the transportation, trade, and service industries to move the materials from the site of production to the construction site. The producers' value of each of the materials, supplies, and equipment was then converted to the primary man-hours needed for their production in the final stage of manufacture. The secondary employment was estimated by determining the interindustry transactions and associated employment needed to produce the materials, supplies, and equipment used in construction.8

On-Site Labor Requirements

The 94 man-hours of on-site labor include about 40 hours for equipment operators and crafts associated with equipment operation, 31 hours for unskilled occupations, 13 hours for miscellaneous construction crafts, 9 hours for professional and managerial duties, and 1 hour for the various clerical and service occupations.

This distribution reflects the widespread use of construction equipment. As shown in the following tabulation, equipment operators and the occupations associated with equipment, such as oilers, mechanics, firemen, and stationary engineers, accounted for over 43 percent of the on-site labor. All other skilled crafts combined (for example, structural steelworkers, carpenters, and concrete finishers) represented only a seventh of the total hours worked. About one-third of the hours were put in by unskilled workers.

operating speeds of earth moving machinery and the development of paving and finishing machines which combine operations previously performed by several machines. Part of the decrease in onsite man-hour requirements may also be due to changes in the design and composition of the roads.

On-site unit man-hours have declined not only absolutely but also, it is probable, in relation to off-site man-hours. Material costs in relation to the other construction cost items have become more important, and the unit man-hour requirements for roadbuilding have declined 47 percent contrasted with the 30-percent decline in the total private economy during the same period. Both the increasing importance of material costs and the more rapid decline of on-site man-hours probably would result in some shift of man-hours off site relative to the amount on site.10

7 The man-hours ratios used were value of shipments and man-hours as reported in the 1958 Census of Manufactures.

The interindustry transactions were obtained by using the Bureau of Labor Statistics interindustry transactions matrix for 1947 with current price levels and man-hour ratios. See W. Duane Evans and Marvin Hoffenberg, "The Interindustry Relations Study for 1947," Review of Economics and Statistics, May 1952, pp. 97-142.

For data from 1947 through 1954, see E. L. Stern and E. J. Coppage, Jr., "Labor Usage in Highway Construction As Influenced by Improved Equipment," Public Roads, February 1957, pp. 151-152. Data for 1955 through 1958 are from unpublished BPR statistics.

10 The data on material costs are unpublished figures from the BPR; the unit man-hour requirements in the private economy are based on hours worked per dollar of real product. See Trends in Output per Man-Hour in the Private Economy, 1909-1958, BLS Bull. 1249 (1960).

Off-Site Employment

Primary and secondary man-hour requirements away from the job site totaled 125 hours per $1,000 of construction contract in 1958, as previously noted. These hours were generated in the industry sectors which produce, transport, and sell the raw materials, intermediate products, and final materials and equipment needed in highway construction, as well as in the off-site activities of the construction industry. (See table 1 and chart 2.) The area of the economy (other than construction) most affected by highway construction was the manufacturing sector. Each $1,000 of construction required 60 hours of employment in manufacturing, and about half of these hours were in the last manufacturing stages before shipment to the construction site. The principal manufacturing industries directly affected were producers of fabricated metal products, construction machinery, cement, and concrete products (table 2). Manufacturing industries less directly affected but representing substantial employment included the iron and steel industries which furnished steel to the metal fabricating industries, and the iron and steel forgings industry which furnished parts to the equipment manufacturers and other fabricators.

The mining industries furnished the crushed rock, sand, gravel, and other aggregate and required 20 hours of employment for this purpose for each $1,000 of highway construction. The mining industries also supplied other affected industries.

About three-fourths of the 15 hours of labor required in the transportation industries resulted from the delivery of products to the construction site. The remaining man-hours were required to

As stated earlier, nearly two-thirds of the roadbuilding contract dollar in 1958 was spent for materials and construction equipment. The kinds of material and equipment purchased are shown in table 3.

Expenditures for construction equipment-the graders, rollers, tractors, cranes, trucks, bulldozers, pavers, etc., which are so essential to modern road building-represented almost oneeighth of the contract dollar.

Purchased aggregate accounted for 7.1 percent of contract cost. In addition, contractors produced aggregate at or near the site, and aggregate was used in many of the other purchased materials, including premixed concrete, bituminous paving materials, and concrete pipe. In total, it is estimated that aggregate represented over 18 percent of the total contract cost.

The steel used, exclusive of that in construction equipment, represented over 12 percent of the construction cost. Steel was used principally for reinforcing both roadways and bridges and for structural purposes in bridges.

Petroleum products for the construction equipment and for use in highway surfacing was a major material cost, representing 10.7 percent of the

contract dollar. However, because of low labor requirements in the petroleum industry, the offsite employment generated by these purchases represented a very small portion of the total labor requirements.

Comparisons With School Construction

A comparison of this study with a similar analysis of school construction discloses that total man-hour requirements for highways are virtually the same as those for schools. The BLS findings show highway construction (1958) requirements totaling 219 man-hours per $1,000 of expenditures compared with 212 man-hours for a like amount of spending for school construction (1959). However, a rough estimate of changes in prices and output per man-hour between 1958 and 1959 indicates that the total man-hour requirements per $1,000 of highways in 1959 was probably very close to that for schools in 1959.

Despite their similarity in total man-hour requirements, expenditures for schools and highways affect individual industries in the economy in a different manner. On-site requirements for highways are somewhat higher than those for

schools even after price and productivity shifts update the highway hours to 1959.

The ratio of off-site man-hour requirements to on-site man-hour requirements for highway construction was 1% to 1. School construction required 11⁄2 man-hours off-site for each man-hour at the site.

Generally, the effect of school building is felt more strongly in manufacturing industries but has less impact on the mining industries than highway construction because of the importance of aggregate.

The principal manufacturing industry groups affected were the same for both types of construction, but they generally received a higher share of the school construction dollar.

-RONALD E. KUTSCHER AND CHARLES A. WAITE Division of Productivity and Technological Developments

Economic Status of Social Welfare Workers in 1960

THE BUREAU OF LABOR STATISTICS, in cooperation with the National Social Welfare Assembly and the U.S. Department of Health, Education, and Welfare, conducted a nationwide survey of social welfare manpower during 1960. The survey, conducted by mail questionnaire, was designed to provide information on annual salaries and working conditions of personnel in the social welfare field as they relate to the personal characteristics of the work force, programs and position levels, and the educational and experience levels of workers. This article gives major attention to nationwide employment and salary levels.1

public assistance programs (one-half) and in child welfare work (a fourth).2 Among 4,000 Federal workers, nearly a third were in recreation work and about a fourth were in psychiatric social work.

Of the 39,600 social welfare workers employed by State or local voluntary agencies, a fourth were engaged in group work-informal education and leisure-time activities (e.g., scouting and YWCA). Child welfare work, family services other than public assistance, and community organization work (including community chests, united funds, sectarian coordinating councils) each accounted for about a sixth of the total. A third of the 2,600 workers in national voluntary agencies were employed in community organization work.

Direct-service positions accounted for 62 percent of the nationwide employment in social welfare agencies in 1960; executive and supervisory positions accounted for 21 and 10 percent, respectively. The remainder of the workers were mainly in teaching, research, and consultation. Case workers accounted for three-fourths of all direct-service workers and for slightly less than half of all workers within the scope of the study.

As shown in the following tabulation, the distribution of workers by position differed substantially among the employing agencies. These differences undoubtedly reflect a variety of factors, including variation in type of program and average size of staff.

1 Greater detail on the survey, including tabulations of regional and metropolitan area salary levels, is provided in Salaries and Working Conditions of Social Welfare Manpower in 1960, published by the National Social Welfare Assembly, Inc., 345 East 46th St., New York 17, N.Y. Price $1.75.

2 The survey did not include about 2,000 vocational rehabilitation counselors employed in State or local government programs.

"Economic Status of Social Workers in 1950," Monthly Labor Review, April 1951, pp. 391-395. Since personnel in strictly recreation settings were not included in 1950, time comparisons exclude recreation workers.

4 See footnote 1, table 5, for definition of western region.

Sex, Age, and Marital Status. Women accounted for 65,750, or 57 percent of the social welfare workers in 1960. Regionally, the proportion of women ranged from one-half in the Pacific region to two-thirds in the Southeast. Excluding recreation workers, women accounted for 59 percent of the employment total in 1960 compared with 69 percent in 1950. During the decade, the number of women and men social welfare workers increased by 23 and 85 percent, respectively.

Distributions of men and women by age and marital status are shown in chart 1. Median age

estimates in 1960 were 42 years for all workers, Ex39 years for men, and 46 years for women. cluding recreation workers, the average age for men was unchanged from 1950, whereas the average for women rose by 5 years.

Among men workers, 82 percent were married, 15 percent were single, and 3 percent were widowed, divorced, or separated. Estimates for women were as follows: married, 45 percent; single, 35 percent; widowed, divorced, or separated, 19 percent.

Men and women were distributed somewhat differently by position level. Excluding recreation workers, for example, 41 percent of the men and 59 percent of the women were case workers; 12 percent of the men and 8 percent of the women were in other direct-service positions; 9 percent of both men and women were supervisors; and 29 percent of the men, compared with 15 percent of the women, were executives. Within each position grouping, the median age for women exceeded that for men; among case workers, the difference was 8 years and amounted to 7 years in most of the other positions.