In this discussion, technological change is regarded as embracing all the many types of innovations which result from the application of scientific and engineering knowledge and techniques to the processes of production and distribution. In general, the purpose of such change is either the production of new or improved goods or services, or greater efficiency of operations and consequent cost savings. The term "automation" has been given widely varied meanings by different authors. This coined word, derived from "automatization," was used originally to denote certain technical innovations in metalworking industries, involving mass-production machinery with a high degree of automatic operation. Later, the term was applied to electronic computers and to automatic control devices which can measure and adjust performance. These developments are often cited as examples of the advanced stage of technology where mechanisms are used in place of human labor to monitor as well as perform work. It is in this general sense that the term is used in the present report.1

Both theoretical and practical difficulties have so far prevented the development of direct statistical measures of technological change-except for a few types of innovations (notably, electronic computers, on which statistics are presented later in this chapter).2 In efforts to assess the pace of such change, reliance has therefore to be placed on analysis of related statistical series, descriptive data, or a combination of the two.

Among the most important indirect indicators of the rate of technological change are the indexes of man-hour productivity issued by the Department of Labor for many years. Though influenced also by a variety of other factorsincluding production levels, changes in capital investment per worker, education, and the skills of the work force-the rates of change in productivity provide valuable insights into the pace of technological progress in American business and industry.

Total private economy

PRODUCTIVITY TRENDS

Productivity (output per man-hour) in the private economy of the United States rose at an average rate of 2.4 percent a year over the half century covered by the index 1909-63.3 During the postwar period 1947-63, the increase averaged 3 percent.

The increase for the past 3 years, 1960-63, averaged 3.6 percent. While this is a relatively high rate for a 3-year period, it should be noted that the private economy has achieved equal or higher 3-year rates several times in the past. In the immediate postwar readjustment years 1947-50, productivity rose at an average rate of 4.3 percent. And in the years 1952-55, the rate of gain was again relatively high, 3.3 percent.

Another reason for caution in drawing conclusions from the productivity gains of the past 3 years is that, if the period covered by the analysis is extended, the findings appear substantially different. For the period 1958-63, the average increase in productivity is estimated at 3.1 percent, little above the 3-percent average for the entire postwar period.

1 In contrast to this meaning, a definition of automation as "a new way of organizing and analyzing production, a concern with production processes as a system, and a consideration of each element as part of the system" was recently given by a leading expert. Furthermore, the meaning of the term is often extended popularly to cover all types of technological change, including labor-saving machinery of conventional as well as advanced design, and the application of scientific management methods. So broadly defined, automation becomes a loose abstraction, standing for the far-reaching impact of technical change on American industry and manpower.

2 See p. 53.

3 Output per man-hour here refers to the constant dollar value of goods and services produced, in relation to the hours of all persons employed (including proprietors and unpaid family workers). The output concept is consistent with that of gross national product and is measured in terms of 1954 dollars. The man-hours are based primarily on establishement reports to the Bureau of Labor Statistics. Another series of estimates of output per man-hour have also been developed, based on man-hour data from the labor force statistics. In general, the two series show similar movements over the postwar period.

Year-to-year changes in productivity are more closely related to changes in production levels than to any other factor. In general, the largest gains are registered in years of upswing in the business cycle. The rise in output per manhour was sharp in the recovery years 1950, 1955, 1959, and 1962. (See table 9.) TABLE 9.-Year-to-year percent change in output per man-hour,1 1947-63

1 Output refers to gross national product in constant dollars. The man-hour data are based primarily on establishment reports to the Bureau of Labor Statistics. 2 Preliminary.

The generally rising trend in productivity which underlies these cyclical swings -which has been a major element in the country's economic growth and the rising standard of living is the outcome of a variety of factors. Among these are technological advances, capital investment, and scientific and engineering research and development. These factors have a complex interrelationship. Advances in technology are made possible by research and development but are realized through capital investment. Growth in capital spending is stimulated, in turn, by the opportunities for more efficient operations and new and improved products provided by advances in technology-with the total effect of increasing productivity.

Over the long run, education is also an important factor in productivity gains and the country's economic expansion-through its contribution to scientific and technological progress and the quality of the labor force. Recent private studies have emphasized education's contribution to our ecomomic advance to date. One such study found, for example, that two-fifths of the increase in real product per worker over the period 1929-57-a gain of 56 percent-could be attributed to increasing formal education."

The influence of education on productivity is longer term and more indirect than that of the other factors mentioned. It is brought to bear, in large part, as highly educated administrators effect improvements in organization and operations, and as the knowledge of scientists and engineers is realized in technological progress.

On the other hand, advances in education and productivity are mutually reinforcing. As productivity rises, the higher incomes that are generated help young people to stay in school longer and make it easier for experienced workers to leave the labor force for further education or training. Through the increased knowledge and skills thus acquired, these workers and potential workers can then contribute to further gains in productivity and economic growth."

Industry sectors

The great differences in productivity trends among the various sectors of the economy reflect, in large measure, the widely varying rates and character of technological advances in these sectors, as well as fluctuations in production levels,

For a more extensive discussion of these factors, see the 1963 "Manpower Report of the President," pp. 73-76.

"The Sources of Economic Growth in the United States," Supplementary Paper No. 13, Edward F. Denison, Committee for Economic Development, January 1962, p. 73; and hearings before the Subcommittee on Education, Committee on Labor and Public Welfare, U.S. Senate, 88th Cong., 1st sess., Apr. 29-30 and May 1, 1963, vol. 1, p. 408.

Closely related to the rising educational level of the population and hence to productivity gains is the increase in total expenditures for education in the United States. Expenditures for education as a percent of national income dropped considerably during the depression and war years of the 1930's and 1940's, but have climbed steadily throughout the postwar period. The 1962 figure of 7 percent of the national income devoted to education compares with a 1929 figure of 4.7 percent.

changes in production capacity, and other economic factors having a differential effect on different industries."

Agriculture has, in the postwar period as a whole, outdistanced the nonagricultural economy in productivity gains. (See chart 12.) The average annual increase in farm output per man-hour from 1947 to 1963 was 5.7 percent, an extremely high rate for such a long period and just under 21⁄2 times the corresponding postwar average (2.4 percent) for all nonfarm private industry.

7 It will be noted that the discussion of productivity in this section, as in the preceding ones, is limited to the private economy since no satisfactory method has been found for measuring productivity in the Federal Government. However, a positive and vigorous effort to devise ways to measure productivity and to improve manpower utilization is being made within the executive branch of the Federal Government, and examples of dramatic increases in productivity in a number of agencies can be cited. See "Examples of Improved Manpower Management Plans, Programs, and Accomplishments Developed by the Departments and Agencies of the Federal Government During the Fiscal Year 1963," Subcommittee on Manpower Utilization of the Committee on Post Office and Civil Service, 88th Cong., 1st sess., Oct. 16, 1963.

Rapid and widespread technological advances of many kinds-in farm machinery, fertilizers, pesticides, crops, and livestock-are the basic reason for this productivity record. But other factors have also had a sharp effect, especially on the short-run fluctuations in farm productivity.

Annual productivity changes tend to be much more extreme in agriculture than in the rest of the economy. (See table 9.) During the past 5 years, the changes have ranged from a gain of 9 percent in 1957-58, to a loss of 0.2 percent in 1958-59, to gains of around 6 percent in the following 2 years, and finally to one of 7.4 percent in 1962-63.

Because of these wide yearly fluctuations, modifications in the underlying trend of farm productivity are not easily discernible. Nevertheless, there does seem to have been some slowing down in the rate of productivity growth in agriculture. During the last 6 years, the average annual rate of increase was 4.9 percent, compared with 5.7 percent over the postwar period 1947-63.

From the manpower viewpoint, however, the most important fact about farm productivity is that it continues to rise much more rapidly than total farm output From 1957 to 1963, agricultural production rose at an annual rate of only 1.3. percent, little more than a fourth as fast as output per man-hour. In consequence, farm employment declined (by about 3.3 percent per year). Aggregate farm man-hours went down slightly more (by about 3.4 percent per year), probably because of a decline in the workweek.

In the nonagricultural sector, productivity growth has moved in much the same way as in the total private economy, of which this sector comprises such a large part. The average annual increase in output per man-hour was high in nonagricultural industries in the early postwar years, then slackened, and has risen again in recent years. For the entire postwar period 1947-63, the average rate was 2.4 percent. From 1957 to 1963, productivity gains in this broad economic sector averaged at least 21⁄2 percent. (See table 10.)

TABLE 10.-Average annual percent change in output 2 per man-hour and related data, 1947-63

1 Computed from the least squares trend of the logarithms of the index numbers. 2 Output refers to gross national product in constant dollars.

Within the nonagricultural economy, some industry groups have achieved much more rapid gains in productivity than others (as shown in table 11). In mining and in communication and public utilities, the rate of increase in output per man-hour averaged more than 3.5 percent per year from 1947 to 1962. (Data for 1963 are not yet available.) In transportation the annual productivity increase averaged between 2.5 and 3.5 percent over the postwar period as a whole, but during the last 5 years, 1957-62, rose to more than 3.5 percent. All these industry groups are notable for extensive and rapid technological advances.

* See the 1963 "Manpower Report of the President," pp. 73–74.

The major technological developments in communication and transportation are discussed later in this chapter. See also the 1963 “Manpower Report of the President,” pp. 74–75.

TABLE 11.-Approximate percent change in output per man-hour, by industry, 1947-621

1 These estimates of output per man-hour are based in part on estimates of output (GNP) published by the U.S. Department of Commerce in Survey of Current Business, September 1963. Exact averages are not shown because the output figures will be benchmarked next year to the 1958 input-output table and are, therefore, still preliminary. 1963 data are not yet available for these industry groups.

The broad field of manufacturing ranks next in rate of gain in productivity, with an average annual increase between 2.5 and 3.5 percent over the past 5 years and also the past 15 years. It should be emphasized, however, that this moderate productivity record for manufacturing as a whole is the outcome of a great diversity of technological developments and of economic and other factors which have a differential effect on different industries. The statistical information so far available does not permit quantitative comparisions of the progress in productivity and technology in the various major branches of manufacturing. However, some industries-for example, the manufacture of synthetic fibers and petroleum refining are known to have been leaders in the introduction of new technology and the achievement of increased output per man-hour, whereas in some others--for example, apparel and shoe manufacturing-technological changes have been limited in recent years.10 Such gains in productivity as have been made in these two industries during the postwar period have largely resulted from better organization of production and other nontechnological factors.

The remaining industry divisions-construction; trade; finance, insurance, and real estate; and service industries have all made relatively small gains in productivity. For each of these industry groups, the average annual increase in output per man-hour is estimated to have been less than 2.5 percent both during the 5 years 1957-62, and over the entire 1947-62 period. It should be noted, however, that these productivity estimates were obtained by output measures consistent with the data on gross national product. For some sectors of the private economy-particularly those producing services rather than goods, where productivity gains are shown to be smallest there are some questions about the adequacy of the data to fully reflect real changes in output.

Automation and other recent technological innovations have made headway in a few individual industries within these broad groups. The insurance industry, for example, has been a pioneer in the application of electronic data processing (EDP) to massive clerical operations, and more banks are now converting to EDP systems. New materials-handling devices are also being widely installed

10 1 he regional conferences on industrial modernization, now being sponsored by the Department of Commerce to stimulate growth in lagging industries, may have important manpower implications.