Whither the Middle Class'? A Dynamic View

Research using cross-sectional survey 'snapshots' of household income taken over the past quarter century reveals a growing inequality in the distribution of annual money income of households in the United States (Thurow, 1987; Levy, 1987; Levy and Michel, 1991; Michel, 1991; Karoly, 1990; Center on Budget and Policy Priorities, 1990; Easterlin, MacDonald and Macunovich, 1990), prompting some to argue that the U.S. middle class is disappearing (Phillips, 1990; Bradbury, 1986). Aggregate data from the National Accounts and from wealth surveys (Wolff, 1989; Eargle, 1991) reinforce this conclusion by showing a growing share of income from capital, a falling share for earnings, and a slightly increasing concentration of wealth among upper-income groups. Also well-documented is greater inequality in the size distribution of earnings and wages in the late 1980s as compared to one or two decades before (GottschaLk and Danziger, 1989; Burtless, 1989; Blackbum et al., this volume). Despite the consistency of these results, their almost universal reliance on data drawn from cross-sectional snapshots leaves unanswered many important questions regarding the nature of the changes taking place in the distribution of income and wealth. Most importantly, cross-sectional snapshots provide information only on net changes in economic position and thus reveal little about the extent and nature of movement into and out of the middle class.. Are increasing numbers of families 'falling from grace', as Katherine Newman (1988) puts it? If so, who are they and what events are linked to their income losses? Or is mobility into the middle class declining? And, if so, does this affect in particular young families? What avenues for upward mobility are disappearing? These are the types of questions we seek to address for adults crossing either the lower or the upper boundary of the middle class. A second set of issues we address involves linkages between changes in income and changes in wealth. We analyze trends in the transitions of prime age (25-54 years old) adults into and out of the middle class using 22 years of data from the Panel Study of Income Dynamics. We begin by reviewing the methodology and measurement procedures that we employ to define the middle class and transitions into and out of middle-class status. Next we present our basic findings which, in fact, show a persistent 'withering' of the middle class since about 1980. We then search for clues as to who moved into and out of the middle-income groups and the source of such changes. Because notions of 'class' are usually based on measures of wealth as well as income, we also investigate longitudinal changes in the wealth distribution in the 1980s for these same individuals. Our findings on wealth reinforce those based on income. The paper concludes with a brief discussion of the policy implications of our findings.