The Oregon Economics Blog: Fred Thompson: Taxation and Inequality

Editor's Note: Once again Fred Thompson rescues this over-burdened blogger with an interesting and detailed discussion of tax policy and income inequality.

The most conclusive evidence showing that that families within the top 1 percent of the income distribution experienced very large gains relative to the average since 1980 comes from Piketty-Saez, who use tax data to construct their inequality estimates.

Piketty-Saez further show that much of the increased concentration of personal income occurred between 1986 and 1988. They accept that much of that increase was a consequence of the 1986 Tax Reform Act, which reduced the difference between personal tax rates and corporate tax rates, thereby causing a shift of income to the personal tax base from the corporate tax base. Indeed, they acknowledge that tax shifting could easily account for two well-noted phenomena of recent decades: the increasing concentration of personal income and the declining rates of corporate profitability. Nevertheless, they argue that if that tax reform was all that was going on then the upper-percentile income share would have shrunk down to its 1986 level over the following decade and, needless to say, the share did not.

However, I think Piketty-Saez are wrong in assuming that shift in income from corporations to individuals caused by the 1986 TRA was a one shot deal. In the first place, corporate profitability (based on the federal flow of funds accounts) dropped significantly in 1982 and in 1986 and has subsequently continued its downward trend, despite the fact that America’s biggest firms earned record profits during this period. Where did the missing income go? Many tax experts believe that much of what is now reported as personal income was previously reported as corporate income.

Not only did the 1986 TRA reduce the difference between personal tax rates and corporate tax rates, it also made it easier for the owners of S corporations and most professional corporations to treat profit from their businesses as personal rather than corporate income and arguably increased the incentives of the owners of closely-held C corporations to treat corporate income as personal income, in part by severely restricting the use of tax shelters and requiring businesses to treat most benefits they provided in-kind to their owners as ordinary personal income (even funds deposited by these businesses in tax-deferred retirement accounts, a sum of $17 trillion in 2007, are now counted as personal income when earned, where once they were treated as a business expense until they were distributed). Consequently, the number of S and professional corporations shot up after 1986 and has continued to grow at a steady rate as a percentage of all corporations in the US ever since.

A lot, maybe most, of the discussion of high earning individuals is misleading. It focuses on the CEOs of big corporations, sports stars, and entertainers. But those folks represent a tiny proportion of the 1 million-plus families in the top 1 percent of the income distribution. Most high-income families have high incomes because they own businesses, farms, or legal, medical, or financial service providers. For example, it has been estimated that the CEOs of the 500 biggest corporations in American and the 500 highest paid athletes and entertainers earn about $25 million on average. That adds up to $25 billion. In contrast, it looks like the 570,000 small-business owners in the top 1 percent of the income distribution may have shifted $300 billion from their corporate accounts to their personal accounts in 1995 more than they would have if the 1986 TRA had never been enacted into law. This implies that they paid about $40-$50 billion more income tax that year than otherwise, but, perhaps, as much as $100 billion less in corporate income taxes, maybe more.

In other words, what the 1986 TRA allowed them to do was keep a lot more of what they earned. Consequently, the effects of the 1986 TRA were reflected in reduced corporate incomes, which in turn reduced corporate income tax payments. Of course, this is not the only reason for declining corporate income tax payments, especially at the state level, but it is probably the main reason. The effects of the 1986 TRA were also reflected in higher reported personal incomes and, despite reductions in marginal rates, increases in personal income tax payments.

This last fact accounts for the claim that Laffer was right, tax rate cuts produced increased government revenue. However, that claim ignores the effect on corporate income tax revenues. Almost no serious student of public finance believes that, tax rate cuts produce increased government revenue at current rates, or is even close to being true.

On the other hand, every serious student of public finance knows that taxes create wedges and most accept that dead-weight losses increase as an exponential function of marginal tax rates. The interesting question is how much of the relative increase in the income reported by Piketty-Saez is due to income shifting, how much is due to reduced corporate tax payments, and how much is due to a reduction in the tax wedge.

The Tax Foundation got blasted for suggesting that the answer to the last part of this question is 20-40 percent. Those do not seem like completely unrealistic numbers to me. Moreover, that is probably also the upper limit of the real increase in the gains in pre-tax income accruing to the top 1 percent of the income distribution relative to the average.

Figure 2. Tax Wedges on Labor Income Showing How an Increase in Tax Rates can Reduce Revenues by Increasing Deadweight Losses

It is a striking fact that the fraction of total wealth held by the rich has not changed noticeably in the last 70-80 years. As Saez observes” “A jump in reported wage income with no change in wealth is consistent with the income-shifting explanation, since all that is changing is where income is reported, not how much income is earned.” We could say the same thing about the consumption of families in the top 1 percent of the income distribution, although this may be an artifact of the methods the census bureau uses to survey consumption levels.

These issues may seem more relevant to federal tax policy than to state tax policy and, indeed, most of the discussion about them is in that context. But Oregon is currently addressing some fundamental issues about its tax structure that reflect the issues addressed here. One prominent proposal involves substituting a transaction tax (probably like a sales tax in its incidence) for the existing state corporate income tax (which this argument suggests is much more like the personal income tax in its). This analysis goes directly to an assessment of the likely consequences, both in terms of adequacy and fairness, of those alternatives, especially given the kinds of changes to federal income taxes we can anticipate under either Obama or McCain.

See Irwin Diewert, Denis A. Lawrence, and Fred Thompson. The Marginal Costs of Taxation and Regulation, in Handbook of Public Finance. F. Thompson and M. Green (eds). New York: Dekker, 1998: 135-173.