An Experiment in Austerity

Bruce Bartlett held senior policy roles in the Reagan and George H.W. Bush administrations and served on the staffs of Representatives Jack Kemp and Ron Paul.

Today’s Economist

Perspectives from expert contributors.

One reason that economics is not a true science is that economists can’t really perform experiments in which certain details are altered to see if they change the results and in what way. Of course, small experiments can be done using student volunteers playing economic games, but we can’t rerun the 1980s without the Reagan tax cut, which was signed into law 30 years ago this Saturday, and see if we would have gotten slower growth. Economists do their best to simulate alternative scenarios, but it’s not the same thing as rerunning the same history under different sets of policies.

Perspectives from expert contributors.

But once in a while economists get to observe a natural experiment where policy is changed in a way that allows a theory to be tested by real-world results. We are about to run just such an experiment by tightening fiscal policy and cutting spending significantly below baseline projections at a time when the economy is weak. As I pointed out in a July 12 post, we performed this experiment once before, in 1937. The budget deficit was sharply reduced and a recession immediately followed.

Since the beginning of our recent economic woes, there has been a theoretical debate among economists about the proper governmental response. I think it is reasonable to say that most supported the idea of fiscal stimulus in principle, although there certainly were different opinions about the size and structure of the February 2009 Recovery Act. But there were also conservative economists arguing that the stimulus was a bad idea for the same reasons they urged Franklin D. Roosevelt to balance the budget in 1937. Funds that the government borrows would be better invested by the private sector, they said.

Conservatives lost the first round, but have launched a counterattack. Their primary argument is that Congress enacted a $787 billion stimulus package and two and a half years later the economy is still in neutral. Stimulus supporters, like this paper’s columnist Paul Krugman, have long argued that the stimulus was too small to do more than moderate the downturn and wasn’t large enough to offset the negative economic momentum that the Obama administration inherited.

To average people, however, $787 billion sounds like a lot of money; more than enough to do the job if the theory on which it was based was sound in the first place. Consequently, many people have concluded that the stimulus experiment has failed and are receptive to the conservative theory that fiscal contraction may do a better job of jump-starting growth.

A June NBC News/Wall Street Journal poll asked if a cut in government spending would help or hurt the economic recovery; 37 percent said it would help, 31 percent thought it would hurt and 23 percent said would have no effect. A July Washington Post/ABC News poll asked people if they thought large cuts in federal spending would do more to create jobs or cut jobs; 47 percent said it would create jobs, with 44 percent saying it would cut jobs.

Space prohibits a full discussion of the theory and data underlying the debate among economists on what the economic effects of fiscal consolidation will be. Prominent economists at respected institutions have been discussing the effects for some time with complicated results.

The European Central Bank has published a number of papers over the last several years strongly endorsing the idea that fiscal contractions can be expansionary. Its latest study was published on July 12 and finds that the economy’s reaction to higher spending depends critically on expectations. If it is expected to be reversed in the future it is expansionary, but if it is expected to continue indefinitely then it is contractionary. If the results are symmetrical, then cuts in spending viewed as temporary may be contractionary while those viewed as permanent may be expansionary.

The International Monetary Fund has also published a considerable amount of research on this topic. Its latest study was published on July 6 and found that analyses which show fiscal contraction to be expansionary have misspecified the relevant changes in the fiscal balance. They lumped deficit reductions resulting from nonpolicy-driven factors, like stock market booms that raised government revenues, together with those explicitly engineered by policy makers. The I.M.F. study concluded that when the analysis was limited to policy-driven fiscal contractions, they were all economically contractionary. It found that a 1 percent of gross domestic product fiscal consolidation reduces real private consumption by 0.75 percent and real G.D.P. by 0.62 percent over the next two years.

A June 16 paper for the Bank for International Settlements found that the principal stimulus from fiscal consolidation came from lower interest rates and exchange rate depreciation, which boosted exports. Therefore, if interest rates are already quite low and depreciation is not possible, then there may not be any mechanism whereby fiscal consolidation can be expansionary.

The Congressional Research Service explained in a June 6 report that the economic effects of fiscal consolidation depend a great deal on the point in the business cycle in which it is carried out. Those occurring when the economy is operating close to its potential were successful by freeing public resources for private use, exactly as conservatives say, but those that took place when there was a large amount of unused capacity were not. The study concluded, “Fiscal adjustments beginning in a slack economy (like the current situation in the United States) appear to have a low probability of success.”

My point is not to exhaust the issues involved in calculating the economic effects of fiscal contraction; only to show that it is complicated. Readers should be skeptical of simplistic studies that cherry pick their examples, fail to control for variables other than the deficit as a share of G.D.P. and don’t make any effort to explain examples that don’t prove their point. A good place to start any analysis is a recently compiled data set of fiscal consolidation episodes by the I.M.F.