"cancellation" or "termination" of the agreement. The following is the standard language found among clauses covering electricians in the construction industry:

Local Union . . . is a part of the International Brotherhood of Electrical Workers, and any violation or annulment of working rules or agreement of any other local union of the IBEW or the subletting, assigning, or the transfer of any work in connection with electrical work to any person, firm, or corporation not complying with the terms of this agreement by the employer, will be sufficient cause for cancellation of this agreement, after the facts have been determined by the international office of the union. One agreement in the apparel industry sanctioned union exercise of an injunctive remedy, as follows:

The employer agrees that he will at no time buy cut goods for caps or hats to be manufactured on his premises nor shall he contract any work to any nonunion shop. Notwithstanding the provisions of paragraph 23, and in addition to the relief provided in said paragraph, a breach of this clause shall also entitle the union, in an action at law or in equity, to judgment for damages for wages lost by its members, employees of the employer, as well as to injunctive relief to restrain a further breach of this clause.

A number of provisions, mostly in the construction and utilities industries, permitted noncompliance with subcontracting provisions under certain circumstances. Commonly, the union waived enforcement if the employer, in fulfilling his obligations under the subcontracting clause, would violate State or Federal statutes. In a few additional situations, the clause could be bypassed if its compliance created economic hardships for the employer. The following provision from an electrical machinery agreement covered both situations:

When building or construction work of the type customarily performed by the building trades unions of the AFL-CIO is contracted out, preference shall be given to qualified contractors employing members of the trade unions affiliated with the AFL-CIO. Nothing herein shall require the company to violate Federal, State, or municipal regulations, to delay the work, to employ a contractor either not readily available or not equipped to do the work, or to bear unreasonable cost. If faced with such contingencies, the company shall immediately take the matter up with a proper representative of the local union.

Labor Requirements for School Construction

EDITOR'S NOTE.-The article that follows is a summary of the first of several studies undertaken by the Bureau of Labor Statistics to determine the labor requirements for various types of public construction. The full report will be published as BLS Bull. 1299. Additional studies currently underway deal with highways, hospitals, public buildings, and public housing. The results of the highway study will be summarized in the next issue of the Review.

THE CONSTRUCTION of public schools in 1959 required 212 man-hours of employment per $1,000 of construction contract, a survey by the Bureau of Labor Statistics indicates. These man-hour requirements consisted of 84 hours for on-site activity in the construction industry and 128 hours for various off-site activities required to produce and deliver the materials used by construction contractors plus their own off-site employment requirements. (See chart on p. ii of this issue.) Since the average cost of the schools surveyed was about $730,000, the construction of each one created the equivalent of a year's employment for 81 workers-38 in construction and 43 in other activities.1

Thus, with about 70,000 classrooms being built each year at a cost of some $2.5 billion (about onesixth of total public construction), it was estimated by the Bureau that such activity has been the source of over 115,000 jobs annually for on-site work in the construction industry and 160,000 in the off-site activities mentioned. Substantially higher employment will be required if the need for an estimated 427,000 new classrooms within the next 5 years is to be met.2

These estimates exclude work on school projects by government and public utility employees and those engaged in planning and designing, as well as work on site preparation and landscaping when

not included in the principal construction contractThey also exclude the multiplier effect of employ. ment created by the spending of the wages of the workers and the profits of their employers. On the projects studied, on-site wages represented 26 percent of the total contract and the cost of materials was nearly 56 percent. The residual 18 percent, while not studied, would have covered such items as overhead and administrative costs, taxes, supplementary wage benefits, and profits.

Scope and Method of Survey

Data for this survey were collected from contractors and contracting authorities on 128 school construction projects, about equally divided between those built with Federal assistance and those financed solely by State and local agencies. Only new school buildings were studied; additions to existing buildings were excluded.

The projects were chosen to represent all public schools built under contracts awarded between July 1958 and June 1959. The sample was stratified by a number of school characteristics which were potential sources of variation in man-hour requirements: type of school (elementary or secondary), amount of contract, and regional location of the project, with adjustments to insure adequate representation of both metropolitan and nonmetropolitan areas and of the full range of costs per square foot. This sample design also yielded representative data by type of framing, exterior finish, number of stories, and other variables which were presumed to affect man-hour requirements.

The study was designed primarily to determine the man-hours implicit in each $1,000 of new school building construction. It covered both onand off-site employment associated with activities resulting from construction contracts, with the exclusions noted.

1 The annual employment estimate for construction workers is based on total man-hour requirements divided by 50 times the 1959 average workweek in contract building construction work of 35.8 hours. For other types of employment, 2,000 hours were considered a year's employment.

2 Ten-Year Aims in Education, 1959–1969 (U.S. Office of Education, January 1961).

On-site man-hours were obtained, in most cases, from contractors' payrolls on an occupational basis. The on-site man-hour averages presented in the accompanying tables were obtained by combining the man-hours reported for individual projects with weights proportionate to the contract amount. The contractors were also queried on their use of mechanical equipment and the quantities and costs of materials used.

Off-site man-hours were estimated from reported purchases of materials, supplies, and equipment used at the construction site. Primary manhours, or those of the employees in the manufacturing, transportation, and trade industries who were directly engaged in the production and distribution of these materials, were determined by using the information on purchases in conjunction with data on production and man-hours of employment from the 1958 Census of Manufactures. Indirect or secondary man-hours, or those of workers employed in activities generated by the transactions incident to the production and distribution of the materials, were determined by using data from the Bureau's study of 1947 interindustry relationships.3 Adjustments were made for changes in prices and man-hour ratios on the basis of the BLS data on wholesale prices and output per man-hour. As indicated earlier, the resulting estimates do not include the employment generated by the spending of wages or profits by either the primary or secondary workers and employers.

Schools constructed in metropolitan areasabout 55 percent of those studied-were generally larger and more expensive than those built in nonmetropolitan areas. (See table 1.) Costs also varied by geographical location both in total cost and per square foot of space, particularly between the Northeast and the South. Part of the regional variation was due to regional differences in the characteristics of the schools.

In spite of certain differences in the type of construction, the great majority of the schools surveyed had many similar features. About 80 percent were one-story structures, without basements. Exterior walls were generally of masonry and the floors were poured concrete covered with some type of soft tile, usually asphalt. Most of the schools (83 percent) had cafeterias. While auditoriums and gymnasiums were generally found only in secondary schools, the others generally had a multipurpose room that also served for assemblies.

On-Site Labor Requirements

While over half of the projects required between 75 and 95 man-hours of on-site labor per $1,000 of contract, requirements on the others varied substantially, reflecting special project circumstances such as the applicable building code, the choice of material, and the building design. Apart from the variation among individual projects, there were marked differences in labor requirements for schools with various characteristics. Thus, average on-site man-hour requirements for secondary schools at 82.6 hours-were about 4 less than for elementary schools (table 2). Requirements for metropolitan locations were similarly lower than for nonmetropolitan areas. Schools constructed in the northeastern part of the country averaged 76 man-hours per $1,000, compared with 99 in the South. Large projects typically required fewer on-site man-hours per $1,000 than small ones.

The primary cause of these variations, however, appeared to be not the differences in characteristics themselves but those in construction methods. While the data on the extent to which mechanical equipment was used, the amount of prefabricated components used, and the relative employment of

See W. Duane Evans and Marvin Hoffenberg, "The Interindustry Relations Study for 1947," Review of Economics and Statistics, May 1952. For specific methods employed, see appendix A of BLS Bull. 1299.

skilled and unskilled labor are inadequate for precise measurement, they do support some inferences about their effect on labor requirements.

1. Use of Mechanical Equipment. The large projects and those built in metropolitan areas frequently made greater use of mechanical equipment, such as cranes, elevators, and conveyors. In general, these projects tended to show lower onsite unit man-hour requirements. Small projects or those some distance from available equipment were not usually able to take advantage of these devices. The degree of mechanization varied also by region.

2. Amount of Prefabricated Components Used. The extent to which prefabricated components are used is one of the more important determinants of on-site man-hour requirements. Although their use reduces on-site labor, much of the saving in man-bour requirements may represent employment transferred from the site to more remote shops and factories.

Prefabrication was used at various stages in con struction on the projects surveyed. Some use prefabricated concrete units in the construction o the frame, others used prefabricated wall panels and many used prefabricated units in the interio finish such as window and door frames, tack and chalk boards, and wardrobes.

3. Employment of Skilled and Unskilled Labor Low man-hour requirements on a project were fre quently associated with greater than average us of skilled craftsmen. Projects employing lowe proportions of laborers and helpers usually had lower on-site man-hour requirements per $1,000 of contract.

Regional Comparisons. Most projects in the Northeast required less than the national average of 84 man-hours of on-site labor. The relatively low average in that region, 76 man-hours per $1,000 is believed to reflect the greater use of equipment, prefabrication, and a higher proportion of skilled

of

Class

Number of

Number

Number

Number

foot (thou

Total, all schools...

Class- of

Class- of

Class

foot (thou

sands)

*See footnote 1, table 1.

craftsmen. The same factors were also noted for projects constructed in the West, where 70 percent of the schools required less than 84 on-site manhours per $1,000 of contract and the average was 80.6. Characteristics of the schools in that region were noticeably different from those in other regions, as table 1 indicates.

> Over 80 percent of the school projects in the South fell in the upper range of man-hour requirements and, in general, reflected more frequent use of laborers and helpers. The average project in the South required 99 man-hours per $1,000 of

1 Insufficient data to warrant presentation.

About the same differential in unit man-hour requirements was noted between the national averages for schools constructed in nonmetropolitan areas and those in metropolitan areas. Projects in the metropolitan areas might have required fewer on-site hours because they were of sufficient. size to warrant the use of certain heavy equipment not usually available outside large cities, as well as prefabricated components, and they could draw upon a more highly skilled labor force. The differences in the labor requirements between these areas were most noticeable in the South and North Central regions; in the West, projects in nonmetropolitan areas required fewer man-hours than those in metropolitan areas.

In general, the smaller projects required more on-site man-hours per $1,000 of contract but the differences in averages were usually small. The variation in man-hour requirements of individual projects, especially within regions, would indicate that, in the absence of procedures feasible with large projects to reduce manpower requirements, such factors as choice of materials and design are more important in determining on-site man-hour requirements per dollar of contract.