Sunday, April 20, 2008

Mark Thoma gives George Will a deservedly hard time, but..

I agree with Mark that for George Will to accuse someone else of elitism is too rich. For Bill Kristol also to do so is the ultimate in Chutzpah (I would call Will guilty of Chutzpah too, but I would guess George doesn't know what it means).

But Will's basic point about the Fed is not as ridiculous as it might seem. Will says:

The Fed's mission is to preserve the currency as a store of value by preventing inflation. ... The Fed should not try to produce this or that rate of economic growth or unemployment


My problem is with the use of the word "should." The more interesting question is whether the Fed can influence long-term economic growth or unemployment. Kydland and Prescott's model of credible commitment and central bank behavior (Journal of Political Economy 1977), which produced a Nobel prize in 2004, suggests that the Fed can only credibly commit to fighting inflation; that issues of time consistency prevent it from both fighting inflation and maximizing employment, and that if it commits to something other than zero inflation, it will produce socially undesirable levels of both inflation AND unemployment. This has spawned a literature on credible commitment on the part of central banks. And while I am not a macroeconomist (and won't play one on TV--when reporters ask me to forecast interest rates, I tell them to go away), my reading of the literature suggests that it is within the realm of possibility that Kydland and Prescott are correct.

1 comment:

Anonymous said...

I continued in comments:

To follow up a little bit:

Economists on all sides, e.g. Real Business Cycle and New Keynesian types, believe that the natural rate of output and employment, i.e. output and employment in the long-run, is independent of monetary policy. The real debate is over the short-run and whether monetary policy can be effective in terms of stabilizing output and employment.

There are three possibilities. First, monetary policy can be used to stabilize the economy in the short-run, second that monetary policy has no effect in the short-run (or long-run), at least not one that's big enough or long enough to worry about, and third, that trying to stabilize the economy actually makes things worse.

You can find proponents of all three views but the first two, that monetary policy is useful, or that it is neutral, are held by most economists, and across those two views, most believe it is useful. There are those who believe monetary policy makes things worse, and though it's not a majority, it is more than a small group. But as we have seen recently, even economists on the right or those who are fairly libertarian have called for the Fed to do something in the current circumstance. They must think it will help.

The second thing that has come up above is the transmission mechanism - how a change in monetary policy works its way through the system and impacts output and employment. There are two views here. The standard view is that there is a liquidity effect - the increase in money causes interest rates to fall, and the fall in interest rates induces changes in I, C, and NX that increase AD. The other view is the credit view, that there is credit rationing, credit market imperfections, etc. that cause monetary policy to have real effects. For example, if there is loan rationing, then an increase in reserves can ease the constraint and stimulate more loans, more investment, and more real activity. Or, another story, there can be a relationship benefit - over time small firms develop a relationship with a bank and are able to secure credit fairly easily since they are known to be good risks (due to the long-standing relationship). If credit dries up these firms, unlike large firms, won't be able to turn to alternative sources of financing and real activity will decline.

The credit market stories are well-developed and believable, but it's been hard to find strong empirical support for them - Bernanke has been one of the people who has taken the lead here and he has had some success - but the lack of more evidence has stopped the credit market channel from playing a more prominent role in policy discussions in the past. I suspect that may change.

Fortunately, however, where monetary policy and stabilization are concerned, the particular underlying model isn't critical, the policies come out qualitatively similar.