The dubious impact of the stimulus

By Philip Zelikow

The roots of the financial crisis may have been the fall in aggregate demand, and globally, the fundamentals for a recovery of aggregate demand may already be there, waiting for the credit hydraulics to recover.  Still, the United States is not the ideal candidate to lead a global spur in aggregate demand. After all, before the crisis, the United States already had weak savings, large and persistent current account deficits, and a debt to GDP ratio which, though it had not yet reached an alarming level, was too large for comfort.

The impact of massive stimulus is highly uncertain. The debaters are not likely to acknowledge how weak the evidence is on this. In October 2008, the IMF summarized that: "Perhaps surprisingly, the empirical literature on the effects of fiscal policy does not provide a clear answer to the simple question of whether discretionary fiscal policy [as opposed to automatic stabilizers like unemployment benefits] can successfully stimulate the economy during downturns." 

If the rise in aggregate demand is diffused globally, the effects will be more diffuse. If the Congress attempts to impose "buy America" restrictions on the money, then another very dangerous set of problems will arises.

Rather than measure impact on aggregate demand, some advocates focus on the use of U.S. government spending to create U.S. jobs. The job creation numbers tend to extrapolate a rule of thumb, that every billion dollars of government spending or investment creates about 10,000 jobs.  So to create 3 million extra jobs in 2009 and 2010 quickly — the goal the Obama team has announced — this back-of-the-envelope arithmetic yields a goal of about $300 billion extra spending in each of those two years. 

I am uneasy about the validity of relying on such gigantic macro-extrapolations. There is, even now, deep and unresolved debate about the effect of fiscal policy on US recovery in the most well-studied, classic case, the U.S. economic revival between 1938 and 1942.

It is important to be fair to the proponents: there is evidence that fiscal stimulus can have positive effects. But the details are important. In a good case, with the right associated conditions, the IMF’s review of the data suggests that spending an extra 2 percent of GDP to revive the economy in 2009 and another 2 percent again in 2010 (for a total of $600 billion) might add about 0.2 percent of GDP spur on impact in the first year, rising to a cumulative injection of 2 percent of GDP improvement after 3 years.  Thus, in this extrapolation, the U.S. economy might be growing at a rate of about 1.5 percent more than would otherwise be the case by 2012, and 2 percent hotter in 2013. This is potentially positive reinforcement for the administration’s case.  

But then notice two things: 

One, the lag effect in the best case, heating up an economy which may already be in recovery. If there is a lot of pent-up global aggregate demand and since, in the current case, the monetary gas pedal has been floored, we could be setting up another "great inflation" in the medium-term. On this, check out Robert Samuelson’s current book, which recounts the important, suggestive story of the 1960s and 1970s.  

And two, partly from anticipating just this danger, if the U.S. moves into high debt to GDP ratios, the IMF data suggests the effects of the stimulus probably won’t be positive at all. Instead, per the IMF, the effects of the stimulus could become "consistently large and persistently negative," as doubts grow about fiscal sustainability.