Larry Meyer, former Governor of the Federal reserve Board, on macro modeling and on Fed policy during the crisis:
Interview with Laurence Meyer, by Mark Sniderman, FRB Cleveland: ...Mark Sniderman, executive vice president and chief policy officer at the Federal Reserve Bank of Cleveland, interviewed Meyer on June 9, 2010, in Cleveland. ...
...Sniderman: What does that tell us about the state of macro modeling?
Meyer: It tells us something very important—something we certainly should have learned—that macro modeling should not be static. It has to evolve over time, and we’re continuously learning. We found holes, and we try to close those holes.
But we know in the future there will be crises coming, or shocks in areas that we didn’t anticipate. We’ll find new holes that we have to fill. In this case, there were really so many. This notion of the financial accelerator wasn’t just a cute idea that the [Federal Reserve] chairman [Ben Bernanke] came up with. It was central to our understanding of how the macro-economy works, particularly when there are intense changes in financial conditions. So you do get these adverse feedback loops that the financial accelerator is all about.
Most of us as macro modelers came out of a tradition in which the transmission of monetary policy, the financial sector, is about real interest rates, about equity values, about the dollar, with virtually no variables that we would call credit variables—they just weren’t there. In milder times, that was OK. That probably got the job done. But when the situation was the drying up of credit markets, dysfunctional credit markets, you simply had to give the model more information than otherwise.
Two things seem valuable that we’ve tried to integrate into our models. First would be “willingness to lend variables” from the senior loan officer survey. Imprecise as it may be, it is a measure of lending terms beyond rates. That’s very important and that wasn’t there, and I think we can integrate that. And the other is credit spread variables—Baa corporate rate relative to, say, a Treasury rate. The reason that’s important is that a risk variable gives an indication of the risk appetites and risk aversion that come into the system when there are financial crises. And that variable tends to be very important in spending equations as well.
Sniderman: Should we expect to be living with our mainstream workhorse macro models for some time, and should we feel good about that? Is there enough progress there?
Meyer: I love that question! So I think we have two kinds of modeling traditions. First there is the classic tradition. I was educated at MIT. I was a research assistant to Franco Modigliani, Nobel laureate, and the director of the project on the large-scale model that was used at the time at the Federal Reserve Board. This is the beginning of modern macro-econometric model building. That’s the kind of models that I would use, the kind of models that folks at the Board use.
There’s also another tradition that began to build up in the late seventies to early eighties—the real business cycle or neoclassical models. It’s what’s taught in graduate schools. It’s the only kind of paper that can be published in journals. It is called “modern macroeconomics.”
We didn’t see the fundamental connection between property busts and collateral in the banking system, bringing the banking system toward insolvency, toward the edge of the abyss. Put on top of that the buildup of leverage in the system—this acts as a multiplier.
The question is, what’s it good for? Well, it’s good for getting articles published in journals. It’s a good way to apply very sophisticated computational skills. But the question is, do those models have anything to do with reality? Models are always a caricature—but is this a caricature that’s so silly that you wouldn’t want to get close to it if you were a policymaker?
My views would be considered outrageous in the academic community, but I feel very strongly about them. Those models are a diversion. They haven’t been helpful at all at understanding anything that would be relevant to a monetary policymaker or fiscal policymaker. So we’d better come back to, and begin with as our base, these classic macro-econometric models. We don’t need a revolution. We know the basic stories of optimizing behavior and consumers and businesses that are embedded in these models. We need to go back to the founding fathers, appreciate how smart they were, and build on that.
Sniderman: Wouldn’t inflation expectations be a counter-example? That has become an important variable in many classical macro models that policymakers use to help them construct their inflation forecasts. Isn’t that at least one place where we see this interplay between the research agenda in macro modeling and the practical use of models?
Meyer: A brilliant question! And you’re absolutely right. This is a good example of interplay between the classic and modern macro approaches. It is true we had a push toward smaller models. This happened because if you want to use these forward-looking expectations, in the form in which modern macro does, forward-looking expectations that are model-consistent, it’s really hard to do if you have a huge macro-econometric model. It’s very easy to do in the smaller, modern macro models. But I think what you saw is exactly what you are suggesting, that it jumped out of those models and became a key area for research and integration into the large-scale macro-econometric models.
But that doesn’t mean policymakers should say, “I like these modern macro models because they treat expectations the way we should.” The Federal Reserve Board’s classic econometric model treats expectations the way you think they should. But it’s a richer, more valuable model for policymakers, number one. And number two, do you really think that you want to model individuals as having their forward-looking expectations based on solving a model out 20 years? I don’t think that makes any sense at all. You need small models to do that, but the reality is that expectations are formed, they’re forward-looking, but we don’t have any idea what the true world looks like.
Our models are caricatures. Everyone has got different model in his head. I think we learn something about trying to get forward-looking expectations into our model. We model the Phillips curve in a way that is very important. We have long-term expectations directly in the model, playing a very important part. That’s something that we didn’t used to do. That’s the way the profession advances in these classical models as they become refined.
Sniderman: One thing models can do is provide different scenarios about what the future might look like; models that provide simulations thousands of times to give us a distribution of outcomes that could help us understand the future possibilities a little more richly. Should we as policymakers be looking for more modeling of that spirit, that spirit of scenario-planning and distributions about outcomes?
Meyer: I think the answer is absolutely yes. It’s not such a simple task to build a sensible, interesting, alternative scenario. I think we should be constantly refreshing and coming up with sensible ideas in each forecast round of what are clearly risks that are on the horizon we want to work into our alternative scenario.
Even more important, we’ve got to sit down every once in a while and say, “Hmm. What’s the worst thing you could think of happening? Tell me something really bad. Find a hot spot.” Maybe it’s something nobody is thinking about. Maybe we would have thought about this incredibly rapid growth in subprime and structured products and said, “Whoa, what could that mean?” Or we could have thought about sovereign debt developments that were going on and were percolating in Europe. It’s not just looking at these incremental things—what happens if this fiscal plan is changed? what happens if oil prices go up?—but looking at these worst-case scenarios.
Sniderman: Of course, that’s not the model itself issue; that’s the human element.
Meyer: Absolutely. You always have to come back to that. So many times people ask me, “What are the rules for forecasting, what are the ingredients?” And I say, “It’s very simple. It’s one part science; that’s the model. One part art, that’s your judgment. And one part luck. That’s how you become a really good forecaster!”
Sniderman: We’ve seen a lot of innovations during the financial crisis in terms of monetary policy. Are there any features in monetary policy design that you think should remain more permanently?
Meyer: To begin to address this question, it’s useful to make a distinction between what I call liquidity policy on the one hand and monetary policy on the other. By liquidity policy, I mean providing enough liquidity when there’s a panic and the market just wants to hold a lot more liquidity. To prevent that from having powerfully negative impacts on the economy, you give it to them.
The Federal Reserve and central banks around the world acted as liquidity providers of last resort. They all found ways to do that. The Fed was extraordinarily creative, very aggressive. You have to give an A-plus to all those operations. They saved the day. You also have to give high marks to the fact that the liquidity programs were designed so they would naturally go out of business as the panic dissipated. And now the Fed has closed the door on them because no one was there anymore.
So that’s gone—beautiful. Central banks all around the world did a great job. Now we’re talking about monetary policy and we say, “That’s just a lot more complicated!” And we have a disagreement about what’s really part of this. Does it matter what the size of the balance sheet is? Does it matter how many reserves you have in the system? Or do you just need to raise rates, using interest on reserves? I’m sure you and I could have a nice debate on that.
We’ve never had this super-abundant level of reserves. We’ve never had this size of a balance sheet. So, for reasons I think we can understand, there’s a desire to do all of these things—shrink the balance sheet, drain reserves, and raise rates. But we’ve never taken these things away. We put them in, and now we’re trying to take them away. We’ve never done that before.
So we don’t know, really, what the impact is if we begin to do asset sales today. How can we unwind that balance sheet without having such adverse circumstances on the markets that we regret it? We’re learning about that, too. I think views have changed dramatically even over the last six months or so with market participants much less concerned about the market consequences of asset sales. There are three things that we have to get done, and we have tools for every one of them. For draining reserves, we have reverse repos and term deposits. For shrinking the balance sheet, we can just let it run off or we can sell assets. And for raising rates, even there we have complementary roles of both raising interest on reserves and managing reserves at the same time.
Does it matter what the size of the balance sheet is? Does it matter how many reserves you have in the system? Or do you just need to raise rates, using interest on reserves? I’m sure you and I could have a nice debate on that.
The Fed was more aggressive and more effective than any other central bank in the monetary policy dimension. That’s because other central banks, whether they admitted it or not, were doing what we call quantitative easing. They were just pushing reserves into the system.
What the Fed did and other central banks didn’t do, because the Fed was in unique circumstances, was make use of the mortgage-backed securities, or MBS, market. The Fed was allowed to hold MBS in its portfolio, and yet MBS was a market that had become illiquid and distressed. It was tied to the housing market, which was under incredible pressure. The Fed was able to go into that market and have big impacts because the market was so distressed and illiquid.
That’s the good news. The bad news is now we’ve still got all those assets on the balance sheet. How do we get rid of them? Being the most aggressive and effective during the stimulus means that you’re the most challenged when it comes to exit.
Sniderman: There’s been a long-running debate about how central banks should deal with asset bubbles. One of the issues that’s come out in the wake of the financial crisis has been the interplay between using regulatory tools and techniques as opposed to, or in conjunction with, monetary policy. Do you have thoughts on that spectrum?
Meyer: This is a very important and evolving area of thought among central banks. We really should start by making a distinction between types of bubbles, between equity bubbles and property bubbles. We lost something like $7 trillion in the bust of the tech bubble. Sounds like a lot, but the economy just shrugged it off—with a very shallow and very short recession.
Equity bubbles are just not a big deal. But property bubbles are absolute killers. We know that from historical experience. The difference is that property is held by leveraged institutions, are the collateral of the banking system, and if you make your banking system insolvent, you’ve got real problems.
The good news here is that although we don’t have good supervision and regulation procedures for dealing with equity bubbles, we do for property bubbles. We’ve got a lot of ways of handling that. We could lower the loan-to-value ratio—essentially increase the down payment that people have to have on their homes to build a better capital cushion. We could do a whole variety of things on the regulatory side. We could increase capital requirements against those properties that seem to be more risky because of bubble-like conditions. We could do a whole variety of things that in principle should be, could be, effective.
The question is, would we recognize that a bubble was emerging in time to implement supervisory and regulatory policies that could have some effect? My views have changed a lot since I was on the Board. I’m a firm believer now that you can always catch bubbles and identify them in time to do something about them before they get dangerous. The question is, what to do? The first line of defense—and this is certainly what the chairman [Bernanke] and others have said—is supervisory and regulatory policies.
I think it’s important for the public to understand two things: the responsibilities of the Fed—what you should be holding it responsible for and what you shouldn’t be holding it responsible for—and then the limits of what any central bank can do.
But we have to be realistic. It might work; it might not. And so the big question for central bankers is therefore—what do you do if it doesn’t work? Do you have to do something in addition? That’s the real issue—do you want to use monetary policy itself, and do you want to lean against bubbles even when the broader macroeconomic conditions would not lead you to, for example, want to tighten? That is a taxing issue.
The issue is less whether you can identify a bubble than what do you do if you think it’s emerging. I’ve come away with a very different understanding of the risks of allowing bubbles to go unchecked. But that’s property bubbles. I’m not so concerned with equity bubbles. Property bubbles—that can be handled to some extent by supervision and regulation, but I think we should be very open minded here. We’re searching, we’re debating, we’re not sure what monetary policies should or could do in those circumstances. If we come to that place again, I’m sure there will be a very good debate in the Federal Reserve System, as there should be, before deciding whether to be more pre-emptive than was the case before.
Sniderman: What is it that you wish the general public would better understand about central banks and their role in the economic system in which we live?
Meyer: What should the public know? First of all, the public has its representatives in Congress. And Congress has a very important job overseeing the Fed. I’ve said this many times—wouldn’t it be good if Congress learned a little bit more about monetary policy and how it works? I’m always amused and distressed about how poor the questions are during Congressional oversight committee hearings. The first part of the public I’d like to see understand more about monetary policy is the Congress, particularly members of the oversight committees.
Other than that, I think it’s important for the public to understand two things: the responsibilities of the Fed—what you should be holding it responsible for and what you shouldn’t be holding it responsible for—and then the limits of what any central bank can do.
It’s partly the limitations of our knowledge. It’s partly the limitations of what central banks’ tools can accomplish in the real world. But I would say to understand what they do, what their responsibilities are, and then understand how they try to achieve those objectives and appreciate that there are limits. When you want to hold central banks accountable, understand that perfection in central banking is no more possible than it is in any other profession.
Sniderman: Maybe you can leave us with some thoughts on things you’ve been reading these days?
Meyer: My wife and son always warned me that if anybody asked me that question, I shouldn’t even answer it because they view my reading list as, shall we say, not intellectual enough to go along with my reputation.
I have two sets of readings on my night table. One is books on the financial system and recent history in particular. Too Big to Fail [Andrew Ross Sorkin], it’s like a story unfolding before you, and I’m in the middle of that one. The Black Swan [Nassim Nicholas Taleb] has fascinating stories about the weight that should be given to improbable events, brainstorming on catastrophic things that could happen, and how to protect yourself in advance from those possibilities. And then I’ve got the book by Michael Lewis, The Big Short, that’s on my list.
Finally, I read mysteries, spy novels, and my current group is by the author from Sweden, Stieg Larsson, The Girl with the Dragon Tattoo and all the ones that followed. Fantastic reading. These books are insanely popular all around the world. This is a series that has really caught my attention, and I’ve got one more of those to go.
Sniderman: Thanks for taking the time to talk with us today.