Life-Course Income Curves in the Era of Middle Class Retrenchment, 1970-2011: Has Aggregate Income Stagnation Coincided with Individual Income Stagnation?

Friday, 3 July 2015: 8:30 AM-10:00 AM
TW1.3.02 (Tower One)
Edo Navot, Columbia University, New York City, NY; University of Haifa, Haifa, Israel; University of Haifa, Haifa, Israel
Since at least 1980, the United States has experienced both increasing top-heavy inequality and stagnation of middle class incomes.  And concern over rising inequality and middle class income stagnation has recently penetrated the public discourse.  Within the progressive and liberal political camps in the U.S., there have been at least two reactions to the dual problem of top-heavy inequality and middle class income stagnation:  First, established Democratic Party political groups have presumed that voters have, and will continue to, vent their dissatisfaction with inequality and middle class economic struggles at the polls.  Most recently and prominently, the Center for American Progress released a report raising alarms about these issues while urging enactment of policies to attenuate both languishing middle class income and rising inequality.  In journalistic accounts, this report was described as an attempt at early articulation of a socio-economic platform for a presumed presidential campaign for Hilary Clinton.[1]

Another response has questioned why, when the U.S. has returned to Gilded Age-levels of inequality while the middle class struggles, there have not been social upheavals comparable to those of the Progressive Era or the 1960s.[2]  By comparison to those periods of mass politics, the Occupy Wall Street movement was relatively small and has achieved little policy change.   

Both of these responses share the presumption that rising inequality has, will, or should have major political ramifications.  Indeed this presumption partly motivates the theme of this mini-conference.  This article focuses on middle class income stagnation in the U.S. but takes a step back from the shared outlook that this phenomenon generates political ramifications.  That is, I raise the possibility that middle class income stagnation as usually discussed by academics and journalists may not be a salient feature of the middle class experience in the U.S.  To be sure, concentration of income at the top of the U.S. earnings distribution coupled with lack of growth at the center of the distribution have been established as unarguable facts.  But these are both structural features of the aggregate of all income earners at any given time.  Yet individuals' experiences of income changes, or lack of changes, are not necessarily connected to the aggregate income distribution.  Rather, individuals move dynamically up and down within the national U.S. income distribution as they age.  Middle class individuals may therefore not experience the middle class income stagnation that researchers observe in the distribution of national income over time.

That is, just because incomes around the median in the national income distribution are languishing, it does not necessarily follow that this also reflects the subjective experience of the typical individual.  In order to distinguish between the national and the individual-level phenomena, I refer to lack of growth in median incomes in the U.S. as a whole as structural stagnation and lack of growth of individuals' earnings as existential stagnation.  To apply these terms explicitly: structural stagnation of middle class incomes in the U.S. is a well-established fact.  Existential income stagnation, in which the incomes of individuals within the middle class fail to increase as they age or increase at slower rates relative to earlier generations, remains relatively unexplored.

The main contribution of this article is that it examines individual income dynamics in a way that more closely reflects the experience of individuals over the life-course.  I show that the existential phenomenon of earnings changes over time bears little resemblance to the U.S. median income. 

The conclusions of this article result from analysis of the Panel Study of Income Dynamics (PSID) and the Current Population Survey's (CPS) March supplement.  I examine earnings levels within and between five successive cohorts of earners in the U.S.  The earlier cohorts began their careers towards the end of the post-WWII economic boom while the later cohorts' work experience occurred in the era of labor market liberalization that began about 1980.

Given the distinction between structural and existential stagnation, this article then tackles a crucial descriptive problem: to what extent do people have the existential experience of the stagnation of earnings that have been observed at the aggregate level?  I proceed by employing the PSID to calculate each individual cohort member's average earnings within their early career phase and several “peak career” stages.  The ratio of individuals' average earnings within each peak to the early career phase is obtained.  This ratio represents a linearized slope describing the rate at which individuals' incomes changed from early to peak career stages of the life-course.  I then examine the distribution of individual life-course income slopes (i.e. rates of change) of each cohort.  Each cohort's median life-course income slope represents the typical experience of the cohort and the interquartile range of slopes is the middle class experience.  The same basic procedure is applied to synthetic cohorts in the CPS.

I find that there has been no statistically significant change in the median cohort earnings slope in the PSID.  Although it is not significant, the observable trend of life-course earnings slopes of successive cohorts is upward, not downward.  In the CPS, which measures ratio of cohorts' median incomes over time rather than the median slope, there have been statistically significant increases in the rate of growth of earnings over the life-course.  Overall, there is no evidence that existential income stagnation is increasingly taking hold in the U.S.  On the contrary, to the degree that the findings from the CPS are correct quite the opposite may be true.

There are important implications to this finding.  While researchers, progressive think tanks, and others may be alarmed by structural middle class income stagnation, existential income stagnation is not empirically observable.  However, the article does not argue that because existential earnings stagnation is not taking place that structural stagnation can be disregarded or is unimportant.  Rather, structural stagnation without existential stagnation may not be sufficiently salient to motivate social movement formation, voting behavior, or other outlets of social grievance.