Economic Sophistries

Eliot Spitzer has written a provocative article for Slate addressing the idea that high marginal tax rates reduce GDP. From the first two words — the title, Tax Fraud — we know how Spitzer feels about common arguments against higher or more progressive taxes.  He writes:

The result of sky-high marginal rates, this anecdote was supposed to prove, was declining productivity and economic growth. Is this true? Let’s look at a graph of the nominal top marginal tax rate in any given year and GDP growth in that year.

Then Spitzer shows the pretty graph, two squiggly lines that don’t seem to have much relationship to each other. Obviously, when two lines squiggle, each with apparent disregard for the squigglings of the other, there is no connection between the things the lines are purported to represent.

A caveat—obvious but critical—is in order. Simultaneity does not equal causation. Annual growth rates are a consequence of many factors, macro and micro, and the isolated impact of marginal tax rates on growth is hard, if not impossible, to discern from these numbers alone.
That said, it’s obvious that there is no correlation between higher marginal tax rates and slowing economic activity.

Let me translate that last paragraph: “Even though this analysis is completely invalid, I’m going to use it anyway.” Spitzer is making an economic argument based on (relatively) complex statistical analysis that he does, apparently, in his head. Economists have to use computers to do what comes so naturally to Spitzer (the numbskulls). He doesn’t actually have to compute the correlation, because it is obvious. He also doesn’t have to do any more detailed study on the issue because it was so obvious. But he does provide academic support for his conclusion:

More sophisticated efforts to analyze this relationship also produce decidedly murky results. An excellent review of this in the Yale Law Journal, “Why Tax the Rich? Efficiency, Equity, and Progressive Taxation,” concludes that there is scant, if any, legitimate academic support for the proposition that moderate, as opposed to dramatic, increases in marginal rates have any impact on the willingness of the wealthy to participate in the economy.

So a lawyer has referred to other lawyers for support for his position on an issue in economics. Furthermore, the referenced article is a book review, not, for instance, a scholarly review of the economics literature. The book, Does Atlas Strug? is a collection of papers, concerned entirely with taxing the rich. If the “rich” are the top 1% of us taxpayers by income, the rich have about a 14% share of US income. So, what Spitzer is telling us is that changing the tax rates by, say +/- 5% on the people who contribute 14% of US GDP, and who don’t really need to work anyway, doesn’t change GDP much. What an amazing insight! What that does not answer is what is the effect of  high marginal tax rates on the other 86% of the economy.

Maybe someone should ask the economists? These days, asking scholars is really easy, using Google Scholar.

Well, Alan Reynolds, an economist, says “Lower Tax Rates Mean Faster Economic Growth”. Reynolds is from CATO, so maybe you’ll want to discount him, but Engen and Skinner also report that lower taxes result in higher growth, although the effect is modest.

The issue of the relationship between tax rates and GDP doesn’t seem to have gotten as much study as the relationship between tax rates and hours worked, a supporting topic in Spitzer’s article. Spitzer notes that “Central to the intellectual debate about marginal tax rates has been the question of whether higher rates discourage people from working.” Let’s see what economists have to say on this topic.

In “Does It Pay to Work?”  and “Does It Pay to Work And Save?” Gokhale  Kotlikoff, and Sluchynsky (economists, not lawyers) write:

First, thanks to the incredible complexity of the U.S. fiscal system, it’s impossible for anyone to understand her incentive to work, save, or contribute to retirement accounts absent highly advanced computer technology and software. Second, the U.S. fiscal system provides most households with very strong reasons to limit their labor supply and saving. Third, the system offers very high-income young and middle aged households as well as most older households tremendous opportunities to arbitrage the tax system by contributing to retirement accounts. Fourth, the patterns by age and income of marginal net tax rates on earnings, marginal net tax rates on saving, and tax-arbitrage opportunities can be summarized with one word – bizarre.

In “Why Do Americans Work So Much More than Europeans?”, Edward Prescott (an economist at the Minneapolis Fed, not a lawyer) notes:

The surprising finding is that this marginal tax rate accounts for the predominance of differences at points in time and the large change in relative labor supply over time.

A study of “Taxes and entrepreneurial risk-taking” reports:

We first show theoretically that taxes can affect the incentives to be an entrepreneur due simply to differences in tax rates on business vs. wage and salary income, due to differences in the tax treatment of losses vs. profits through a progressive rate structure and through the option to incorporate, and due to risk-sharing with the government. We then provide empirical evidence using U.S. individual tax return data that these aspects of the tax law have had large effects on actual behavior.

In “The Effect of Taxation on Human Capital”, Alan Reynolds, (an economist, not a lawyer) says:

This study finds a significant negative effect of proportional income taxation on human capital. Of the few earlier studies to address this issue, most suggested a negligible effect of taxation on investment on human capital. This earlier conclusion is shown to be incorrect by using a model that is more general in several respects than the models used previously.

In “Workforce 2005: The Future of Jobs in the United States and Europe.” Reynolds (still not a lawyer) writes:

In many European countries, as the OECD Observer article quoted above points out, “additional work effort leads to little or no increase in net (after-tax) income because incremental gross earnings are largely, or even fully, offset by marginal income taxes and the reduction, or complete loss, of benefit payments” (OECD Observer, 1993).

The Showalter and  Thurston (both economists !)  study “Taxes and labor supply of high-income physicians” found that high taxes are disincentives for the most entrepreneurial doctors:

We use the 1983–1985 Physicians’ Practice Costs and Income Survey, supplemented with federal and state tax rates, to estimate the effect of variation in marginal tax rates on work hours for high-income physicians. We find that self-employed physicians are much more sensitive to the marginal tax rate than would be suggested by previous labor-supply studies, while those who are employees have no discernible sensitivity to marginal tax rates.

In their study of the 1998 tax flattening in Canada, Sillamaa and Veall (an economist and a statistician) write:

…Navratil (1995) finds evidence that tax-price responses are higher for high income individuals for both episodes of United States tax reform he studies….

…However our results are consistent with a tax price response by the self-employed and those with high incomes that is much larger than the overall response….

…Using that approximation, these estimates suggest that for a population of these higher income individuals, revenue would be maximized by a marginal tax rate of about 45% for the working age population, which is slightly less than the top marginal rate at the time….

So, in less than an hour, I’ve found 7 papers, written by actual economists, that disagree with Spitzer on the relationship between taxes and incentives to work.

Of course, on any complex subject, there is disagreement, and it is reasonable to supposed that I have preferred to include only the papers that support my conclusion. (I’ll let you guess what that conclusion is.) But I didn’t find much research that  indicated that tax rates don’t have an effect on hours worked.  The authors of Taxation, Human Capital, and Uncertainty suggest that tax policy has little effect on the number of hours worked by middle-aged male workers and  the blog Angry Bear would support Spitzer’s argument.

My impression is that there is substantial debate on the relationship between taxes and growth, with researchers finding everything from no relationship to a substantial relationship. This isn’t too surprising: GDP is a nation-wide measure that encapsulates all of the things going on in the entire economy, so it should be hard to pick out any one correlation. But economists generally support the statement that increased taxes are a disincentive to work, contradicting Spitzer.

Supporters of  Darwin’s theories or Anthropogenic Global Warning base the bulk of their arguments on the scientific consensus, the preponderance of scientists that support these theories. But the same people (well, people in the same ideological camp) have a complete disregard for any consensus in economics about how people respond to incentives, preferring to take their lessons in economics from lawyers rather than  economists.

(The portrait image is of Frédéric Bastiat, the author of  an insightful and entertaining refutation of protectionism, Economic Sophistries, in 1845. )