• Email
  • Single Page
  • Print

How to End This Depression

Interim Archives/Getty Images
An unemployed man selling apples during the Great Depression, circa 1930s

In short, the experience of Obama’s first term suggests that not talking about jobs simply because you don’t think you can pass job-creation legislation doesn’t work even as a political strategy. On the other hand, hammering on the need for job creation can be good politics, and it can put enough pressure on the other side to bring about better policy too.

Or to put it more simply, there is no reason not to tell the truth about this depression.

A Moral Imperative

It has been more than four years since the US economy first entered recession—and although the recession may have ended, the depression has not. Unemployment may be trending down a bit in the United States (though it’s rising in Europe), but it remains at levels that would have been inconceivable not long ago—and are unconscionable now. Tens of millions of our fellow citizens are suffering vast hardship, the future prospects of today’s young people are being eroded with each passing month—and all of it is unnecessary.

For the fact is that we have both the knowledge and the tools to get out of this depression. Indeed, by applying time-honored economic principles whose validity has only been reinforced by recent events, we could be back to more or less full employment very fast, probably in less than two years. All that is blocking recovery is a lack of intellectual clarity and political will.

But one question remains. I have argued that in a deeply depressed economy, in which the interest rates that the monetary authorities can control are near zero, we need more, not less, government spending. A burst of federal spending is what ended the Great Depression, and we desperately need something similar today.

Yet how do we know that more government spending would actually promote growth and employment? After all, many politicians fiercely reject that idea, insisting that the government can’t create jobs; some economists are willing to say the same thing. So is it just a question of going with the people who seem to be part of your political tribe?

Well, it shouldn’t be. Tribal allegiance should have no more to do with your views about macroeconomics than with your views on, say, the theory of evolution or climate change. The question of how the economy works should be settled on the basis of evidence, not prejudice. And one of the few benefits of this depression has been a surge in evidence-based economic research into the effects of changes in government spending. What does that evidence say?

Before I can answer that question, I have to talk briefly about the pitfalls one needs to avoid.

The Trouble with Correlation

You might think that the way to assess the effects of government spending on the economy is simply to look at the correlation between spending levels and other things like growth and employment. The truth is that even people who should know better sometimes fall into the trap of equating correlation with causation. But let me try to disabuse you of the notion that this is a useful procedure by talking about a related question: the effects of tax rates on economic performance.

It’s an article of faith on the American right that low taxes are the key to economic success. But suppose we look at the relationship between taxes—specifically, the share of GDP collected in federal taxes—and unemployment over the past dozen years. What we find is that years with high tax shares were years of low unemployment, and vice versa (see Figure 1). Clearly, isn’t the way to reduce unemployment to raise taxes?


Even those of us who very much disagree with tax-cut mania don’t believe this. Why not? Because we’re surely looking at spurious correlation here. For example, unemployment was relatively low in 2007 because the economy was still buoyed by the housing boom—and the combination of a strong economy and large capital gains boosted federal revenues, making taxes look high. By 2010 the boom had gone bust, taking both the economy and tax receipts with it. Measured tax levels were a consequence of other things, not an independent variable driving the economy.

Similar problems bedevil any attempt to use historical correlations to assess the effects of government spending. If economics were a laboratory science, we could solve the problem by performing controlled experiments. But it isn’t. Econometrics—a specialized branch of statistics that’s supposed to help deal with such situations—offers a variety of techniques for “identifying” actual causal relationships. The truth, however, is that even economists are rarely persuaded by fancy econometric analyses, especially when the issue at hand is so politically charged. What, then, can be done?

The answer in much recent work has been to look for “natural experiments”—situations in which we can be pretty sure that changes in government spending are neither responding to economic developments nor being driven by forces that are also moving the economy through other channels. Where do such natural experiments come from? Sadly, they mainly come from disasters—wars or the threat of wars, and fiscal crises that force governments to slash spending regardless of the state of the economy.

Disasters, Guns, and Money

As I wrote, since the crisis began there has been a boom in research into the effects of fiscal policy on output and employment. This body of research is growing fast, and much of it is too technical to be summarized in this article. But here are a few highlights.

First, Stanford’s Robert Hall has looked at the effects of large changes in US government purchases—which is all about wars, specifically World War II and the Korean War. Figure 2 on this page compares changes in US military spending with changes in real GDP—both measured as a percentage of the preceding year’s GDP—over the period from 1929 to 1962 (there’s not much action after that). Each dot represents one year; I’ve labeled the points corresponding to the big buildup during World War II and the big demobilization just afterward. Obviously, there were big moves in years when nothing much was happening to military spending, notably the slump from 1929 to 1933 and the recovery from 1933 to 1936. But every year in which there was a big spending increase was also a year of strong growth, and the reduction in military spending after World War II was a year of sharp output decline.


This clearly suggests that increasing government spending does indeed create growth and hence jobs. The next question is, how much bang is there per buck? The data on US military spending are slightly disappointing in that respect, suggesting that a dollar of spending actually generates only about fifty cents of growth. But if you know anything about wartime history, you realize that this may not be a good guide to what would happen if we increased spending now. After all, during World War II private-sector spending was deliberately suppressed by rationing and restrictions on private construction; during the Korean War, the government tried to avoid inflationary pressures by sharply raising taxes. So it’s likely that an increase in spending now would yield bigger gains.

How much bigger? To answer that question, it would be helpful to find natural experiments telling us about the effects of government spending under conditions more like those we face today. Unfortunately, there aren’t any such experiments as good and clear-cut as World War II. Still, there are some useful ways to get at the issue.

One is to go deeper into the past. As the economic historians Barry Eichengreen and Kevin O’Rourke point out, during the 1930s European nations entered, one by one, into an arms race, under conditions of high unemployment and near-zero interest rates resembling those prevailing now. In work with their students, they have used the admittedly scrappy data from that era to estimate the impact that spending changes driven by that arms race had on output, and have come up with a much bigger bang for the buck (or, more accurately, the lira, mark, franc, and so on).

Another option is to compare regions within the United States. Emi Nakamura and Jon Steinsson of Columbia University point out that some US states have long had much bigger defense industries than others—for example, California has had a large concentration of defense contractors, whereas Illinois has not. Meanwhile, defense spending at the national level has fluctuated a lot, rising sharply under Reagan, then falling after the end of the cold war. At the national level, the effects of these changes are obscured by other factors, especially monetary policy: the Fed raised rates sharply in the early 1980s, just as the Reagan buildup was occurring, and cut them sharply in the early 1990s. But you can still get a good sense of the impact of government spending by looking at the differential effect across states; Nakamura and Steinsson estimate, on the basis of this differential, that a dollar of spending actually raises output by around $1.50.

So looking at the effects of wars—including the arms races that precede wars and the military downsizing that follows them—tells us a great deal about the effects of government spending. But are wars the only way to get at this question?

When it comes to big increases in government spending, the answer, unfortunately, is yes. Big spending programs rarely happen except in response to war or the threat thereof. However, big spending cuts sometimes happen for a different reason: because national policymakers are worried about large budget deficits and/or debts, and slash spending in an attempt to get their finances under control. So austerity, as well as war, gives us information on the effects of fiscal policy.

It’s important, by the way, to look at the policy changes, not just at actual spending. Like taxes, spending in modern economies varies with the state of the economy, in ways that can produce spurious correlations; for example, US spending on unemployment benefits has soared in recent years, even as the economy weakened, but the causation runs from unemployment to spending rather than the other way around. Assessing the effects of austerity therefore requires painstaking examination of the actual legislation used to implement that austerity.

Fortunately, researchers at the International Monetary Fund have done the legwork, identifying no fewer than 173 cases of fiscal austerity in advanced countries over the period between 1978 and 2009. And what they found was that austerity policies were followed by economic contraction and higher unemployment.

There’s much, much more evidence, but I hope this brief overview gives a sense of what we know and how we know it. I hope in particular that when you read me or Joseph Stiglitz or Christina Romer saying that cutting spending in the face of this depression will make it worse, and that temporary increases in spending could help us recover, you won’t think, “Well, that’s just his/her opinion.” As Romer asserted in a recent speech about research into fiscal policy:

The evidence is stronger than it has ever been that fiscal policy matters—that fiscal stimulus helps the economy add jobs, and that reducing the budget deficit lowers growth at least in the near term. And yet, this evidence does not seem to be getting through to the legislative process.

That’s what we need to change.

  • Email
  • Single Page
  • Print