Downward mobility: is it a growing problem?

American Journal of Economics and Sociology, The,  Jan, 1994  by Patricia K. Smith

• E-mail
• Print
• Link

I

Introduction

FROM THE END OF WORLD WAR II to the early 1970s, Americans enjoyed steady increases in average income and living standards. Upward income mobility seemed virtually automatic. In 1973, however, average real wages began stagnating and during the 1980s the growth in the inequality of family income accelerated (Levy and Murnane, 1992; Karoly, 1993). Academics and the popular press expressed increasing concern that the middle class was disappearing (Thurow, 1987; Duncan, Rodgers, Smeeding, 1991 and 1992) and that "The American Dream" was vanishing (Dentzer, 1991; Samuelson, 1992). The expectation of upward mobility seemed to be dimming, and the fear of downward mobility growing (Vobejda, 1991; Koretz, 1992; Brownstein, 1992).

Studies of the U.S. income distribution typically find considerable mobility, both upward and downward (Duncan et al., 1984; Duncan, 1988). There are always some downwardly mobile persons; the question is whether the number is increasing. If so, there is cause for deep concern. On a personal level, downward mobility inflicts both material and psychological hardship (Newman, 1988; Ehrenreich, 1989). Large drops in income can stress individuals' mental health and the stability of families.(1) A myriad of social problems--depression, divorce, interruption of schooling, and violence--can ensue.

Increasing downward mobility can also cause serious economic problems. The downwardly mobile must reduce consumption, investment, and savings. Such cuts by a substantial segment of the population could dampen future economic growth. Even those not downwardly mobile could feel more at risk, possibly depressing consumer confidence. An increase in the number of downwardly mobile could also contribute to income inequality. For example, if more persons experience very large drops in income while fewer maintain their income level or enjoy income gains, income inequality increases. Lastly, if many of the downwardly mobile become poor, the demand on already strained public assistance resources increases.

The social and economic consequences of increasing downward mobility also have political ramifications. Phillips (1993) argues that the middle class' recent economic losses contributed to the Republicans' defeat and Perot's strong third party showing in the 1992 Presidential election. More importantly, history records numerous occasions when groups disproportionately suffering large economic losses instigate violent, political upheavals.

To determine whether the problem of downward mobility is growing, this paper uses data from the Panel Study of Income Dynamics (PSID) to measure and compare the extent of downward mobility over two periods of modest economic growth: 1976-1978 and 1984-1986. Two types of downward mobility are examined. Absolute downward mobility is measured as a decline of fifty percent or more in the real family income-to-needs ratio. Relative downward mobility is measured by drops to lower quintiles in the distribution of real family income. The characteristics of the downwardly mobile, including their initial class status, are also determined.

The results show that relative downward mobility occurred somewhat less frequently during the 1984-1986 period than in the 1976-1978 period. However, absolute downward mobility occurred more frequently in the 1984-1986 period. The characteristics of the downwardly mobile do not differ much between the two periods, however, the risk for some demographic groups did change. For example, the incidence of downward mobility increased for persons whose household head was neither Caucasian nor African-American. Finally, the analysis finds weak evidence that a larger proportion of the downwardly mobile came from the middle and lower classes in the 1984-1986 period than in the 1976-1978 period. II

Downward Mobility and Changes in the U.S. Income Distribution

DUNCAN ET AL (1984) USe the PSID to examine family income mobility over the period 1971-1978. They find substantial mobility up and down the quintile distribution of family income. Of particular interest, nearly one-third of the sample individuals dropped at least one quintile and eleven percent dropped two or more. Duncan (1988) finds that approximately one-third of Americans experienced a drop of fifty percent or more in their real income-to-needs ratio at least once from 1974 to 1983.

Duncan (1988) relates downward mobility to "life events" such as unemployment and divorce. Burkhauser and Duncan (1989) study life-cycle patterns of downward mobility from 1974 to 1983. They find that among women the risk is highest in the age 46 to 55 cohort, while among men the risk is highest in the 66 and older cohort. The risk for women is either the same or higher than the risk for men in all cohorts. Their results suggest that labor market events impact men and women's risk of downward mobility similarly, but family composition changes have a greater impact on women. The present analysis will expand the set of demographic variables to include region, race, and education in addition to age and gender.