Earners and Their Dependents in the Population in April 1948

male labor-force participation rate after age 20 reflects in part the progressive increase up to about age 35 in the proportion of married women in the female population. The rate for single women remains relatively high at all ages. The extent to which women stay in the labor market is affected also by home responsibilities. In April 1948, married women with no children under age 18 were almost twice as frequently in the labor force as married women with one or more children under 18. Among widowed and divorced women, labor-force membership was more common, for reasons related to age differences, for women with children; for example, widowed and divorced women without children under 18 were older on the average and less able to And or hold jobs. The influence of sex, age, marital status, and home responsibilities on labor-force membership is illustrated in table 4. S OCIAL insurance programs to an increasing extent have been making provision for the payment of benefits to dependents or survivors. Estimates of beneficiaries and costs under such programs, present and proposed, necessarily involve assumptions concerning the relation of the number of dependents to the number of earners. Basic sources of information on this relationship are the family data of the Bureau of the Census. Considerable interest attaches therefore to the averages and ratios to be derived from the Census Bureau’s April 1948 family-composition survey. The purpose of the present article is to present the results of this survey as they bear upon the workerdependent ratio and to indicate the changes that took place between 1940 and 1948 in underlying family relationships. It should be made clear at the outset that the estimates of earners, dependents, and insured persons offered here have a general rather than a specific application. They provide a basis for an estimate of the numbers who would have certain kinds of social insurance protection under certain conditions; they do not constitute estimates directly applicable to any particular program now in effect or proposed. For broad social insurance analytical purposes, the civilian population of 145 million in April 1948 may be divided into three groups: 59 million earners, defined as persons in the labor force that month (excluding unpaid family workers) ; 64 million primary dependents of earners, defined as the nonearner wives and chil-