Uses and Users

The major price statistics programs of the U.S. Government are designed to provide comprehensive and detailed measures of price change at different levels of the U.S. economy. The data provided by these programs are used in analyzing and evaluating the Nation's economic health, in the formulation and evaluation of public policy, and in the implementation of public laws. Price statistics also form the basis for the escalation of wages, incomes, and other costs and are used to deflate current costs and values to constant dollar terms. The data form the basis for analyses of change from one time period to another and in some cases from place to place, item to item, or from industry to industry.

The programs of the Bureau of Labor Statistics (BLS) in the Department of Labor, which are the sole subject of this chapter, produce monthly measures of price change for the consumer and industrial sectors of the economy and rank among the most important statistical series produced by the Federal Government. The Consumer Price Index (CPI) is the principal source of information on trends in prices. paid by consumers. The CPI is used in the formulation and evaluation of economic policy; in wage negotiations and collective bargaining; and for escalation of wages, pensions and a variety of other transfer payments in both the public and private sectors. For example, retirement annuities of Federal civilian and military employees and of social security recipients, and compensation payments to the families of Federal employees who are disabled or killed while working are escalated by changes in the CPI. In addition, Federal support for other programs, such as the school lunch program, the nutrition program for the elderly, and the Food Stamp program are also adjusted on the basis of CPI changes. Moreover, the income levels in the official

'Federal civilian and military pension adjustments on the basis of CPI changes are required by 5 U.S.C 8240, 10 U.S.C 1401, 22 U.S.C 1121, and 50 U.S.C 402. Social Security and Railroad Retirement payments escalation on the basis of CPI changes are required by 42 U.S.C 415 and 45 U.S.C 231. Other statutes mandating changes in Federal programs based on CPI include: 42 U.S.C 3045, and 7 U.S.C 2016. The WPI's use in establishing sugar price parity is required by 7 U.S.C 1111.

poverty statistics are updated each year by changes in the CPI.

The BLS producer price program is the major source of information on price changes in primary markets of the economy; that is, price change at the first commercial transaction for products sold in large quantities in open markets, and additionally is used as a precursor of changes in the CPI. The most widely known measure produced by this program is the Producer Price Index (PPI), formerly called the Wholesale Price Index (WPI), which provides detailed monthly data on price changes of products grouped by commodity type, stage of processing, and industry origin. Industry Sector Price Indexes are a second major output from the producer price program. These data, which measure price change for products produced by industries within the Standard Industrial Classification framework, are used to estimate "real output" in the system of national accounts prepared by the U.S. Department of Commerce. Aside from overall use as a macroeconomic indicator of price trends and in the evaluation and formulation of economic policy, data produced by the producer price program are widely used in escalation of long-term purchase and sales contracts and in market analysis in both the public and private sectors. Further, the PPI is used in establishing parity for sugar prices.

Data from the international price program are used in international negotiations, in analyzing the effect of import and export price changes on the U.S. economic situation, and in the assessment of the effects of U.S. Government policy decisions on balance of payments and trade.

The BLS price statistics program also includes periodic surveys and studies of consumer expenditures providing information on spending patterns, income, assets and liabilities of families grouped by different characteristics. Such data are used to determine and analyze variations in consumer spending habits and to form the basis for selection of items to be priced and weighting patterns required for periodic revisions of the Consumer Price Index. On an annual basis, the BLS price statistics program produces hypothetical family budgets for a four-person

city worker's family and for a retired couple at three levels of living-lower, intermediate, and higher. The program includes a comparison of the costs of these budgets in different areas of the country which provide an indication of differences in living costs among areas. Currently, 39 metropolitan and 4 nonmetropolitan areas are included in the program. Consumer expenditure and income data are used by a wide variety of public and private agencies and groups, and by individual firms in market research studies, program planning, development and evaluation, and in general economic research.

Important Federal Government uses include the use of these data: (1) by the Internal Revenue Service in updating the Optional State Sales Tax Tables; (2) by the Department of Health, Education, and Welfare in determining the poverty level and related analyses; and (3) by the Department of Energy in studying the impact of changes in energy costs on consumers. Family budget data are also used in both the public and private sectors to assess variations in area living costs for administration of government programs, by private individuals comtemplating a change of residence, and for labor-management negotiations involving regional or local adjustments in wages and salaries. As an indication of costs to purchase goods and services at three hypothetical levels of living, the data for this program are also used in the administration and design of a variety of Federal as well as State and local income maintenance programs.

The CPI and WPI (now called PPI) are specifically referenced in one or more Federal statutes. The WPI's legal basis dates back to 1891 when, according to the BLS Handbook of Methods (Bulletin 1711), a U.S. Senate resolution "authorized the Senate Committees on Finance to investigate the effects of the tariff laws upon the imports and exports, the growth, development, production, and prices of manufactured articles at home and abroad." The BLS Family Budget Program is referred to by statute (29 U.S.C. 882) which provides for the development of "methods to establish and maintain more comprehensive household budget data at different levels of living...in regions and localities, both urban and rural." Other programs, such as the BLS International Price Indexes (IPI) and Tuesday Spot Market Index programs were originated in response to the programmatic needs or specific requests of policymakers and analysts for the data. An example of such a request is that which was made by the Department of the Treasury which resulted in the issuance, in January 1934, of a daily commodity price index-the precursor of the spot market price index.

The statistics on prices paid by consumers and on producer prices (including the industry sector prices and stage of processing indexes) are regularly used by the Joint Economic Committee of the Congress and by the Labor, Education, and Public Works Committees of both the House and the Senate as well as other standing and special committees. Virtually every Federal executive branch agency uses these data in preparing forecasts of economic events, in their analysis of program impact and in the process of analyzing proposed legislation. In addition, the data from the BLS series are used by the Bureau of Economic Analysis in its work on the national accounts-specifically in deflating the accounts. (For more discussion of this point, see the report of the Advisory Committee on Gross National Product Data Improvement.) In addition, the Council on Wage and Price Stability uses all of the series, but particularly the detailed data on industry sector prices and the CPI in analyzing specific price movements and their anticipated effect on the

economy.

The CPI is used by the public and the private sector in wage and salary negotiation, administration, and escalation. The PPI and CPI are used in commercial negotiations, and in formulating business strategy. Academics use the data in their analyses of the functioning of the economy and the interrelationship of its components. State and local governments use the data in much the same way as does the private

sector.

The BLS price statistics are the principal, directly collected price data used for the purposes previously noted and are principal inputs to the construction of the implicit price deflators of the gross national product. The GNP deflator and other price measures are discussed in the chapters on construction statistics, agricultural statistics, economic accounts, and financial statistics. It should be recognized that wages, salaries, and compensation are special forms of prices. These data, however, are discussed in the section on labor statistics. In addition, a new project for preparing constant-dollar estimates of national defense purchases is now being conducted by the Bureau of Economic Analysis (BEA). This project, undertaken on a reimbursable basis for the Department of Defense (DOD), is based on a sample of DOD's records of purchases of goods and services and from data collected directly by BEA from DOD contractors. The data are expected to be used for a number of purposes including the construction of the GNP accounts.

Advisory Groups

The Bureau of Labor Statistics advisory groups are discussed in the chapter on labor statistics. A special review of the Government's price statistics program was undertaken in 1961 by the National Bureau of Economic Research under contract with the Office of Management and Budget (then the Bureau of the Budget). That review resulted in a report on The Price Statistics of the Federal Government, known as the Stigler Committee report (which was reproduced as part of U.S., Congress, Joint Economic Committee, Government Price Statistics, Hearings..., January 24, 1961). The report described the programs as they existed and made a number of recommendations for change-some of which, such as expanding the CPI to cover the entire urban population and moving from import-export trade unit value indexes to specification pricing have only recently been implemented.

Some of the recommendations, such as the one to obtain PPI data from buyers, not sellers, so as to obtain actual transactions prices may have been overtaken by new technical approaches (subsequently discussed), while others are now under active consideration. Still other recommendations, such as the one to produce a broadly based index of transportation rates, have been accepted in principle and work has progressed on their implementation but additional work is needed. More recently, a review of the PPI, with the cooperation of BLS, was arranged by the Council on Wage and Price Stability. This review, recently completed by the National Bureau of Economic Research, coincided with a full-scale ongoing examination of the PPI concepts and procedures by BLS staff.

Data Issues

The conceptual and methodological bases of the Bureau of Labor Statistics' price statistics series are described in the BLS Handbook of Methods (Bulletin 1711) and Major Programs: Bureau of Labor Statistics (BLS Report 459). Major changes were introduced into the CPI as an integral part of the recently completed revision program. Therefore, the description of the CPI program published in BLS Bulletin 1517 and in BLS Report 459 describes the program that was revised and not the new program.

Consumer Price Index (CPI)

The CPI, which is compiled by the BLS, was recently revised. This revision, the first in almost 15 years, was completed early in 1978 with the publication of a revised index for urban wage earners and

clerical workers (who account for about 40% of the noninstitutionalized urban civilian population) and a new index covering the entire noninstitutionalized urban civilian population in urban areas (about 80% of the population). Rural areas within SMSA's are also included.

One of the major problems encountered by BLS in the course of the CPI Revision program was the question of population coverage for the CPI. Historically, the index has been oriented toward the urban worker and the old CPI, in fact, covered only the city wage earner and clerical worker population. As a result of demographic and economic changes, and the growing importance of the CPI as a means of adjusting payments to individuals (such as retired persons) who are not included in the wage earner and clerical worker population, pressures to both broaden the population base of the CPI and, in addition, produce indexes for specific components of the population have greatly increased. In view of the strong trade union interest in maintaining the existing Urban Wage Earner and Clerical Worker Index as an instrument which is "absolutely essential for effective, responsible, and rational collective bargaining" and the concomitant need for a more comprehensive index, the CPI Revision program resulted in the production of two indexes-a revised index for Urban Wage Earners and Clerical Workers and a new CPI for All Urban Households. Both indexes will be calculated and published for a 3-year period at which time it is expected that a determination will be made as to whether one index is adequate or both should be continued. While this particular issue concerning CPI population coverage has been at least temporarily resolved, the broader issue concerning the need for a family of indexes has not yet been decided.

BLS has received numerous inquiries from various congressional committees, individual Members of Congress, and from other public and private groups about the possibility of producing indexes for the aged, the poor, and the rural population. Generally, these inquiries stem from concern that the national CPI may not adequately represent the price experience of a particular population subgroup, and that the price change experienced by, for example, retired individuals or the aged may vary from that which is experienced by the working-age population. This concern has resulted in the introduction of bills to produce consumer price indexes for particular

'Excerpt from statement of Leonard Woodcock, President, United Automobile, Aerospace, and Agricultural Implement Workers of America (UAW) before the Subcommittee on Priorities and Economy in Government of the Joint Economic Committee, April 5, 1974.

groups such as the aged, retired individuals, and the poor. These bills have not yet been seriously considered by the Congress. Whether actual price experience does vary significantly and whether collecting these data would make a significant difference in the efficient and equitable allocation of Federal resources (particularly when balanced against the cost of collecting the data) are issues which require careful and thorough examination.

As in the case of consumer price indexes for specific population subgroups, local area consumer price indexes are not mandated by specific statutory requirements. However, for the revised Consumer Price Index program, price data is being collected in 85 separate pricing areas selected on the basis of the 1970 Census of Population. Of the 85 areas, 28 are self-representing and 57 are representative of the balance of the Nation's Standard Metropolitan Statistical Areas (SMSA's) and the remainder of the urban population. As a product of the revised CPI program, monthly or quarterly indexes are published for 28 urban areas as compared with the 23 local area indexes under the old program. The 28 indexes published as part of the revised CPI program represent a small fraction of the total number of requests which BLS has received, and continues to receive, for State, county, and city consumer price indexes. In partial response to this demand for more geographic detail, the revised CPI program has been designed to provide, for the first time, regional indexes for cities of different population-size classes. While this will provide individual areas for which a CPI is not compiled with a measure which is broadly representative of their particular situation, the issue of specific local area indexes still remains. This specific issue is, of course, just one indication of a much broader issue concerning the extent to which national statistical programs should be designed to yield local area detail (see chapter on Federal/State/local cooperative programs). Despite the efforts of numerous high-level commissions and study groups, no general solution to this difficult question has been developed.

The question of the inclusion of rural areas in the CPIs coverage-thus further expanding its scope to 100% of the population-is also related to the issue of population and area coverage. When rural America was functionally separated from urban America, the rural population did not generally purchase many of the goods and services consumed in urban areas, and to the extent that they did, did so from markedly different types of purveyors. Times have changed and there is reason to believe that the purchasing and expenditure patterns of rural and urban America are now more alike than was formerly the case. It is

noteworthy that the Department of Agriculture concluded that changes in prices paid by farmers for nonproduction items, such as food, clothing, recreation, and so forth, as determined through Agriculture's prices paid by farmers survey program, are similar to price changes as reported by BLS. Accordingly, the Department of Agriculture dropped its collection of these data and is using the BLS CPI data as inputs into Agriculture's parity price program.'

Present public policy, as established by Federal statute, does not require price data for local areas or for specific components of the population. Moreover, the collection of such data is extremely costly. Further, there is no mechanism currently in place with which to gauge the extent to which different parts of the population have different purchasing patterns-if in fact there are important differences. Fortunately, a mechanism for doing so is now being developed. Moreover, that mechanism, a continuing consumers expenditure survey, is designed to include the rural population. In addition, as noted before, the cost of collecting data for a consumer price index is very expensive and there is a significant question as to how far the Federal Government should go in providing CPI's for local areas when the Federal Government does not need such data for Federal purposes. These issues and questions and their implications for public policy decisions are all interwoven and should be examined. Until these issues and questions can be examined in some detail-in both analytic and policy terms-the policy that suggests itself is that the CPI coverage should not be expanded beyond that covered by the program as it now exists. With respect to data for component parts of the population, however, another factor needs to be considered. Present public policy, as expressed by statute, does not require data for population components. Moreover, most Federal uses of the data affect all population groups. Therefore, indexes for various population components should not now be created unless and until public policy, as expressed by statute, requires such an approach.

Frequency of Pricing

It has been argued that the frequency of pricing for the CPI should be expanded from that being used for the revised index (i.e., 52.8% of the value of the index items, based on the old weights which were derived from the 1960-61 Survey of Consumer Expenditures, are priced monthly with virtually all the rest being

'Items used by farmers in the agricultural production process continue to be studied separately by the Department of Agriculture. These items might appropriately be included in the BLS's industrial sector pricing program and consideration should be given to this possibility.