Starving the Beast

Does cutting taxes to reduce the size of government work? In today's LA Times, Jonathan Chait takes obvious glee in Bill Niskanen's claim that it does not work and that, in fact, it backfires. According to Niskanen,

In a professional paper published in 2002, I presented evidence that the relative level of federal spending over the period 1981 through 2000 was coincident with the relative level of the federal tax burden in the opposite direction; in other words, there was a strong negative relation between the relative level of federal spending and tax revenues. Controlling for the unemployment rate, federal spending increased by about one-half percent of GDP for each one percentage point decline in the relative level of federal tax revenues.

Some of Niskanen's regressions were recently presented and critiqued by economist Mark Thoma.

Here's a puzzle: Niskanen's claims seem patently inconsistent with this study by Henning Bohn from the Journal of Monetary Economics (one of the premier academic journals of macroeconomics):

The paper provides a historical perspective on the issue of whether budget deficits are typically eliminated by increased taxes or by reduced spending. By examining U.S. budget data from 1792–1988, I conclude that about 50–65% of all deficits due to tax cuts and about 65–70% of all deficits due to higher government spending have been eliminated by subsequent spending cuts, while the remainder was eliminated by subsequent tax increases.

Niskanen's results suggest that the "starve the beast" hypothesis is completely wrong, whereas Bohn's results suggests there is a large degree of truth to it.

There are many differences in the two studies, including both data and methodology. Regarding data, Bohn includes much more of it, which is a plus, but he ends earlier, omitting the most recent experience, because his study was published in 1991. Regarding methodology, Bohn explicitly incorporates the government's intertemporal budget constraint into the analysis, whereas Niskanen apparently does not. This seems like a shortcoming of Niskanen's approach.

Until someone sorts out these apparently conflicting results, it is (as Thoma suggests) premature for anyone (like Chait) to conclude that Niskanen has the last, or even the most persuasive, word on the topic.