are never sufficient or that the market must always be established through detailed studies.⁽²⁸⁾

Here, in contrast to the casual, contradictory, and conclusory statements found insufficient in H.J., plans and reports generated over several years focused, with respect to readers, on competition between the Morning News and the Times to the virtual exclusion of all else. Although, as Appellants note, see NAT Br. at 30-31 n.29, George Smith at trial attempted to recharacterize these statements as morale-building rhetoric, the district court correctly rejected Smith's attempt to walk away from "the contemporaneous, prelitigation" record. 892 F. Supp. at 1156. Moreover, in light of detailed and consistent evidence, spanning several years, indicating that the papers did not regard other media as competitive threats, the district court permissibly inferred that other media did not supply viable alternatives for readers.

In any event, this is not a case in which market delineation depends solely on the perceptions of current market incumbents.⁽²⁹⁾ The government's expert testified that other media do not belong in the market, and contrary to Appellants' contention, see NAT Br. at 34-35, he based that conclusion not only on the documentary evidence described above, but also on his analysis that "the reduction in profitability of [the Times and the Morning News] after the [outbreak of] competition . . . around 1990 basically amounted to a ten percent transfer away from the shareholders toward the consumers" (T. 918, GA 137). This appraisal reflects that, although the Times and the Morning News undertook numerous competitive reactions to one another, there is no evidence of them taking steps to combat the competitive moves of other media.⁽³⁰⁾ If, as Appellants contend, see NAT Br. at 31-32 n.30, other media had the capacity to take readership away from the Times and the Morning News, one would expect to see such reactions. But there is no evidence of the Times responding, for instance, to the dramatic growth in the cable industry.⁽³¹⁾ Supporting Baseman's analysis is testimony that in Little Rock the surviving local daily paper was "freer to set circulation rates or circulation prices higher" than when it faced competition from a recently vanquished competing local daily (T. 495, GA 53) despite the existence of other media.

3. Finally, Appellants contend that the court failed to conduct the fact-based inquiry required by this Court's market-definition precedents but instead "proclaim[ed] that the majority of other courts have held that daily newspapers constitute a definitive market from other media." NAT Br. at 33-34. This contention is meritless. The district court quite clearly "made its own findings of fact," 892 F. Supp. at 1157, and simply observed that precedent supported the conclusion it reached. See id. at 1156-57.⁽³²⁾

3. THE DISTRICT COURT CORRECTLY DETERMINED THAT THE TIMES AND THE MORNING NEWS COMPETE FOR ADVERTISERS IN THE SAME LOCAL DAILY NEWSPAPER MARKET

Because the Times and the Morning News compete for readers in the same local daily newspaper market, this Court may affirm the district court's holding that the acquisition violates section 7 in at least one relevant market. However, the district court also correctly determined that the Times and the Morning News compete for advertisers, and that other advertising vehicles do not belong in the relevant market.

1. The Times And The Morning News Compete For Advertisers Through The Feedback Loop

The district court found only a "small amount of evidence that the [Times and the Morning News] compete `directly' for advertisers." 892 F. Supp. at 1159. However, it accepted testimony that the papers compete indirectly for advertising through the "negative feedback loop." This phenomenon, recognized to pervade newspaper markets,⁽³³⁾ explains why, as demonstrated below, the Times and the Morning News undertook a number of competitive actions aimed at obtaining advertising while at the same time denying it to the other paper. The feedback loop can operate in the following manner: Even if two papers, because they have few common readers, are not viewed by advertisers as direct substitutes, competition between the papers for readers nonetheless generates incentives for both papers to keep advertising rates down. If one paper raises advertising rates, it will sell less advertising. Diminished advertising may cause readers, who purchase papers in part for their advertising content (T. 678, PA 121), to switch to the competing paper; and reduced readership, in turn, makes the paper less attractive to advertisers (T. 877-80, GA 111-14). Because of the threat that a price increase, or quality decrease, might initiate this unprofitable cycle, the Times and the Morning News, as the government's expert testified (T. 876-82, GA 110-16) and the district court found, constrain one another from exercising market power over advertisers and therefore belong in the same market.

Appellants' various criticisms of the district court's reliance on the feedback loop are unsound. They contend that the feedback loop "cannot get started" because the Times and the Morning News are not substitutes for readers. NAT Br. at 37 & n.37.⁽³⁴⁾ As explained above, Appellants' premise is incorrect; the two papers belong in the same market for circulation. Appellants also contend that the theory is refuted by historically distinct advertising rates between the papers, pointing in particular to evidence that the Times' CPM⁽³⁵⁾ "approached three times that of the [Morning] News." NAT Br. at 38. However, differing CPM rates may reflect different demographics of the papers' current subscribers. The CPM disparity alone does not answer the critical question: whether the papers could increase rates or decrease quality from existing levels without fear of initiating the deleterious effects of the feedback loop.

On this issue, the record demonstrates that the Times and the Morning News were particularly concerned with losing key advertisers to the other paper, and that, as a consequence, the papers offered promotions and discounts designed to prevent advertisers from reducing linage.⁽³⁶⁾ Advertisers, in turn, testified that they would purchase greater linage as circulation increased (T. 714-16, GA 78-80). The papers, moreover, well understood the dynamics of the feedback loop. For instance, a Morning News monthly report recognized that increased circulation yields additional advertising.⁽³⁷⁾ And, the Morning News' publisher explained: "We need to plan [on] keeping the [advertising] rate low enough to get the [Morning News] established as the main advertising buy in Northwest Arkansas. When we raise rates, we want our advertisers to cut their lineage in the Times and Daily Record, not the News" (PX 44 at DONR-10702, PA 500).

Appellants finally state that "[t]he `negative feedback loop' theory simply cannot apply in this case where the overwhelming evidence proves that the Times and the [Morning] News are not substitutes." NAT Br. at 39-40. To the extent Appellants mean that direct substitutability of one paper for another from the perspective of advertisers is a prerequisite to the feedback loop's operation, they offer no basis for this assertion, and both their expert and the government's expert disagreed (T. 877-81, 2148-49, GA 111-15, 163-64; T. 2065-67, A 278-80). To the extent Appellants imply that the constraining effect of the feedback loop is insufficient to place two papers in the same market in the absence of present substitutability, they ignore the Supreme Court's admonition that market delineation must strive to "conform to competitive reality," United States v. Continental Can Co., 378 U.S. 441, 457 (1964); see also Archer-Daniels-Midland, 866 F.2d at 246 (explaining that "[r]easonable interchangeability and cross-elasticity of demand are not used to obscure competition but to recognize competition . . . to the extent such exists" and merely constitute tools that "help evaluate the extent competition constrains market power" (emphasis added)).⁽³⁸⁾

2. The District Court Correctly Excluded Other Advertising Vehicles From The Market

Appellants briefly contend, relying primarily on precedent they elsewhere discard as irrelevant, that the district court erred in excluding other media from the advertising market. See NAT Br. at 46-47 (citing Paschall v. Kansas City Star Co., 727 F.2d 692, 701 (8th Cir.), cert. denied, 469 U.S. 872 (1984)). However, ample evidence in the record supports the district court's finding. Advertisers testified that television and radio provide poor substitutes for local daily newspapers and instead properly are viewed as complements to them (T. 440-42, 471-72, 702-03, 714-15, GA 34-36, 51-52, 74-75, 78-79; T. 1363-65, PA 255-57). They similarly testified that direct mail (T. 446-47, GA 40-41), "shoppers" (T. 396-97, 407-08, 444-45, 703-05, GA 22-25, 38-39, 75-77; T.1365, PA 257), weekly papers (T. 409-11, 714-15, GA 26-28, 78-79; T. 1365, PA 257), and state-wide and out-of-state papers (T. 455-56, 659-60, GA 49-50, 63-64; T. 1372, PA 264), are not viable substitutes in the Fayetteville area market. Indeed, one advertiser testified that, if confronted with fifteen to twenty percent increase in newspaper advertising prices, his probable response would be to "cut out some radio and television," rather than to switch to them (T. 714, GA 78).

As noted above, a number of the Times' documents mention the advertising rates for, and the advertisers gained or lost by, other media. However, these documents primarily demonstrate that the Times engaged in what Appellants define as "benchmarking,"⁽³⁹⁾ a practice that Appellants themselves argue does not evince competition. See NAT Br. at 40 & nn.43-44. Even if some of the documents show more than that, the advertiser testimony described above is sufficient to support the district court's exclusion of media other than local daily newspapers from the relevant advertising market.

4. THE DISTRICT COURT CORRECTLY CONCLUDED THAT THE ACQUISITION MAY SUBSTANTIALLY LESSEN COMPETITION

1. The District Court Properly Found The Acquisition Presumptively Unlawful And That The Defendants Failed To Rebut The Presumption

The district court, after defining the relevant markets, found the acquisition presumptively unlawful because it resulted in the Stephens family controlling "in excess of 84% of circulation and 88% of advertising revenue." 892 F. Supp. at 1168.⁽⁴⁰⁾ Appellants contend that the district court's flawed market definition renders infirm the district court's reliance on these market shares. See NAT Br. at 47-48. However, as explained above, at a minimum the court correctly concluded that the Times and the Morning News compete in the same local daily newspaper market for readership (as well as for advertising), the market that supports the court's alternative holding, and the Stephens family's post-acquisition share of that market is even higher than in the broader "Northwest Arkansas" market. See supra p.7. Accordingly, the presumption of illegality from undue market share properly applies to an assessment of the acquisition in the narrower market that underlies the district court's alternative holding, and the acquisition is presumptively unlawful in that market.

This leaves Appellant's one-sentence objection that they "adduced evidence of low barriers to entry, low concentration, and likely pro-competitive benefit to consumers, sufficient to rebut the presumption." NAT Br. at 49.⁽⁴¹⁾ The record, however, amply supports the district court's finding of "formidable," 892 F. Supp. at 1168, barriers to entry. To wrest readers from established local daily papers is a high risk undertaking, requiring a new entrant to defeat existing loyalties and to incur considerable, unrecoupable sunk costs (T. 455-56, 839, 868-69, GA 49-50, 92, 102-03). Moreover, until substantial readership is established, attracting advertising is both difficult and expensive (T. 683-84, 743, GA 70-71, 85). The evidence, including the testimony of Appellants' own expert (T. 2226-29, GA 191-94), also supports the court's conclusion that these already high barriers to entry would be raised by the acquisition -- specifically by rendering much less likely entry into the market by the Little Rock-based Arkansas Democrat-Gazette (T. 503-07, 663, 670-71, 735-38, 769-70, 774, 838-40, GA 54-58, 65, 68-69, 81-84, 86-87, 88, 91-93; T. 515-18, 542-43, 787-89, 825, 869, PA 108-11, 114-15, 154-56, 163, 166; PX 174 at WEHC-20120, GA 366). The evidence additionally supports the court's judgment that, even if the Democrat-Gazette entered the market with a regional "zoned" edition, its impact would not be sufficient to alleviate the acquisition's anticompetitive effects (GXs 363-64, GA 352-57; T. 564-67, 668-69, 694-95, 738, 828-29, 887-93, 973-74, 1559-60, GA 59-62, 66-67, 72-73, 84, 89-90, 121-27, 140-41, 151-52; T. 454-55, 855-56, 1556-57, PA 99-100, 164-65, 312-13).

Finally, Overstreet's assertion that the papers, under common ownership, would offer lower rates to advertisers (T. 2054-55, 2228, A 267-68, 330) did not demonstrate, by clear and convincing evidence, see United States v. Rockford Mem. Corp., 717 F. Supp. 1251, 1288-89 (N.D. Ill. 1989), aff'd, 898 F.2d 1278 (7th Cir.), cert. denied, 498 U.S. 920 (1990), that the merger would result in significant efficiencies that could not be achieved but for the acquisition. Presumably, the papers could offer some joint advertising even if independently owned.⁽⁴²⁾ And, in light of the Stephens family's professed intentions not actually to merge the operations of the two papers (T. 1102, A 137), Appellants did not show that the acquisition would result in significant cost savings (T. 869, GA 103), as Overstreet claimed (T. 2092-93, A 305-06). In any event, even if Appellants established these efficiencies, there is no basis for concluding that they would outweigh, as they must, see Rockford, 717 F. Supp. at 1289; FTC v. University Health, Inc., 938 F.2d 1206, 1222 (11th Cir. 1991), the acquisition's demonstrable anticompetitive effects.⁽⁴³⁾

2. The District Court Correctly Aggregated The Stephens Family's Shares in Donrey And NAT

Appellants argue that, even if the district court correctly applied standard section 7 analysis, the district court erred in "reviewing this case" under that provision, NAT Br. at 4; Donrey Br. at 41, because the court improperly aggregated the Stephens family's interests in NAT and Donrey. The Court should reject Appellants' attempt to elevate the form of a transaction over its economic substance.

1. The district court's decision to aggregate the Stephens family's shares of Donrey and NAT rested on a finding that the relevant members of the Stephens family, at least with respect to the family's investments, will not pursue separate economic interests.⁽⁴⁴⁾ This factual finding, which is reviewed for clear error,⁽⁴⁵⁾ is amply supported by the record.

Jack Stephens and his sister-in-law, Bess Stephens, together own a majority interest in Stephens Group, Inc. ("SGI") (T. 1015, PA 170), which Bess Stephens described as "the parent company of all of the Stephens' interest[s]" (B. Stephens Dep. 22, GA 389). SGI owns 99% of Donrey, which owns 100% of the Morning News. Jack Stephens' son, Warren Stephens, is SGI's president and CEO (T. 1015, GA 142) and serves at his father's and Bess' will (T. 1016, PA 171). The Stephens family members regularly make joint investments (B. Stephens Dep. 22, 24-25, GA 389, 391-92), and the precise percentage of a particular investment held by each family member simply depends on the number who participate in the venture (id. at 22-23, 25-26, GA 389-90, 392-93).

The Stephens family's investment philosophy, as Warren Stephens testified, has always "focused on the long run"; the family "think[s] in terms of generations. . . and . . . in terms of building value for the next generation" (T. 1336, 1342, GA 149-50). Consistent with this objective, his subordinate position in SGI, the general structure of the family's investments, and his position as principal beneficiary of Jack Stephens' estate (T. 1016-18, PA 171-72), Warren Stephens never has competed directly against his father in a business venture (T. 1014, 1278, PA 169, 240).

The genesis of NAT, Bess Stephens explained (B. Stephens Dep. 24, GA 391), was typical of how the family operates its investments. In February 1995, SGI distributed a dividend to various Stephens family trusts, and some of these funds were exchanged for shares in NAT; additionally, trusts controlled by Jack and Bess Stephens loaned $13 million, obtained via the dividend, to NAT (PX 120, PA 556; T. 1068-73, PA 188-93). These loans were unsecured (T. 1086-87, PA 197-98), and the precise interest in NAT held by members of the Stephens family were not determined until following NAT's acquisition of the Times (B. Stephens Dep. 15-16, GA 386-87). NAT was formed, moreover, only after Jack Stephens indicated to Donrey that it should back off from its own plans to acquire the Times (T. 1857-60, GA 153-56).

These facts alone are sufficient to sustain the district court's finding. Even if, as NAT proposed to the court after consummating the transaction, Warren Stephens replaced Jack Stephens as NAT's CEO, and the family's equity interests in NAT were reshuffled, the Stephens family members still would lack incentives to compete.⁽⁴⁶⁾ Not only do the family members possess multiple common investments that in and of themselves might dampen intra-family competitive rivalry elsewhere, but also they stand to benefit financially from maximizing the wealth of the family, an objective inconsistent with the Times and the Morning News engaging in a robust rivalry. The Stephens family members plainly do not treat one another as arms-length investors, but willingly act, as Warren Stephens described, in what is perceived as the family's best interest, a "common sense economic" result (T. 943, A 114), as even Appellants' expert had to concede, see T. 2210 (GA 190) (explaining that, "[s]etting everything else aside" "people will act in the self-interest of the family unit"). Thus, although Appellants claim that the district court merely presumed common economic interests from the existence of family ties and thereby created "new antitrust law," Donrey Br. at 38, the district court plainly had an adequate basis to conclude on this record that "the Stephens family members have little, if any, incentive to compete aggressively against themselves." 892 F. Supp. at 1169.

2. Appellants contend that the district court's factual finding, even if correct, is legally irrelevant. Combining the market shares of separate firms to presume a substantial lessening of competition is appropriate, they claim, only when "a single person or entity" controls both firms. Donrey Br. at 25-28. They further argue that overlapping minority interests cannot be considered the equivalent of a "single person or entity" unless those possessing such interests (1) have expressly joined together to act in concert, see id. at 29-30; (2) constitute a single enterprise under Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984), or its "logic," see Donrey Br. at 32-34, 38; or (3) profess an intent to use, or actually use, a minority interest in one firm to accomplish anticompetitive consequences, see id. at 31-32, 35-36.

This analysis fundamentally misapprehends the rationale underlying the presumption of illegality from undue market share employed in a Clayton Act section 7 case. The presumption is not justified merely by a single individual's or entity's control of the relevant assets. Rather, it reflects a view that, when an undue percentage of the market can be assumed to act in the furtherance of a single interest or in a manner that otherwise seriously impairs preexisting rivalry, deleterious market consequences are sufficiently likely so as to justify shifting the burden of proof to the defendants. Nothing in section 7 or the case law relied upon by Appellants suggests a limitation on the range of factual circumstances in which that assumption might be warranted, other than, as the district court recognized, see 892 F. Supp. at 1171, that the consequences feared must be related in some way to an "acquisition."

To be sure, the assumption that firms will act in furtherance of a single interest arises in the ordinary case because two previously separate firms arrive, via the acquisition, at what under principles of corporate law constitutes "common control." See, e.g., United States v. Philadelphia Nat'l Bank, 374 U.S. 321, 363, 365 (1963); Grumman Corp. v. LTV Corp., 665 F.2d 10, 11 (2d Cir. 1981). Sometimes, however, less than what corporate law might consider control suffices. See, e.g., United States v. Tidewater Marine Serv., Inc., 284 F. Supp. 324, 332 (E.D. La. 1968) (50% controlled subsidiary). And, consistent with the underlying rationale described above, courts have applied the presumption when only a significant minority interest has been obtained through an acquisition, on the theory that substantial anticompetitive effects are probable in such circumstances even if control is not ultimately attained and even when anticompetitive effects have not yet occurred. See, e.g., F. & M. Schaefer Corp. v. C. Schmidt & Sons, Inc., 597 F.2d 814, 815-16, 818 (2d Cir. 1979) (per curiam).⁽⁴⁷⁾

The case law similarly recognizes that, in appropriate factual settings, familial relationships provide a basis for inferring that minority interests are likely to act in furtherance of a common interest.⁽⁴⁸⁾ For instance, in Consolidated Gold Fields PLC v. Minorco, S.A., 871 F.2d 252 (2d Cir. 1989), the court upheld, for the purposes of analyzing an acquisition under section 7, the district court's aggregation of market shares of several entities based on the "Oppenheimer family['s]" partial ownership and "intertwined relationship[]" with them. Id. at 255, 261. And in American Crystal Sugar Co. v. Cuban-American Sugar Co., 152 F. Supp. 387 (S.D.N.Y. 1957), aff'd, 259 F.2d 524 (2d Cir. 1958), the court appeared to attribute to the defendant not only his shares but also those, inter alia, of his "immediate family." Id. at 392. Appellants claim the attribution in American Crystal hinged on "[d]efendants' avowed objective" of "bring[ing] about [a] `closer connection' between the two companies, either by merger or common control." Id. at 392. See Donrey Br. at 31. However, they confuse this -- the basis on which the court, given the mere minority interest collectively possessed by the defendant and others, found a section 7 violation -- with the reason for the aggregation: that immediate family "were likely to accept [defendant's] advice." American Crystal, 152 F. Supp. at 392. The degree of ownership of a company necessary to invoke the presumption of harm to competition is distinct from the question of whether, in determining the percent owned, it is proper to aggregate family members' interests.⁽⁴⁹⁾

Here, there is no doubt the shares of Donrey and NAT collectively possessed by the Stephens family rise to the level that warrants invoking the presumption; the only question is the validity of the district court's basis for the aggregation. And, as described above, the ground on which the court found market share attribution appropriate -- that the Stephens family acts with a single economic purpose -- comports with the harm-to-competition rationale for applying the presumption.⁽⁵⁰⁾

3. Finally, Appellants claim that market share attribution is improper because "[a]t best" this case presents "a premature claim under § 1 of the Sherman Act" for "[n]o anticompetitive consequences possibly arise as a result of NAT's acquisition of the Times unless the Times and the Morning News . . . are combined or the respective minority owners otherwise engage in some anticompetitive, concerted conduct." Donrey Br. at 39. In short, Appellants contend that because all the anticompetitive effects of the acquisition may be vanquished by the antitrust laws if and when they occur, there is no need presently to invoke section 7. See id. at 23 & n.22, 42-43. At a minimum, this argument is factually flawed; as a proposition of law, it represents a frontal assault on section 7's fundamental policies.

As the district court found, see 892 F. Supp. at 1170, competition may be diminished by NAT's acquisition of the Times in many ways that do not depend on members of the Stephens family engaging in conspiracies in restraint of trade. Knowing that their employers' collective wealth is maximized by diminished competition, the key managers of the Times and the Morning News may refuse to compete vigorously. In recognition of the same economic realities, the Stephens-family-dominated boards of Donrey and NAT similarly may choose to scale back their investments in each paper.

These consequences, contrary to Appellants' contention, see Donrey Br. at 23 & n.22, are not merely hypothetical. As found by the district court, see 892 F. Supp. at 1170, substantial evidence exists that the papers' constraining effect on one another diminished following the acquisition.⁽⁵¹⁾ While the Stephens family controlled the Times, Warren Stephens obtained competitively sensitive information concerning the Morning News from Donrey (PXs 42-43, PA 489-97; T. 1346-47, PA 246-47). Moreover, during this period, the papers' previously robust competitive rivalry substantially abated. Prior to the acquisition, Thomson's partially implemented strategic plan for the paper called for establishing the Times as the dominant newspaper in Northwest Arkansas (GX 83 at NAT01-00083, -85, GA 226, 228). But that competitive goal was abandoned following the acquisition by NAT (T. 257-58, 347-48, PA 62-63, 92-93). Moreover, although the Times previously sought to hire staff from the Morning News (Westerfield Dep. 45, GA 395), the Times hired no new employees from the Morning News following the acquisition (id.). These anticompetitive effects did not result from express agreements among members of the Stephens family; instead they largely reflect that the papers' managers know who their bosses are (1 T. P.I.H. 234, GA 19).

But even if, contrary to the record, the antitrust laws may be invoked to condemn all the potential anticompetitive effects of the acquisition when they manifest, that provides no basis for refusing to hold the acquisition unlawful under Clayton Act section 7. That provision condemns any "acquisition" the effect of which "may be substantially to lessen competition." 15 U.S.C. 18. Section 7 thus incorporates, as Appellants concede, an "incipiency standard." See, e.g., United States v. E. I. Du Pont de Nemours & Co., 353 U.S. 586, 589 (1957) ("Section 7 is designed . . . to arrest in their incipiency restraints . . . in a relevant market which, as a reasonably probability, appear at the time of suit likely to result from the acquisition . . . ."). Although an important purpose of section 7 is to prohibit acquisitions that may facilitate anticompetitive conduct not otherwise within the ambit of the antitrust laws, cf. United States v. Penn-Oil Chem. Co., 378 U.S. 158, 170-71 (1964), that exhausts neither the reach of the statute on its face nor its underlying purposes. For instance, in American Crystal, the anticompetitive effects feared depended on the defendant's future acquisition of additional stock, see 152 F. Supp. at 400, conduct plainly within the reach of the antitrust laws. The court nonetheless found a Clayton Act section 7 violation based on acquisitions to date, explaining that the Clayton Act "permit[s] courts to forestall [anticompetitive results] in limine rather than waiting until events proceed to such a point that there has been actual violation of the Sherman Act," id. at 396, a point reiterated by the court of appeals, see 259 F.2d at 527 (explaining that "conduct may fall under the ban of amended § 7 before it has attained the stature of an unreasonable restraint of trade"); see also e.g., Gulf & Western Indus., Inc. v. Great Atlantic & Pac. Tea Co., 476 F.2d 687, 693-95 & n.9 (2d Cir. 1973) (rejecting "as a matter of law" the contention that section 7 does not condemn an acquisition when control has not yet been obtained even assuming that all the anticompetitive effects of the acquisition depend on the defendant's attaining control, which the court anticipated would be accomplished through future stock acquisitions); Hamilton Watch Co. v. Benrus Watch Co., 114 F. Supp. 307, 316 (D. Conn.) (finding a violation when the defendant sought control through stock acquisitions but ceased purchasing stock when a voting trust established effective control), aff'd, 206 F.2d 738 (2d Cir. 1953).

It accordingly would flout Congress' intent to refuse to find an acquisition to violate section 7 merely because of the possibility that the anticompetitive effects feared might be remedied by other provisions of the antitrust laws when they arise, even if it were guaranteed -- which it is not -- that such violations would come to the attention of antitrust enforcement authorities. Here, of course, when actual and anticipated competitive harm resulting from the acquisition might not otherwise be reached by the antitrust laws, the same conclusion applies with even greater force. Although proclaiming that the district court created "new antitrust law," Donrey Br. at 38, it is Appellants who seek to nullify the law that Congress wrote.

The district court's Judgment and Order of Rescission should be affirmed.

Respectfully submitted.

Of Counsel: Craig W. Conrath Philip R. Malone Allee A. Ramadhan Alexander Y. Thomas Brigid L. Thomas    Attorneys    Antitrust Division    U.S. Department of Justice    1401 H Street, N.W.    Washington, D.C. 20530

Anne K. Bingaman    Assistant Attorney General Joel I. Klein    Deputy Assistant Attorney General Robert B. Nicholson Mark S. Popofsky    Attorneys    Antitrust Division    Appellate Section    U.S. Department of Justice    10th & Pennsylvania Ave., N.W.    Washington, D.C. 20530    (202) 514-3764

December 1, 1995

CERTIFICATE OF SERVICE

I hereby certify that on December 1, 1995, I caused a copy of the foregoing BRIEF FOR APPELLEE UNITED STATES OF AMERICA to be served upon the following counsel in this matter by overnight courier:

Philip S. Anderson, Esq. Williams & Anderson 111 Center Street, 22nd Floor Little Rock, Arkansas 72201 W. Bassett, Jr., Esq. Bassett Law Firm P.O. Box 3618 Fayetteville, Arkansas 72702 Thomas Burke, Esq. Burke & Eldridge 201 N. East Street Suite 102 Fayetteville, Arkansas 72701 W. Jackson Butt, II, Esq. Davis, Cox & Wright 19 East Mountain Street Fayetteville, Arkansas 72702

James M. Dunn, Esq. Warner & Smith 214 North 6th Street P.O. Box 1626 Fort Smith, Arkansas 72902 Jeremy G. Epstein, Esq. Shearman & Sterling 153 East 53rd Street New York, New York 10022 Jerry C. Jones, Esq. Rose Law Firm 120 East Fourth Street Little Rock, Arkansas 72201

__________________________________ Mark S. Popofsky    Attorney    Antitrust Division    Appellate Section    U.S. Department of Justice    10th & Pennsylvania, Ave., N.W.    Washington, D.C. 20530    (202) 514-3764

FOOTNOTES

1. In the two Fayetteville zip codes encompassed within the relevant market alleged by the United States, the Morning News' daily circulation is 3,986 compared to 8,007 for the Times; in the two Springdale zip codes so alleged, the Times' daily circulation is 1,856 compared to 10,136 for the Morning News (GX 361, GA 349). These four zip codes comprise almost the entirety of the two communities and include some surrounding territory.

2. The Morning News also circulates in Rogers, a town located to the north of Springdale. Indeed, until November 1994 the Morning News published two separate papers, one based in Springdale and one in Rogers. The Benton County Daily Record ("Daily Record") circulates primarily in Bentonville and Bella Vista, located to the northwest of Rogers, but has some Rogers circulation.

3. The court found it appropriate to aggregate the Stephens family's holdings in NAT and Donrey because the facts showed that members of that family would not vigorously compete against one another. See 892 F. Supp. at 1169-72.

4. The court also determined that both private plaintiffs possessed standing under the Clayton Act to pursue their claims. See id. at 1165-67.

5. We recognize that the district court did not determine the precise geographic contours of a local daily newspaper market that excludes the Daily Record, but as implicitly acknowledged by the reference to the government's market in the court's alternative holding, determining that market's precise boundaries down to the last zip-code was unnecessary. See, e.g., United States v. Continental Can Co., 378 U.S. 441, 456 (1964) ("`[T]he "market," as most concepts in law or economics, cannot be measured by metes and bounds. . . . Obviously, no magic inheres in numbers.'" (quoting Times-Picayune Pub. Co. v. United States, 345 U.S. 594, 611-12 (1953))); United States v. Philadelphia Nat'l Bank, 374 U.S. 321, 360 n.37 (1963) (explaining that some "fuzziness would seem inherent in any attempt to delineate the relevant geographic market"); United States v. Empire Gas Co., 537 F.2d 296, 304 (8th Cir. 1976) (same), cert. denied, 429 U.S. 1122 (1977); cf. FTC v. Freeman Hosp., Nos. 95-1448, -2882, 1995 WL 638354, at *11 (8th Cir. Nov. 1, 1995) (rejecting the contention that the acquisition was unlawful "within any of the three alternative markets" identified when the agency failed to establish facts necessary to define any of these markets). Any alternative local daily newspaper market encompassing only the Times and the Morning News would yield, in view of the district court's other findings, virtually 100% post-acquisition Stephens family market dominance. This is true whether the market consists of the four Fayetteville and Springdale zip codes identified by the government, the slightly broader market of Washington County, in which the district court expressly found "that the Times and the Morning News strongly compete against each other for readers and advertisers," 892 F. Supp. at 1158, or any comparably sized market.

Moreover, as the government's expert noted (T. 899-900, GA 133-34), there is no theoretical inconsistency in concluding that the market found by the district court and the market advanced by the United States both are properly delineated. Under the 1992 Merger Guidelines, markets may be defined for each merging firm, see Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines § 1.0 (Apr. 2, 1992) ("Guidelines"); in some instances, this analysis may result in distinct yet overlapping relevant markets. In this case, defining the relevant markets for the Times may yield, as the government argued, markets for readers of, and advertising in, local daily newspapers that include only the Times and the Morning News. However, defining the markets from the perspective of the Morning News may well yield the conclusion, reached by the district court, that the Daily Record is in those relevant markets, even if the Daily Record does not belong in the relevant markets delineated for the Times. Because the government believed that NAT's acquisition of the Times violated § 7 in the narrower market alleged, it had no need to challenge the acquisition's legality in a broader market.

6. Although Appellants contend that the Times and the Morning News belong neither in the same geographic market nor in the same product market, a single overarching argument underlies both objections: that the papers are not substitutes from the perspective of consumers. See NAT Br. at 12, 25-29, 36-46. Appellant's overlapping analysis reflects, as the Guidelines explain, see Guidelines § 1.1 n.8, the interrelated nature of the product and geographic components of the market. We accordingly do not address these dimensions separately.

7. Market definition presents a question of fact reviewed for clear error. See Alexander v. National Farmers Organization, 687 F.2d 1173, 1192 (8th Cir. 1982). "`Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous.' Likewise, a factual finding that is supported by substantial evidence on the record cannot be clearly erroneous." Ricks v. Riverwood Int'l Corp., 38 F.3d 1016, 1018 (8th Cir. 1994) (quoting Anderson v. City of Bessemer City, 470 U.S. 564, 574 (1985)).

8. If the Morning News' Fayetteville distribution for 1989, 1,611 daily and 2,011 Sunday (A 421), is used instead of the 1990 figures, the Morning News' gains in Fayetteville were even more impressive. These numbers do not include the Morning News' Rogers edition.

9. By the time the increase was implemented in January 1994, the Times, as discussed below, increased the price to $1.00 (A 412).

10. We use "elasticity" of demand to encompass both the "cross-elasticity" test, which measures the sales gained by only one product in response to a price increase by another, see United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377, 400 (1956), and the "own" elasticity test of the Merger Guidelines, which takes into account all sales lost by a producer from its imposition of a price increase or equivalent quality reduction. See United States v. Eastman Kodak Co., 63 F.3d 95, 108 (2d Cir. 1995) (defining own elasticity). Both these approaches have been endorsed by this Circuit as permissible tools of market delineation. See United States v. Archer-Daniels-Midland Co., 866 F.2d 242, 246, 248 (8th Cir. 1988) (applying both the Guidelines' "hypothetical monopolist" approach and the cross-elasticity test), cert. denied, 493 U.S. 809 (1989); H.J. Inc. v. International Tel. & Tel. Corp., 867 F.2d 1531, 1537-38 (8th Cir. 1989) (same). The difference between them is immaterial for the purposes of this appeal because Appellants concede that both the Guidelines and the cross-elasticity test supply appropriate market delineation principles. See NAT Br. at 10-11, 16-17, 45-46. We note, however, that Appellants erroneously appear to equate the two tests. See id. at 17, 36, 46.

11. Appellants rely on Sun Newspapers, Inc. v. Omaha World-Herald Co., 1983-2 Trade Cas. (CCH) ¶ 65,522 (D. Neb. June 14, 1983), aff'd in part and modified in part, 713 F.2d 428 (8th Cir. 1983), for the proposition that a newspaper with only a 37% penetration of the relevant market is not competing in that market. See NAT Br. at 13-15. Sun Newspapers stands for no such principle. There, the court was required to determine, in a Sherman Act § 2 case, the proper geographic market in which to assess whether the World-Herald possessed monopoly power. In making this determination, the court included counties in which the paper's penetration approximated 80%, for in those counties the paper "total[ly] lack[ed] strong competitor[s]." 1983-2 Trade Cas. at 68,590. This same conclusion could not be reached for other counties in which the World-Herald achieved less success; thus those counties were excluded from the relevant market. See id. at 68,586. Put otherwise, the county in which the World-Herald achieved a penetration of only 37% was excluded not because of a finding that the paper did not there compete, but because a narrower area existed in which other papers did not constrain the World-Herald's power. Having found the World-Herald to possess monopoly power in a relevant market, the court had no occasion to define others.

12. For the Sunday editions, the switch need number only 1,800 (GX 361, GA 349).

13. Appellants, in arguing that the Times and the Morning News are not substitutes, point to the testimony of two Fayetteville residents. NAT Br. at 26. But their testimony cannot possibly require ignoring that the Morning News has a Fayetteville circulation of 4,000. See also infra note 14 (noting testimony that both papers cover the entire region). Appellants also rely on Dr. Overstreet's assertion that the Times and the Morning News "are distinct products with distinct hometown newspaper flavors." NAT Br. at 26. The district court permissibly rejected this testimony. Dr. Overstreet offered this opinion largely as a means of explaining calculations that demonstrated, in his view, the absence of competition between the Times and the Morning News (T. 2158-59, GA 167-68). But these calculations, as the district court recognized, were flawed. See infra pp.17-19. Moreover, as the district court observed, Dr. Overstreet's testimony "leaves unexplained why everyone involved with these papers thought they were competing and made numerous business decisions and took innumerable competitive actions as if they were." 892 F. Supp. at 1162.

14. See, e.g., GX 21 at TC 004477 (PA 802) (news and advertising sharing agreement between the Times and the Daily Record); GX 26 at TC 001941 (PA 827) (same); GX 83 at NAT01-00084-86 (GA 227-29) (explicating the Times' aspiration to appeal to readers in both Washington and Benton counties); T. 1220 (PA 235) (Morning News' publisher agrees that the paper's goal is "to be a true regional newspaper for all of northwest Arkansas"); T. 412-16 (GA 29-33) (noting the Times' and the Morning News' coverage of the entire region); T. 439-40 (A 62-63) (explaining that both the Times and the Morning News "cover the whole northwest area").

15. Although Appellants claim that "[i]t cannot be overemphasized" that the 100% price increase, which represented "not a `small but significant' price increase" but instead "a doubling of price," "failed to reflect a significant cross-elasticity of demand," NAT Br. at 27 n.26, they ignore the simple mathematical implications of having such a large price increase. With a 100% price increase, even a 100% quantity reduction would yield an elasticity of only 1. It was accordingly mathematically impossible for Dr. Overstreet's exercise to yield an elasticity nearly as great he required.

16. Conceptually, the effect of Dr. Overstreet's error was to overstate greatly the Times' critical elasticity and to understate the actual elasticity for the price increase examined.

17. See infra p.30.

18. Dr. Overstreet testified that he attempted to estimate the Times' elasticity of demand through econometrics but was unable to do so (T. 2187-89, GA 172-74).

19. Appellants also point to Dr. Overstreet's testimony that if the Times and the Morning News were close substitutes, the Times would not have had "the luxury" (T. 2101, A 314) of waiting for two years before following the Morning News in converting to morning delivery. See NAT Br. at 28 n.28. However, Dr. Overstreet did not explain why the Morning News' substantial circulation gain in Fayetteville after converting to morning delivery did not evince a fairly high elasticity of demand.

20. Indeed, one court presented with estimated elasticities found the evidence "completely useless, primarily because we have no basis for evaluating what a particular elasticity coefficient means." United States v. Mrs. Smith's Pie Co., 440 F. Supp. 220, 227-28 (E.D. Pa. 1976).

21. This includes, inter alia:

(1) evidence that buyers have shifted or have considered shifting purchases between products in response to relative changes in price or other competitive variables;

(2) evidence that sellers base business decisions on the prospect of buyer substitution between products in response to relative changes in price or other competitive variables . . . .

Guidelines § 1.11. The same nonexclusive factors apply to geographic market definition. See id. § 1.21.

22. As Baseman testified, there was "information in the record of a quantitative nature" that allowed him "to be comfortable with [his] conclusion that these firms are in the same market under a 5 or 10% merger guidelines test. You can get there with econometrics; you can get there other ways" (T. 913, A 95).

23. See GX 26 at TC 001942 (A 530) (discussing the demise of a "long-term weekly alternative newspaper and Thrifty Nickel" in the wake of the Times' "[t]arget[ing of] non-advertisers that participated in competition"); GX 24 at TC 002226 (A 527) (noting that the Thrifty Nickel "went belly up"); GX 78 at NAT06-00611 (A 534) (mentioning the Star Shopper, a publication that, as its publisher testified (T. 397, 408, GA 23, 25), is not a substitute for readers because it contains only advertising); GX 79 at NAT06-00234 (A 542) (noting that electronic media also seek the advertising dollars sought by the Times); id. at -00249-50 (A 557-58) (Star Shopper); GX 80 at TC 000017-18 (A 571-72) (same); id. at -000021 (A 575) (discussing other advertising vehicles); GX 81 at NAT06-00114-15 (A 584-85) (noting the Star Shopper); GX 85 at NAT08-00142 (A 612) (noting "market share" for advertising); GX 84 at NAT08-00148 (A 606) (same); GX 87 at TC 001797 (A 619) (same); GX 89 at NAT08-00132 (A 626) (same); GX 96 at NAT08-00110-11 (A 629-30) (same); GX 99 at NAT08-00102 (A 633) (noting a new coupon book product and the "ad count" of a new "alternative" Fayetteville paper); GX 14 at NAT08-00230 (A 525) (noting advertising rates); GX 78 at NAT06-00610 (A 533-34) (noting non-local daily papers, television, and radio in a "Competition Analysis" but discussing in detail only state-wide and out-of-state papers in relation to circulation competition and describing these papers, essentially, as complements).

24. See GX 84 (A 603) (exhibit year misdated 1992); GX 85 (A 609-14); GX 87 (A 615-21); GX 89 (A 621-26); GX 90 (GA 232-37); GX 92 (GA 244-50); GX 99 (A 631-38); GX 102 (PA 867-71); GX 106 (PA 872-76); GX 108 (PA 879-84); GX 110 (GA 251-56) (1993); GX 111 (GA 257-62); GX 114 (GA 263-69); GX 117 (GA 270-78); GX 119 (GA 279-84); GX 121 (GA 285-90); GX 123 (GA 291-97); GX 125 (GA 298-304); GX 127 (PA 885-90); GX 129 (GA 311-16); GX 130 (GA 317-23); GX 132 (GA 326-30); GX 134 (GA 331-37) (1994).

25. See, e.g., GX 78 at NAT06-00611 (A 534).

26. See, e.g., GX 80 at TC 000019 (A 573) (noting that the Times competed "head-to-head" with Community Publishers, Inc.'s Herald Leader in Siloam Springs, a small community approximately 20 miles to the west of Fayetteville/Springdale); GX 81 at NAT06-00118 (A 588) (noting a new "alternative" Fayetteville paper); GX 99 at NAT08-00102 (A 633) ("ignor[ing]" the "alternative" paper); GX 83 at NAT01-00112 (A 602) (noting the demise of the alternative paper and merely the existence of a Fayetteville weekly).

27. Appellants' reliance on Flegel v. Christian Hosp., 4 F.3d 682 (8th Cir. 1993), is wholly misplaced. There, in stating that "[t]he plaintiffs' perspective on the market has no bearing," id. at 690 n.7, the court, as the very next sentence and citations reveal, simply meant that if consumers have suffered no injury, an injury to competitors is irrelevant, see id.

28. See, e.g., Freeman, 1995 WL 638354 at *10 (upholding the district court's refusal to credit fully testimony of market participants that consumers would not turn elsewhere in the absence of "significant economic or statistical data" but "hesitat[ing] to require particular evidence, especially in litigation involving complex industries . . . in light of the broad range of evidence that may be of value in determining a geographic market" (emphasis added)).

29. Even if it were, the detailed and consistent documentary record in this case would be sufficient to sustain the district court's exclusion of other media. Cf. Freeman, 1995 WL 638354 at *10-*11 (upholding the district court's refusal to "fully credit" testimony of market participants that differences between care available at one hospital and others removed the latter from the market but emphasizing the district court's discretion in this regard).

30. For instance, the Times monthly manager's reports from January 1993 through February 1995 include no references to any competitive responses undertaken by the Times to combat threats to circulation from radio or television. See supra note 24 & GX 137 (GA 338-43).

31. Appellants accordingly are wrong that Baseman's testimony "failed to offer any basis" for inferring an elasticity of demand that properly excluded other media. NAT Br. at 35. Indeed, as the district court noted, see 892 F. Supp. at 1156, Appellants' own economic expert testified that his elasticity calculation demonstrated that other media did not belong in the same relevant market as the Times (T. 2164, GA 169; T. 2168, A 326). Appellants contend that the district court's reliance on his testimony was "egregious[]" because, once Dr. Overstreet determined that each paper belonged in a separate market, he had no need to make any further inquiry of substitutes. NAT Br. at 32-33. But whether Dr. Overstreet could have testified to less, he expressly excluded other media from the market, and not merely "arguendo," id. at 33.

32. Appellants also complain that the court improperly "took judicial notice" of "`inherent' differences between local daily papers and other media." NAT Br. at 31-32. Even if this were true, ample other evidence supports the district court's finding that other media do not belong in the market. Moreover, the record contains some testimony that other media offer no close substitutes for the particular and desirable combination of physical characteristics and content found in local daily papers (T. 407-08, 441-43, GA 24-25, 35-37).

33. Appellants' expert conceded that he had "no quarrel" with the theory (T. 2065-66, A 278-79; T. 2149-51, GA 164-66)

34. Appellants further assert, without support, that the feedback loop, like the "private Plaintiffs' `downward spiral theory'" is "a theory only of how CPI will be injured by the acquisition" and is "not a valid basis for actually determining the scope of the markets." NAT Br. at 37. Although the feedback loop assumes circulation competition, and cannot alone exclude other media from the advertising market, it is a valid basis for finding that the Times and the Morning News belong in the same advertising market. Moreover, it is distinct from the private plaintiffs' "downward spiral" theory, which posits that a dominant paper's imposition of an advertising rate increase will result in advertisers not cutting linage.

35. CPM is the "cost per thousand readers" to advertisers.

36. See T. 378-79 (GA 20-21) (Smith); T. 448-51 (GA 42-45) (Lewis); T. 710-711 (PA 128-29) (Watson); GX 28 at TC 004788 (GA 213) (Times introduced a discount coupon book in an effort to produce such a product before the Morning News); GX 32 (GA 214) (explaining that the Times "[c]reated a lower `auto' rate in 1994 to try to entice dealers to use [the Times] vs. [the Morning News]"); GX 48 (GA 217) (Times offered 4 ads for the price of 2 in response to the Morning News' offer of 3 for the price of 2); GX 59 (GA 219) (Times advertisement, directed to advertisers, extolling the Times' greater circulation base in Fayetteville compared to the Morning News); GX 66 at TC 02347 (GA 220) (Times considered having telemarketers call people who placed classified ads in the Morning News); GX 86 at TC 001802 (GA 231) (explaining that the Times embarked on a "[b]ig [p]ush" for auto dealers); GX 130 at NAT07-00041 (GA 320) (explaining that the Times was worried that the Morning News was "going after big advertisers" and responded by printing fliers showing the Times' larger Fayetteville subscriber base); GX 131 at NAT08-00037 (GA 325) (similar); GX 230 at NAT08-00218 (GA 344) (Times advertising editor explains that she performed an "advertiser count and analysis of the [Morning] News and the Times" "[e]ach day," and asks for a "push on any Springdale advertiser that runs" ads, inter alia, in the Morning News"); GX 14 at NAT08-00229 (A 524) (comparing the Times' and the Morning News' promotions for classified advertising).

37. See GX 386 at DMG06-00423 (GA 358) ("Due to the increased circulation in the Fayetteville area, we were able to almost double the amount of ads which we would normally have run for local [political races].").

38. Because the district court's reliance on the feedback loop theory is sufficient to sustain its finding that the papers belong in the same advertising market, we do not address the other factors relied upon by the district court in reaching this conclusion or Appellants' criticisms of them. See NAT Br. at 40-46.

39. See GX 14 at NAT08-00230 (A 525) (comparing advertising rates); GX 78 at NAT06-00613-14 (A 536-37) (same); GX 84 at NAT08-00147-48 (A 605-06) (same); GX 85 at NAT08-00142 (A 612) (noting "market share" for advertising); GX 87 at TC 001797 (A 619) (same); GX 89 at NAT08-00132 (A 626) (same); GX 96 at NAT08-00110 (A 629) (same); GX 99 at NAT08-00102 (A 633) (noting revenue generated by advertising vehicles). One document also generally compares the costs among a number of papers, see GX 78 at NAT06-00616 (A 539), an activity squarely fits Appellants' definition of benchmarking: "look[ing] at another producer for ideas on how to improve [one's] own operation." NAT Br. at 40.

40. As Baseman demonstrated, the market shares would not differ materially if the calculation included the state-wide Democrat-Gazette's circulation in the readership market and the linage for "shopper" publications in the advertising market. See GXs 361-62 (GA 347-51) (figures do not include the Daily Record).

41. This argument is made in such a cursory manner that it should be considered waived.

42. If such joint advertising raised antitrust concerns under Sherman Act § 1, the acquisition would not, under Appellants' theory, solve the problem, for Appellants strenuously insist that the papers are not a "single entity" under Copperweld v. Independence Tube Corp., 467 U.S. 752 (1984), and its progeny. See Donrey Br. at 32-35.

43. Because the Stephens family possesses an even higher share of the narrower market defended above, the above findings apply a fortiori to the district court's alternative holding finding the acquisition unlawful in such a market.

44. The district court actually rendered two alternative holdings: first, that NAT's acquisition of the Times was likely to lessen competition substantially; and second, that Donrey indirectly acquired the Times. See 892 F. Supp. at 1171. But whether the relevant transaction is NAT's acquisition of the Times or Donrey's indirect acquisition of the Times through NAT, the analysis hinged on a single rationale: the propriety of aggregating the Stephens family's interests in Donrey and NAT. The district court also found evidence of present and anticipated harm to competition that might sustain the first holding in the absence of market share attribution. See infra p.45. Because the district court appropriately aggregated the Stephens family's interests, however, this Court need not reach that question.

45. Appellants, citing Besta v. Beneficial Loan Co., 855 F.2d 532, 533 (8th Cir. 1988), suggest that plenary review of the aggregation question is appropriate. See Donrey Br. at 22. Besta makes clear, however, that factual findings underlying mixed questions of law and fact are reviewed for clear error. See Besta, 855 F.2d at 536; accord Sherron v. Norris, No. 95-1265, 1995 WL 672335, at *4 (8th Cir. Nov. 14, 1995).

46. Under NAT's proposed ownership structure, Bess Stephens' three children (B. Stephens Dep. 7, 25, GA 385, 392), presumably the beneficiaries of her estate, and Warren Stephens would possess equity interests in NAT. See 892 F. Supp. at 1169 n.20. Donrey would remain controlled by Jack and Bess Stephens.

47. Contrary to Appellants' contention, see Donrey Br. at 35-36, a partial stock acquisition is not unlawful only when there is expressed intent to use the shares to achieve anticompetitive effects, or actual evidence of concerted action resulting in such effects. United States v. E.I. du Pont de Nemours & Co., 353 U.S. 586 (1957), cited by Appellants, certainly established no such rule. The issue, as framed by the Court, was whether the acquisition reasonably threatened anticompetitive effects at the time of suit. See id. at 597, 607. That the Court had before it evidence of actual anticompetitive intent and effects resulted from the fortuity that the suit postdated the relevant acquisition by several decades.

48. Other areas of the law recognize that family members, depending on the facts, may jointly control distinct enterprises. See, e.g., Reich v. Bay, Inc., 23 F.3d 110, 115 (5th Cir. 1994) (finding two companies under "common control" for the purpose of the Fair Labor Standards Act when members "of the [same] family owned both companies" and three brothers held positions of control).

49. A violation of Clayton Act § 7 may be established in ways that do not depend on the presumption, such as when a partial stock acquisition is involved. See, e.g., Hamilton Watch Co. v. Benrus Watch Co., 114 F. Supp. 307, 317 (D. Conn.), aff'd on other grounds, 206 F.2d 738 (2d. Cir. 1953). However, as demonstrated by F. & M. Schaefer, 597 F.2d at 815-16, depending on the facts, the presumption might still apply to a partial stock acquisition. See also 5 Phillip Areeda & Donald F. Turner, Antitrust Law ¶ 1203d, at 320-22 (1980). Because, as discussed below, the presumption properly applies here, this Court need not decide whether the district court's holding may be sustained in its absence. See also supra note 44.

50. Appellants suggest, see Donrey Br. at 35 n.32, that aggregation for § 7 purposes may be appropriate under a common economic interest rationale only if the Stephens family members lack conspiratorial capacity for § 1 purposes under Copperweld and those cases that have extended its analysis beyond the parent/wholly owned subsidiary context. See, e.g., City of Mt. Pleasant v. Associated Elec. Co-op., 838 F.2d 268 (8th Cir. 1988); Century Oil Tool, Inc. v. Production Specialties, Inc., 737 F.2d 1316 (5th Cir. 1984). The law requires no such symmetry. Market shares have been attributed even when conspiratorial capacity plainly remains, see, e.g., F. & M. Schaefer, 597 F.2d at 816, and it is well established that "[a] company need not acquire control of another company in order to violate the Clayton Act." Denver and Rio Grande Western R.R. v. United States, 387 U.S. 485, 501 (1967). Although lack of conspiratorial capacity provides sufficient justification for aggregation, cf. Consolidated Gold Fields, PLC v. Anglo American Corp. of South Africa, Ltd., 698 F. Supp. 487, 502-03 (S.D.N.Y. 1988) (finding lack of conspiratorial capacity when aggregation appropriate), aff'd in relevant part, 871 F.2d 252 (2d Cir. 1989), it is not a prerequisite.

51. In denying Thomson's request for a stay of the Order of Rescission pending appeal, the district court reiterated that allowing NAT to "keep the Times during the appeal" would "likely diminish competition and result in monopoly." Mem. Op. 3 (Aug. 2, 1995) (GA 7).