The share of income earned by the “top 1%” can be misleading. When it comes to the income distribution, the “top 1%” consists of the 1.4 million tax returns with the greatest amount of adjusted gross income reported in a given year. Data on single-year income distributions are a lot less meaningful than commonly supposed because it’s not clear who those 1.4 million returns belong to, or what share of households remain in the top 1% from year to year. As a result, the statement that “most of the income growth over the past 30 years has gone to the top 1%” is really confusing because the households in the top 1% in 1981 were not the same as the households in the top 1% in 2011. All we know is that the income of today’s top 1% of taxpayers is higher than the income earned by the top 1% a generation ago.
The shift in income in the top 1% should not be ignored or treated as an irrelevant, however. The challenge is to analyze and interpret the data in a way that makes sense and does not depend on the mistaken assumption that “the 1%” is a fixed category of households over time. For example, Senate Majority Leader Harry Reid recently argued that those who “benefited from Bush tax cuts” should now pay higher taxes to reduce the deficit. But households currently in the top 1% differ from households that benefitted from passage of the 2001 and 2003 tax cuts. For every 5 households in the top 1% in a given year, only 2 remain in the top 1% a decade later. As the U.S. Treasury explains:
This statistic illustrates that the top income groups as measured by a single year of income (i.e., cross-sectional analysis) often include a large share of individuals or households whose income is only temporarily high. Put differently, more than half of the households in the top 1 percent in 2005 were not there nine years earlier. Thus, while the share of income of the top 1 percent is higher than in prior years, it is not a fixed group of households receiving this larger share of income.
There is an obvious divergence between what the data mean and what proponents of tax increases interpret them to mean so as to make the case for higher taxes. If the “top 1%” is thought to be a fixed category of households, then single-year income tables leave the impression that this small number of households collects more and more in national income each year. If 20% of taxpayers were expected to be in the top 1% of the income distribution in any given year, increasing tax rates on the “top 1%” would impact one-in-every five households rather than one in every 100.
The notion that one-in-five households could find itself in the top 1% at some point may seem to be overly optimistic, but most of the churn in the top 1% over the course of 30 years is related to life cycle trends. Many households that were part of the top 1% in 2005, for example, have since retired and been replaced by households whose primary earners were in more junior positions in their company, attending graduate school, or yet to start their own businesses in 2005. A large number of taxpayers who benefitted from the Bush tax cuts are likely out of the labor force. Raising taxes on the top 1% is therefore not going to impact them.
Beyond life cycle, another issue that’s behind the churn in the top 1% of the income distribution comes from non-recurring sources of income, like asset sales. Every year, some portion of the top 1% sell family businesses, large stock portfolios, or real estate. The capital gains on these sales is non-recurring; the capital gains income from these sales comes in a single year. These sales may push up the average income of the top 1% for that year, but they have no impact on the average earnings of the group. The chart below looks at the share of the income of the top 1% that comes from non-wage income like capital gains, small business earnings, dividends, and interest. In 2007, only 36.3% of the income of the top 1% came from wages (1 - 0.637), while 33.2% came from capital gains (elective asset sales).
Sources of Income for the Top 1% (IRS)
The composition of income is clearly significant because discussions about income often concern how much someone “makes.” Implicit to this formulation is that the income comes from salary or recurring income. Generally speaking, the lower on the income distribution, the larger the share of earnings that come from wages and the more likely the previous year’s income predicts income the following year. But for the top 1%, the previous year’s income may be inflated by a one-time event, which means that the household could contribute meaningfully to the average income of the top 1% for a single year and then drop out of the top 1% entirely the following year.
Rather than the average income, the chart below looks at the income threshold for households to be in the top 1%. In 2009 (most recent data available), the income of the 1.4-millionth household from the top was $343,927, down 10% from the previous year. Since passage of the Bush tax cuts in 2003 (when the 35% top rate and 15% capital gains rate went into effect),the income of the “theoretical” household on the cusp of the top 1% rose by just 2.6% per year in nominal terms. This household’s income has failed to keep pace with inflation. At 3% annual growth between 2003 and 2009, the $295,495 income would equal $352,836 in 2009. Thus, on an inflation-adjusted basis, the earnings of the household at the bottom of the top 1% have actually fallen by more than 2% since passage of the Bush tax cuts.*
Income Threshold to be in the Top 1% (IRS)
Of course, this household doesn’t really exist – which is why it has been labeled theoretical. There is virtually no chance that the tax return on edge of the top 1% in 2004 belonged to the same household on the edge of the top 1% in 2008. In reality, many households fall in and out of the top 1% based on the aforementioned asset sales and “flow through” small business income of other households. Yet, this is the same type of misleading analysis that is used to show how far other segments of the income distribution are falling behind. Estimating a household’s net worth or lifetime income using a single year’s income makes is problematic to say the least. But this is actually much more reasonable than tracking the change in income of a certain percentile of the distribution and pretending as though the same household falls in that percentile each and every year.
What does it mean if the income of the household at the 47th percentile of the income distribution is flat over the course of ten years even if it is a different household each year? It’s not clear – or there does not appear to be any consensus conclusion amongst economists. Some of the stagnation in wages could come from low-skilled immigration, for example. If a lower skill worker replaces a retiring middle class worker in the income distribution, the result could be an apparent decline in income for the middle class even as the incomes of the families formerly in the middle might have actually increased. Educational attainment and economic opportunity are important issues as well – and the shifting income distribution suggests that labor skills are a key driver now more than ever. That said, what should be crystal clear is that the confusing data in this area is often conveyed in the context of a political agenda that promotes class warfare. At minimum, policymakers and economic analysts would be smart to tone down the rhetoric and focus in on arguments and conclusions that are truly supported by the data.
*This paragraph was updated on March 26, 2012.