Agency use these statistics for their statistical programs.

The Federal Reserve Board compiles and publishes the Industrial Production Index. This index is a comprehensive measure of the physical output of mines, manufacturing plants, and utilities of the United States. It is made up of 235 separate production series which are combined with value-added weights to a total output. For the sectors it covers, the Industrial Production Index provides an alternative measure of the goods producing part of the economy. It is a monthly index and is available usually 15 days after the end of the month, thereby providing a very current measure of the nonagricultural commodity output. Because the goods producing sectors are more volatile than the service producing sectors, this measure is useful in pinpointing the source of important changes in the economy.

An important proxy measure for estimating some Industrial Production Index series, and an industrial measure in its own right, is the monthly electric power consumption data by industry. These data are collected by 3- or 4-digit SIC detail from most utilities and a majority of the large self-generators of electricity by the Federal Reserve Banks and forwarded to the Federal Reserve Board.

The Federal Power Commission (FPC) collects data on the total amount of electricity generated, the amount used by the power companies themselves, as well as power donated, the line loss, the amount sold for resale, and that sold to ultimate consumers. The FPC also collects substantial data on fuel used, fuel stocks, and the type and capacity of generating equipment and transformers as well as the number of miles of line by voltage carried. Similar data are collected for gas utilities. (See chapter on energy statistics for additional details.)

The U. S. Forest Service in the Department of Agriculture collects data on the production of logs for lumber and pulp and to a lesser extent for poles, posts, chips, Christmas trees and firewood. The U. S. Fish and Wildlife Service in the Department of the Interior collects data on catches of fish and wild animals. The Nuclear Regulatory Commission collects data on mining and refining of uranium and other radioactive materials (see chapter on energy statistics). The Small Business Administration collects some information on small businesses.

The Federal Trade Commission collects data with their Corporate Patterns Survey, Line of Business Survey, Quarterly Financial Report and Working Capital of Nonfinancial Corporations. These surveys are discussed in the chapter on financial statistics.

Many other agencies collect data in specialized areas which are described in other sections. Specialized agencies focus to a much larger extent on activity rather than the industry data for their areas of interest. For example, if a university runs a printing and publishing activity, the employees, receipts, capital expenditures, and so forth of that activity may not be counted in either education or printing. (The BLS employment figures should include it in education, though.)

International trade statistics are used in many programs by the Departments of Agriculture, Interior, Commerce, Treasury, Labor, Energy, and State as well as by the U.S. International Trade Commission, the Special Trade Representative and others. These programs rely heavily on statistics on imports and exports, but production, employment, and prices are also used.

International merchandise trade statistics are compiled and published by the Census Bureau from information provided by importers and exporters. Data on imports are collected as part of the documentation of the importation of goods and are handled administratively by the U.S. Customs Service in the Department of the Treasury. Data on all imports over $250 and a sample of imports of less than $251 are processed and tabulated on a monthly basis with cross-tabulation of commodity by country of origin and in other arrangements.

Export data are also compiled by the Bureau of the Census using an export declaration form or machinereadable monthly reports. The U.S. Customs Service has administrative responsibility for the collection of the export declaration. This declaration is required. for all exports of over $250, and is supplemented by calculated estimates of the exports of less than $251. For exports to Canada, consideration is being given to the use of the Canadian figures on U. S. imports from the United States. Duplication of effort in the collection of data concerning exports on one side of the border and imports on the other makes for extra burden and incompatible figures. The objective is to reduce the incompatibility between our exports and "their" imports to a minimum. Arrangements should then be made to use only one source for trade data. A standard international commodity classification system would help minimize the differences (see chapter on a program of standards development). Efforts to achieve a "harmonized" international system are now taking place through the efforts of the Customs Cooperation Council.

County Business Patterns (CBP) is a publication of the Census Bureau tabulating the number of establishments, first quarter payroll, annual payroll,

and March 12 pay period employment to the 4-digit SIC level by county. This program utilizes the Census Bureau's annual Report of Company Organization to obtain establishment data for multilocation firms. Data for single unit firms are obtained from administrative records of the Internal Revenue Service and the Social Security Administration. More specifically, payroll data are obtained from Internal Revenue Service form 941 and employment data from the Social Security Administration form 941, schedule A. Excluding the self-employed, County Business Patterns covers approximately 75 percent of the employment in the private sector. Major exclusions are government employees, farm workers, and domestic service workers. Also, railroad employment subject to the Railroad Retirement Act is excluded. CBP coverage of hospitals, schools, and public utilities is limited to those operated by the private sector. Even though the coverage is poor in these areas (due to the predominance of government establishments in these industries), CBP is a very useful detailed profile of the sectors of the economy, especially at the county level.

Comprehensive measures of personal income and employment by industry for States and local areas (including counties) are prepared by the Bureau of Economic Analysis as part of the regional economic accounts.

Important Gaps

Many areas that need strengthening in production and distribution statistics are outlined in the chapter on national economic accounts and are not covered here.

There is very little knowledge of size, receipts, and organization of some service industries. Statistics about industries which are not included in the economic censuses, especially finance, insurance, real estate, social and miscellaneous services, charitable organizations, and truck transportation, are especially lacking. (Some of these industries have been included in the 1977 Economic Censuses.) The statistics on government operating or nonadministrative establishments are not very well integrated with private sector statistics, for example, local transportation or highway repair.

Inventories and capital expenditures data are weak, but improvements are now being explored by the Census Bureau. (See the chapter on national accounts, particularly the Data Improvement Project.)

The Current Industrial Reports (CIR) mentioned above cover about half of the manufacturing sector. With the elimination of the Annual Survey of Manufacturers question on the quantity of produc

tion in 1975, no quantity data on many items is collected between the Economic Censuses. For many production items, quantity is not meaningful, but in view of the requirements of the 1974 Trade Act, a study of the CIR program in relation to the needs is necessary.

Programs Which Should Be Discontinued

The Industrial Mobilization Surveys that the Bureau of Domestic Business Development in the Department of Commerce does for the General Services Administration (GSA) should be evaluated for the feasibility of incorporating them in the Capacity Survey or Current Industrial Report programs of the Census Bureau. This review should include the GSA requirement for individual plant data. The Capacity Survey is still in need of continued work, both conceptually and in data gathering techniques, before it can substitute for mobilization needs.

The employment data collected by the Census Bureau could be dropped when the Industrial Directory (see chapter on the Industrial Directory) is operating. A single data item of overlap might be needed, at least for a while, but eventually it too could be dropped. Data on number of employees collected by the BLS could be incorporated into the data system that serves as the background to the Industrial Directory. This program needs to operate in parallel for a while before the differences in definition are even verifiable.

Recommended New Programs

A Broader Census Publication Program

Title 13 of the U.S. Code forbids the Census Bureau to collect data from industries that are required to provide similar data to other agencies. But title 13 authorizes Census to tabulate and publish that data. This effort of tabulating and publishing the data already in the hands of the Federal Government in a format consistent with the quinquennial census would increase the knowledge available to analysts, without any increased industry reporting burden. Much of the data are already in machine readable form. Although much of the data can be assembled by a perservering analyst, they are not in comparable form, units, or coverage. Some data cover a complete universe, but other data are uninflated sample totals. This program is of special relevance now, since the Census Bureau's Report of Company Organization allows much greater control over the industry classification and minimizes the possibility of duplication or omission of establishments.

An annual economic data publication that would be an expansion of the present County Business Patterns which provides mid-March pay period employment, first quarter and annual payroll, and number of establishments should provide a comprehensive source of statistics about the Nation's businesses. This publication now includes establishments with about 75 percent of the private sector workers. The inclusion of Federal and State

and local government workers as well as all private sector workers would substantially increase the coverage of CBP. The government sector represents a total of approximately 14 million employees or over 15 percent of the labor force. The addition of sales and receipts data to this series would provide an important measurement of the economic spectrum and would increase the value of the series. These data would provide comparable statistics for the 4-year interval between economic censuses.

Specialized agencies collect little data on the secondary activities of the establishments they cover and have uneven coverage of data concerning the secondary activities of industries that are outside their responsibility. The need for data on activities rather than establishments is not fully met by most data collected on an establishment basis. There also is very little integration or uniformity of data collected by regulatory agencies with that collected by the Census Bureau. There are data related to activities, (maybe more than can be utilized in a statistical program) but they are far flung and difficult for the analyst to put together. This compilation could and should be done by the Census Bureau. The Report of Company Organization provides the control to incorporate these statistics into a unified, comprehensive statistical data program.

One way to try to fill the gaps between activities and industries would be to ask all establishments to complete a checklist of activities. Because of the wide range of activities engaged in by some companies, this list could not cover all activities, but it could answer questions such as: Does the establishment have a printing capability, a health unit, serve meals, have construction capability, legal counsel, or advertising staff? The Census of Manufactures question on "selected metal working operations" could be used to provide one set of such "yes-no" activity check boxes. This checklist would be

expensive, but combined with the information in the Industrial Directory there would be a substantial payoff. A few inquiries to firms of various sizes and industries that conduct a specific activity could provide a ballpark estimate of the extent and size of these secondary activities.

An example of the use of this checklist would be to evaluate a particular concern about establishments industry covers only about 40 percent of that activity, serving food away from home. The "Eating Places" since schools, hospitals and other health institutions, airlines, drug stores, department stores, the Armed Services and even some farms serve meals to nonfamily members.

Transportation Statistics—Integration of Data with Industry Shipments Data

There is little knowledge of activities of truck transport and some other transportation industries. This is a difficult industry to obtain data about because of the large number of small establishments, and the frequency of many operators going in and out of business. A truck owner may be in business for himself one day and not the next. An entrepreneur will get a hauling contract, lease a truck to fulfill it, but be back as an employee of another business before long.

Collecting data for the transportation census by getting it from manufacturers is a roundabout approach to the collection of transportation data. This does not mean we ought to abolish the existing collection effort until we have an improved one in place. Even then we may want to collect some data each way. At least for the present, the survey forms that go to manufacturers to collect production statistics could be combined with those to collect shipping statistics into one package. A standard commodity classification which would provide enough detail for freight rate calculation would be a small start in collecting such data from shippers. (See chapter on a program of standards development.)

Comparable Public and Private Sector Data.Statistics on the business activities of States, local governments, and even the Federal Government are quite limited for comparison to private sector data. Wherever private sector data are collected, similar kinds of data should be collected for comparable government establishments.

Science and technology (S&T) are implicitly embedded in much, if not most, of the data collected by the Federal Government. This is not surprising, considering that science and technology are deeply ingrained in our culture and thus pervade almost all elements of our national fabric. Therefore, it would be conceptually incorrect to place all statistics related to science and technology into a single functional area. Science and technology provide the tools with which we as a Nation achieve our objectives and these objectives are and should be the appropriate functional categories used to describe the Federal statistics program.

However, there is one set of science and technology statistics which is a distinct, conceptually cohesive, functional group, namely the one dealing with the magnitude and nature of American science and technology. These statistics cover primarily the inputs such as human resources, funds, and institutions as well as some of the outputs attributable solely to science and technology activities.

As will be seen from the subsequent enumeration of Federal science and technology statistical programs, the National Science Foundation (NSF) is the principal generator and compiler of such data. This is in line with the Foundation's statutory authority, the National Science Foundation Act of 1950, and subsequent amendments which authorize and direct NSF:

... to maintain a current register of scientific and technical personnel, and in other ways to provide a central clearinghouse for the collection, interpretation and analysis of data on the availability of, and the current and projected need for, scientific and technical resources in the United States.

to initiate and maintain a program for the determination of the total amount of money for scientific research received by each educational institution and appropriate nonprofit organization in the United States from agencies of the Federal Government, and to report annually thereon to the President and the Congress.

By Executive order, NSF has also been assigned focal agency responsibility for the "development and analysis of information on the supply, demand, utilization, and education of scientific and technical personnel."

Users

Science and technology statistics, as defined above, are utilized by a variety of organizations. The Federal Government is one of the principal users since this information is required for the development of Federal science and technology policy. Organizations in the Executive Office of the President that extensively use S&T statistics include the Office of the President's Science Advisor, the Science and Technology Policy Office, and the Office of Management and Budget. In addition, agencies and departments that are heavily involved in science and technology related activities make extensive use of S&T statistics. These include the National Institutes of Health, the National Aeronautics and Space Administration, the Department of Defense, the National Science Foundation, and the new Department of Energy. Congress is also a heavy user of S&T statistics, especially the committees that cover the activities of the agencies and departments listed above and those that deal with S&T-related topics such as energy or the environment.

The private sector requires and uses S&T statistical information. The higher education community, which has the responsibility for the training of new scientists and engineers, requires continually updated data pertaining to scientific and engineering human resources-supply, utilization, graduate student support, enrollments, and so forth. Furthermore, since this sector receives a significant amount of Federal research and development (R&D) funds, it is also interested in statistics relating to these financial resources. Industry, the principal performer of R&D, is a heavy user of S&T statistics, especially those dealing with R&D funding. Besides these institutional users, individual researchers in the science policy area depend on federally generated S&T statistics.