A recent book review in Foreign Affairs, perhaps the most well-respected of all international relations journals, noted of the recent recessionary difficulties of many Americans:
This was not an anomaly but rather a continuation of a 40-year trend of ballooning incomes at the very top and stagnant incomes in the middle and at the bottom.
This statement is just one example of what is perhaps the most talked about economic and political topic outside of healthcare reform. The Occupy Wall Street movement and its protest is only the most prominent form of public outrage at what has been often termed “The New Gilded Age.”
But are these dismal claims true? Though you probably have a canned response ready to flip assuredly off or your tongue, in this post, and the ones that follow, I want to challenge these notions and mainly argue that the answer is “It depends.” It depends for a lot of reasons that I will discuss, but one of the most surprising is that many of the terms we think of as quite precise are actually inexact. Even the word “income” does not, in and of itself, allow for easy comparisons. Are we talking about pre- or post-tax income? Does income include benefits such as 401K matching, vacation, and, most notably, healthcare coverage? Is it pre- or post-government transfer we are talking about? Are we considering individual income, tax unit income, or household income? How are we accounting for changes in any of these variables over time when making our comparisons?
The assumptions used—the answers to those questions above and similar ones—changes the results of any analysis about differences in income over time. For now I want to put these questions to the side, however, and instead talk about broad trends in income inequality. Though differences in the assumptions above can change the degree to which inequality is measured—middle-class income may, for instance, improve from a roughly 3% increase since 1979 to a 35% increase—the assumptions are not large enough to override the broader patterns in income distribution. It is still important to remember, however, that, even putting aside these technicalities, the data that follow are not the last word on inequality. Some economists, for instance, argue that inequality of wealth is more important than that of income, others consider consumption a better relative measure of quality of life, and still others point out, correctly, that macroindicators that consider populations as a whole do not tell us about the mobility between income quintiles—the top and bottom quintiles in 1979 and 2012, respectively, do not contain the same individuals (we can think of obvious examples of “upper mobility,” for instance, if we consider entrepreneurs such as Mark Zuckerberg or Sergey Brin, both of whom jumped from lower income brackets to higher ones in a relatively short amount of time). I will return to all of these considerations in later posts, including the more technical aspects.
So on to the data. Roughly speaking two patterns are true: (1) income inequality within the bottom 99% remained relatively unchanged since 1979, while inequality between the bottom 99% and top 1% increased drastically and (2) incomes among all quintiles increased during that same period. That is—to use the famous pie analogy—the pie itself got much larger since 1979 and every person’s slice of the pie also got larger. The top 1% of “pie eaters” got a slice that had grown very, very much, while the other 99% each got a slice that had grown only moderately. This does not, however, imply that the top 1% somehow “stole” the portion of pie that the rest of us were “owed” or “deserved,” a point I will return to at the end of this post.
One great starting point for high-quality data is this Congressional Budget Office (CBO) report from late 2011. In it, the CBO used Internal Revenue Service (IRS) tax return data in conjunction with the Current Population Survey (CPS) from the Census Bureau. Because of the long delay in getting high-quality data from government sources, the CBO report only includes incomes through 2007. Below is the share of incomes by quintile.
You can see that the lowest through the fourth quintiles lost, in percentage terms, shares of overall US income between 1979 and 2007, the 81st to 99th group of income earners have roughly equal shares in both years, and the top 1% enjoyed a large increase in their share. Again, getting back to the pie analogy, this chart says that if we compare the pie from 1979 with the one from 2007, most people are getting a smaller slice. But, and it is an important but, the pie has gotten much larger. Effectively, then, most people are enjoying a smaller slice of a much larger pie (and thus overall are getting to “eat” more pie). Here, I’m staying agnostic about whether the rich have too much pie relative to the rest of us (that will be discussed in a later post).
This increase in the pie can be seen if we look at income growth from 1979 to 2007, which is shown below.
Here we see what I mentioned earlier: first, that all quintiles have seen increases in income since 1979; second, that most income brackets have enjoyed roughly equal increases in income (although clearly the higher the income, the more gains have been realized); and third, that the top 1% has seen a very large increase in income gains relative to the other 99%. Below is the same data displayed differently:
The CBO also investigated this gain by breaking income down into its constituents. They find that from 1979 to 1988 the shift in income shares to top earners resulted primarily from a higher concentration of labor income accruing to this group. From 1991 to 2000 inequality increased as a result of an overall shift in the economy toward monies from capital gains and and increasing concentration toward top earners in both labor and capital income. From 2002 to 2007 the increase is largely due toward further shifts to capital gains and these gains being further concentrated among top earners.
Though much public rhetoric pins the increase in inequality on exploitation by top earners, the fact is that few if any economists believe that increases in the incomes of the top one, five, or twenty percent came at the expense of the rest of us. As the CBO notes:
The precise reasons for the rapid growth in income at the top are not well understood, though researchers have offered several potential rationales, including technical innovations that have changed the labor market for superstars (such as actors, athletes, and musicians), changes in the governance and structure of executive compensation, increases in firms’ size and complexity, and the increasing scale of financial-sector activities.
It is important to recognize, however, that while gains in inequality may not have come at our expense, the resulting effects may well be. Most economists think some amount of inequality is actually good for society because it provides incentives for hard work and fiscal prudence. The challenge, though, is identifying the inflection point where inequality ceases to provide positive incentives and starts to become discouraging to workers, increases political and societal frictions between classes, and, perhaps, even becomes dangerous to the extent that increased inequality offers channels for a small elite to enjoy unhealthy amounts of political power. These issues, as well as others, will be taken up in my continued series on inequality in America.