We spoke on the sidelines of a Boston Fed conference devoted to the many economic puzzles confronting policy makers. I began with the most immediate: Is the economy strong enough for the Fed to resume hiking interest rates?
Q. In your statement explaining your September dissent, you said the progress of the economic recovery “might, by itself, be sufficient to justify” a rate hike.
A. We’re at 5 percent unemployment. By many people’s estimates that’s full employment. My own number is a little below that, so I would like to see the unemployment rate come down a little bit more. And if you look at the inflation rate, the core rate is 1.7 percent. It’s not at all unusual to have an inflation rate half a percent higher or lower than what you’re expecting. It says that we’re relatively close on both full employment and on where our inflation target is, and we have an interest rate that is well below what we think the interest rate will be in the long run. Roughly around 3 percent is where people think the Fed funds rate is going to settle, and we’re well below that. So if you think you’re near both elements of the dual mandate, and the policy rate is well below where you think it’s going to be in the long run, I think there is some justification for taking action on that basis.
Q. Your main argument for raising rates, however, is a little more complicated. You’re basically worried the Fed is on track to help too many people find jobs.
A. That is the actual rub. I actually would like to run the economy a little bit hot, and I view a 4.7 percent unemployment rate as running the economy a little bit hot. We would be actually probing how low full unemployment is. But our own forecasting model is expecting that there is going to be enough growth in the economy going forward that it’s quite likely that we’ll be below 4.7 percent and, depending on what the path of interest rates is, potentially well below 4.7 percent, and that would concern me.
Q. Low unemployment sounds like a good thing.
A. During periods when the unemployment rate has gotten to the low 4s, we haven’t stayed there for a real long time. And that’s because we do start seeing wages picking up, and we do see prices start picking up, and we do see asset prices picking up. In that environment we start to tighten and when we tighten we’re not so good at getting it exactly right.
The problem is the dynamics of how firms and individuals start thinking about the tightening process. Those dynamics make it very hard to calibrate the monetary policy process. People understand tightening. But convincing them of how much you’re going to tighten and that you’re going to hit it exactly right — particularly given that you haven’t hit it exactly right in the past, it’s pretty tough to convince people of that. Not surprisingly, they start worrying about: “Well, they’re starting to tighten, they may tighten too much. What do I do? I start pulling in in terms of my own spending.” Firms start pulling in, saying, “We want to be prepared in case they don’t get this quite right.” Those kinds of actions — which are very hard to predict, and individually everyone behaves a little differently — in aggregate, cause a problem where we sometimes slow down the economy more than we intend.
So you don’t see instances where we go from 4.2 percent to 4.7 or 5 percent and level off. What you actually see is when we start tightening we end up with a recession.
Q. How do you avoid that by tightening sooner?
A. Ideally, what I’d like to see is us staying around full employment. The argument I was making is we want to make sure we have as long a recovery as possible. I am concerned that if we wait too long we will end up tightening more quickly and possibly more over all. The result may be that the recovery we are trying to maintain would actually be much harder to maintain with a policy that waits much longer and then tightens versus tightening at a more gradual pace. You can’t wait too long to start if you want to make sure it’s going to be gradual. If we wait too long to start raising rates, I don’t think we will have the luxury of moving as gradually as I would like.
Q. In her speech here on Friday, Janet Yellen said that driving down unemployment could have benefits. Do you think she overstated those benefits?
A. I would view it more in terms of risks. Our own forecast would have us fall below the sustainable level of unemployment by 2018, and probably by enough that I would be concerned. I view that as a risk. I don’t know if that’s going to happen. But that risk is sufficiently large that I would take some insurance out against it and make sure that we’re going to settle in around full employment and not go well below it. And it’s not that there aren’t benefits. I think there are benefits from running hot and there’s some benefits from running real hot. The problem is when you run real hot you get to the point where it will presumably show up as inflation or asset prices, and we will end up reacting to that and it’s not so easy to calibrate that reaction.
Q. A number of the academics at this conference said they don’t think you should be trying to raise rates. What do you make of their hesitations?
A. We haven’t hit 2 percent inflation for a while. Some of them have argued that we should in effect be price-level targeting, which is to say that the misses that we’ve had in the past ought to be made up in the future. So they have a different model than we actually are using for monetary policy. We have an inflation target right now. If we wanted to move to price-level targeting, as was advocated by a number of the academics at the conference, we should have that discussion. We should announce it publicly. I don’t think we should do it without telling the public.
I think also the inability of so many central banks to hit their 2 percent inflation target has caused some people to say, “I want to actually see evidence that you can hit 2 percent, and since we’ve just seen the consequences of hitting the zero lower bound, I want to take out some insurance against hitting the zero lower bound more quickly.” I think both concerns are credible. My concern with those arguments would be that the very scenario that causes the next recession might be that we overshoot.
Q. You pointed to some evidence of asset bubbles, but you didn’t mention it in your September statement. How worried are you?
A. It’s a concern. For stock prices, there’s evidence it’s a little bit elevated. It’s not widespread in asset markets. I do think commercial real estate is one of the asset markets where I would say there is more evidence that there is a building up of momentum, that relative to history it’s a little bit unusual. So both the 10-year Treasury and the commercial real estate capitalization rate are low by historical standards, and we should think about why is that and what do we think would happen if that unwound quickly. We have had commercial real estate problems in the past, both domestically and globally. It’s not like there’s no history of commercial real estate causing problems in the past. Why some people are more sanguine than I am is that it’s distributed more widely among financial institutions and nonfinancial institutions. It’s not concentrated only in the largest institutions. But when we had the problems in New England, it was mostly small and medium-sized banks that failed in that period, and I think it had a real impact on the economy at that time.
Q. Are you concerned the presidential election could be economically disruptive?
A. Our internal model does not have an election dummy in it. It’s saying that we’re not necessarily expecting that our long-term forecast is going to be affected by this election. That may be right or may be wrong.
We had an instance with Brexit where a vote went very differently than people anticipated. It’s a little bit different than what we’re experiencing now. People are already voting now, and so far it doesn’t look like we’re going to get an outcome dramatically different than what people are expecting. And so I’m not expecting it to have a big impact on the overall economy. I could be wrong, and if conditions change then we’ll have to factor that into our model.
Q. There is a widespread belief that the Fed will not raise rates at its next meeting in November, yet the Fed continues to insist every meeting is live. Why bother?
A. I can imagine that there would be people who would at least consider dissenting at this meeting. I don’t know what the economic conditions will be or what I’ll actually be voting on when it comes to the next meeting. But I could imagine conditions under which it would make sense to do it. I was willing to do it three weeks ago. The difference between September and November is not dramatically different.
Now that argument also holds in the other direction. Right now the market has a pretty high probability for December. And the difference is only six weeks. No econometric model would say that time span would make that much of a difference.
I don’t think the fact that we’re actually having an election — if that election were to change the way we’re modeling the economy or thinking about the progress of the economy, that’s something that should play a role. If we don’t think the election is going to materially impact our forecast going forward, then it shouldn’t play a large role. We should be doing it on the economic fundamentals. Now, some people may argue that the risk around that election is sufficient that it will affect economic fundamentals and it should be taken into account. We’ll have to see as we get closer whether that seems like a likely outcome.
Q. So you might vote to raise rates in November?
A. I’m open-minded. We’ll see how the data comes in. We’ll see exactly what the recommendation is that we’re considering at the time. And I’ll listen to my colleagues and what their views are. I have no idea what the political conditions will be. I’m not predetermined on what I would do or wouldn’t do. I can imagine a situation where I would dissent and a situation where I wouldn’t dissent.
Q. This conference has posed a lot of good questions about our understanding of the economy. Not a lot of good answers. In introducing Ms. Yellen, you said it was a great time to be an academic, but a difficult time to be a policy maker.
A. Our whole conference has been about anomalies. Some of those anomalies are pretty fundamental. Why has G.D.P. growth been slow? Why has the labor force participation rate come down so much? Why haven’t we hit 2 percent inflation more quickly? Those are all really interesting questions for an academic. For a policy maker you actually have to make a determination with imperfect information about what the answers to those questions are. So an academic gets to work on exciting ideas and there’s not much cost to being wrong. Anybody who’s in a policy-making role knows there is cost and so that makes it more challenging.
Q. One conclusion I often hear is that fiscal policy makers should be doing more. Do you think the Fed should be making that case more forcefully?
A. There are times where the chair of the Fed does talk about fiscal policy. You need to tread lightly on that because we have a different set of responsibilities than they do. There can be costs to being too vocal about what other people should be doing when it’s their responsibility. So I don’t think we should get too engaged in discussions, and that’s particularly true if it’s of a very partisan nature. One of the most valuable features of the Federal Reserve is that it’s a nonpartisan organization. We’re trying to do what makes sense given the fiscal policy that we’re given. And fiscal policy obviously isn’t done that way. We have a partisan electoral process. I think in broad brush, saying that the monetary-fiscal policy mix may not be exactly right is probably appropriate. But we have to take fiscal policy as it is.