By Philip I. Levy Thursday, February 18, 2010
Filed under: Government & Politics, Economic Policy
This week marks the first anniversary of President Obama’s stimulus legislation, one of the more controversial—and unpopular—bills in recent memory. In a New York Times piece marking the anniversary, David Leonhardt attempts to explain to the unwashed masses how foolish they are to scorn the stimulus. He attempts to enlighten the benighted public, but his efforts are marred by a straw man and twisted logic.
First, the straw man: “The case against the stimulus revolves around the idea that the economy would be no worse off without it.”
Actually, for $862 billion, the hurdle would be a bit higher than merely having some positive effect. In a time of looming fiscal crisis, with all sorts of pressing national needs, we care about bang for our buck.
There is no doubt that if you “make it rain” with billions of borrowed dollars, it will have some effect. Suppose we spend billions to shift car sales from the fourth quarter of 2009 to the third. One independent estimate found that the “cash for clunkers” program cost $24,000 per additional car sale induced. Was that money well spent?
This brings us to the heart of the difficulty with assessing the stimulus. You need to grapple with economic models. What is an “additional car sale induced”? It’s a car sale that would not have taken place without the program’s incentives. But how do we know what would have happened? We model it. We come up with some projection based on past data and our understanding of how the world works.
Leonhardt puts forward some models that sing the praises of the stimulus bill and the millions of jobs it has created. He is dismissive of the misguided “hard-core skeptics” whose models show a much smaller effect (he points out that they work in an institution with the name “Hoover.” Hoover! Get it?). Here’s one way to restate this section of Leonhardt’s argument: the Keynesian macro models that predicted the stimulus would work still predict that the stimulus worked. These are not new, independent verifications.
We have at least two reasons to doubt these models. Remember that we are relying upon them for their forecasting accuracy. We need them to tell us just what would have happened had we failed to act. The Keynesian macro models used by the Obama administration predicted that without stimulus, the unemployment rate would be around 9 percent right now; with it, the rate would be below 8 percent. The actual rate has hovered around 10 percent. The administration has countered that conditions were worse than they had anticipated. Here’s one way to restate this argument: Don’t blame us; our models weren’t very good. If a model is accurate to within a couple of percentage points, that means that with a labor force of roughly 154 million, it should be able to measure jobs created, give or take 3 million or so.
The second, older reason to doubt the Keynesian models goes back to the stagflation of the 1970s. In the Keynesian approach, you either pushed your economy too fast and got inflation, or you let it slow too much and got unemployment. If unemployment was caused by excessively weak demand, how could you have inflation at the same time? And yet, in the late 1970s we had both—high unemployment and high inflation. This was a big part of what originally interred the Keynesian approach so that it required revival. Its failure set off a flurry of macroeconomic work that tried to provide firmer foundations for understanding how the economy functions. The fruit of some of this research underlies the “hard-core skeptics.”
As a savvy journalist, Leonhardt knows how little the average reader likes to wallow in questions of economic modeling. So, he moves on to reasons to embrace the stimulus “above and beyond those complicated, independent economic analyses.” He offers two. He says you can look across the economy and see where the stimulus is having an effect, such as in state and local government spending. What’s more, if we look across the history of financial crises, we would have expected things to be even worse than they are now. It must be that “the stimulus package, flaws and all, deserves a big heaping of credit.”
This is the twisted logic part. In stepping away from complicated economic modeling, Leonhardt just turns to facile economic modeling. He implicitly assumes that none of the effects he sees in the economy would have happened otherwise. He implicitly assumes that there are no ill effects occurring elsewhere, such as government borrowing crowding out private borrowing or concerns about future tax increases. And he makes the seriously flawed assumption that the Obama stimulus bill was the only major economic force acting at the time.
In fact, there were multiple forces hitting concurrently in late 2008 and into 2009. Oil prices dropped dramatically. On the New York Mercantile Exchange, the price per barrel went from $145 in July 2008 to below $37 in mid-January 2009. The Federal Reserve undertook an enormous expansion of its balance sheet to boost the money supply. The Bush and Obama administrations both worked to unfreeze the banking sector. And there was fiscal stimulus through “automatic stabilizers”—those tax cuts and spending increases already baked in through previous legislation. Each of these would be expected to provide a major boost to the economy, though not instantaneously.
To argue that the grab bag of tax cuts, disbursements, and long-term spending projects in the American Recovery and Reinvestment Act of 2009 was the economy’s salvation, or even just money well spent, one would need to do much more than collect anecdotes in which a billion dollars buys you something. It is tempting to argue that in a time of slack demand government spending was the common-sense solution. But as someone once said, common sense is what tells us the world is flat.
Phil Levy is a resident scholar at the American Enterprise Institute.
Image by Darren Wamboldt/Bergman Group.
Read Full Article »