Sunday, October 5, 2008

The Fed's Contribution to the Financial Crisis

Mark Thoma points us to an interesting take by Nick Rowe on the origins of the financial crisis. Among other things, Rowes' account speaks to the role the Federal Reserve played in creating the financial crisis. Here is a key excerpt:
So why did bubbles happen in so many countries at about the same time? First, because world interest rates were low. And world interest rates were low not because Greenspan set them low (central banks cannot set interest rates below the natural rate without causing accelerating inflation), but because world savings were high.
So Rowe has drank the "saving-glut" Kool Aid and sees no culpability for the Fed. His justification for this conclusion is that since inflation was not accelerating in the early 2000s it must be that the Fed did not push interest rates below the natural interest rate level. A stable inflation rate, however, is not a sufficient condition for macroeconomic stability. This is evident when one considers that the inflationary effects of an overly accommodative monetary policy--a positive aggregate demand shock--can be temporarily masked by gains to aggregate supply. A close look at the data seems to indicate this was case in 2003-2005: productivity growth accelerated just as the fed pushed interest rates to floor. The figure below illustrates this point. It shows the year-on-year growth rate of productivity versus the real (ex-post) federal funds rate (click on figure to enlarge):

It is worth noting that higher productivity growth rates are typically associated with higher real interest rates, not lower ones. When productivity growth--which provides a proxy for profitability--diverges so far from the real federal funds rate--the cost of borrowing to invest in this profitability--as it did during this time a credit boom is inevitable. Given the Fed's monetary hegemon status, this credit boom went global and created a global liquidity glut.

Clearly, there was more than just loose monetary policy behind this financial crisis (see Barry Ritholtz for a nice summary). But as I have argued here on this blog and elsewhere, inordinately loose monetary policy between 2003-2005 played a key part in stoking the coals of the housing boom during that time.


  1. You might be right (though a different explanation for the low interest rates would not affect the rest of my analysis). But I still think the savings glut was the main cause.

    First, housing bubbles seem to have been just as big in countries that are not part of the US dollar bloc. A temporary reduction in US interest rates, below the world natural rate, should cause a smaller reduction in interest rates in those other countries (and thus smaller or fewer bubbles).

    Second, the world natural rate is determined by world productivity growth (and other things), not just by US productivity growth. An increase in productivity growth in one country does not mean that the central bank in that country should raise interest rates. Rather, an increase in world productivity growth (regardless of where it occurs) would mean that all central banks should raise interest rates.

    I'm not married to the savings glut vs loose world monetary policy as the main cause of low interest rates. But you haven't quite convinced me yet.

    Nick Rowe

  2. Nick:

    I agree to make my case there needed to have been a positive global productivity shock,or more generally a positive global aggregate supply (AS) shock. I believe there were a series of positive global AS shocks buffeting the world at this time: (1) the opening up of China, India, and other formerly closed economies and (2) the ongoing technological gains. These positive global AS shocks should have resulted in a higher global natural rate and lower global price level. Instead, the Fed inordinately eased and, given its monetary hegemon status, caused a global liquidity glut. This development pushed interest rates below the natural rate across the globe.

    I have an article forthcoming in the next issue of the Cato Journal that more fully develops this argument. You may also want to see this posting where a similar argument is made.

    This interpretation, of course, does not change the outcome of your analysis as you note. Best wishes in your presentation.