obviously unsuited for women. Among them were such cities as Pittsburgh, Youngstown, Steubenville, Weirton, and Wheeling. Job opportunities for women, as for men, were also limited by the poor economic conditions and the small area population growth (or loss). As mentioned earlier, the unemployment rate in half of these areas was in excess of 6 percent. Earnings of men in the areas with low participation rates for married women were greater than in the other areas because of higher paying industries. At the same time, the annual earnings of women appear to be somewhat smaller in the low participation areas. (These earnings, of course, reflect amount of time worked as well as wage scales.)

Relatively few married women were in the work force in such places as Detroit (Mich.),

Earnings are for 1959, a year which included a severe strike in steel.

Gary-Hammond (Ind.), and Gadsden (Ala.), even though many were nonwhite. This is another example of the limitations of the white-nonwhite factor in explaining rates of labor force participation.

The need for separating the effects of the individual factors mentioned in this section, as well as assessing the importance of other factors, is evident. Work of this type is planned in a more extensive study of the 1960 Census data which will cover all of the Standard Metropolitan Statistical Areas.

In order to see to what extent areas with high or low rates of labor force participation for married women remain in the same category over time, the 15 (out of 151) Standard Metropolitan Areas with the highest rates in 1950 were compared with those in the highest decile (21 out of 212 SMSA's) in 1960 and a similar comparison

was made for areas in the lowest deciles. (See table 11.) For the high rate areas, only 2 of the original 15 areas were no longer in the highest decile 10 years later. Their rates increased,

although already high in 1950, by about the same amounts as did those in the lowest decile. Both periods included a disproportionate number of areas in Massachusetts and North Carolina,

where manufacturing of textiles and other nondurable goods or service industries were important. Of the 15 areas with the lowest rates in 1950, all but 4 were still in the lowest decile in 1960.

Among the areas which areas which showed the largest increases in rates were Buffalo and Boston, where employment in service industries rose substantially, and several areas in Pennsylvania (e.g., Altoona,

39.6

25.6

40.9

36.7

29.3

57.8

21.8

22.1

44.7

29.7

26.8

60. 1

45.7

20.2

48. 4

Dallas, Tex.

24.5

24. 4

42.6

73.5

service industries, which offer better work possibilities to women. Despite these employment changes, most of the 15 areas continued to have large concentrations of mining and heavy goods manufacturing-and the relatively high employment associated with these industries.

The fact that fewer married women worked in these areas of high unemployment, despite the tendency for wives of unemployed men in general to have high participation rates, underscores the extent to which the types of industries and the economic conditions of an area determine suitable job opportunities and worker rates for married

women.

It is unfortunate that the areas where men are most likely to be unemployed are also the areas where their wives have the least opportunity of finding suitable jobs which would provide needed family income. The small proportion of married women working in these areas of high unemployment should also be noted by those who feel that married women are taking away men's jobs an who attribute much of the unemployment among men to the employment of married women. The areas where unemployment is highest are the very areas where women are least able to replace men in jobs because of the nature of the work. Few, if any, unemployed steelworkers and coal miners, for example, could claim that married women took their jobs and probably very few would be interested in the kinds of jobs held by married women. Moreover, in places where married women are most likely to be employed, job competition between them and men is obviously reduced by the availability of jobs which are of interest predominantly to women and by the overall high levels of economic activity in these

areas.

Data for the country as a whole from both decennial censuses and monthly surveys of the labor force show that women whose husbands are unemployed are more likely to be in the labor force than wives of employed men. This does not contradict the finding that women are less likely to be in the labor force in areas of high unemployment, since data on wives of all unemployed men are for the entire country, including those areas where unemployment is relatively low and wives can more easily find work. Furthermore, within individual areas of high unemployment and areas of low unemployment, the wives of unemployed men probably have the highest rates of labor force participation.

Self-Employment in the United States, 1948-62

JOHN E. BREGGER*

IN THE POST-WORLD WAR II PERIOD, self-employment has diminished in significance as a source of livelihood, largely as a result of the sharp decline in agricultural employment and the gradual trend toward the corporate form of enterprise and away from the small owner-operated farm or business. Between 1940 and 1962, the number of selfemployed dropped from 22 to 13 percent of total employment. As a proportion of nonagricultural employment, self-employment has leveled off at about 10 percent since 1952. In the nonfarm occupations of the self-employed, there has been a pronounced rise in the number of professionals and sales workers and a corresponding decline in the number of proprietors.2

The self-employed are a very heterogeneous group, ranging from the highly skilled and highly paid professional worker to the marginal worker with very low earnings. In agricultural selfemployment, farmers (including sharecroppers) predominate. In nonagricultural employment, the self-employed range from high income groups such as doctors and lawyers operating their own practices to small businessmen selling merchandise or offering specialized services and to more marginal forms of self-employment such as street-corner peddlers or women who take in laundry or sewing.

This article describes employment trends of the self-employed between 1948 and 1962 with particular emphasis on the characteristics of selfemployed workers compared with those of wage and salary workers. Trends and characteristics of unpaid family workers, a group closely associated with the self-employed, are also examined.

Definition of Terms

A self-employed person, according to the monthly labor force survey, operates his own unincorporated farm or business and derives his income as profit from his enterprise or fees for services rendered. For the purpose of this definition, a business exists when one or more of the following are present: (1) investment in

670039-63- 4

machinery or equipment of substantial value which is used by the owner or his employees, (2) operation of an office, store, or other place of business, and (3) some type of advertisement of the business or profession.

Examples of the self-employed include farmers, shopkeepers, doctors and lawyers operating their own practices, and free-lance writers. In an unincorporated business owned and operated by more than one person, each partner would qualify as being self-employed. However, the sole owner of a corporation would be classified as a wage and salary worker because he is employed by the corporation and not by himself.

An unpaid family worker is one who works at least 15 hours a week without direct monetary compensation in the farm or business owned and operated by a member of his immediate household who is related by blood or marriage. An unpaid family worker who works less than 15 hours during the survey week is classified as not in the labor force.

Certain work situations may be misclassified in the enumeration of class-of-worker categories. Most difficulties arise in distinguishing between self-employed and wage and salary workers in

"Of the Division of Manpower and Employment Statistics, Bureau of Labor Statistics.

1 The statistics in this article are derived from the monthly labor force survey, conducted and tabulated each month for the Bureau of Labor Statistics by the Bureau of the Census and reported in the Monthly Report on the Labor Force and Employment and Earnings, both Bureau of Labor Statistics publications. These statistics are based on sample surveys and are therefore subject to sampling variability, which may be relatively large in the case of smaller estimates or small differences between estimates. Annual and quarterly averages, which are more reliable than figures for a given month, are used throughout most of the article. An exception was made in the case of hours worked where data for the month of May were used because of their relative freedom from the effects of holidays and unusual weather conditions.

Although there are other sources of self-employment data, the monthly labor force survey is the only current series which specifically reflects the activities of individual workers. One series, "Persons Engaged in Production, by Legal Form of Organization," published by the Department of Commerce in the Survey of Current Business (November 1959 and May 1962 issues), is based on types of businesses with estimates derived from a variety of sources, including the U.S. Business Tax Returns of the Internal Revenue Service, the Census of Business taken by the Census Bureau, and the monthly labor force survey. In spite of major conceptual differences, the trend of the Department of Commerce series on active proprietors has roughly paralleled that of the nonfarm self-employed as measured in the monthly labor force survey during the past decade. However, because the Department of Commerce estimates are on an annual basis, they are more comparable in level to self-employment data from surveys of work experience of the population on a calendar-year basis. The Internal Revenue Service data are not comparable with the data presented in this report, primarily because the statistics relate to number of businesses rather than workers. Statistics on sole proprietorships and partnerships in the quadrennial Census of Business include only wholesale and retail trade and the various service industries.

2 The term "proprietor" is used in reference to persons who primarily manage their own business.