Dylan Matthews discusses a new theory of how changes in tax rates might impact the distribution of income:
The theory comes from a paper by economists Emmanuel Saez, Thomas Piketty and surprise guest star Stefanie Stantcheva. They hypothesized that lowering marginal tax rates increases the incentive for high earners to bargain up their wages — and pocket more of their raise. Moreover, because the higher earners are bargaining for greater pay without actually becoming more productive, the additional money they’re taking in is coming form other workers, generally those lower down the income scale.
Sure enough, Saez, Piketty, and Stantcheva found that there’s a strong correlation between the size of countries’ tax cuts on the rich and increases in the income shares of the rich, even before taking taxes into account. That is, the rich are getting richer even before you take into account that they’re paying less in taxes:
What’s more, increases in the rich’s income share didn’t coincide with increases in the rate of GDP growth. That fits the Saez/Piketty/Stantcheva theory that the rich aren’t getting more productive; they’re just getting better at bargaining because lower tax rates give them the motivation to get better at bargaining.
Much more research has to be done before we can accept the Saez/Piketty/Stantcheva findings as proven. But their research does point to a plausible way that lowering income tax rates for the wealthy — and in particular lowering rates on capital income — could worsen pre-tax inequality without any economic benefit.
Two issues immediately arise. The authors draw on earlier work of Piketty and Saez on the changing tax burden for high-earners, yet Arpit Gupta has raised questions about the validity of some of their assumptions, e.g., Piketty and Saez assume that the entire burden of corporate tax rates falls on investors, despite the fact that the emerging consensus among economists is that labor bears at least some of this burden. If corporate taxes do indeed lower wages for rank-and-file employees, the tax burden on high-earners in the U.S. has decreased by less since the 1960s than Piketty and Saez assume.
And then there is the question of whether Piketty and Saez are correct in their assessment of the increase in income inequality. This month, Phillip Armour, Richard V. Burkhauser, and Jeff Larimore released a new working paper addressing this issue, the abstract of which reads as follows:
Recent research on United States levels and trends in income inequality vary substantially in how they measure income. Piketty and Saez (2003) examine market income of tax units based on IRS tax return data, DeNavas-Walt, Proctor, and Smith (2012) and most CPS-based research uses pre-tax, post-transfer cash income of households, while the CBO (2012) uses both data sets and focuses on household size-adjusted comprehensive income of persons, including taxable realized capital gains. This paper provides a crosswalk of income growth across these common income measures using a unified data set. It then uses a more consistent Haig-Simons income definition approach to comprehensive income by incorporating yearly-accrued capital gains to measure yearly changes in wealth rather than focusing solely on the realized taxable capital gains that appear in IRS tax return data. Doing so dramatically reduces the observed growth in income inequality across the distribution, but most especially the rise in top-end income since 1989.
Recall our discussion of tax systems that rely on annual assessments of the market value of financial assets vs. those that are realization-based. The beauty of a realization-based system, from the perspective of an investor, is that she can pick and choose when she realizes a capital gain, and she will presumably do so with an eye towards minimizing her tax burden. Some argue that one reason for the recent surge in federal tax revenues, for example, is that taxes on capital income are scheduled to increase under the post-cliff ATRA legislation. But this doesn’t change the fact that an owner of a financial asset benefits when the asset in question increases in value, whether or not she decides to sell it, e.g., rising home values tend to increase consumption. When we rely taxable realized capital gains in our analysis, we include asset appreciation that may have occurred over several years in the income for one particular year, which will tend to make the income for that year look much higher that it would otherwise. And just as importantly, this method leads us to exclude the accrued gains that occur in years when individuals don’t sell their assets, which will tend to make the income for that year look lower than it would otherwise. There is an obvious reason why we might rely on taxable realized capital gains: this is the data that is most readily accessible. That does not mean, however, that it gives us the clearest picture of how income changes over time.
Armour et al. find that accounting for accrued gains changes the income inequality picture dramatically:
[W]hen we include yearly-accrued capital gains excluding housing gains and private business gains, instead of taxable realized capital gains, the inclusion of these gains slows income growth in all but the bottom quintile of the distribution. Thus, when using this measure that is more in line with Haig-Simon’s income principles, the top quintile of the distribution had the least growth in income from 1989 through 2007 while the bottom quintile of the distribution had the most. Measured in this way income inequality fell between 1989 and 2007.
How is it possible that the choice of treatment of capital gains can have such a dramatic difference? It results from both the timing of realizing gains and from the likelihood of assets appearing in taxable accounts for individuals at different points in the income distribution.
One driver of this outcome is that mean equity investment holdings in the bottom quintile grew over 7-fold while increasing by a factor of 3.5 in the top quintile between 1989 and 2007. One assumes that the post-crisis economic environment has darkened this picture considerably, but this development appears to be significant all the same. And limitng our analysis to taxable realized capital gains obscures this development because a disproportionately large share of this increase in equity holdings has occurred in tax-sheltered accounts. And if we assume that low-income and middle-income households hold a larger share (not a larger amount) of their equity holdings in tax-sheltered accounts, we can see how this might impact income distribution.
Here is another thing to keep in mind: stock market gains were higher in the 1980s and 1990s than in the 2000s. If a hypothetical worker allowed her equity holdings to increase in value over those decades and started to sell them in the 2000s, as she approached retirement, we would fail to capture increases in her income in the 1980s and 1990s while in effect overstating her income gains in the 2000s.
There are, to be sure, some weaknesses in the Armour et al. account, as they freely acknowledge, and their paper hardly represents the last word on the subject. But if taxes decreased by less on high-earners than Piketty and Saez suggest, and if relying on taxable realized capital gains has led them to overstate the increase in income inequality, we might have to return to the drawing board.