Introduction and Overview

The need for statistics on the aggregates and size distribution of household income, wealth, and consumption is to depict over time the variations of income and living conditions among the Nation's population. This information is basic to much of the discussion of public policy issues related to welfare, tax reform, aggregate economic demand, and, indirectly, business capital formation. Involved in each of these issues are considerations of shares of income (flow of claims on resources), consumption (execution of claims on resources), and wealth (stock of resources or of claims on resources) for persons, families and households, and of how these shares relate to performance and need.

In predicting that there would be over the next few years a major public policy debate on income shares and the distributional effects of public actions, Alice Rivlin' listed four sets of questions that the public and politicians might reasonably ask of economists:

1. What is happening? How great is income inequality and which way is it moving? What are the effects of current government programs?

2. What would be better? Presumably no one wants economists telling them what the distribution of income ought to be, but they might reasonably expect them to be helpful in defining alternative norms and discussing how varying degrees of equality might be expected to affect other objectives.

3. Why are things as they are? Economists worth their pay ought to provide a workable theory that explains why the distribution of income has the shape it has and why it is changing or not changing.

4. What can we do about it? Useful economists ought to spell out some policy options and estimate their effects, not only on the

'Alice M. Rivlin, "Income Distribution-Can Economists Help?," American Economic Review, Vol. LXV, Number 2 (May 1975).

distribution of income, but on the size of the pot to be distributed.

The fundamental importance of income, wealth, and consumption statistics to describe conditions, compare alternatives, test theoretical hypotheses, and estimate impacts of existing and potential programs is self-evident and a justification of their need in the broadest sense. (The reader should also see related materials in the chapters on labor statistics, price statistics, economic accounts, financial statistics and income maintenance and welfare statistics.)

Income, wealth, and consumption data are regularly used in the Congress by the Joint Economic Committee, the House Committee on Ways and Means, the Senate Finance Committee, various other standing and special committees, and the budget committees of both Houses in the consideration of economic, tax, income maintenance, and budget issues. In the Executive Branch, the Council of Economic Advisors, the Department of the Treasury, the Office of Management and Budget, the Department of Health, Education, and Welfare, the Department of Labor, and other agencies use these data in consideration of economic policy, social wellbeing, taxation and revenue measures, and the Federal budget. The Federal Reserve Board uses these data in formulating monetary policies. Outside the Federal Government, these data are used by business firms, labor unions, trade associations, research organizations, universities, State and local governments, and regional public agencies.

Data concerning income, wealth and consumption currently are compiled from three broad approaches: (1) direct survey data (e.g., the Census Bureau's household Current Population Survey); (2) administrative records (e.g., Internal Revenue Service's tabulations of individual tax returns in Statistics of Income); and (3) synthesis of all available information into a comprehensive national economic accounting framework (e.g., Bureau of Economic Analysis' measures of personal income and its size distribution). The availability of the data on the components of income, consumption and wealth varies markedly in terms of frequency, distributional detail, and comprehensiveness. Wealth data typically

are compiled only for the aggregate (i.e., very limited data on size distributions) of all households; these are available annually for housing and consumer durables stocks, bank deposit ownership and only sporadically for portfolio investments. Data concerning liquid assets are available quarterly. Income data are available monthly as an aggregate total for all households and annually for size distributions among households. Consumption data, which in concept equate to the using up of goods and services in a designated period whether they are obtained by purchase, receipt of in-kind payment or gift, or through the using up of assets, are generally not available. For many applications expenditure data are an adequate proxy. Expenditure data are provided quarterly for the aggregate of all households, but up until now only about once every ten years for size distributions.

Responsible Agencies and Core Programs

The Bureau of the Census collects and compiles comprehensive data on money income and its distribution among persons, households, and families in the Decennial Census of Population and the March Current Population Survey (CPS). The income data are for a calendar year and are associated with the pertinent social and demographic characteristics of the income-receiving units (education, labor force status, age, etc.) as of the following March.

Annual and periodical statistical reports based on tax returns are prepared by the Internal Revenue Service. Reports in the Statistics of Income series which are derived from individual tax returns show size distributions of adjusted gross income crossclassified by types of returns, deductions, and exemptions. Periodic statistics from estate tax returns, which are used in studies of individual wealth, include total estate by type of property, lifetime transfers, and deductions by size of estate and tax status. However, due to changes in the tax law, tax returns for estates below $175,000 will no longer be available after 1980.

The Bureau of Economic Analysis (BEA) in the Department of Commerce provides nationwide estimates of monthly personal income by industry; quarterly estimates of personal income by State and industry; quarterly estimates of disposable and per capita disposable personal income in current and constant dollars, personal consumption expenditures, and personal saving; annual estimates of the size distribution of personal income; annual estimates of personal income by county and metropolitan area; and annual wealth estimates of the stocks of housing

and consumer durables in historical-cost, constantcost, and current-cost valuations. These measures are prepared as part of the BEA work on the national economic accounts (see the chapter on economic accounts). They are based on a synthesis of primary economic data mainly provided by the Census Bureau, Bureau of Labor Statistics, IRS, and Department of Health, Education, and Welfare. The synthesis reflects a comprehensive crosschecking and reconciliation of the data provided from the various sources as well as statistical adjustments to be consistent with the conceptual framework of the economic accounts. For example, the crosschecking process quantifies the tendency of understatement in the income data reported in Census household surveys as compared with income reported on IRS tax records; and conceptual adjustments include the addition of imputed income for the rental value of owner-occupied housing and food furnished to employees.

The Department of Health, Education, and Welfare (DHEW) provides information on income transfer payments to individuals. The Social Security Administration compiles data on beneficiaries and benefits by type which are published monthly and quarterly in the Social Security Bulletin. An annual supplement presents more detailed data on benefits awarded by the demographic characteristics of the beneficiary. Also, the annual continuous work history sample based on social insurance tax records constitutes a longitudinal data base for the study of earnings. In addition, the Office of the Assistant Secretary for Planning and Evaluation (ASPE) in DHEW is developing a household Survey of Income and Program Participation (SIPP), which is scheduled to be operational by the early 1980's. When this survey is implemented, it is expected to provide data on quarterly as well as annual income from a broad range of sources and to include some of the more important income in kind. Data on wealth will be included to establish program eligibility and to address broad policy questions. The SIPP will also oversample low income populations which are the primary beneficiaries of DHEW programs. Data from administrative records will be used to upgrade estimates of types of income which are generally poorly reported in household surveys. Considerable effort will be made to measure income in kind as well as in cash. ASPE also sponsors the income dynamics panel, a longitudinal survey whose main purpose is to study the factors that affect the changes in the economic well-being of families over time.

Estimates of aggregate personal income from farm sources on a monthly basis nationally and quarterly by State are prepared jointly by the Economics,

Statistics and Cooperatives Service (ESCS), Department of Agriculture, and by BEA from farm and nonfarm sources. ESCS also prepares annual wealth estimates of the value of farm dwellings and of household equipment and furnishings on farms, and conducts surveys of farm production expenditures and farm family living expenditures. The Agricultural Research Service publishes data on the economic aspects of family living including estimates of household food consumption, the cost of food at home, the cost of clothing for individuals, and the cost of raising a child, as well as the income, expenditures, and consumption of rural families.

The most detailed recent data on consumer expenditures were collected by the Bureau of the Census for the Bureau of Labor Statistics (BLS) in the 1972-73 Consumer Expenditures Survey (CES). The previous CES was conducted for 1960-61. Data were collected on expenditures with related information on income and savings. The results of the survey, with varying amounts of expenditure detail, classified by socioeconomic characteristics of families, will be summarized for the Nation, geographic regions, and selected urban areas. BLS is developing the Current Consumer Expenditures Survey, for which full funding was provided in the 1978 budget. This survey will provide a rotating system for collecting up-to-date information on consumption and income data, classified by socioeconomic characteristics of the population, for use in analyzing changing patterns of consumption. This will be a major improvement over the previous periodic surveys conducted approximately every 10 years. BLS also compiles a variety of regular data series on wages, real spendable earnings, and standard family budgets. (Also see the chapters on labor statistics and price statistics.)

The Board of Governors of the Federal Reserve System compiles aggregate quarterly estimates of household liquid assets and the amount of mortgage debt outstanding, by holder and property type. For short- and intermediate-term consumer installment credit, the Board compiles monthly estimates of the amount of credit extended, liquidated, and outstanding, and the net change in outstanding credit, by holder and credit type, based partly on data collected by the Board from commercial banks and finance companies. In addition, the Board gathers and publishes statistics on finance rates for selected types of consumer loans offered by commercial banks (monthly) and finance companies (monthly or quarterly). The Board also has been compiling a series on household net worth, on a preliminary basis, and although these figures have not yet been published, the Board expects to put out a more

refined version of a household balance sheet in the near future along with comparable balance sheets for all other sectors.

Advisory Groups

Four advisory groups impact directly on income, wealth, and consumption statistics. The Interagency Committee on Poverty and Income has an important role in coordinating practices and setting standards for all affected agencies. A subcommittee of this group focuses on statistics of wealth. The Advisory Committee on Gross National Product Data Improvement recently completed a comprehensive evaluation study of data improvements needed for the preparation of the national income and product accounts. Several recommendations were made in the committee's report that also would strengthen the measures of income and expenditures discussed in this chapter (see the chapter on economic accounts). Two Census Bureau public advisory committees, the Census Advisory Committee of the American Economic Association and the Census Advisory Committee on Population Statistics, provide valuable guidance on content and methodology of income statistics programs in that agency.

Adequacy

As described in the previous section, the Nation has a highly developed set of data systems designed to monitor the performance of the economic system and to meet the more specialized policy concerns in individual subject areas. All of these systems in one way or another are related to statistics of income, wealth, and consumption (or expenditures). They include the National Economic Accounts (NEA), measures of unemployment and employment, Social Security and other retirement system program data, income maintenance program data, and consumer expenditure surveys. Virtually all other Federal program areas are concerned with the economic status of potential recipients.

As the Federal Government has been assigned policy responsibility in each of the three areas, supporting data and analysis programs have been developed. Thus, we have the NEA; the development of the Current Population Survey (CPS) designed to produce current unemployment estimates; statistics of income developed from tax records; and the Consumer Expenditure Survey designed to produce weights for the Consumer Price Index. Each of these systems is highly complex and a number of them approach the limits of what can be collected from a single household survey or administrative record program.

Yet even as the data programs have been improved, the importance of the interrelationships of income, wealth and consumption within each policy area has grown. However, integration of the individual data sets has not kept up with needs and the growing technical means for joint analysis of separate microdata sets.

The most general shortcoming is in the coordination of agency planning. The single greatest difficulty is that interest in statistics relating to the economic well being of households is not the primary focus of any one agency. This divergence of primary responsibilities has the result that no one agency is concerned with the preparation of a comprehensive statement of data requirements for statistics of income, wealth or consumption. At the same time, the program data of one agency need to be analyzed with those of many others, as well as with general purpose statistics.

As a result of the lack of focal points for statistics of income, wealth or consumption, major improvements have come about as adjuncts to data programs developed to meet specific agency missions. Thus, data giving the distribution of money income are produced annually from the CPS program which has as its main focus the estimation of monthly employment and unemployment. Statistics on consumption expenditures are available from the Consumer Expenditure Survey (CES) which is designed to produce information needed to provide weights for the Consumer Price Index. Improved income data are anticipated from the Survey of Income and Program Participation, which is designed to improve estimates of income and wealth of the population eligible for DHEW programs.

Program agencies necessarily have to place major emphasis on their primary missions and therefore the pattern just described can be expected to continue. As a counter balance to the individual agency requirements, strong coordinating efforts are needed in the area of economic data concerning households.

Improved coordination should build on existing efforts. A considerable analytical and developmental effort with respect to income statistics is already in place in the Bureau of the Census in support of the development of the DHEW Survey of Income and Program Participation and to a lesser extent the CPS and other Census programs for the collection of income data. This effort should be continued. The DHEW Office of the Assistant Secretary for Planning and Evaluation has invested the major share of effort for the development of the SIPP program and arranged for involvement of the academic world. Provision for a continuing analytical effort on the

part of DHEW is essential to ensure sensitivity to emerging policy needs. Communication with other agencies having a strong interest in the development and analysis of income statistics should be improved through the establishment of a committee which would advise the Bureau of the Census concerning priorities for statistics and developmental work on income.

A developmental program for the measurement of household income should continue in the Bureau of the Census with joint funding by the Bureau and agencies having a major interest in the improvement of income statistics. Provision for input from the principal agencies engaged in the production or analysis of income data should be made through the establishment of a policy group to advise on priorities.

The Department of Health, Education, and Welfare should continue to provide for an analytical program and a modest developmental effort focusing on the measure of the income, wealth and consumption, in order to ensure the sensitivity of data collection efforts to emerging policy concerns.

A similar need exists for an organizational unit focused on broad developmental efforts with respect to statistics of household wealth. However, with respect to wealth data, there is no existing unit which has an ongoing program focused on improvement of wealth data to meet the priority requirements of all agencies. The most active work on statistics of household wealth currently is centered in the SIPP developmental program, but the focus of the program is on wealth data needed to estimate DHEW program eligibility. For the longer run, the Bureau of Economic Analysis (BEA) would appear to be the agency that would sustain a continuing analysis of the broad spectrum of requirements for data on household wealth. A unit within BEA could, with the assistance of an advisory group of agencies concerned with the production and analysis of wealth data, provide the focus of continuing developmental work on the measurement of household wealth.

A continuing developmental program for the measurement of household wealth should be established in the Bureau of Economic Analysis. Provision for input from the principal agencies engaged in the production or analysis of wealth data should be made through the establishment of a committee to advise on priorities.

Concern for statistics of expenditure is focused in the Bureau of Labor Statistics (BLS). The other major agency activity in this area is in the Department of Agriculture, which is concerned with household and individual food consumption. As