In his book Capital in the Twenty-First Century, Thomas Piketty did us the great service of bringing the problems of wealth and income inequality to the fore. In the process, however, he also may have performed a bit of a disservice – making those problems seem simple, a mere function of the inequality r > g, where r is the rate of return to capital and g is the rate of economic growth. The solution, he suggested, was equally simple: a tax on wealth.
Bariş Kaymak and Markus Poschke, in The Evolution of Wealth Inequality over Half a Century: The Role of Taxes, Transfers and Technology, offer a more complex picture. They construct a general equilibrium model of the U.S. economy over the past half-century, incorporating (1) reduced income taxes on top earners (from a 45% effective rate for the top 1% in 1960 to a 33% effective rate in 2004, and from a 71% effective rate for the top 0.1% in 1960 to a 34% effective rate in 2004), (2) expansion of government transfers from 4.1% to 11.9% of GDP over the same period, and (3) higher pre-tax wage inequalities, which they attribute to technological change. (For these purposes, effective rate is defined as income taxes paid as a percentage of taxable income.) The question they ask and attempt to answer is: To what extent were the observed increases in wealth and income inequality over that period attributable to each of these changes or trends?
Income: Their answer with respect to income inequality is unequivocal: After taking into account standard general equilibrium adjustments, cuts in top rates had almost no effect on after-tax income distribution. Instead, the authors find that during the period studied, after-tax income inequality increased almost entirely because pre-tax income inequality increased.
Wealth: Their answer with respect to wealth inequality might strike some as counterintuitive. They find that an increasingly robust safety net (principally Social Security and Medicare) appears to have reduced incentives to save for the bottom 90%, resulting in increased wealth concentration at the top. Taken together, changes in U.S. tax and transfer systems explained nearly half of the observed rise in wealth concentration over the past half-century; the remainder was explained by increases in pre-tax income inequality.
Interactions: In addition, the two inequalities interacted in complex ways, intermediated by interest rates and prices:
Accumulation of additional wealth in response to tax cuts leads to a decline in the interest rate and an increase in the wage rate. The fall in the equilibrium interest rate discourages savings by lower wealth groups and exacerbates the direct effect of tax cuts on wealth inequality. As for income, the lower interest rate mitigates the rise in top incomes, while a higher wage rate benefits lower income groups as they live mainly off labor income.
By email to this reviewer, Prof. Kaymak explains the relationship between wealth inequality and wage rate increases as follows:
[T]he link from capital accumulation to the wage rate is an equilibrium effect …. As new wealth is channeled to production through investment, it generates a demand for additional labor since labor and capital are complements in production. Higher demand for labor then raises both employment (job creation effect) and wages. Of course, tax cuts could also change the labor supply behavior …. But we find this [latter effect] to be much weaker ….
Bottom line: greater inequality in wealth reduced income inequality by reducing the return to capital and increasing wages.
Finally, because general equilibrium adjustments take time, the authors predict that two or three more decades of increasing wealth concentration will result from policy and economic changes that have already occurred, at which point the top 1% will own about half of all U.S. wealth, up ten percentage points from their current share.
What should we make of all this?
First, general equilibrium analysis, although absolutely essential, is notoriously difficult and sensitive to assumptions. The authors note that their conclusions are inconsistent with those reached by some (Atkinson (2011) and Mertens (2013)), but consistent with those reached by others (Saez (2012)). Nonpartisans may want to take all such conclusions with at least a small grain of salt.
The authors might have distinguished between taxes on income from labor and taxes on income from capital, although these are difficult to tease apart. A priori, at least, we would expect different kinds of taxes to have different effects on savings and interest rates. Optimal tax theory, for example, conventionally assumes that taxes on income from capital depress savings, but that taxes on income from labor do not.
One wonders also whether some aspects of the U.S. tax system might have offset the results the authors describe through depressed demand for lower-wage labor. Accelerated and bonus depreciation may have encouraged the automation of less skilled jobs, for instance, and/or a largely territorial multi-national corporate tax system may have encouraged the offshoring of those same jobs as tariffs and other trade barriers were lifted.
With respect to the effect of offshoring, Prof. Kaymak observes, again by email:
For wages to rise through the equilibrium effect…, investment has to stay at home. We had run some simulations allowing for capital flight in early versions of our paper. What that does essentially is to mute the effect of top tax cuts on wealth inequality (interest rate does not fall in this case, so the bottom wealth groups do not curb savings as much). Income inequality in turn increases somewhat, because part of additional investment goes abroad, hence no wage gains for labor.
In other words, capital flight reduces wealth inequality but increases income inequality.
In any event, the paper makes an important contribution; its thoughtful analysis should persuade the reader that problems of wealth and income inequality are more complex than Piketty claims. Increases in the safety net that reduce disparities in consumption may exacerbate disparities in wealth. Ed Kleinbard, in his book We Are Better Than This, has urged that we worry less about progressivity in taxation and more about progressivity in spending. Doing so, the current paper suggests, may actually result in further concentrations of wealth in the already wealthy.
This, in turn, raises fundamental normative questions: Do we really care about disparities in wealth? Why? Do we care more about disparities in income? Or is our ostensible concern about inequality really a concern about poverty? If Kaymak and Poschke are even partly right, it may not be enough to say: “Equality good. Inequality bad.” We may actually have to do some hard normative work.