Escalation Guide for Contracting Parties
Business firms in search of effective methods of coping with inflation
often employ price adjustment (escalation) clauses in long-term
sales and purchase contracts. A conservative estimate is that
contracts with a lifetime worth of $200 billion are currently
escalated using the Producer Price Index (PPI) family of indexes,
either alone or in conjunction with other sources of economic data. 1
Because they measure price changes objectively, both in general
and for particular products, free from possible manipulation by
either of the contracting parties, the producer price indexes
calculated by the Bureau of Labor Statistics (BLS) are widely recognized
among business people, economists, statisticians, and accountants
as useful in price adjustment clauses.
This guide provides guidance on the development of escalation
clauses in contracts that are to be tied to PPI
data. Such clauses should be written with great care to avoid
serious problems when contract adjustments are implemented. The
information in this guide is based upon BLS staff experience
in handling issues that have been brought to their attention in
connection with actual escalation clauses.
The role of the BLS is to provide requested
data and to explain their underlying methodology and limitations.
The Bureau does not encourage or discourage the use of price adjustment
measures in purchase and sales agreements. The Bureau does not
directly assist in writing contracts nor does it provide advice
on disputes arising from contract interpretation. Because index
methodology and publication conventions could be crucial in developing
escalation clauses, this guide is intended to alert users to
potential problems arising in these areas. 2
This guide is divided into three sections. First, an overview
of the PPI system describes the major categories
and groupings of the several thousand indexes that are published
each month. Then, guidelines for assisting in the development of
escalation clauses are outlined. Finally, a practical example
of provisions that might be incorporated into a contract is presented,
based upon the guidelines discussed, along with an example of
the price adjustment calculations that would be needed to implement
these provisions.
The structure of producer price indexes
Producer price indexes measure the average change in prices received
by domestic producers of commodities in all stages of processing.
A PPI is an output price index, that is, it measures price changes
received by manufacturers or service providers. It is neither a buyer's
index nor an input price index, that is, it does not measure the
cost of producing that item. PPI data are based on selling prices
reported by establishments of all sizes selected by probability
sampling, with the probability of selection proportionate to size.
Individual items and transaction terms from these firms are also
chosen by probability proportionate to size sampling methods. PPIs are based on a
monthly sample of about 100,000 quotations, resulting
in publication of about 10,000 different indexes each month.
Indexes are organized in three major structures:
(a) Stage of processing (SOP) — products are organized by class
of buyer and degree of fabrication, that is, finished goods, intermediate
goods, and crude goods;
(b) Industries and their products — products are organized by producing
industry as defined in the North American Industry Classification System (NAICS); and
(c) Type of commodity — products are organized by similarity of
end-use or material composition.
(For a more detailed description of these three index structures,
see the appendix.)
Indexes are available at different levels of aggregation and detail
within each of the three major structures. There are broad SOP,
industry, and commodity groupings, and there are indexes for specific
product groups or individual items, for example, electronic components,
diesel fuel, or raw cotton.
Guidelines for developing escalation clauses
(1) Establish the base selling price subject to escalation.
The item whose price is subject to escalation should be specified
as precisely as possible. State whether the base price refers
to a per-unit quantity or a certain volume of units. Give the effective
month or year of this base selling price; this time period is
often called the base period. Indicate the length of time it will
remain in effect. (Note that BLS no longer publishes any dollar
unit prices for any item within the PPI system.)
(2) Select an appropriate index or indexes.
The Finished Goods Price Index may best indicate the general trend
of inflation for goods sold in primary markets. The PPI for finished
goods excluding foods may be more appropriate
for users wishing to exclude the effects of volatile movements
in food prices. The Intermediate Materials Price Index or the
Crude Materials Price Index may best indicate price trends for
semifinished or raw materials in general. Again, indexes excluding
food-related materials may be more appropriate for many applications.
Indexes for commodities or detailed commodity groupings may best
indicate price trends for specific commodities.
Contracting parties may want to escalate the base price of a product
by a single PPI series. Often, however, users may prefer to escalate
on the basis of several data series, including some from other
government statistical programs, to reflect changes in costs of
a variety of inputs. In some contracts, for example, costs of major
materials and supplies are escalated with one or more PPIs, while costs of labor
are escalated with other BLS series such as the Employment Cost
Index. 3
In such cases, the escalation clause should specify the percentage weight given to
each index in calculating the total escalation amount. (See detailed
discussion under guideline 9d.)
Contracting parties should choose an index or indexes representing
the costs for providing a particular product or service, rather
than an index for the product itself. For example, if an apparel
manufacturer were contracting for long-term purchases with a producer
of finished fabrics, it would be more advisable to tie the escalation
clause to a PPI for synthetic fibers than to a PPI for a type
of finished fabric. Otherwise, the parties may find themselves
in a serious problem that could be difficult from which to escape. 4
Regarding the level of index aggregation or detail that might
be chosen, it should be understood that while detailed indexes
may target costs more specifically, they are also more likely
to be permanently discontinued by BLS, or to have occasional gaps
in data. Contracts should provide for these contingencies, and
may minimize them if they cite a commodity index that does not
go below the 4- or 6-digit level of detail, or a product code (industry-oriented)
index that does not go below the 7-digit level. 5
Even with the PPI program's full coverage of the mining and manufacturing
sectors, not all products are included directly in the sample
or published in the PPI system. Sometimes indexes must be chosen
as proxies to estimate the price movements of materials or products.
(3) Clearly identify the selected index and cite an appropriate
source.
The escalation clause of a contract should identify the index
selected by its complete title and any identifying code.
Please note that there is no single index entitled "The Producer
Price Index." The term "Producer Price Index" refers
to a family of indexes compiled by the Bureau of Labor Statistics.
A specific index should be cited in the contract by referring
to "the Producer Price Index for..." followed by the
exact title and any identifying code number.
The clause should also cite an appropriate source for the index
selected. The primary official BLS source of PPI data
is the monthly periodical, PPI Detailed Report. It contains
all indexes and is available online the day each month's data is released (www.bls.gov/ppi/ppi_dr.htm).
Current PPI data in print may also be obtained from the Producer Price Index News Release, and the
Monthly Labor Review (MLR). The Producer Price Index News Release is available without
charge through an e-mail subscription on the BLS website (subscriptions.bls.gov/accounts/USDOLBLS/subscriber); however,
it contains only a limited number of indexes. The MLR contains
aggregate rather than detailed PPI data and is posted about two weeks after the PPI data are first available. The MLR may
still be a convenient data source if a very broad aggregate PPI
category is called for, other BLS series are also included
in the escalation provision, and quick availability of data is
not necessary.
PPI data can also be obtained from LABSTAT, the BLS online data retrieval tool (www.bls.gov/ppi/data.htm). One-Screen Data Search and Multi-Screen Data Search are form-based query applications for both Industry Data and Commodity Data designed for users unfamiliar with the PPI coding structure. These applications guide a user through the PPI classification system by listing index titles and does not require knowledge of commodity or industry codes. Data retrieved are based on a query formulated by selecting data characteristics from lists provided. Two options are available to create customized tables, depending on a user's browser capability. The one-screen option is a JavaScript application that uses a single screen to guide a user through the available time series data. The second option is a multiple-screen, non-Java-based application. Both methods allow a user to browse the PPI coding structure and select multiple series codes. Users can modify the date range and output options after executing the query using the reformat button above the data output table.
Contracting parties should not cite table numbers and/or table
titles in their escalation contracts because they are subject
to change. BLS sources are preferable to secondary sources such
as other government publications or private firms. If contracting
parties agree to accept updated index values on the telephone
from BLS staff members, the escalation clause should specify
appropriate procedures and whether subsequent verification from
a published source is necessary.
(4) Specify whether seasonally adjusted indexes or unadjusted
indexes are to be used.
In general, seasonally adjusted indexes are not appropriate in
escalation agreements. Because price adjustment clauses usually
are intended to capture actual price changes, contracting parties
normally would not want to remove seasonal price movements from
their adjustment calculations.
(5) State the frequency of price adjustment.
The escalation clause should specify whether price adjustments
are to be made at fixed intervals, such as quarterly, semi-annually,
or annually, or only at the expiration of the contract. To conform
to the procedure described in guideline (9), price adjustments
have to be calculated over an interval whose beginning point is
the contract's base period. (This is the time period associated
with the chosen base price; for a discussion of base price, see
guideline (1).)
Difficulties will be encountered with those contracts which do
not designate a specific frequency for price adjustment, but rather
state that the latest data available as of a certain date should
be used for adjustment. In this case, or for any other case that
does not cite a specific time interval, problems will arise unless
the designated procedure corresponds with the version of the data
to be used, and the date on which the price adjustments will be
made. Guideline (7) expands upon these issues.
Note that PPI data are published as monthly indexes and
as annual averages for calendar years. Monthly PPIs
are representative of the entire month and do not refer
to a specific date of the month. Avoid wording such as "the
index for aluminum mill shapes as of September 30," since
several different and equally plausible interpretations are possible
for such language. It could mean the index that was available
on September 30, which would be the August figure; it could mean
the September index; or it could mean the October index, since
the September index would be based on information supplied to
BLS before September 30.
(6) Provide for missing or discontinued data.
Occasionally any given PPI may be unavailable
for a particular time period, usually because price information
was not supplied by a sufficient number of survey respondents
to meet BLS publication standards. Highly detailed indexes are
more susceptible to this problem than indexes for broader groupings.
For example, the Producer Price Index for laminated veneer lumber, code
08-22-01-07 was temporarily unavailable from March 2007 to December
2007; during that period, contracting parties had to use data
for the product class prefabricated structural members, code 08-22-01, or some other series of
their choosing. Escalation clauses should provide procedures to
be used when required data are missing.
Sometimes an index is permanently discontinued when a commodity
declines in market importance; this most commonly occurs as a
result of periodic resampling by BLS of industries and their output.
Escalation clauses may provide for successor indexes if original
indexes are discontinued, or for contracting parties to renegotiate a
successor index. A default provision that calls for using the
next higher-level series might be included in the contract.
Note that if BLS merely changes the title or recodes an index,
it is considered to be the same series and therefore, presumably,
should not necessitate any contract renegotiation. The monthly
periodical PPI Detailed Report routinely provides lists
of recoded indexes each time there is a sample change; normally,
these lists appear in the January and July issues each year.
(7) Specify that calculations of price adjustments shall always
use the latest version of the PPI data published
as of the date specified for such calculations; this requires
that contracting parties explicitly agree on the date the price
adjustment calculations are to be made.
Adherence to this principle and its implications should prevent
many potential problems. Contracts that fail to incorporate this
guideline will instead need to specify which version of PPI data
should be used, because: (a) BLS routinely revises PPI data 4
months after initial publication; (b) PPI data are rebased at
infrequent intervals; and (c) on rare occasions, PPI data may be
corrected.
Among other advantages, following guideline (7) should resolve
any ambiguities arising due to the fact that all PPI data are routinely
subject to revision once and only once, 4 months after their original
publication, to reflect late reports and corrections by respondents
in the PPI survey. Revisions are usually small at the higher
levels of index aggregation, but may be relatively large for
detailed indexes. The version of any PPI published 4 months
after its initial publication is considered final and will not
change again (barring corrections, and rebasing — a separate matter
addressed in guideline (8)). It is not appropriate to refer
to the first-published version of a PPI as "preliminary,"
and neither the first-published nor the final version of a
PPI should be labeled "actual," a term that might mean
different things to different contracting parties and which has
no official meaning in PPI terminology. 6
To follow guideline (7) effectively, it is essential to specify
the date on which the price adjustment is to be made. Currently,
PPI data are usually first published between the 9th and the 18th
day of the month following the reference month in question. Thus,
the earliest day for price adjustment that a contract ought to
specify needs to be after the 18th day of the month following
the designated data month. All first-published indexes for a given
month, as well as final indexes for the fourth previous month,
are considered officially published and are available on the day
of release of those data. The contracting parties' selection of the date on
which the price adjustment is to be made should be made only after
they have agreed on, first, the reference month and, second, on
whether their calculations are to be based upon the first-published
version or the final version of that month's index. The date for
calculating the price adjustment can then be selected so that
the desired data will be available.
It is vital to address these matters before a contract
is ready for signature. Otherwise, disagreements may arise when
the first-published and final versions of the selected index are
different, and there will be no criterion for selecting either
version.
If contracting parties do not specify an exact date for making
price adjustments, the contract should at least specify whether
first-published or final data should be used for calculations.
If this is the case, the final version of the data should be specified
whenever feasible, because only final data will be rebased retroactively
whenever BLS may update the PPI reference base.
Any procedure that departs from guideline (7), by failing to specify
the version of the data or the date when the price adjustment is
to be made, needs to be constructed so that it will be in harmony
with the frequency of price adjustment, as specified elsewhere
in the contract. This is discussed in guideline (5).
A contract should not refer to an index value associated
with a base price, but instead to its month and year alone. That
is, what should not be written into the contract is language such
as the following: "Divide the current index value by 103.9
(which is the value of the index for the base period January 2010)
and then...." Rather, it should be written: "Divide the
current index value by the index value for January 2010, which
represents the base period, and then...." Contract clauses that
incorporate specific index values will become problematic when
the PPI reference base is later changed by BLS; the
index value incorporated into the contract will be incompatible
with current official data after BLS has implemented the rebasing.
(Guideline (8) discusses reference base issues.)
(8) Avoid locking indexes used for escalation into any particular
reference base period.
Contracting parties should simply follow the principle of guideline
(7) by calculating percent changes using indexes expressed on
the reference base period in effect when the contract escalation
is carried out. For example, if a contract called for a price
adjustment to be made in December 1987 (just prior to the rebasing
that became effective on February 12, 1988), then indexes expressed
on the old reference base of 1967 = 100 would be used. In general,
relying upon a new index reference base period as set by BLS should
not affect calculations (except for rounding differences), as
long as all percent changes are derived solely from indexes expressed
on the official base period. Because rounding may indeed make
a substantial difference when the dollar amount of a contract
is very large, it will be doubly important for such contracts
to rely only upon official data on the current base as determined
by BLS.
Comprehensive base period changes in the PPI system have been
routine although infrequent. The switch to the current standard
reference base period of 1982 = 100 in early 1988 was the first
such rebasing since BLS adopted 1967 as the standard in 1971,
and that in turn was the first rebasing since the 1957-59 base
was adopted in 1962. Previously, the standard reference base
period was updated roughly every 10 years. 7
When the new 1982 = 100 standard reference base was adopted, BLS
advised contracting parties and other PPI data users to calculate
index percent changes using officially rebased data. As with all
other changes to new standard reference base periods, BLS had
taken all PPI final data that had been expressed on the 1967 base
and officially released these figures retroactively on a 1982
base. Tables of official historical data for each PPI series
from its beginning to the present on a consistent 1982 = 100 are available from BLS via LABSTAT.
Official PPI data for current time periods are not available on
previous reference bases after a base change has been implemented
by BLS. Further, as a general rule, estimating a conversion of
PPI data to an old base for the purpose of contractual price adjustment
is inadvisable because such a method could well be challenged
for referencing something other than official government data.
Rebasing factors are only made available by BLS to convert data
on the current standard reference base period to the immediately
preceding one. Thus, for example, there are no official rebasing
factors to convert data on the 1982 = 100 base back to 1957-59 =
100 base.
Rebasing is not considered "revising," because the relative
movements of any series over time are not affected. Users must
recognize that the absolute level of any index has no intrinsic
meaning other than relating a measurement to the base year, which
is itself arbitrary to a degree.
Older contracts may already specify use of originally published
indexes, particularly since this was recommended by BLS in the
September 1979 version of this guide (BLS Report 570). BLS is now
strongly discouraging such language in escalation contracts, in
accordance with guideline (7) recommending that the latest available
version of index data be used. In addition, BLS does not generally maintain
records for originally-published indexes. As a result, no official
rebased versions of such originally published indexes may exist.
(9) Define the mechanics of price adjustment.
(a) Simple percentage method. One method of price adjustment
is to have the base price changed by the same percentage as that
calculated for the selected PPI. To illustrate,
suppose that the contract escalation clause refers to the Finished
Goods Price Index. Also suppose that the Finished Goods Price
Index was 110.0 when the base price was set. A year later when
the first adjustment is made, the figure is 115.5. This represents
an increase of 5.0 percent in the Finished Goods Price
Index as shown.
Index at time of calculation ................................... 115.5
Divided by index at time base price was set .............. 110.0
Equals ......................................................... 1.050
This means that the base price should be increased by 5.0 percent.
To proceed:
Base price ..................................................... $1,000
Multiplied by ............................................ 1.050
Equals adjusted price .......................................... $1,050
In later years, this procedure would be applied again by taking
the current index value and dividing by the index value
at the time the base price was set, and then proceeding just as
described above.
(b) Escalation of a portion of the base price. Another
procedure sometimes employed changes the base price so that only
part of it is escalated by a selected PPI, while
the balance remains fixed. This may be done by changing the base
price by a certain dollar amount for each 1-percent movement in
the selected index.
To illustrate, suppose that an item has a base price of $1,000,
of which $700 is to be escalated by the index while the other
$300 remains unchanged. To determine the "certain dollar amount"
that is needed for citation in the contract, simply divide the
designated variable portion of the base price ($700) by 100, which
in this case would yield $7. The escalation clause is written
so that it provides that the base price of $1,000 shall change
$7 for each 1-percent movement in the index.
Using this approach, the base price would rise to $1,035.00 for
a 5-percent rise in the finished goods price index as shown:
Base price ..................................................... $1,000.00
Plus 5.0 times $7 ........................................... 35.00
Equals adjusted price .......................................... $1,035.00
(c) Index points. Relatively few escalation clauses which
rely on PPI data adjust contract prices on the basis of changes
in index points. (In the earlier example, the index-point change
would be 5.5.) When prices are adjusted by a percentage on the
basis of a change in index points, the value of an index point
will fall in percentage terms as the index level rises, and vice
versa. For example, a 1-point increase in an index from 105.5
to 106.5 represents an advance of 0.9 percent, but a 1-point increase
from 205.5 to 206.5 represents an upward movement of only 0.5 percent.
Conversely, a 0.9-percent increase in an index of 205.5 would
raise the index 1.8 points to 207.3.
Thus, if the base price is adjusted by a dollar amount according
to a change of index points, the procedure is then vulnerable
to changes in the index base period. Index point values would
differ for an index rebased to a later year or expressed on a
1967 = 100 base.
In contrast, adjusting a base price by a percentage change in
an index, as in approaches (a) and (b) above, will not result
in these discrepancies.
(d) Composite indexes. Some contracts describe construction
of a composite index based on several PPI series. The advantage
of a composite index is that it may more accurately identify the
appropriate change for a base price (see guideline (2))
since it will refer to several of the costs involved in producing
the product or service in question. However, a composite index
entails more calculations at the time of adjustment than the simpler
procedures described earlier. Composite indexes constructed by
the contracting parties are not official BLS data.
One procedure for specifying a composite index is illustrated
by the following steps:
(i) Choose the indexes that will represent the different costs
involved in producing the item (such as a fuels index, a machinery
index, or whatever is appropriate);
(ii) Choose the appropriate weights for these indexes, in accordance
with the proportion of the production budget which may be devoted
to these various categories. The list of chosen weights should
sum to 100 percent.
(iii) Clearly specify the time period that these relative weights
are supposed to represent. The weights should be chosen to represent
the time period associated with the base price. (This will be
referred to as the base period.)
(iv) The first step necessary for the calculation of the special
index is to rebase all of the original index data to the contract's
base period. This is done for each series by dividing the indexes
by the index value for the base period, and then multiplying the
result by 100. (For this and following steps, note the detailed
example at the end of this guide.)
(v) Then derive values for the composite index by multiplying
the relative weights by the rebased index values for each index
series and summing the results. (This calculation must be done
for each month, or other time period, needed for determining the
current adjustment.)
(vi) Using the composite index values created in step (v), calculate
the current adjustment in standard fashion, that is, by using
the procedure described in (a).
(e) Limits for price adjustment. Escalation clauses sometimes
contain a floor, a ceiling, or both, to limit the total
price adjustment during the life of the contract. If the upper
or lower limit is reached, the parties may renegotiate prices
for the duration of the contract. Some contracts specify that
no price adjustments are to be made until a minimum change in
the selected index has taken place. Contracts may also provide
that an escalation is to apply in both an upward and downward
direction, or in one direction only.
Example of escalation procedures
Suppose a manufacturer of widgets enters into a long-term sales
contract with a customer. The buyer and the seller agree to include
an escalation clause which will adjust the selling price once
a year to account for changes in labor and materials costs. The
following is an example of the terms which might be incorporated
into such an escalation clause. The example assumes the use of
the special index method, discussed in section (d) of guideline
(9).
(A) The base selling price for a lot of 10,000 type A widgets
is set at $768,450.00 as of December 2009, to remain in effect
for 1 year. December 2009 is hereafter called the reference base
period.
(B) The base selling price shall be adjusted in accordance
with the percent changes of the special index which is described
in (D) below. The special index shall be derived from the following
index series:
(i) The Employment Cost Index for total compensation, private industry manufacturing, not seasonally adjusted, as it appears
in the periodical Monthly Labor Review as published by
the U.S. Department of Labor, Bureau of Labor Statistics; this
series shall be referred to as the labor index.
(ii) The Producer Price Index for special industry machinery
and equipment, commodity code 116, not seasonally adjusted,
as it appears in the PPI Detailed Report as
published by the U.S. Department of Labor, Bureau of Labor Statistics;
this index shall be referred to as the materials index; and
(iii) The Producer Price Index for number 2 diesel fuel, commodity
code 057303, not seasonally adjusted, as it appears in the periodical,
PPI Detailed Report as published by the U.S. Department
of Labor, Bureau of Labor Statistics; this index shall be referred
to as the fuels index.
(C) The selling price shall be adjusted on February 20 of each
subsequent year, based upon the percent changes (whether up or
down) in the special index specified below, between the reference
base period December 2009 and December of the most recent year.
All calculations for the special index shall be based upon the
latest versions of the Producer Price Index and Employment Cost
Index data published as of February 20 each year.
(D) The special index shall be derived in the following manner:
(i) The values for the current period for each of the three BLS
index series specified in (B) above shall be rebased to the reference
base period December 2009; this shall be done by dividing the
current value of each index by its value for the reference base
period, and then multiplying the result by 100.
(ii) The rebased labor index shall be assigned a relative weight
of forty (40) percent; the rebased materials index shall be assigned
a relative weight of forty (40) percent; the rebased fuels index
shall be assigned a relative weight of twenty (20) percent; these
relative weights represent the base period of December 2009.
(iii) Multiply the rebased current value for each of the three
indexes by its relative weight.
(iv) The sum of these three figures shall be the value of the
special index for the current time period;
(v) Multiply the current value of the special index by the original
base price, and then divide by 100; this final figure shall be
the adjusted price for the current time period.
(E) If December ECI data are not available for any year,
the ECI for the immediately preceding September shall be used
as the basis for adjustment of the labor index. If December PPI
data are not available for any year, the PPI data for the immediately
preceding November, October, or September, whichever is the most
recent month which has published data, shall be used as the basis
for adjustment of the materials and fuels indexes. If no ECI or
PPI data have been published for those months, then the contracting
parties shall agree upon substitute series by February 20.
With these terms in effect, table 1 shows some hypothetical data
and calculations which might have been made on February 20, 2011
to determine the new selling price for a set of 10,000 type A widgets
as of December 1, 2010.
Table 1. Example of calculation procedures
|
Labor |
Materials |
Fuels |
Composite |
Base price = $768,450 |
- |
- |
- |
- |
Current period series values (December 2010) | 110.0 | 190.2 | 259.2 | - |
Divide by the base period series values (December 2009) | 107.0 | 189.5 | 205.1 | - |
equals: | 1.028 | 1.004 | 1.264 | - |
Multiply by 100 to yield the converted series values | 102.8 | 100.4 | 126.4 | - |
Multiply by assigned weight (Labor 40%, Materials 40%, Fuels 20%) | 41.12 | 40.16 | 25.28 | - |
Add the three figures to get the current value (December 2010) for the special index | - | - | - | 106.6 |
Multiply by original base price ($768,450) | - | - | - | 81,916,770 |
Divide by 100 to yield the adjusted price | - | - | - | $819,168 |
Pitfalls to avoid
- Vague citation of "the Producer Price Index" rather
than a reference to a specific index by its title and any identifying
code number. See guideline (3).
- Citation of the all commodities index or the industrial commodities
index rather than an index that does not include multiple counting
of price changes. See the discussion of commodity indexes in
the appendix.
- Use of unofficial estimates derived from rebasing factors
rather than relying upon official BLS data. See guideline (8).
- Ambiguous reference to dates ("index as of May 30").
See guideline (5).
- Lack of a provision for a successor index should the designated
index be dropped from the PPI program, or if it should
become temporarily unavailable. See guideline (6).
- Locking index into a specific base period. See guideline
(8).
- Using ambiguous terms. For example, referring to "actual"
indexes. See guideline (7).
Appendix: Three Index Structures: A Brief Overview
Stage of Processing (SOP) indexes
The Finished Goods Price Index measures price changes for goods
that will not undergo further processing and are ready for sale
to the final demand user, either an individual or a business firm.
Consumer foods include processed foods such as bakery products
and meats. Other finished consumer goods include durable goods
such as automobiles, household furniture, and appliances, and
nondurable goods such as apparel and home heating oil. Capital
equipment includes producer durable goods such as heavy motor
trucks, tractors, and machine tools.
The stage-of-processing category for intermediate materials, supplies,
and components consists in part of commodities that have been partly
processed but require further processing. Examples of such semifinished
goods include flour, cotton yarn, steel mill products, and lumber.
The intermediate goods category also encompasses nondurable, physically
complete items purchased by business firms for their operations.
Examples include diesel fuel, belts and belting, paper boxes,
and fertilizers. Several sub-category indexes are available, such
as an index for intermediate goods less foods and energy.
Crude materials for further processing are products entering the
market for the first time that have not been manufactured or fabricated
and that are not sold directly to consumers. Crude foodstuffs
and feedstuffs include items such as grains and livestock. Examples
of crude nonfood materials include raw cotton, crude petroleum,
coal hides and skins and metal scrap.
Industry indexes
The entire output of various industries is sampled to derive price
indexes for the net output of industries and their products. Such
indexes are grouped according to the NAICS and Census product code
extension of the NAICS. Industry price indexes are compatible with
other economic time series organized by NAICS codes, such as data
on employment, wages, and productivity. This is especially convenient
if indexes reflecting cost inputs other than PPIs also are used in the escalation procedure.
Commodity indexes
The commodity classification structure evolved over many years; its greatest
usefulness is the availability of a large amount of historical
data. The coding system used for these indexes is unique to the
PPI program; no other governmental statistical program uses it.
Commodities are grouped according to similarity of material composition
and end use, regardless of the industry of origin.
Unlike SOP indexes, some of the traditional commodity grouping
indexes such as the all commodities index and the industrial commodities
index exhibit a multiple counting problem in reflecting price
changes. This occurs because many products go through successive
stages of fabrication or processing and have their price changes
counted separately at each stage. SOP indexes largely offset the
defect of multiple counting of price changes.
Multiple counting of price change can arise as follows: Suppose
that a price for steel scrap results in an increase in the price
of steel sheet and then an increase in the price of automobiles.
The all commodities index would increase as a result of all three
changes, whereas the typical end-use purchaser would only note
the price increase for automobiles. The grouping of products by
stage of processing eliminates double counting of commodity price
changes as they pass through different stages. The SOP structure
would reflect the increase in the price of steel scrap only in
the Crude Materials Price Index, the rise in steel sheet only
in the Intermediate Materials Price Index, and the rise in automobile
prices only in the Finished Goods Price Index.
Footnotes
1
See, The BLS Industrial Price Program: A Survey of Users, Report 509 (Bureau of Labor Statistics, 1977).
2
Data requests and technical questions concerning the PPI may be addressed by the PPI Section of Index Analysis and Public Information. They can be reached at telephone number 202-691-7705, or by e-mail at (ppi-info@bls.gov). Please refer to the desired series by title and code number, exactly as cited in the contract.
3
The Employment Cost Index (ECI) is based upon a quarterly survey and is available only for the months of March, June, September, and December each year. Because the ECI has relatively little industry detail at present, data users may have to use a higher level of aggregation than they do with PPI data. However, the Employment Cost Index is a highly useful measure of labor costs because it covers all workers (not just production and nonsupervisory workers) and because it includes not only wages and salaries but also employer costs for employee benefits. Like the PPI, the ECI is a fixed-weight index and this is not influenced by employment shifts among industries and occupations with different wage and benefit levels. But unlike the PPI, ECI data are final when they are first published and are not subject to revision (except on a seasonally adjusted basis).
4
From the seller's point of view, a contract which escalates the price of a product based on the change in the PPI for that same product might not provide an appropriate basis for changing the base price. In those cases where most companies reporting a product's price to BLS are tied to escalation clauses using the PPI for that same product, these firms would be unable to raise their price until the PPI advances; there could be no advance in the PPI until the companies are able to raise their price. From the buyer's point of view, a reverse circularity is evident when the price of a product purchased is escalated by the PPI for the same product. A rise in the contract price may be reflected in a rise in the PPI, which would trigger yet another rise, etc.
5
Sometimes, however, government agencies, laws, or regulations may dictate which index or level of detail must be cited.
6
As an example of PPI practices, first-published PPI data for December 2010, as well as final data for August, were released on January 13, 2011. Final data for December were released on May 12, 2011 with the first release of April data. Final data for all indexes appear in each issue of the PPI Detailed Report and are available online through LABSTAT. Contracting parties who want to use other BLS series for escalation in addition to PPIs should be aware that each BLS program has its own revision and correction policies.
7
For SOP and most commodity indexes, the base year currently is 1982; indexes introduced into the system since then are based on the month they were first calculated, usually either December or June. Industry and product indexes currently have no standard base but are based on the month of their first publication.
Last Modified Date: May 14, 2012