 anyone left in the room. A economist who has a Nobel Prize and a J.B. Clark Medal, but even better than that, he's been consistently right about the nation's economic problems and how to solve them. Since the mid-1990s, he's been worried, maybe even obsessed precisely with the idea that a combination of financial crisis and the zero lower bound on interest rates could flummox policymakers into making mistakes that could lead to years of unnecessary pain and misery and waste. And this caution wasn't only relevant to developing countries. He warned in a 1999 book that even rich countries could fall prey to this. I think we all wish he had not been right about this one. But given this background, I can't imagine anybody better to talk about what policymakers and economists should be doing going forward from here. So please join me in welcoming Paul Krugman. Okay. So I missed all but the very tail end of the last session. So I'm sure I'll be repeating some stuff. But okay. Let me just get straight to it. I'm not trying too much framing here. For the moment, at least, the question is when, under what circumstances, should the Fed start to tighten? Now, that's an enviable problem. I mean, you'd much rather be Janet Yellen than Mario Draghi right now because we seem to have a recovery that's gathering some steam. And the question is how not to mess it up. Obviously, in Europe, it's trying to get a recovery of any kind. And Japan is trying to finally break out of years and years of this stuff. Those are important morals for us, part of what we need to take into account in our own policy. And I'm getting increasingly worried by the kind of tone of the discussion. What you hear, certainly in a lot of the press reporting, a lot of the commentary, and a little bit what you're hearing from some people at the Fed. And I don't have any inside knowledge at all. But I think that there's a possibility that we may be about to see something like a trichet moment, an unjustified, a rise in interest rates or a rick's bonk moment if you like, where the Fed starts to lift off too soon, tries to normalize when things are not at all normal. And I want to make that case. And the way I want to do it is, first, by saying that if you want to know what ideally would the path, should the path of a Fed policy of interest rates be from here? If we had full information, what would we be doing right now? And the answer to that is I don't know. But equally important, you don't know. Helen doesn't know. Nobody knows. All of this has to be undertaken in the knowledge that we are in a time of, I guess, things are always uncertain, but maybe even more uncertain now than usual. Let me give you three. This is going to be kind of a sub-list talk, but forgive me for that. It's kind of like, I have no slides, but it's kind of like a bullet points and sub-bullet points talk, because I think all of this needs to be laid out. So let me give you three crucial issues of uncertainty that are relevant even to monetary policy in the very near future. The first is we don't know how far we are from full employment. If somebody tells you he knows, he doesn't. I keep on getting papers come across my desk that use a different set of variables, different set of instruments. Maybe it's only short-term unemployment that matters. Maybe it's all unemployment. Maybe we should be using lagged inflation. Maybe we should be using survey inflation expectations. And at this point, I think to myself, fine, but I don't think I'm going to be convinced by any one of these papers, because the fact of the matter is that there's a lot of different things that could fit the data. We have mysteries. The unemployment rate is not that high, although it is worth remembering that in the early 2000s, an unemployment rate about where we have now, we thought was terrible. So there is a strong element to have been down so long, it looks like up to me. But in any case, there are mysteries around that. We've had a very sharp drop in labor force participation. Some of that is demography. Maybe much of it is demography, but we don't know how much. You can go around and around. You can pick your numbers and come up with stories that seem to suggest, well, we're not that far from full employment, or stories that suggest that we're quite a long ways from full employment. Worth remembering that even in what seems in retrospect like a less confusing time in the 1990s, circa 1994, 1995, it was very widely believed by reasonable people that the non-accelerating inflation rate of unemployment was 6%. And then it turned out that we ran down to 4, and nothing terrible happened. So even under those circumstances, we were wildly uncertain about what we were really capable of doing. The second source of uncertainty is we don't even know what full employment means, not at this point. So I said Nehru. We have the Nehru. It's in all the textbooks, mine included. The idea that if you have unemployment below some level, then you see accelerating inflation. Above it, you see decelerating inflation. That's a very good story for making sense of the stagflation of the 70s. Seems to fit the disinflation of the 80s also, temporary period of high unemployment brought inflation down. There are a lot of reasons to think that it's not the right model when inflation is very low. If you actually just did a raw look at the data right now, they look, if anything, more like an old fashioned Phillips curve with a simple slope of unemployment versus inflation. A lot of the papers that I'm getting now that are trying to analyze inflation don't have lagged inflation, don't have or they do have expected inflation, but instead of using lagged inflation, they use the survey of professional forecasters number or something which hasn't really changed at all since the late 1990s. And so it doesn't look at all. The data, nothing that's happened in the past 20 years would lead you to believe in an acceleration of Phillips curve. Which means that we're not even sure that there's a unique number that constitutes full employment right now. It may be that there is, in fact, the kind of trade off economists used to believe there was in the 60s, where we really have some choice. We don't know that, but we're not just uncertain about the parameters. We're even uncertain about the model. At low inflation rates, I could go on, there was a really interesting study from the IMF of plugs, prolonged large output gaps, which almost never actually lead to actual deflation. It's interesting, according to the standard Phillips curve they should, but they almost never do. Japan is an exception and even there it's always been crawling deflation, no more than that. So good reason to think that the model we use to define full employment doesn't apply in the current situation. Finally, suppose that we decide something that we were going to call full employment. How high would interest rates want to go after we've had full recovery to stabilize the economy around there? Where does the natural rate of interest, the neutral rate of interest, whatever term you want? We have very little idea. It might be that things are the way they always were. It might be that secular stagnation is very real. I kind of am in the second camp. I think that if you look at the evidence it suggests that equilibrium or average across the cycle real interest rates seem to want to be a lot lower than they were in the past. I could make you an argument about demographic changes looking forward with slower growth of the working age population, further depressing investment demand. In any case, we don't even know if you say, well, the Fed should normalize interest rates. Where is normal? And the answer is we don't know. So you have to make policy in the face of this uncertainty. The range of the uncertainty, I would argue, is very large. What you need to do, what the Fed ought to be doing, what those of us who are kibitzing at the Fed ought to do, is talk about what makes sense to do given all of that uncertainty. At that point, what I would say is that there's overwhelming reason to consider that the risks are extremely asymmetric, that the damage that we know that the Fed is going to raise rates either too soon or too late. Problems we just don't know when that is. The risks of raising rates too soon are vastly, vastly greater in this environment than the risks of raising them too late. There is just no comparison in terms of what can go wrong. So let me start with what the people who are urging the Fed to start moving are saying, which is that you don't want to wait too late. And the reason you don't want to wait too late is that inflation could rise and it could rise above target. Okay. First of all, we do know what to do about that. It won't be pleasant, but if the Fed decides inflation has gotten too high, we know very well what to do about it. The Fed just raises rates and squeezes the economy. And when it thinks that we've suffered enough, it relents again. That's unpleasant. It's, I shouldn't say it's just unpleasant. That would be a bad experience for a lot of workers. But it's not something where we have real doubts about the ability to do it. When you see people say, oh, does the Fed have the tools or the will to control inflation? I have to think, well, it's kind of a Rip Van Winkle thing. They went to sleep when Arthur Burns was still running the Fed and haven't noticed the last 40 years. It's just not a problem. And even if you start to say, talk about the Fed has this complicated balance sheet. Yeah, but also the Fed has pays interest on reserves. And so they have other tools and there's no need for the technical issues about managing it are just that. They're technical issues and they're not a problem. What are we worried about in any case? I guess we're worried that the inflation rate could rise above the target, which is 2% at the moment. Now, first of all, 2% is meant to be a target means it's supposed to be in the center of what you hit. And it's really bad that it gets treated as a ceiling. That's a mistake. But also, why exactly are we so concerned about 2%? How did we get at 2% as an inflation target? And that's actually, it's amazingly undiscussed. I did some archaeology. I actually did a paper for the ECB last summer. And I did some archaeology trying to figure out how 2% became the sacred number. And it's an interesting thing because it's a confluence of three different lines of argument, two of which it's almost an accident of seeming to make sense to people and doesn't stand up very well in retrospect. So there were three things. One was there was concern even 20 years ago that you could find yourself in a situation where zero is not a low enough interest rate. So the possibility of hitting the liquidity trap with a zero lower bound was definitely out there. It's not news to monetary types. But they looked at it, there was some statistical analyses. They thought that they were reasonably sure that as long as you had 2% underlying inflation, that was going to be a minimal threat. It was going to happen very rarely. And you can find there are explicit analyses showing that hitting the zero lower bound should happen less than 5% of the time in an economy with a 2% inflation target. Worth pointing out that it now turns out to be utterly, utterly wrong. We actually hit roughly 2% inflation in circa 1994. So we've been in that sort of 2% inflation range for about 20 years now, of which we have spent six at a zero interest rate. So using a maximum likelihood estimate, we would say that at a 2% inflation rate, the chance of hitting the zero lower bound is not 5% but 30%. Okay, that's a pretty big difference and it completely changes. It completely guts that argument saying that 2% was a good target. There was a second argument involving price and wage stickiness. Arguing that there's lots of changes in relative prices that have to happen all the time and that we know that it's very hard to cut prices and cut wages but it's much easier to simply allow them to decline in real terms because you have underlying inflation. And again, there was estimates that suggested that most of the gains from having a little bit of lubrication from positive inflation would take place with just 2% inflation. There was no point really in getting more than that. That's people like George Akerlof making estimates like that. It's not as dramatic as the zero lower bound but I think I would argue that there's substantial evidence that that was overly optimistic as well. That if we look now, if you look at it, there's a really quite amazing fraction of wage changes that are exactly zero in the United States and even more amazing if you look in Europe that it's looking like that's a bigger issue. Finally, 2% was a number that enabled you to sort of buy off the price stability fanatics. There were a lot of people who said, no, the way the argument was couched in the early to mid 90s was, well, should we have price stability or should we have 2% inflation? And you were able to buy off some of the price stability types by saying, well, technological progress, whatever seemed like a big deal at the time. Internet, there's this thing called the internet. That changes everything so that what measured 2% inflation could arguably be price stability, which is really silly all around. But it did allow, Island Green spent, for example, to be assuaged with the argument that 2% measured inflation is actually price stability. So all of these things, 2% seemed to be a number that satisfied several different constituencies, was adequate for several different constituencies. And then it became locked in as this is the sacred trust, 2% inflation. Very hard to move. But at the very least, we should understand that there is, in fact, no compelling economics that suggests that anything bad happens if inflation goes above 2% for a while. That's one side. Yeah, so if you ask, what is the downside? If it turns out that the recovery is stronger than I think it is, that we're closer to full employment than I think we are, it might be that inflation rises above 2% for some time. Very unlikely to be either something that does significant economic damage or something that the Fed can't control. Very hard to see how it's more than that. You really have to, the leap from a year or two of 3% inflation to the kinds of apocalyptic inflation warnings that you hear is a very large one. I don't see why you should want to make it. Okay, risks of tightening too soon. First of all, if the recovery is weaker than we think and we're further from full employment than we think, you tighten rates because you think things are improving and there's some inflation risk, and one day you wake up and discover that you're on the edge of deflation and you've cut rates back to zero and you're still on the edge of deflation and the economy is weak and you are slowly sinking into quicksand. And that is, of course, that might have sounded hypothetical 20 years ago. Now, Japan, of course, for a long time, and Europe. At this point, the question is not whether it's possible that the euro area will sink into a Japanese-style trap. The question is, can you come up with any plausible roots by which it doesn't? It's extremely hard to come up with a story where that doesn't happen. So this is not abstract. It's very real. And the point is it looks like it's a very, very hard thing to get out of. I think if you go back to circa 2000 when, actually, I've made this remark, right? When I started at Princeton, there was actually a little group of people who were Japan warriors. Mike Woodford, Lars Fenson, who's been having his own saga, myself and Ben Bernanke. And we were all worried that something like a Japanese scenario could unfold in the United States as it has. But we were all pretty confident that, of course, we would handle it much better. That we would use aggressive monetary and fiscal policies to get out of it. And I propose that all four of us should go to Tokyo and apologize to the emperor. Because it turns out that we're handling it worse than they did in many ways. And it turns out that both the economics of unconventional monetary policy and the political economy of doing the right thing, as opposed to getting pulled off into worrying about the wrong thing like budget deficits, are very, very difficult. You really don't want to get into that position. So this is a very, very serious thing. You really don't want to risk getting into that position. Second point, if the Phillips curve is not what we thought it was, if this is not our father's Phillips curve, but our grandfather's Phillips curve or something, if there really is no accelerationist thing, then there's a real danger, I think, of falling into a policy, I've been calling it the price stability trap, where we have, in fact, a lot of unnecessary unemployment, but we don't have accelerating deflation. We have slow inflation. And there's a way too easy for central bankers to say, oh, we're doing our job. We have price stability. What more do you ask us for? And so there's a real possibility that you can end up trapped with an economy persistently operating below potential. Not something you want to get into. And that's a really good reason to push to make sure that you find out where the potential really is, not preemptively cut off recovery before it gets anywhere close to full employment. Hysteresis in English, depressed economy doesn't just damage the present, it damages the future. At this point, the estimates of potential output all around the advanced world are way down from where they were expected to be. It's very difficult to avoid the conclusion that the slump is what did it, that a depressed economy through a variety of channels reduce business investment, wasted skills of young college graduates, people, discouraged workers, erosion of skills from long-term unemployment, that a depressed economy now casts a long shadow over the future of the economy, which means that the costs of tightening too much and having an economy that does not recover as much as it should are not just one year, but extend far into the future, are very much larger. And again, that's a reason to really not want to do that. Finally, to the extent that we have a choice about low pressure or high pressure labor market, to the extent that we may well have a choice, that the level of unemployment that we're going to face, the average rate of unemployment over the next 10 years, I would say almost truly is not something that's given to us by the structure of labor markets and institutions and so on. It's not that we're going to have a Nehru that's going to be, we can fluctuate around it, but that's what's going to be. I would say that almost certainly we have a lot of choice about what the unemployment rate is going to be over the next 10 years, and I know Josh Bivens has been saying, there's every reason to think that our problems of inequality have a lot to do with persistent slack in labor markets. The only time we've had anything that looked like serious wage gains for a lot of workers was during the 90s boom, which was not inflationary, that if we're going to have, there are lots of things we should be doing to reduce inequality, but probably the biggest single thing, and something that would make everything else we might do work a whole lot better, would be to make sure that we really do have something that feels like a full employment economy by tightening too soon, the Fed would be risking our chance of getting there. I've now given you four reasons to really be very, very worried about the effects of tightening prematurely. They seem to me to be collectively at least in order of magnitude more serious than the costs of waiting too long to tighten. It's just very hard to tell a really bad story about what happens if the Fed waits a year too long. It's depressingly easy to tell a story about what happens if the Fed moves and it shouldn't have. I would have thought that that balance of risks was overwhelming in favor of letting it rip, not doing anything until you see the whites of inflation's eyes and maybe a little bit after that. I know that there's going to be some talk about financial stability, I guess that's the later. I should say something about that because that is the other argument that comes up here, the argument that says that, well, aside from all that low interest rates lead to financial instability and therefore we need to raise rates even in the face of all of these other arguments. So it's an interesting point of view. What always gets me about the financial stability arguments for raising interest rates is that normal, there seems to be a tremendous loosening of normal standards of logic and evidence that goes into these things. The willingness to take seriously assertions that low interest rates lead to financial instability without demanding clarity or historical data to validate the point seems to me to be really striking. First of all, I mean, yes, people who can borrow cheaply might make bad risky investments. Also, people who have to pay a lot of money to borrow might make bad risky investments because they're trying to earn enough to justify the cost of the loan. You can tell the story either way. Either way, you're requiring that there be some kind of irrationality on the part of people out there in the financial markets, and I'm certainly willing to believe in that, but why should I believe in one type as opposed to the other? Where did this certainty that low rates are the type that encourage bad, destructive irrationality on the part of financial markets? Where does that come from? Hard to see. And you get even stranger arguments. I mean, I shouldn't, but if you just track, I mean, my favorite is to read successive reports from the Bank for International Settlements, which just happens to be the hotbed at this stuff, where each year they come up with a different argument about why interest rates should go up, but there's always a reason. And what's amazing is the creativity that goes into the arguments. It's funny, you might think that the easy money people like me are wild and crazy, and that the stern responsible managers are the ones who are defending orthodoxy, but in fact, I'm going with more or less straight textbook economics. They're inventing new theories every few months to justify their position. I just heard a new one, I guess I can't say, but a senior European official explained to me that low interest rates are self-perpetuating because low interest rates lead to people to run up more debt, and we know that debt has a depressing effect on the economy, and therefore you end up lowering interest rates even more or two. And sort of isn't there an intermediate step along the way there where people borrow and you get an economic expansion, but just skip that step. And it's amazing stuff, and yet this was not a guy off the street, this was someone with some real influence on policy inventing this, I'd say, bonkers theory to justify rate hikes despite the weakness of the economy. You might say, well, what about the lessons of history? And I guess I don't quite see those. It seems to be one observation, which was that the U.S. had low interest rates in the mid-2000s and had a destructive housing bubble. Of course, Europe did not have interest rates as low as we did. They also had an enormously destructive housing bubble, basically the same size. You try to look otherwise. I've been spending a fair bit of time lately looking at the great Scandinavian financial crisis of 1990, Sweden in particular, which is interesting in a couple of ways. I actually used it last week as a caution against believing that inequality is at the roots of our financial crisis. Because Sweden had a crisis that looks a lot like what we had. They had a runaway banking system, huge housing bubble, big increase in debt, crash, very nasty, slow recovery, all taking place in a society with a genie coefficient not much more than half of ours. So that was showing that you don't have to have high inequality to have a big financial mess. But also, all of that took place in an interested environment where short-term interest rates never got below 8.6%. So it's also showing you don't need low interest rates to have a runaway financial system and a housing bubble and a banking crash. That doesn't prove the case, but since there's so little evidence being used to support the claim that financial instability is enough of a risk from a low interest rate policy that we need to act in defiance of everything else we see, I think that is important. We're thankful to the Swedes once again, I guess, not just for having a financial crisis that puts the lie to some of this stuff, but also for, despite that, going and buying into the financial stability argument for interest rate hikes, doing in effect what people on the other side of this argument say the United States should have done circa 2003-2004, should have raised rates significantly, the economy was still weak and inflation was low to preempt financial instability, and of course what's happened is they are now at deflation and not too happy, but they're working very energetically to denounce the people who've criticized them for their missteps. If you've been reading the Financial Times the last day or so. So basic story, again, huge uncertainty. We may be able 10 years from now to look back and say well if we had known then what we know now, interest rates should have started to increase on November 17th, 2015. Or we may not, I think the odds are we'll be confused about this in perpetuity. What's clear is that we don't know now. And then the question is what do you do in the face of that uncertainty? And the downside, if the Fed raises rates too soon, is potentially enormous. The downside if it waits too late is fairly minor. And I don't know, I hope we can get this through to them. And normally I would think that the channels of communication were pretty good since... Finally these are, well not just finally at the Fed, in general, these are people I know and they're not, it's not like trying to talk to the house majority. But I'm not sure, there is this sort of gathering sense of inevitability about an early rate hike and it really needs to be fought. End of story, thanks. How do we do this? I lost Josh. Yeah, we can take a few questions. I have a public transit which George Wilt has us as evil and socialistic as until four so I've got a little time. Yeah, oh, yeah, right. I'm curious, now a huge segment of the public is not going to have access to capital. Right, but the indirect, so sure, maybe I should say something about this. How does monetary policy affect most people? A little bit if they're buying houses or refinancing houses, but mostly not. But it does, we think it affects employment. Traditionally, a lot of that comes through housing and even now, by the way, even though housing, you would say, well, we're not going to get another housing boom, but we actually do have the beginnings of a housing recovery because we built hardly any houses since 2006. We're actually housing short right now and if there's a rise in mortgage rates, that'll choke off that and housing employs workers directly and then indirectly leads to further employment. A little bit the value of the dollar. If you want to think about, right now, the dollar has risen quite a lot because the U.S. seems to be recovering and because there's all this muttering about the Fed raising rates while Europe is flat on its back. That's not good for the competitiveness of U.S. manufacturing, so that's going to cost. So there's a bunch of channels. I mean, it would be nice if you ask, shouldn't we have policies that would directly deliver jobs, livelihoods to the people to Main Street or Martin Luther King Boulevard and all of that as well? Sure, but those are all things. The reason we talk so much about monetary policy is that all the things that we should be doing on fiscal policy are blockaded politically. So the Fed is the only game in town. I wish it weren't, but that's why we do it. But you should not think, I get this all the time, so easy money just benefits Wall Street. But it's not true. Let's put this way. I think easy money filters down, trickles down. It does do something for ordinary workers and tightening monetary policy definitely would do a lot of damage to Main Street. Yeah. Oh, not exactly. Thank you, Mike. I think we need to be a little bit more cautious about the interestability risks associated with low interest rates. I think we do observe sources of concern. And I think it's a little bit stylized. I agree with you that monetary policy is not necessarily the right tool. But then that's, it always leads into the second panel, which is what are some other policies that we considered? No, I can tell financial instability stories heading forward. There's no reason to think that we solved that problem. I would argue that very little of that is caused by low interest rates. It may be that the market's gotten too complacent believing that rates will stay low forever. I don't know what you could do about that. And you think about the idea that what we need to do is sort of do the wrong thing from a macroeconomic point of view in order to break the complacency of the market seems funny. I am finding it a little hard to tell a story about a major financial shock, maybe famous last words right now. Give us time. Something can go wrong. And I guess we should be concerned a little bit that if we need another tarp, it ain't going to happen. So in some ways we're vulnerable to future crises because of the paralysis of Washington. But yeah, I think... I mean, we could go through... But maybe I should say this, that I do think that being too close... This is self-serving, because I'm not too close to it. I'm not close to it at all. But being too close to the mechanics, the plumbing of the financial markets can make you start to think that these kinds of things are really first-order issues for economic policy. I would argue that they're not. But yeah, I'm willing to believe that there are... Maybe I should say that I'm willing to believe there are some funny things going on, partially in response to low interest rates as opposed to the different funny things that would be going on if interest rates weren't so low. But I know, fair enough, fair enough point. Well, and subprime auto loans. Is that a low interest rate thing or is that an income inequality thing or something? But something's going on there. It's hard to match the amount of damage you can do by screwing up homelending, but we're working on these other things a little bit. Although it's partly those that they just haven't... It's just the bankers are still the same bankers as they were, and they kind of did the maximum damage to one sector. So now they're looking for other sectors to ruin. So I think in order to get the second panel started, we're going to cut it off there. Sorry. No, thank you so much.