Do We Want Civil Engineers to Predict Bridge Failures?

No.  We want them take whatever steps necessary to prevent bridge failures. At least if given the tools and resources to do so. We now have an economist leading the Federal Reserve, and have given the Fed almost unlimited power to print currency and buy tens of trillions of dollars worth of various assets. We have also given them the responsibility stabilizing the macroeconomy using the best economic advice available. Do we really want economists to be able to predict recessions? As James Hamilton noted:

“You could argue that if the Fed is doing its job properly, any recession should have been impossible to predict ahead of time.”

Strictly speaking, Hamilton’s observation only applies to recessions caused by demand shocks, but in practice, supply shocks are also pretty unpredictable.

I ran across Hamilton’s observation in early 2008 while working on a paper arguing that economists would never again be able to forecast recessions. I had assumed the Fed now followed Svensson’ criteria–i.e. always set monetary policy in such a way that the forecast equals the policy target. And since the Fed presumably doesn’t target recessions, they should be unforecastable as well.

Little did I know that events would discredit my hypothesis within months. Not because economists successfully forecast this recession (the consensus didn’t see a recession until a couple of months after it had begun) but because the Fed clearly no longer follows Svensson’s criteria–by late 2008 nominal growth forecasts for 2009 were far below what the Fed would like to see.

Hasn’t this always been true of recessions?  No.  In the new Keynesian era (1982-2008) the few recessions we did have were almost over before we realized they had even begun. Once economists recognized the recession was underway, there was confidence that gradual recovery would occur over the next 12 months. This is different.

The public always complains about economists not being able to predict recessions. Be careful what you ask for. Now we are predicting 2009 nominal growth far below anything seen in decades.

When people used to ask me whether Fed policy was too easy or too tight, I usually said “I don’t know.” By October 2008 I was able to answer that question with confidence. That’s when I knew we were in trouble.

Tags: Forecasting

This entry was posted on February 03rd, 2009 and is filed under Forecasting, Monetary Policy. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response or Trackback from your own site.