Fiscal stimulus, beloved by Keynesians, is not only expensive but causes long-term harm to the economy by distorting business incentives. The hundreds of billions of dollars pumped into the economy go, often as not, to cronies and industries chosen by politicians, propping up politically connected businesses at the expense of more efficient ones.
This practice is not sustainable.
A Keynesian will attempt to justify all of these costs—decisions made by elites at the expense of the consumer—and say that they’re worth it. Why? Because fiscal stimulus cures recessions. Paul Krugman, addressing the just-breaking Great Recession in late 2008, said, “Increased government spending is just what the doctor ordered.”
But the best reason to oppose fiscal stimulus is that it does just the opposite of what Krugman claims. It doesn’t cure recessions; it exacerbates them.
Making Recessions Worse
Libertarians haven’t explored this angle enough, because up until recently the research just hasn’t been available to support the assertion. But as I explain here, 100 years of history show that stimulus quantitatively makes recessions worse.
In that paper, I start with research done by Christina Romer, former chair of Obama’s Council of Economic Advisers and coauthor of Obama’s 2009 plan for recovery. In 1999, Romer created a measure of the severity of recessions. The idea, in simple terms, is to add up how much industrial production was lost from one peak until the economy got back to that level. Add up the shortfall for each month between those two points, and you have one number—percentage-point months (PPM) lost—that tells you how deep that recession cut.
Since she published the paper in 1999, she did not include data for the 2000–2001 recession nor the 2008 recession. I was able to ballpark the former and I used Krugman’s own figure (which even he says is probably a little low) for the latter.
What I found was that Keynesian thinking has made recessions less frequent, but more painful and durable.
The Body Economic
If you imagine that the economy is like a person, then a recession would be our need for sleep. It’s natural and normal to sleep, just like recessions are a natural market self-correction. Fiscal stimulus works like downing a Red Bull every time you need to sleep. Doing so lets you stay awake a little longer. But eventually you’re going to have to sleep, and your crash will be much worse than if you had just let your body rest instead of trying to counter that instinct with a stimulant.
In the real world, here’s how that looks: Back before the Federal Reserve and Keynes’s ideas, the United States was a virtual Austrian paradise (at least by comparison). The approach to recessions was hands off, for the most part letting the market self-correct. If the body is tired, let it sleep. Back then, the United States averaged about 1 recession every 3.9 years. They were frequent, but they were also pretty minor and over fast.
But in the 1930s, stimulus became the medicine of choice to handle economic downturns. Successive presidents, from Eisenhower to Kennedy to Obama, have used variations of fiscal stimulus to try to counter every recession. And the result has, admittedly, been fewer recessions: one every 5.7 years, compared to one every 3.9 in the “Austrian” years (before 1913). But these recessions have been longer, deeper, and more damaging than anything we encountered in the pre-government intervention years. The 2009 crash, for instance, is already over 1.5 times worse than the worst recession on record before 1913.
In my paper, I examine the net effect of all of these recessions on the United States economy. By looking at three factors (unemployment rate, lost industrial production, and GNP), it’s possible to objectively measure most recessions in terms of the damage they do to the country. These measurements aren’t perfect, but they’re good. And it turns out that, objectively measured, recessions in the Keynesian years do 12.5 percent more damage per year (not just years of recession, but total years in history) than those in the “Austrian” years.
So to go back to the sleeping metaphor, the Keynesian approach means you sleep less often, but you sleep so much more when you do finally crash that you spend about 12.5 percent more time overall asleep than if you’d avoided stimulants in the first place.
Cure or Disease?
Fiscal stimulus is also promoted to us as a way to cure recessions, not just fend them off. In 2008, an economist at the Daily Kos proposed a stimulus of $1 trillion to $1.6 trillion to combat the looming Great Recession. Keynesians across the board argue that fiscal stimulus is so important that it justifies going deeper into debt, because it’s a cure-all for our economic woes. It is significant, then, to show mathematically that fiscal stimulus makes worse the very problem it aims to solve.
Even if stimulus were free, it would still be a bad idea because it makes recessions as a whole more painful. And, of course, it’s not free. The New Deal, the original stimulus package, cost $542 billion (in 2013 dollars). The stimulus passed by Presidents Obama and Bush to cope with the 2009 recession totaled over $1.1 trillion. The “American Jobs” act President Obama proposed in 2011, if passed, would have cost another $447 billion.
There is a second cost of fiscal stimulus: It distorts incentives. Any stimulus bill is composed of bloated contracts to make bridges or new energy or—as Keynes famously suggested—to dig and fill holes in the ground. Government takes it upon itself to pick and choose which industries and which companies should receive these contracts, which are often big enough that they can make or break a company. These contracts go to firms that are well-connected politically, rewarding companies that spend money on lobbyists instead of making a better product. In the long run, this malinvestment hurts the economy overall.
Fiscal stimulus already has significant costs. Defenders will acknowledge these, but justify them as the cost of making the economy better. Unfortunately for us all, stimulus just makes the economy worse.