On Monetary Policy

Will Monetary Policy Become More of a Science?

Recent Financial Market Developments And Implications For Monetary Policy

The Revolution in Monetary Policymaking Institutions

Does a downward-sloping yield curve predict a recession?

Interview with Frederic Mishkin;

Stern: The European Central Bank arguably doesn't have a dual mandate. At the end of the day, do you think that's going to matter to economic performance in Europe?

Mishkin: The Congress has given us a dual mandate; that is, the Federal Reserve seeks to promote the two equal objectives of maximum employment and price stability, so that's what we have to execute. Even if the Congress hadn't given us such a mandate, the basic structure of the dual mandate is what I would feel is appropriate, and so we should be aiming to pursue such an objective anyway.

A hierarchical mandate says that first we focus on price stability and if we're successful then we'll focus on other concerns, particularly output fluctuations. If you interpret a hierarchical mandate as focusing on price stability in the long run, making sure that long-run inflation expectations are grounded—and we've seen tremendous success not just in the United States but in Europe in terms of grounding inflation expectations—then the dual mandate and the hierarchical mandate are identical.

Some people have said to me that the dual mandate versus hierarchical mandate dichotomy is a red herring. I don't agree, because I think it is an important issue in communications strategy. It's important to make it clear that you care about output fluctuations, but you're going to look at this from a long-run context and never take your eye off the inflation ball. That's the right way to do the dual mandate.

Similarly, with the hierarchical mandate, you should not be an “inflation-nutter,” as Bank of England Governor Mervyn King has expressed it. That is, you shouldn't be focused solely on inflation control. You must also worry about the fact that if you act too quickly to get inflation down to your long-run objective, you might have excessive, unnecessary fluctuations in output. So I think modern monetary theory, in writing down a hierarchical mandate or a dual mandate, will write exactly the same loss function, exactly the same kind of optimization theory for a central bank.

In some contexts it may be better to discuss monetary policy in terms of the hierarchical mandate. I think the reason it's been done in Europe is because they have had so much worse monetary policy in many countries. To make sure that people understood that they would really control inflation, they had to do it by talking about it as a hierarchical mandate. While in the United States, which has actually never had a hyperinflation and has had much more successful monetary policy, it's more appropriate to talk about it in terms of a dual mandate.

Stern: Maybe we're less prone to the time consistency problem.

Mishkin: Exactly. The time consistency problem is a central issue in thinking about how to do central banking—and also in terms of bank supervision. It's really the same issue. You want to make sure that you're doing the right thing in the long run and not pursuing short-run strategies that end up with very bad long-run outcomes. It's extremely important—in order to deal with the time consistency problem—to say that in the long run, price stability is absolutely going to happen. And that means that you can actually exercise "constrained discretion," the phrase Ben Bernanke and I coined in our earlier work. The idea is that you do need some discretion to deal with the shocks in the economy, but you want to make sure that that discretion is constrained in the sense that you don't ever get into the time consistency problem of allowing the nominal anchor to be weakened. And that's really what the whole concept of constrained discretion is.

This also relates to bank supervision. In my research on this, I felt that the distinction between rules and discretion is too stark. We know with discretion you can get into the time consistency problem. The way I think about this is, suppose it's New Year's Eve and I say I'm going to go on a diet. Then, of course, at the next meal I see a beautiful piece of cake and I can't resist: I've got to eat it. But I say to myself, It's no problem because I won't eat it tomorrow. Well, the next day comes and I can't resist again and keep on eating that cake, and I end up being obese. So we know that one of the ways to solve that problem is to set yourself a rule: Thou shalt not eat cake.

The problem is that there are always going to be unforeseen circumstances where actually you may need to use discretion. It's something you couldn't predict beforehand. If you have a rigid rule, you may find the rule no longer applies, and if you stick to it you will get very bad outcomes.

In terms of bank supervision, in my initial work on this I looked at prompt corrective action strategies. Originally, the idea was that PCA should be a hard and fast rule. No matter what, it has to be done. When you hit particular triggers, you automatically have to do X, Y and Z. What the Congress did in the FDICIA legislation of 1991, which I thought was very smart, was to say, “Look, there is a norm, and that's what should usually be done. But there could be unforeseen circumstances where we need to allow for deviation from that rule.” They did this by saying that there would be a presumption that the rule should be followed but did give the supervisory agencies some discretion to deviate from the rule.

Why then aren't we back in a time inconsistency view of the world? Because Congress constrained the discretion. How? Through transparency. FDICIA requires a mandatory review of any bank failure that imposes a cost on the FDIC. The result report on what actions the supervisory agencies took must then be made available to any member of Congress and to the general public upon request, and the Government Accountability Office must do an annual review of these reports. Opening up the actions of the supervisors to public scrutiny will make it far more likely that they will follow PCA unless they have a very good reason for doing otherwise.

So it's exactly this constrained discretion kind of idea. Constrained discretion says that for most cases you want to operate according to a rule. On the other hand, there are going to be circumstances we can't predict where you may have to deviate from the rule. But in that case we don't want to let you do whatever you want. We want to have some check-and-balance on the system. In fact, my view is that this is also what our Constitution is all about. Having an institutional framework to deal with some of these time consistency problems is something that we see in the political sphere as well.