Paul Krugman Still Befuddled by Supply-Side Success

The Congressional Budget Office reports that from October through June, "net [tax] collections from individuals were up by $105 billion, or about 18 percent. Almost two-thirds of that increase, or $66 billion, resulted from higher receipts from nonwithheld taxes, most of which were recorded in April and May." You might think a big fan of big government like New York Times columnist Paul Krugman would be delighted. But he seems to look for a cloud behind every silver lining.

Mr. Krugman thinks "this revenue boost looks like a temporary blip," which "probably reflects mainly capital gains on stocks and real estate, together with bonuses paid in the finance and real estate industries." Even as wild guesses go, that was remarkably wild.

A soaring stock market explains soaring tax revenues in 1997-2000, but not this year. The S&P 500 stock index in June was just 1193, barely higher than its level of 1131 in January 2004.

The housing bubble is nearly as implausible, because capital gains from the sale of a primary residence have been virtually tax-free since 1997. Second homes are a small fraction of the market and generally sell at much lower prices than primary homes; they couldn’t have accounted for much of the revenue surge this year as compared with last year (when home prices were also soaring).

Mr. Krugman’s theory that executive bonuses in finance and real estate could account for much of the $66 billion is politically correct but mathematically impossible. Top executives in the top financial firms received average bonuses of $2 million a year in 2004, according to Mercer consulting. But only the increase in such bonuses could account for increased tax collections. Even if a hundred executives received an extra $300,000 apiece, that would lift the nation’s taxable income by only $30 million. And a 35 percent tax on that sum would amount to only $10.5 million, which can’t begin to explain a $66 billion revenue gain.

Lacking any explanations, Krugman gets personal. "The usual suspects on the right," he writes, "are already declaring victory over the deficit, and proclaiming vindication for the Laffer Curve — the claim that tax cuts pay for themselves because they have such a miraculous effect on the economy."

I have not declared victory over the deficit because doing so would amount to declaring I’m content with the fact that federal spending has grown from 18.4 percent of GDP in 2000 to 20.2 percent, accounting for 69 percent of the deficit. Meanwhile, revenues are already back up to at least 17.4 percent of GDP — virtually the same as the 17.6 percent figure in 1993, after President Clinton greatly increased tax rates on upper incomes, gasoline and Social Security benefits.

As for Mr. Krugman’s hasty dismissal of the Laffer Curve, he has much to learn from the new paper "Dynamic Scoring" by N. Gregory Mankiw, a recent chairman of the President’s Council of Economic Advisers, and Matthew Weinzierl, also of Harvard ( This paper uses a well-established "neoclassical growth model to examine the extent to which a tax cut pays for itself through higher economic growth." The authors explicitly "ignore any short-term effects of tax cuts that arise from traditional Keynesian channels."

Assuming quite conservatively that tax rates on labor and capital are only 25 percent, and employing a conventional model of economic growth, Mankiw and Weinzierl find that "a capital tax cut has a long-run impact on revenue of only 47 percent of its static impact. That is, growth pays for 53 percent of the static revenue loss. A labor tax cut has a long-run impact on revenue of only 83 percent of its static impact, and growth pays for 17 percent of the tax cut."

Some economists estimate that actual taxes on capital are closer to 40 percent, in which case a cut in capital taxes would lose only a fourth as much revenue as conventional budget estimates predict. Some economists figure labor supply is far more responsive to tax rates than commonly assumed, in which case added growth would pay for 30 percent of a cut in labor taxes. Others emphasize "positive externalities" (more investment in widgets fosters related investments in gadgets and gizmos). In that case, Mankiw and Weinzierl estimate that "growth pays for 74 percent of a capital tax cut and 19 percent of a labor tax cut."

As impressive as this is, there is nonetheless much more to the impact of taxation on economic growth than can be captured by a mechanical model designed in 1928. Research by another former CEA chairman, Glenn Hubbard of Columbia University, emphasizes the effect of marginal tax rates on entrepreneurship. Yet another former CEA chairman, Martin Feldstein, demonstrates that income reported to the IRS by high-income taxpayers is extremely sensitive to changes in marginal tax rates. Yet entrepreneurship and tax avoidance are not all that is missing from the Mankiw-Weinzierl calculations.

When I borrowed the phrase "supply side fiscalism" from Herb Stein in March 1976, and suggested it to Jude Wanniski at The Wall Street Journal, we had in mind a number of incentives that have since become associated with a half dozen Nobel laureates in economics, even aside from Wanniski’s brilliant mentor Bob Mundell (who won the 1999 Nobel).

Among other Nobel Laureates, Ed Prescott (2004) emphasizes the impact of labor taxes on work incentives. Bob Lucas (1995) emphasizes tax incentives to invest in physical capital. James Heckman (2000) and Gary Becker (1992) emphasize the way progressive tax rates weaken incentives to invest in schooling and on-the-job training. And the optimal tax theory of James Mirrlees (1996) and Joe Stiglitz (2001) emphasizes both social welfare and tax-revenue (Laffer Curve) gains from low marginal tax rates on highly skilled individuals.

The original supply-siders combined all these effects, not just one or two. Paul Krugman has yet to figure out even one.