As part of the further conceptual development of the accounts, one area that should be given special attention is the need and feasibility of providing more gross data on transactions, such as developing more detail on the sources and uses of funds to further elaborate the "from-whom, to-whom" transactions which show lending and borrowing between and within sectors. The commission could use the limited recommendations for the flow of funds in the GNP Data Improvement Project as one reference for delineating the issues and scope of its study.

Statistics of Income

The annual Statistics of Income (SOI) data published by the Internal Revenue Service are based on a sample of Federal income tax returns for individuals, corporations, and unincorporated business. The data include gross income, deductions and taxable income for individuals and business, as well as balance sheet information on assets and liabilities of businesses, including tabulations by size of businesses and income size class of individuals. Supplementary reports typically are done every 2 or 4 years-estate taxes (4 years), fiduciaries (4 years), individuals sales of capital assets (4 years), individual small areas (2 years), domestic international sales corporations (annually), corporate foreign tax credits (2 years), and U.S. multinational corporations (2 years).

Preliminary data for the annual publications are available about 1 1/2 years after the reference year, and final data one year later. The domestic supplementary data are available typically 2 to 2 1/2 years and the international information 4 years after the reference year.

The SOI data have both program and statistical uses. They are important for assessing the private income and tax revenue impacts of proposed and actual changes in tax and welfare laws, including tax and welfare reform. They are also used for analyzing programs and legislation associated with assisting small business. Statistically, they are direct components of the data base of the national economic and flow of funds accounts on income of individuals and businesses, and indirectly as well. Thus, they are used as the sampling frame for the Quarterly Financial Report and as the benchmark for the quarterly series on Working Capital of Nonfinancial Corporations of the Federal Trade Commission (discussed below), and for the reconciliation of company and establishment data in the Census Bureau's enterprise statistics, all of which are used in preparing the economic and financial accounts as well as for other purposes. Annual interagency meetings are held jointly by IRS and the

Office of Federal Statistical Policy and Standards to discuss user needs for the SOI program.

Data Gaps

Because of the importance of the SOI in preparing the national economic accounts, an extensive review of these data was conducted as part of the GNP Data Improvement Project. This led to the following recommendations for strengthening the SOI pro

gram:

1. Annual data on the income and expenses of retirement, welfare, health and thrift savings plans collected under the Employment Retirement Income Security Act should be tabulated as national industry aggregates by mid-May for the plan year covering the previous July 1 to June 30 period.

2. The method for valuing inventories should be collected on business tax schedules and tabulated annually.

3. For corporations and partnerships associated with farm enterprises, business expenses associated with business farm receipts should be specified and tabulated as is done on the tax returns for sole proprietorships (Schedule F).

4. Data on corporate tax returns that reconciles taxable profits and balance sheets with stockholder reports (Schedule M) should be tabulated.

Beyond the needs of the national economic accounts, there are at least two general areas about which more systematic information is needed for users of the SOI. One relates to the needs of the small business program for economic information. The Small Business Administration has made a start in this direction by undertaking a broad review of its program needs for economic information. This ongoing review of the SBA data base needs has identified several types of beneficial information which could be provided as supplementary data from the SOI program. They include tabulations of small business income tax returns for those companies receiving SBA loans, cross-tabulations of corporate dividend payments by size of company and income size of recipient, and more detailed size class and industry classification categories. These would provide the basis for better evaluating the effects of the SBA loan programs, developing more explicit assessments of proposed changes in tax laws for their impact on investors in small business, and more accurately identifying the universe of small business firms.

After an assessment by the Internal Revenue Service of the feasibility and costs of providing these data for the Small Business Administration is made, they should be ranked in a system of priorities for obtaining additional funding in future years. Because of the close link between the small business and minority business programs, this ranking of priorities by SBA should be coordinated with the Office of Minority Business Enterprise in the Department of Commerce.

Another broad area of the SOI program that requires attention are the data uses and adequacy of the periodic supplemental reports. These have developed over the years in response to specialized data needs, and as adjuncts to the regular SOI program, they have a lower priority on the data processing resources of the overall program. Although individual supplemental reports are reviewed periodically, a more comprehensive assessment is needed of how they relate to needs of the entire program.

The Internal Revenue Service should conduct a survey of users of the supplemental reports to Statistics of Income to assess the need for augmenting and/or modifying their content and frequency.

Federal Trade Commission Financial Reports

The Federal Trade Commission (FTC) provides statistical reports on industry income and balance. sheet information, diversification, mergers and concentration. Four periodic reports are summarized here: Quarterly Financial Report, Working Capital of Nonfinancial Corporations (also quarterly), Line of Business (annual), and Corporate Patterns (quinquennial).

Quarterly Financial Report

The Quarterly Financial Report (QFR) provides financial data on manufacturing, mining, retail and wholesale trade corporations. These data on income, expenses, dividends, retained earnings, assets, liabilities and equity represent economy wide industry totals. They are shown in selected subindustry detail and by asset size for manufacturing, and as industry aggregates for mining and trade, both in dollars and as percent distributions. The information is collected from a probability sample of surveyed firms, with estimates provided for universe totals. The QFR is a mandatory report.

The QFR is used in developing estimates for the national economic accounts, for investor analyses of the profitability and financial health of American

industry, and in assessing the position and prospects of small business for administering the small business assistance program.

Two program improvements for the QFR were recommended in the GNP Data Improvement Project:

1. The Federal Trade Commission should carry out a survey of annual profits (audited by public accounting firms) based on the domestic operations of corporations in its sample of Quarterly Financial Reports. The tabulations of survey results should be completed in time for use in the first July GNP estimates.

2. Explicit figures on employer contributions to private pension, health, and welfare benefit plans on an accrual accounting basis should be provided in the Quarterly Financial Report.

In order to reduce reporting burden on American businesses, the QFR sample was reduced in 1977 by eliminating smaller sized firms. Indirect estimates for the smaller firms are now provided by extrapolating the statistical relationships between small and large companies over the 1974-77 period. This in turn has decreased the number of size class categories for smaller sized firms, information which is of interest for administering the small business program.

The need for obtaining more detailed asset size classifications in the Quarterly Financial Report should be reviewed after 2 or 3 years of experience is gained with the reduced sample of surveyed firms. The review should include the needs of the Small Business Administration for these data, alternative methods for obtaining such information, and the reliability of the indirect estimating technique for small firms as a whole.

Working Capital of Nonfinancial Corporations

This quarterly series provides information on current assets (e.g., cash, receivables, U.S. government securities, inventories) and current liabilities (e.g., short-term bank loans, payables, Federal income taxes accrued, current installments on long-term bank loans) for nonfarm industries. The data are obtained from the Quarterly Financial Report for manufacturing, mining and trade, surveys of electric and gas utilities, transportation company reports to the Civil Aeronautics Board and Interstate Commerce Commission, and for other industries as indirect estimates. The figures are benchmarked from time to time to Statistics of Income data.

The working capital series was transferred from the Securities and Exchange Commission (SEC) to the

Federal Trade Commission in FY 1978. The data are published in the Federal Reserve Bulletin (the additional early release by SEC has been discontinued by FTC pending its more thorough review of the program). Other program changes initiated by FTC were to reduce the reporting burden by replacing sample surveys of the construction and service industries with estimates based on extrapolating relationships between these and other industries over the 1974-77 period, updating the benchmarking to the 1974 Statistics of Income, and improving the data processing.

These data are used in preparing the national economic and flow of funds accounts and for investment analyses. The following recommendation for strengthening the series was made in the GNP Data Improvement Project:

A review of the tabulation schedule of the series on Working Capital of Nonfinancial Corporations should be made in a few years for the possibility of speeding up the availability of these data in time for use in the 75-day GNP estimates.

The recent transferral of the working capital series to the Federal Trade Commission has, as previously noted, led to an overall review of the program as well as program changes. Further assessment of the program will be needed as more experience is gained.

The following recommendations are made for improving working capital data:

1. The need and alternative methods for an earlier release of the working capital data should be examined, including recommendations for the most appropriate dissemination vehicle.

2. The indirect estimating techniques for the nonsurveyed industries in the working capital series should be reviewed in a few years to assess its reliability, including recommendations for continuing or modifying the existing methodology.

Line of Business

The new line of business program is designed to provide annual data on profits, loss, and assets for relatively homogeneous classes of products for approximately 500 of the largest diversified manufacturing companies. The definition of lines of business provides for flexibility on the part of the respondent. They are built up from basic components which can be establishments, product lines, profit centers, or other organizational units, whichever strikes the best balance between accuracy of reporting and compliance burden. Summary

reconciliations of these data will be made with the Quarterly Financial Report information.

Data collection for the line of business began for the year 1973. Due to legal challenges to the program which are still pending in court, some companies have not been reporting data to the Federal Trade Commission, and no reports of the data collected to date have been published, although some results are likely to be published in 1978.

Information from the line of business program will be presented as statistical industry aggregates. Individual company data will not be published. To prevent possible disclosure, at a minimum, data for four companies will be aggregated in a given industrial category.

The program has two major uses. One is to assist the Federal Trade Commission in identifying areas of investigation for its antitrust and consumer protection law enforcement functions which are industry wide in scope. The other is to obtain a clear picture of financial developments in specific industries for analyses of market structure, industrial performance and market entry by companies, investment analysts, FTC economists, and outside scholars.

Corporate Patterns

The new survey of corporate patterns is aimed at providing periodic information for approximately 1,000 of the largest manufacturing companies on value of shipments by product class, nonmanufacturing sales of the diversified manufacturing companies, company sales and assets, and characteristics of joint ventures and minorityinterest companies in which the parent companies have ownership interests. Data were collected initially for 1972 and planned to be repeated on a 5-year cycle to coincide with the quinquennial economic census years. Because of legal challenges to the program which are pending in court, the planned data collection is behind schedule. The last such survey of corporate patterns was conducted for 1950.

Examples of the tabulations expected in the report include the following: (1) share of product class output accounted for by leading firms; (2) concentration ratios of the number and size of firms in a product class; (3) mobility and changing composition of leading firms by characteristics such as aggregate size and extent of diversification and vertical integration; (4) identification of industries according to level of merger activity; (5) relationship to merger activity of factors such as industry size and growth; (6) relative use of mergers and internal

expansion in entering new product lines; and (7) advertising to value of shipments ratios by product class.

Publication of data for individual companies will not be made available for at least 5 years after the data year. The Federal Trade Commission believes that this lag is adequate to prevent disclosure of proprietary information that could benefit rival companies.

The principal use of the corporate patterns data will be to assist the Federal Trade Commission in its enforcement efforts, economic studies and policy planning activities. Due to the lack of consistent and comprehensive published data, there has been no systematic empirical analysis of the significance and impact on competition of the merger movements and increased diversification of large corporations over the last three decades. Such information is considered essential for a basic understanding of competitive conditions and the formulation of sound and effective antitrust policy.

Banking and Credit

A number of Federal agencies provide statistics on banking and credit. The more prominent ones are summarized below.

The Federal Reserve Board publishes weekly and monthly banking statistics in the monthly Federal Reserve Bulletin and as special releases. These include data on the central banking system, money stock, interest rates, commercial bank balance sheets, consumer and mortgage credit, governmental and international financial transactions.

The Federal Deposit Insurance Corporation (FDIC) publishes data on commercial banking in the form of "all bank" statistics from tabulations by the Comptroller of the Currency, Federal Reserve, and FDIC call reports-detailed reports of assets and liabilities tabulated four times each year for the banks under the agency's supervision-and from similar reports semiannually covering the small number of noninsured banks. Annual data are published also on the earnings, expenses, and profits of all insured commercial and mutual savings banks. In addition, the FDIC collects data on mortgages made on new and existing single-family homes by commercial and mutual savings banks for incorporation with similar data from savings and loans institutions into the Federal Home Loan Bank Board monthly series on interest rates and terms of nonfarm conventional first mortgages, and for use in Department of Housing and Urban Development data on gross mortgage flows.

The Federal Home Loan Bank Board provides data on the operations of the Federal Home Loan Banks, savings and loan associations, and general information on the mortgage market.

The Department of Housing and Urban Development provides monthly data on originations of mortgage and construction loans, purchases and sales of loans for residential and nonresidential properties, commitments for mortgage loans, and quarterly information on gross flows of mortgage loans.

The Securities and Exchange Commission provides current and/or annual data on the volume and dollar value of activity in the securities markets, institutional purchases and sales of common stocks, ownership of common stocks, condition of security brokers and dealers, and noninsured private pension funds.

The Department of Agriculture provides current and/or annual data on agricultural credit, including agricultural loans by financial and lending institutions, farm mortgage debt and interest rates.

The Department of the Treasury provides monthly data on the national debt, Federal securities, and international Government transactions.

The Bureau of the Census provides annual data on debt transactions and cash and security holdings of State and local governments. It also provides quinquennial information on farm finances as part of the census of agriculture.

The Bureau of Economic Analyses provides quarterly information on international financial transactions as part of the balance of payments accounts (see chapter on the economic accounts).

Data Gaps and Further Development

Data weaknesses of banking and credit statistics may be classified in four broad categories: money stock, consumer credit, mortgage and construction loan interest rates, and small business data needs.

Money Stock.-The money supply aggregates composed alternatively of various categories of cash and deposits and referred to as M1 through M5 are a basic source of information used by the Federal Reserve Board in setting monetary policy.

In its 1976 report to the Federal Reserve Board, the Advisory Committee on Monetary Statistics recommended a number of improvements for strengthening the analytic usefulness and statistical reliability of the

money stock figures.' These include money stock definitions, nonmember bank deposits, consolidation of financial institution accounts, foreign deposits, seasonal adjustment, short-run transitory variations, and negotiable and payment order of withdrawal accounts.

A program to improve the measures of deposits of nonmember banks of the Federal Reserve System has begun with an experimental survey by the Federal Deposit Insurance Corporation to collect weekly data on deposits of nonmember banks to supplement the existing quarterly call report data. A decision will be made in 1979 on whether to make this survey permanent. Reliable measurement of nonmember banks is important because of their effect on monetary policy. They account for approximately 25% of all commercial bank deposits, and since this proportion is growing, it will be still more important in the future to have an accurate assessment of their role in the economy. The FRB also established a committee of outside experts in 1978 to review the seasonal adjustment techniques presently used and to recommend the most appropriate methods for adjusting the monetary aggregates and other financial data.

To further strengthen the money stock figures, the following is recommended:

1. The Federal Reserve Board should conduct a retrospective review of the recommendations of the Advisory Committee on Monetary Statistics for improving the monetary aggregates to take stock of which recommendations have or are being implemented, and to develop a systematic plan for addressing the remaining recommendations.

2. The experimental program by the Federal Deposit Insurance Corporation to collect weekly data on deposits of nonmember banks of the Federal Reserve System should be made permanent.

Consumer Credit.-Consumer purchasing power is composed of personal income and consumer credit. Recommendations for strengthening the current estimates of total and disposable personal income were made in the GNP Data Improvement Project Report. These include improvements for the components of wages and salaries, labor income from wage and salary supplements, income from selfemployment and rentals, and deductions of Federal income and social security taxes.

'Report of the Advisory Committee on Monetary Statistics, Improving the Monetary Aggregates, Board of Govenors of the Federal Reserve System, June 1976.

The most prominent weaknesses in consumer credit information are the data on loans by credit unions and the use of mortgage borrowing on existing homes for consumer purchases.

Current data on loans by credit unions are derived from monthly surveys of the National Credit Union Administration (Federal and State chartered unions) and from annual surveys of the Credit Union National Association (State-chartered unions). These in turn are developed into national totals by type of loan (automotive, other consumer goods, personal loans) by the Federal Reserve Board. The basic problems with the underlying data are the lack of current monthly information on new loans and repayments by State credit unions (monthly data are available only for loans outstanding), and weaknesses in the sample of Federal credit unions from which data on new loans and repayments are obtained.

The difficulty with the information on mortgage loans on existing housing is the lack of direct information on the use of these borrowed funds-for example, refinancing existing mortgages, downpayment for second home, home improvements, purchase of consumer durables, or payment for medical and educational expenses. A recent study suggests that such borrowing has bolstered consumer expenditures in the mid-1970's as the inflation in real estate values has been used by households to finance additional consumer outlays by increasing household debt." However, it is likely that the use of such funds for consumer expenditures varies with economic and financial conditions. For example, during periods of financial stringency, it is possible that a larger proportion of mortgages on existing homes would be used to build up liquid asset balances. Thus, a tightening of the availability of mortgage funds could have an intensified impact on consumer outlays. The absence of information on the use of mortgage loans on existing housing is an important missing part of data needed for monetary policymaking.

To improve the estimates of consumer credit, the following is recommended:

1. The National Credit Union Administration and the Federal Reserve Board should review the needs, feasibility and costs of obtaining more reliable information on monthly loans and loan

'David F. Seiders, "Mortgage Borrowing Against Equity in Existing Homes: Measurement, Generation, and Implications for Economic Activity," Paper presented at the Eighth Annual Conference of the Society of Government Economists, March 27, 1978.