economic activity is manufacturing. This segment employed a little over 20 percent of the total manpower which was employed in 1935. More manpower was engaged in agriculture than in manufacturing, and nearly as much in wholesale and retail trade, and also in such services to the consumer as education, professional, personal, and domestic services, recreation, and amusement. The small amount of employment in construction reflects very much lower than average activity in that field in the particular year. On the whole, this picture of manpower employed in the different segments of the economy is the best single guide to the role of each segment in the national economy.

A further guide is provided by the volume of physical capital employed in each segment. A very crude estimate of the value of the land, buildings, equipment, and inventories employed in each segment is given in chart III. Estimates are hard to make in this field, partly because of inadequate data but more especially because of the subjective character of the whole process

of valuation. For agriculture, the estimates are those of the census of 1935, while for utilities and manufacturing estimates have been made from income tax returns based on the book values reported by corporations. Estimates for the other segments are only crude approximations. The classification of industries is identical with that in the preceding chart, and the blocks of manpower and physical capital are roughly comparable except for Government, services to the consumer, and, construction. In the case of the first, an important proportion of the physical capital is made up of the public domain and the public highways. These bear little relation to the manpower currently active in the production of government-rendered services. In the case of services to the consumer, residential housing accounts for the bulk of the capital values in this segment and renders services to the consumer with relatively little manpower currently employed, while the manpower engaged in serving the consumer in professional, personal and domestic services, education, and recrea

tion uses relatively little associated physical capital. The estimate of physical capital in the construction industry is very unreliable.

Table I shows the value of the physical capital per equivalent full-time worker for the six segments where the basic figures are sufficiently accurate for the ratio to have significance. The table should be regarded as only suggestive partly because of the crudity of the estimates of physical capital and partly because the proportion of the physical capital which was actually in operation in 1935 is not known. It does indicate the larger physical investment per worker in the transportation, communication, and power fields, the relatively small investment in lands, buildings, and equipment per worker in trade and consumer services, and the relatively similar investment per worker in agriculture and manufacturing.

A third guide in giving the picture of productive activity is the contribution to production made by

each segment of the economy. Chart IV gives for each segment an estimate of the income produced by that segment. The proportionality is approximately the same as that of manpower in the different industries except that the value of the agricultural contribution was smaller and the contribution of government and of the utilities was greater than the corresponding manpower ratio.

TABLE I-Value of Land, Buildings, Equipment and Inven-
tories per Equivalent Full-Time Worker, 1935
Public Utilities -
Mining
Agriculture..
Manufacturing.

$11, 900

8, 700

3, 900

3, 700

3, 700

2,000

4, 600

The three charts above give a very clear picture of the proportionality of the different types of economic activity as they were carried on in 1935. The real structural characteristics of production, however, are only to be found by examining economic activity through time. A first crude picture of the changing relative importance of different segments can be obtained from Charts V and VI which are derived from the Census of Occupations. The first shows for each census year the absolute number of persons reporting themselves as gainfully employed in each of the main types of activity while the second indicates the proportion of the total in each segment.

The most striking indication of the charts is the relative decline of the role of agriculture in the national economy. While the number of persons gainfully employed in agriculture increased gradually from 1870 to 1910, in the 60 years from 1870 to 1930 agriculture dropped from 53 percent of the total gainfully occupied

to little more than 21 percent. The great increases have come in trade and transportation and manufacturing and clerical so that their proportion of the total has increased from 32 percent in 1870 to 57 percent in 1930. Thus from 1870 to the present time the national economy has shifted from an economy which had been dominantly agricultural and in which more than half the workers were agricultural to one which is predominantly industrial. This has shifted the whole character of the productive structure away from that associated with agricultural production and toward that associated with industry.

Estimates of employment and the trends of activity since the World War are shown in terms of employment for each segment in chart VII. The trends are adjusted for depression activity and represent the trends of change in the manpower employed in each segment which could have been expected if reasonably full economic activity had been maintained. The two extractive segments, agriculture and mining, show a gradually declining trend of employment even after adjustment for depression, while manufacturing and the utilities show only a slightly rising trend in the post-war period. The areas of expanding employment have been in the rendering of direct services to consumers, in the field of trade, and in the fields of government and finance. No trend can be drawn for employment in construction for reasons which will become apparent in the discussion of production in relation to durability. Leaving construction out of account, it is apparent that the recent trends of employment have shifted the relative emphasis in productive activity away from the extractive and manufacturing industries and have increased the proportion of the available manpower engaged in rendering services either to the consumer or to the whole economy.

To the structure of production indicated here through time, another dimension must be added, the pattern of sensitivity to changes in consumer income and in the level of productive activity. In Table II below, the segments are arranged in order of the increasing sensitivity of their production to depression as revealed by the sensitivity of their employment. Agriculture and government, the least sensitive, are at the top; construction and mining, the most sensitive, are at the bottom. A rough measure of sensitivity can be obtained by comparing the level to which employment had fallen in 1932 with the level of employment called for in that year by the post-war trend. In Table II the ratio of actual employment to the level called for by the postwar trend is given for each segment. Here the essential stability of agriculture and government and the great instability of construction, manufacturing, and mining are clear.