Printer Friendly

How do demographic changes affect labor force participation of women?

How do demographic changes affect labor force participation of women?

Since World War II, U.S. labor force participation rates among women have almost doubled, reaching about 55 percent in 1985.(1) Increases in labor force activity have been pervasive for all groups, especially married women and women with young children.

Changes in the demographic composition of the female population, particularly during the past decade or so, have had great potential for altering overall participation rates.2 For example, William Johnson and Jonathan Skinner have reported that the rise in divorce rates between 1960 and 1980 may explain up to 17 percent of the rise in labor force participation rates of women during that period.3 Similarly, Ralph Smith has concluded that between 1971 and 1975, the changing demographic composition (for example, marital and family status changes) of women in the labor force accounted for 28 percent of the increase in their rates.4 Compositional changes are likely to be small over a short time period, however, and therefore should not be expected to greatly affect overall female labor force participation rates.

By examining data covering the 15-year period between 1970 and 1985, we provide evidence on the link between changes in demographic composition and labor force participation rates among women.

Specifically, we ask: To what extent have changes in fertility rates, marital status, educational levels, and age structure accounted for growth in labor force participation rates of women since 1970?

Demographic composition

Fertility. The labor force participation rates among married women with children, particularly young children, have been steadily increasing since 1970. In 1985, nearly half of all women with children under age 18 were in the labor force, compared with less than 40 percent in 1970.(5) Moreover, the declines in fertility rates, as well as declines in family size, increasing childlessness, and delayed childbearing have freed many women to pursue employment opportunities outside the home. Completed family size, for example, decreased from 2.4 children in 1970 to 1.7 in 1984 among white women, and from 3.1 to 2.2 children among blacks.6 Recent fertility declines are thus a potentially important demographic source of post-1970 increases in overall female labor force participation rates.

Marital status. Substantial variation exists by marital status, with married women exhibiting labor force participation rates much lower than those of the overall female population.7 Changes since 1970 in the marital status composition of the female population have provided a potentially significant demographic source of growth in female labor force participation. The incidence of divorce, for example, increased from about 14 per 1,000 married women in 1970 to nearly 22 per 1,000 in 1984.(8) In addition, the proportion of never-married women has risen rapidly, especially among young adults, reflecting delayed marriage. For example, the median age at first marriage among women in the United States rose from 20.6 in 1970 to 22.8 in 1984.(9)

Education. The educational upgrading of the female population has been a major facet of social change in the United States. For women age 25 or over, median years of schooling increased from 12.1 to 12.6 years between 1970 and 1980, and the percent graduating from high school grew from 52.8 to 65.8.(10) Changes in the educational composition of the female population must be included in any demographic or structural explanation of rising participation rates among the female population. Indeed, increasing educational attainment alters the relative importance of home work versus the labor market for many women. This is clearly revealed in female labor force participation rates that tend to accelerate with increasing educational attainment.

Age. Age composition is a major structural aspect of the labor force.11 Market-related activities are clearly associated with age. The age profile of women in the labor force is curvilinear, reaching its nadir during the child-bearing years and after age 40 or so, when labor force exits begin to rise. One significant facet of labor force age structure can be linked directly to the post-World War II baby boom. That is, the baby-boom cohort of the 1950's entered the labor force in large numbers during the 1970's. As this cohort aged between 1970 and 1985, declining proportions of women were concentrated in the age categories that typically exhibit lower than average rates of participation (say, those in mid-to late 40's). The "maturing' of the baby-boom cohort thus represents another potentially significant demographic component of change for women in the labor force.

Accounting for change

We restrict this analysis to women ages 25-49.(12) For most women, schooling has been completed by age 25, and labor force exit rates begin to accelerate significantly after 45 or so.

The extent to which changing demographic composition accounts for the increases in labor force participation rates among women can be evaluated using standard demographic methods of decomposition or components analysis.13 It is well known that the difference between two crude rates is attributed to differences in both status-specific rates and population composition. Differences in rates between 1970 and 1985 can thus be decomposed into parts attributed to changing propensity to participate (that is, a so-called true or rate effect) and parts attributed to changes in the distribution of women by number of children, marital status, education, and age (composition effects). The categories of population composition we consider here are provided in table 1 for blacks and nonblacks.

The results of the decomposition analysis are presented in table 2. Total labor force participation rates of women increased from 47.90 percent to 71.01 percent between March 1970 and 1985. Of the 23.11-percentage-point increase in labor force participation rates, 12.48, or about 54 percent, is attributable to the changing propensity to participate. (See the "rate effect.') Simply put, a majority share of the increase over this 15-year period is attributed to changes in behavior rather than changes in demographic composition. This further implies that labor force participation rates would have increased during 1970-85, even if the demographic composition of the female population had not changed during this period. The increase in labor force participation rates for women cannot be explained away with compositional arguments.

This conclusion, however, should not be interpreted to mean that changing demographic composition or changes in the supply of women are unimportant facets of change in labor force participation rates. Indeed, 46 percent of the increase since 1970 is directly attributable to changing demographic composition. (See "composition effect,' table 2.) Although past studies reveal that compositional effects are not dramatic over a short time, the effects of changing demographic composition are considerably more apparent over a longer period, such as that examined here. Moreover, when we examine the relative importance of each compositional component, data reveal that, on the one hand, changing fertility rates, as measured by number of children, account for 4.33 percentage points (or nearly 20 percent) of the overall post-1970 increase in labor force participation.14 Marital status and education changes, on the other hand, account for smaller but roughly similar shares (about 13 percent) of the increase. Changing age composition has virtually no effect on labor force participation rates of women. As these results suggest, while not solely responsible for recent increases in labor force activity among women, changing composition nevertheless is clearly an important and too frequently ignored source of growth in labor force participation rates.

As shown in table 2, limiting the analysis to the total (or nonblack) female population also tends to mask substantial racial variations in the mix of compositional and rate effects. In contrast to nonblack women, our analysis reveals that changing composition is primarily responsible for the increase in labor force participation rates for black women, accounting for 10.93, or nearly 85 percent, of the 13.00 percentage point increase since 1970. This sizable change is mainly attributable to educational upgrading among black women. Indeed, increased education accounts for about two thirds (or 7.25/10.93) of the overall compositional effect and about 55 percent of the overall increase in labor force participation rates for black women during the 1970-85 period. The only other compositional component of any significance is the changing number of children, a demographic component that accounts for about 30 percent of the increase since 1970.


The period since 1970 has revealed a continuing pattern of increase in rates of female labor force participation. Rising wage rates and changing attitudes regarding work have clearly contributed to this increase.15 Our results nevertheless suggest that demographic explanations cannot be entirely dismissed. A substantial share--almost half--of the increase has roots in ongoing patterns of demographic change, especially recent fertility declines, shifts in patterns of marriage and divorce, and educational upgrading. The changing mix of women across various population subgroups thus provides an important demographic explanation of changing female labor force participation rates, particularly for black women.

The results also imply that prospects are good for continuing high labor force participation rates for women. Demographic changes are likely to counterbalance any dampening effects of slowing wage increases or changes in family or work attitudes. Indeed, the changing demographic supply of potential female workers may account for an increasing share of future growth in labor force participation among women.16

1 See William G. Bowen and T. Aldrich Finegan, The Economics of Labor Force Participation (Princeton, Princeton University Press, 1969); Glen G. Cain, Married Women in the Labor Force (Chicago, University of Chicago Press, 1966); and Employment and Earnings (Bureau of Labor Statistics, August 1986).

2 Elizabeth Waldman, "Labor force statistics from a family perspective,' Monthly Labor Review, December 1983, pp. 16-20.

3 William R. Johnson and Jonathan Skinner, "Labor Supply and Marital Separation,' American Economic Review, June 1986, pp. 455-469.

4 See Ralph E. Smith, "Sources of growth in the female labor force, 1971-75,' Monthly Labor Review, August 1977, pp. 27-29.

5 Howard Hayghe, "Rise in mothers' labor force participation includes those with young children,' Monthly Labor Review, February 1986, pp. 43-45.

6 National Center for Health Statistics, Monthly Vital Statistics Report, Advance Report of Final Natality Statistics, 1984 (Hyattsville, MD, Public Health Service, July 18, 1986).

7 Howard Hayghe, "Working mothers reach record number in 1984,' Monthly Labor Review, Dec. 1984, pp. 31-34.

8 National Center for Health Statistics, Monthly Vital Statistics Report, Advance Report of Final Divorce Statistics, 1984 (Hyattsville MD, Public Health Service, Sept. 25, 1986).

9 National Center for Health Statistics, Monthly Vital Statistics Report, Advance Report of Final Marriage Statistics, 1984 (Hyattsville, MD, Public Health Service, June 3, 1987).

10 U.S. Bureau of Census, General Social and Economic Characteristics, United States Summary PC 80-1-C1 (U.S. Government Printing Office, 1983.)

11 James P. Smith and Finis Welch, "No Time to be Young: The Economic Prospects for Large Cohorts in the United States,' Population and Development Review, March 1981, pp. 71-83.

12 Data for this analysis are from the March 1970 and 1985 machinereadable files of the Current Population Survey, Bureau of Census.

13 See Prithwis Das Gupta, "A General Method of Decomposing a Difference Between Two Rates Into Several Components,' Demography, February 1978, pp. 99-112. Methods of decomposition have a long history in demographic research. Any comparison between two crude rates is affected by differences in population composition (for example, age composition). To eliminate compositional differences, standardized rates are often calculated, which eliminate the confounding effects of differences by assigning a similar composition (that is, a "standard' age composition) to each population. Methods of decomposition represent a simple extension of this analytic technique by enabling us to gauge the relative effects of more than one compositional component on crude rate differences.

The general method described by Das Gupta has three primary advantages over other methods of decomposition: (1) the method can be applied to data cross-classified by any number of compositional factors (for example, in the analysis presented here, we use a four-factor model); (2) results are independent of the order in which compositional factors are considered; and (3) the procedure avoids problems with the allocation and interpretation of "interaction' effects among the compositional factors. With regard to the latter point, this is accomplished by calculating the effect of one compositional factor, holding other factors constant at an average level. As a result, a "total' effect (that is, the difference in crude rates) can be uniquely partitioned into a "rate' effect (the difference between two standardized labor force participation rates, using as the "standard population' the weighted average of the 1970 and 1985 female labor force populations, aged 25-49), and "compositional' effects (in this case, one each for changing fertility, marital status, education, and age).

14 In addition to our examination of the effects of changing numbers of children, we also evaluated the effects of changes in the age composition of children. Because labor force participation rates are lowest among mothers with young children, we replicated our decomposition analysis with women separated into three categories: 0 children less than age 18; some or all less than age 6; and all children age 6-18. This analysis produced results that were similar to those reported in table 2. Changes in the age composition of children accounted for about 14 percent of the overall increase in rates for women.

15 Given the results reported here, we are unable to partition sources of the "rate' effect, but surely rising real wages and changing attitudes account for a sizeable share of this effect. See David Shapiro and Lois B. Shaw, "Growth in the Labor Force Attachment of Married Women: Accounting for Change in the 1970s,' Southern Economic Journal, October 1983, pp. 461-473.

16 See George Masnick and Mary Jo Bane, The Nation's Families: 1960-1990 (Cambridge, The Joint Center for Urban Studies of MIT and Harvard University, 1980). They project labor force participation rates of women to the year 1990.

Table: 1. Percent distributions of women ages 25-49, by race and selected characteristics, 1970 and 1985

Table: 2. Components of change in labor force participation rates for women, by race, 1970-85
COPYRIGHT 1987 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1987 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lichter, Daniel T.; Costanzo, Janice A.
Publication:Monthly Labor Review
Date:Nov 1, 1987
Previous Article:White collar salaries vary widely in the service industries.
Next Article:Furniture workers' wages higher under incentive systems.

Related Articles
Labor force projections: 1986 to 2000.
Women's work plans: contrasting expectations and actual work experience.
Employment characteristics of older women, 1987.
Women and the labor market: the link grows stronger.
Effects of intermittent labor force attachment on women's earnings.
The 2005 labor force: growing, but slowly.
Developments in women's labor force participation.
Marriage, children, and women's employment: what do we know?
The February Review.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters