Why Were Poverty Rates So High in the 1980s?

This paper explores the unexpectedly slow decline in poverty that occurred over the expansion of the 1980s. We present evidence on the "stickiness" in the poverty rate in the past decade, compared to earlier decades. The following section investigates several potential non-earnings-related explanations for this fact. There is little evidence that the slowdown in the response of poverty to economic growth is due to problems with the measurement of poverty, to changes in transfer policy in the early 1980s, to the regional distribution of the poor during the 1980s expansion, or to changes in family composition among the poor.

The final section of the paper investigated the decreased responsiveness of income and earnings to the rnacroeconomy among low-income households in the 1980s. A growing body of literature has recently began to explore the widening in wage differentials among less-skilled and more skilled workers over the 1980s. That literature indicates that substantial real wage declines occurred among low-wage workers throughout the expansion of the 1980s, while substantial real wage increases occurred among higher-wage workers. These trends are clearly correlated with the trends in poverty. Declining real wages will make it harder for low-income families to escape poverty. The point of this paper is not to describe that wage decline further, but to investigate how important this decline was relative to other factors that were operating at the bottom of the income distribution.

The lower responsiveness of poverty to economic growth is not due changes in labor market responsiveness over the 1980s expansion. In fact, labor market involvement was more responsive during the 1980s: the unemployment rate fell more rapidly, and earners in the bottom quintile of the population increased their work effort more sharply in the 1980s than in the 1960s. The lower responsiveness of income among low-income households to the economic expansion of the 1980s is entirely due to declining real wages, which offset the increase in labor market effort, resulting in slower income growth.

The implication of these results is that the changing wage structure of the l980s made economic growth a far less effective it was in the expansion of the 1960s. It is still an open question whether these trends will continue into the l990s. If they do, economic growth cannot be expected to produce substantial declines in the poverty rate.

Working Paper No. 57(2.53 MB)