Discussion: (41 comments)
Comments are closed.
A public policy blog from AEI
Washington’s tax hike on wealthier Americans won’t accelerate economic growth, won’t create jobs, and won’t lower the debt by more than a rounding error. So what was the point of all that debate about the fiscal cliff? Why did President Obama insist on those upper-income tax increases, especially when the economy continues to struggle?
Simple: It was a way — even if mostly symbolic — of addressing what President Obama views as America’s biggest problem: Rising income inequality. The purest expression of the president’s economic views came during a 2011 speech in Osawatomie, Kansas, where he spoke at length about how rising inequality “hurts us all” and “gives lie to the promise that’s at the very heart of America: That this is a place where you can make it if you try.”
But does the data back up Obama’s rhetoric? Most claims that it does are based on two inequality studies, one by economists Thomas Piketty and Emmanuel Saez, the other by the Congressional Budget Office. Both studies are frequently cited in mainstream media stories about income inequality. But new research by Philip Armour and Richard Burkhauser of Cornell University and Jeff Larrimore of the Joint Committee on Taxation raises questions about those two studies and their conclusions. Here are their two key findings:
1. All income levels saw expanding incomes from 1979-2007. There’s been no income stagnation for the middle class.
2. While inequality has risen, it was mostly if not almost entirely a 1980s phenomenon.
Let’s briefly walk through the numbers with the help of a few tables:
1. Piketty and Saez’s data shows that from 1979-2007, the mean pre-tax, pre-transfer cash market income of the bottom 20% of tax units fell 33%, the middle quintile rose just 2.2%, while the top quintile rose 32.7%. The top 5% did even better, with their market income rising 37.9%. The poor got poorer, the middle stagnated, the rich got richer. You’ve heard this story.
2. But Burkhauser and Larrimore, in a study last year along with Indiana University’s Kosali Simon, take issue with how Piketty and Saez measure income and make a number of common-sense changes. The Burkhauser-Larrimore-Simon data reflects a) the rise in cohabiting couples and adult children living with parents who can share resources, b) cash-transfers from government, c) the impact of the tax code, d) employer and government-provided health benefits. The result is a measure that better captures the actual consumption potential of US households.
Under this formula, the income of the bottom quintile from 1979-2007 rose 31.8%, middle quintile 34.4%, and top quintile 54.0.%. While inequality did increase over the span of the period — income for the top 5% rose 68% — everyone was better off.
3. But then the CBO took a crack at the inequality debate, adding a new twist. They added income from realized capital gains. And that makes a big difference. When you factor that additional income into the Burkhauser-Larrimore adjustment to Piketty-Saez, you get the following results: Income rose 31.1% for the bottom quintile from 1979-2007, 36.7% for the middle quintile, and 83.1% for the top quintile. For the top 5%, income rose a whopping 136.7% over those three decades. Everyone did better, but clearly upper-income folks did much, much better.
One other point: As the chart below shows, no matter how you slice the numbers, you find that “the majority of that inequality growth occurred in the 1980s—with somewhat less in the 1990s and very little in the 2000s.”
4. But in their new study, Armour-Burkhauser-Larrimore question the CBO’s choice of using realized capital gains income:
Since individuals can choose when to realize them for tax purposes through the timing of transactions … income recorded as taxable realized capital gains this year may not be due to increases in net-wealth this year. Additionally, taxable realized capital gains exclude accrued gains this year from assets that are not recorded on this year’s tax returns, either because the asset was not sold, was sold but held in a tax-sheltered account, or was carved out of the tax code (e.g. primary housing).
To capture changes in net worth, it’s better, the researchers argue, to include capital gains as measured by the increase or decrease in the value of capital assets — including those in tax sheltered accounts — in each year regardless of whether that asset was sold for a taxable realized gain.
Now for data collection reasons, only the years 1989-2007 are included, but that’s enough to show that inequality was falling, not rising, over those two decades and business cycles. Mean income rose 32.2% for the bottom quintile, 20.2% for the middle quintile, and 9.3% for the top 5%.
Armour-Burkhauser-Larrimore sum up:
For those focused on taxable income of tax units excluding taxable capital gains—a Piketty and Saez (2003) measure of market income, based on tax returns—undoubtedly income inequality has grown substantially in recent years, and the middle-class is struggling.
The inclusion of taxable realized capital gains in income measures that are more inclusive—like the CBO (2012) measure using both tax returns and CPS data—will reinforce this view of rising inequality, but it does so by including a measure of capital gains that by definition misstates the timing of gains and misses the increasingly important capital gains in tax-sheltered accounts.
In contrast, when using our comprehensive income definition that mirrors the CBO (2012) report but excludes all capital gains, we observe that incomes have risen throughout the distribution and since 1989 have largely risen uniformly throughout the distribution.
Alternatively, when we include capital gains in our series but do so on a yearly accrued basis that is more in line with Haig-Simons principals, it increases the volatility of income trends but demonstrates slower growth throughout the income distribution than when capital gains are excluded. This reflects lower capital gains accrual rates in the most recent business cycle than in the proceeding ones. But it also illustrates that inequality growth has not risen in recent years, as the top quintile of the income distribution had the slowest income growth from 1989 through 2007 while the bottom quintile had the fastest. …
Doing so provides evidence that contradicts the notion that income inequality fueled by capital gains at the top end of the distribution has dramatically increased over the past two business cycles.
To be clear, we shouldn’t ignore the market income numbers from Piketty-Saez which likely reflect, among other things, an inability of the US education system to deal with the demands that rising automation and globalization have made on workers. But the research of Armour, Burkhauser, and Larrimore suggests that making an inefficient, lop-sided tax code even more distorted — which Washington just did — is nothing more than an ideological indulgence based on a misinterpretation of economic data.
Comments are closed.
1150 17th Street, N.W. Washington, D.C. 20036
© 2015 American Enterprise Institute for Public Policy Research