Commentary

Hello Supply Side

In Goodbye Supply Side, Kevin D. Williamson writes, “Properly understood, there were no Reagan tax cuts. In 1980 federal spending was $590 billion and in 1989 it was $1.14 trillion; you don’t get Reagan tax cuts without Tip O’Neill spending cuts. Looked at from the proper perspective, we haven’t really had any tax cuts to speak of — we’ve had tax deferrals.”

Williamson hopes to “develop a rhetoric in which ‘spending’ and ‘taxes’ are synonyms, so a federal budget with $1 trillion in new spending means $1 trillion in new taxes — levies on Americans today or on our children tomorrow, with interest.” I am sympathetic to the description of deficit spending as deferred taxation. But equating supply-side tax policy with lower revenues (“tax cuts”) is incorrect in theory and fact.

The trouble with what Williamson calls “the bottom-line question of balancing the budget” is the implication that it makes no difference whether the budget is balanced by curbing spending or increasing taxes. A 2002 study by Alberto Alesina of Harvard and three colleagues found that the surest way to make economies boom is through deep cuts in government spending. Higher tax rates had the opposite effect. Although “growth isn’t going to make the debt irrelevant,” as Williamson says, the ratio of debt to GDP can become much larger and more troublesome if GDP stagnates.

[E]quating supply-side tax policy with lower revenues (“tax cuts”) is incorrect in theory and fact.”

Williamson credits the 1985 Gramm-Rudman-Hollings Act with cutting the deficit from 5.2 percent of GDP in 1986 to 2.8 percent in 1989. Yet the largest contributor to that deficit reduction was actually the Tax Reform Act of 1986, which increased income-tax revenue from 9.3 percent of GDP to 10.2 percent even as revenues from the higher capital-gains tax fell by $17.7 billion. Defense spending (which had risen sharply in 1980-86) was also trimmed from 6.5 percent of GDP to 5.8 percent from 1986 to 1989. Non-defense spending accounted for only a third of the deficit reduction.

Deficit reduction under President Clinton happened mainly because defense spending fell from 5 percent of GDP to 3.1 percent, and the Fed’s low interest rates reduced federal interest expense by another 1 percent of GDP. Clark Judge rightly noted that most of the revenue gains in 1997-2001 were the result not of the higher tax rates that went into effect in 1993 but of the lower capital-gains-tax rates that went into effect in 1997. From 1987 to 1995, revenues from the 28 percent capital-gains tax accounted for only 7.3 percent of individual tax receipts. From 1997 to 2000, revenues from the 20 percent capital-gains tax accounted for 11.9 percent of individual tax receipts. The dot-com boom certainly helped. Yet the evidence is overwhelming that a 20 percent tax on realized capital gains generates more revenue than a 28 percent tax.

Williamson says “tax cuts aren’t really the problem” — but the real problem is the threat of European-style tax increases. If we accepted his advice to ignore all the evidence behind supply-side tax reforms, we would be left with no credible defense against Obama’s plans to pile surtaxes on top of surtaxes for investors and high-income families, or against a VAT. If lower tax rates in the 1980s were useless, then higher tax rates in the next year or two must likewise be harmless.

Williamson claims the “proper perspective” looks only at federal spending, treating spending and taxing as synonymous, regardless of tax rates or even tax revenues. This leads him to define money-losing “tax cuts” as a “poorly applied supply-side analysis.” On the contrary, George W. Bush’s advocacy of tax credits and rebates to “put money in peoples’ pockets” was demand-side, Keynesian analysis.

Supply-side economics in 1971-87 was largely about a policy mix to end stagflation: Using monetary policy to reduce the growth of nominal GDP while using tax incentives and deregulation to raise the growth of real GDP. The tax side was always focused on microeconomics — the incentive effects of marginal tax rates.

The burden of government depends on spending, but the “excess burden” or “deadweight loss” of tax distortions depends on marginal tax rates. The size of this excess burden grows with the square of the tax rate, making a 40 percent tax rate 16 times more damaging than a 10 percent rate. Recent estimates find at least a dollar of damage to the private economy for each additional dollar raised through today’s progressive taxes.

Supply-side tax policy is about raising tax revenues in ways that do the least damage to the private economy. Spending cuts are entirely consistent with supply-side theory because (1) transfer payments that phase out with rising income are a disincentive to work and (2) government purchases divert real resources into unproductive uses. One reason David Stockman invited me to join the OMB team in 1981 was that I had enumerated $51.4 billion in spending cuts (10 percent of the budget) in an August 1978 article in Fortune, “Curbing the federal spending spree.”

Supply-side economics has nothing to do with tax cuts per se. In his book Impostor, Bruce Bartlett said that I “spoke for most supply-siders” in 2001 when I wrote in the Wall Street Journal that the primary objective of the Bush tax cuts “seems to have been to maximize revenue loss rather than to minimize tax distortions and disincentives.” Supply-side economists criticized the Bush tax cuts as wasteful, and the phasing-in of lower tax rates as positively harmful. University of Michigan economists Christopher House and Matthew Shapiro, in the American Economic Review of December 2006, “attribute the slow recovery from the 2001 recession, in part, to declines in labor supply stemming from the phased-in nature of the tax cuts. Additionally, the rebound in economic activity in mid-2003 coincides with the removal of the phase-ins enacted in the 2003 tax bill. A comparison of the simulated and actual time series over this time period shows that about half of the rebound in GDP in mid-2003 can be attributed to the elimination of the phase-in of the tax cuts.”

Revenue from top-bracket taxpayers, dividends, and capital gains all soared after those tax rates came down in mid-2003. Offsetting the supply-side revenue gains, however, were big losses from reducing the lowest rate to 10 percent and adding various tax credits. Yet revenues in 2006-2007 were nonetheless 18.2-18.5 percent of GDP — the same as in 1977-79, when the top tax rate was 70 percent, or 1987-89, when to top tax rate was 28 percent.

Reynolds’ Law finds that, except during recessions and the tech-stock boom, individual income taxes are a nearly constant fraction of GDP (8-9 percent), regardless of whether top tax rates are 70 percent or 28 percent. That means future growth of real income tax revenues will depend entirely on the growth of real GDP. If spending grows faster than GDP, then deficits and debt will indeed rise. Because higher tax rates on the rich damage GDP growth without helping the Treasury, however, they do not provide (as Obama pretends) a realistic alternative to serious spending restraint.

In a recent Wall Street Journal article, I showed that the current consensus in economic research suggests that raising marginal tax rates on high incomes and capital gains is unlikely to raise revenue — regardless of the effect on GDP growth. The higher the tax rate, the less taxable income ends up being reported to the IRS.

Williamson says, “When the Reagan tax cuts were being designed, the original supply-side crew thought that subsequent growth might offset 30 percent of the revenue losses. That’s on the high side of the current consensus.” On the contrary, the current consensus is that the typical effect of lower marginal tax rates on revenues will offset at least 40 percent of the losses, and perhaps more than 100 percent at very high incomes. And that is without taking into account any specific impact on economic growth, which would make the offset even greater.

Williamson relies on the unreliable Congressional Budget Office (CBO) for evidence “about how large supply-side revenue effects are.” The CBO does not do revenue forecasting, so they borrowed two Keynesian forecasting models which, the CBO confessed, “are designed primarily to capture short-run business-cycle developments.” The results, I showed, were contradictory and confused.

Federal spending is indeed approaching crisis proportions, but Williamson may not entirely right that “nobody wants to touch it.” Wisconsin congressman Paul Ryan, a favorite supply-side ally of our good friend Jack Kemp, is not shy about proposing large and specific federal spending cuts. Neither is Indiana governor Mitch Daniels, who once recruited me to the Hudson Institute.

If cutting top tax rates from 70 percent to 28 percent under President Reagan did not constitute a tax cut because defense spending went up for a few years, then raising top tax rates under President Obama would be equally irrelevant if spending were unchanged. The only “real” tax cut, in this view, was the post-Soviet peace dividend under Bill Clinton. If that is “the stuff that a broad-based political movement is going to put at the center of its campaigns,” it doesn’t sound too promising.

Alan Reynolds, National Review’s economics editor from 1972 to 1976, is a senior fellow with the Cato Institute.