Conservatives need a better argument than Laffer

Bruce Bartlett

3/29/2006 12:04:00 AM - Bruce Bartlett

Belatedly, Republicans in Congress have become concerned about the federal budget deficit. But this is making it harder for them to find the votes to extend previously enacted tax cuts, some of which expire in 2008, and all which disappear after 2010. Tax-cut supporters are now arguing that failure to extend the tax cuts will actually cause revenues to fall, thereby defeating the goal of deficit reduction.

As the Wall Street Journal put it on March 23, "The revenue data are further proof of the success of the Bush tax cuts of 2003. The fastest way to stop this revenue windfall, and blow an even larger hole in the deficit, would be to fail to extend the 15-percent tax rate on capital gains and dividends through 2010, thus assuring a huge tax increase after 2008."

The flip side of this argument is that the 2003 tax cuts are said to be raising federal revenue because of a Laffer-curve effect. The Journal cited the "astonishing" 14.6-percent increase in federal revenue in 2005 over 2004, which it rounded up to 15 percent. It also noted that revenues in the first five months of fiscal year 2006, which began last Oct. 1, are up 10.3 percent over the same period in fiscal year 2005. No other evidence was offered.

But how likely is it that the Laffer curve is causing revenues to rise, as opposed to normal operation of the business cycle? Not much, in my opinion.

First of all, the Laffer curve came to prominence during a period when the top tax rate on dividends was 70 percent, and the rate on long-term capital gains was 40 percent. Economist Arthur Laffer correctly pointed out that a 100 percent tax rate would raise no revenue and that rates close to this would reduce revenue below what a lower rate would bring in. Given the tax rates in existence, it was plausible to argue that a reduction in the top rate and capital gains tax would raise revenue.

However, when President Bush took office, the top rate on dividends was down to 39.6 percent, and the rate on long-term capital gains was just 20 percent -- far below the rates Ronald Reagan inherited. It is very implausible that these rates were in the "prohibitive" range of the Laffer curve, such that a rate reduction would raise revenue.

But even if we grant the theory, how likely is it that the recent rise in revenue owes anything to this effect? Again, not much.

The fact is that it is only in very exceptional circumstances that there would even be the possibility of a tax cut that would so stimulate growth that it would pay for itself. Even the Bush Administration admits this. The 2003 Economic Report of the President  says, "Although the economy grows in response to tax reductions … it is unlikely to grow so much that lost tax revenue is completely recovered by the higher level of economic activity."

Recent academic research suggests that feedback effects would offset only a fraction of the static revenue loss, that which would result from no effect on consumption or incentives. A 2004 study by Harvard economists N. Gregory Mankiw and Matthew Weinzierl found that a cut in taxes on capital might recoup 17 percent of the static revenue loss in the first three years and a cut in taxes on labor could recoup 13 percent. Mankiw served as chairman of the Council of Economic Advisers under President Bush. A study by the Congressional Budget Office in December 2005 found that a tax-rate cut would recoup at most 20 percent of the static revenue loss in the first five years.

Overall, federal revenues are just barely back to where they were five years ago in nominal terms. According to the CBO, federal receipts were $2,025.5 billion in 2000 and $2,153.0 billion in 2005. Revenues fell 1.7 percent in 2001, fell another 6.9 percent in 2002 and fell again by 3.8 percent in 2003. They didn't start to bounce back until 2004, when they rose by 5.5 percent.

Revenues as a share of the gross domestic product fell every year from 2000 to 2004, from 20.9 percent to 16.3 percent. The 2005 increase only raised revenues to 17.5 percent -- still well below their historical average of 18.1 percent of GDP. It seems to me that the normal cyclical expansion after the end of the recession in 2001 has done far more to raise revenue than any Laffer curve effect. Revenues are simply returning to trend, nothing more.

In short, there is very little likelihood that revenues are rising because the 2003 tax cuts or would fall if they are not extended. The case for extending them must be made on other grounds.