In lieu of an abstract, here is a brief excerpt of the content:

  • Selected Issues in the Rise of Income Inequality
  • Robert J. Gordon and Ian Dew-Becker

Increased American income inequality, in particular the increased skewness at the very top of the income distribution, has received enormous attention. This paper surveys three aspects of rising inequality that are usually discussed separately: inequality within the bottom 90 percent, inequality within the top 10 percent, and international differences in inequality, particularly among top earners.

We begin by examining data from the Current Population Survey (CPS) on income ratios between the 90th, 50th, and 10th percentiles, both for men and women separately and for the two sexes combined. We then examine several proposed explanations of changes in relative incomes within the bottom 90 percent, including the impacts of unions, free trade, immigration, the real minimum wage, and top-bracket tax rates. We also assess the hypothesis that the primary driver of increased inequality is skill-biased technological change.

We then tackle the most controversial issue, namely, why American incomes at the very top have increased so much, both relative to incomes below the 90th percentile in this country and relative to top incomes in Europe and Japan.1 We distinguish three types of top-level income earners: [End Page 169] "pure superstars" in sports and entertainment, a second group of professionals whose compensation also appears market driven but not amplified by access to a mass audience, and corporate executives, whose incomes are arguably not market driven but who instead can manipulate their pay to their own advantage. Our examination of cross-country differences finds that inequality at the top has increased more in the United States and the United Kingdom than in continental Europe, for what seems to be a combination of reasons.

Facts about the Bottom 90 Percent

Our 2005 Brookings Paper reported that during 1966-2001 only the top 10 percent of the income distribution enjoyed gains in real wage and salary income equal to or above the economy-wide rate of productivity growth.2 Accordingly, here we analyze the lower 90 percent of the income distribution separately from the top 10 percent. Drawing the line at the 90th percentile is convenient: the widely used CPS data apply best to income groups at and below that threshold, since top-coding limits what the CPS data can tell us about income shifts within the top 10 percent.

Much of the literature on changes below the 90th percentile emphasizes their exact timing. The top panel of figure 1, which depicts average hourly wage trends for men and women combined, shows differences in timing when different points in the distribution are compared: the ratio of wages at the 90th percentile to that at the 50th (the 90-50 ratio) increases more or less steadily after 1979, whereas the 50-10 ratio shows a sharp increase during 1980-86 followed by a slow and partial reversal. The 90-10 ratio, which combines these trends, shows a distinct increase between 1980 and 1988 followed by a plateau at a level between 20 and 25 percent above that of 1979.3

However, the depiction of trends for both sexes combined blurs some sharp differences in the patterns for men and women taken separately. For men (middle panel of figure 1), the 50-10 ratio has returned by 1998 to its 1979 level, after a temporary jump in 1979-86. Meanwhile the 90-50 ratio increases at almost a constant rate, rising 14 percentage points between 1979 and 1990 and another 11 percentage points between 1990 and 2005. The data for women (bottom panel) are more surprising. Their 90-50 ratio follows the [End Page 170]

Click for larger view
View full resolution
Figure 1.

Changes in Ratios of Average Hourly Wages at Various Percentiles of the Distribution, 1973-2005a

[End Page 171]

same path as that for men, but their 50-10 ratio increases much more, and the increase is sustained. This is consistent with the fact that women are roughly twice as likely to be paid the minimum wage as men.4 Perhaps most surprising is that the overall increase in the 90-10 ratio for women is almost double the increase for men: 44 log percentage...


Additional Information

Print ISSN
pp. 169-190
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.