There's been a bit of chatter in the blogosphere about a recent post on Ezra Klein's blog, featuring estimates from various economists about the revenue-maximizing tax rate. It won't come as a surprise that people on the right tended to give lower estimates and folks on the left had higher guesses. Donald Luskin of National Review estimated 19 percent, for instance, while Emmanuel Saez, Dean Baker, Bruce Bartlett, and Brad DeLong all gave answers around 70 percent.
There are two things that are worth noting.
First, every single answer is to the right of the Joint Committee on Taxation. The revenue-estimators on Capitol Hill assume that taxes have no impact on overall economic performance. As such, even confiscatory tax rates have very little impact on taxable income. The JCT operates in a totally non-transparent fashion, so it is difficult to know whether they would say the revenue-maximizing tax rate is 90 percent, 95 percent, or 100 percent, but it is remarkable that a mini-bureaucracy with so much power is so far out of the mainstream (it's even more remarkable that Republicans controlled Congress for 12 years, yet never fixed this problem, but that's a separate story).
Second, very few of the respondents made the critically important observation that it should not be the goal of tax policy to maximize revenue. After all, the revenue-maximizing point is where the damage to the overall economy is so great that taxable income falls enough to offset the impact of the higher tax rates. Greg Mankiw of Harvard and Steve Moore of the Wall Street Journal indicated they understood this point since they both explained that the long-run revenue-maximizing rate was lower than the short-run revenue-maximizing rate. But Martin Feldstein of Harvard explicitly addressed this issue and hit the nail on the head.
Why look for the rate that maximizes revenue? As the tax rate rises, the "deadweight loss" (real loss to the economy) rises. So as the rate gets close to maximizing revenue the loss to the economy exceeds the gain in revenue.... I dislike budget deficits as much as anyone else. But would I really want to give up say $1 billion of GDP in order to reduce the deficit by $100 million? No. National income is a goal in itself. That is what drives consumption and our standard of living.
For more information, I think my three-part video series on the Laffer Curve is a good summary of the key issues. I posted them in May 2009, but Cato-at-Liberty has been growing rapidly and many people have not seen them. Part I addresses the theory, and explicitly notes that policy makers should target the growth-maximizing tax rate rather than the revenue-maximizing tax rate. Part II reviews some of the evidence, including analysis of the huge increase in taxable income and tax revenue from upper-income taxpayers following the Reagan tax-rate reductions. Part III looks at the Joint Committee on Taxation's dismal performance.
The Wall Street Journal has an excellent editorial this morning on the obscure -- but critically important -- issue of measuring what happens to tax revenue in response to changes in tax policy. This is sometimes known as the dynamic scoring versus static scoring debate and sometimes referred to as the Laffer Curve controversy.
The key thing to understand is that the Joint Committee on Taxation (which produces revenue estimates) assumes that even big changes in tax policy have zero macroeconomic impact. Adopt a flat tax? The JCT assumes no effect on the economic performance. Double tax rates? The JCT assumes no impact on growth.
The JCT does include a few microeconomic effects into its revenue-estimating models (an increase in gas taxes, for instance, would reduce gasoline consumption), but it is quite likely that they underestimate the impact of high tax rates on incentives to work, save, and invest. We don't know for sure, though, because the JCT refuses to make its methodology public. This raises a rather obvious question: Why is the JCT so afraid of transparency? Here's some of what the WSJ had to say about the issue, including some comparisons of what the JCT predicted and what happened in the real world.
...it's worth reviewing whether Joint Tax estimates are accurate. This is especially important now, because President Obama and Democrats in Congress want to allow the 2003 tax cuts to expire on January 1 for individuals earning more than $200,000. The JCT calculates that increasing the tax rates on capital gains, dividends and personal income will raise nearly $100 billion a year. ...we are not saying that every tax cut "pays for itself." Some tax cuts—such as temporary rebates—have little impact on growth and thus they may lose revenue more or less as Joint Tax predicts. Cuts in marginal rates, on the other hand, have substantial revenue effects, as economic studies have shown. ...So how well did Joint Tax do when it predicted a giant revenue decline from the 2003 investment tax cuts? Not too well. We compared the combined Congressional Budget Office and Joint Tax estimate of revenues after the 2003 tax cuts were enacted with the actual revenues collected from 2003-2007. In each year total federal revenues came in substantially higher than Joint Tax predicted—$434 billion higher than forecast over the five years. ...As for capital gains tax receipts, they nearly tripled from 2003 to 2007, even though the capital gains tax rate fell to 15% from 20%. Yet the behavioral models that Mr. Barthold celebrates predicted that the capital gains cuts would cost the government just under $10 billion from 2003-07 when the actual capital gains revenues over five years were $221 billion higher than JCT and CBO predicted. ...Estimating future federal tax revenues is an inexact science to be sure. Our complaint is that Joint Tax typically overestimates the revenue gains from raising tax rates, while overestimating the revenue losses from tax rate cuts. This leads to a policy bias in favor of higher tax rates, which is precisely what liberal Democrats wanted when they created the Joint Tax Committee.
All of the revenue-estimating issues are explained in greater detail in my three-part video series on the Laffer Curve. Part I looks at the theory. Part II looks at the evidence. Part III, which can be watched below, analyzes the role of the Joint Committee on Taxation and speculates on why the JCT refuses to be transparent.