May 17–18, 2021
Day 2 Keynote Transcript
Dave Altig: Good morning, everyone, and welcome back to day two of the 25th annual Atlanta Fed Financial Markets Conference. We are very pleased today to be able to offer as our first event the keynote fireside chat with Dr. Larry Summers.
There is an extended bio on the website for the conference, but if you are not familiar with who Larry Summers is and his biography then you may have wandered into the wrong webcast. But stick around; you're in for a treat. Suffice it to say that Dr. Summers has been one of the most influential and powerful voices on the world stage for decades now. He has occupied the highest offices in both academia and government and is one of the most respected voices on policy in the world, let alone the country. With that: welcome, Professor Summers. It's good to have you.
Lawrence H. Summers: Glad to be with you. Glad to be back at the Atlanta Fed conference. Sorry I won't be able to play golf with some of the participants this year, because we're all virtual, but look forward to that occasion in the future.
Altig: Great; let's jump right in. Yesterday's sessions had an international flavor to them, in large part. Vice Chair Clarida started off discussing the interconnection of financial markets across the globe, bond markets in particular, and noted that the synchronicity, if you will, of bond yields and prices was not entirely accounted for by a correlation in economic conditions. Later on in the day in our panel, Willem Buiter suggested that maybe in the United States policymakers don't pay sufficient attention to those impacts that might exist on the rest of the world.
Let me start off by asking you: What's your assessment of the state of financial market stability in the world, and how do you think about the responsibilities and connection of U.S. policymakers to maintaining and fostering stability?
Summers: I've always believed that the objective of U.S. policy should be to promote the welfare of American citizens, that American citizens benefit from a strong economy, along with price stability, along with having substantial purchasing power, which is why I always supported, as treasury secretary, a strong dollar. I believe that those objectives can best be pursued by taking a cooperative posture; that working to contain financial crises in other countries serves to promote general financial stability, which benefits American citizens. That's why I, in government and out, have urged strong and aggressive responses to various emerging markets' financial crises.
I believed, as I said on many occasions, that the global economy cannot fly on a single American engine, and therefore have thought that the United States has a very strong interest in the success of the global economy because the global economy is, after all, a major market for the ~12 percent of our economy that represents exports. But it seems to me that American policy needs to be anchored in American interests if it is to be politically sustainable.
I have felt, though, as I say, that that requires global cooperation. I was very proud to have the chance to work with [former Canadian Prime Minister] Paul Martin, during the time I was treasury secretary, to establish the G20 as a financial group because I felt very strongly that the G7 could no longer be the dominant group shaping directions for international financial policy in an economy that was rapidly changing.
Altig: Is it your view that, by and large, there is a correspondence between the sorts of actions that have to be taken in the event of disruptions in financial markets offshore? Or are there instances...I mean, you have a lot of experience in this; there are instances where sometimes there are really significant trade-offs that we face and what would those trade-offs look like?
Summers: Let me make sure I understand precisely the terms of the question: certainly the United States has a legitimate interest in resisting financial exchange rate manipulation by other countries. I think that while one can question the way in which the discretion has been used, the ability of the Treasury Department to single out exchange rate manipulators has, I think, been a useful thing, both in the context of particular instances of manipulation, and probably even more as a deterrent to exchange rate manipulation for commercial advantage.
I think in general there is much more risk in the present context that the United States will do too little on the global front than that it will do too much. I think that President Biden's step to make available 20 million vaccines was an important and valuable step, but I think the United States has been substantially insufficient in its support for a strong global response to COVID, which in light of the fact that the main dangers from COVID right now come from evolution, and evolution of the virus is as likely in Lagos as in Los Angeles, I think we have a great stake in an aggressive global response to containing the virus. I think we have a great stake in assuring what's not now in place, which is a satisfactory architecture for dealing with debt problems if and when they come. I think there's a good chance that we will see very substantial pressures on emerging market debts sometime in the next few years.
Altig: Let me kind of push on both of those comments. With respect to COVID, obviously there were fits and starts in our own dealing with COVID in the country. What would a more aggressive and appropriate international policy have looked like?
Summers: It would have involved vastly greater funding for COVAX [COVID-19 Vaccines Global Access], for ACT-A [Access to COVID-19 Tools Accelerator], for the various efforts to support developing countries through this and to make vaccination pervasively available. It would, as I warned years ago, have involved very substantially greater budgets for pandemic preparation than the world engaged in. I am concerned today that we are going to focus an enormous effort on putting COVID in the rear-view mirror, and that we're not going to make a sufficient effort to be ready for the next pandemic.
Looking at the record of COVID, MERS, Ebola, N1H1, the flu, other SARS, and other diseases, I think it's a reasonable judgment that we're going to face a COVID-like threat at least once a decade going forward. That means the world needs to be in a very different place than it is right now with respect to the availability of resources to provide for testing, to provide for masking, to provide for vaccine production. I think it's very difficult in times of emergency to adequately be thinking about the next emergency, and yet I think that is absolutely essential right now.
I would say the central banking community has to date been roughly 50 to 100 times more focused on issues of climate finance than on issues of pandemic finance and of readiness to deal with the next pandemic when it comes. I think that's been an error on the part of the central banking community and I hope that the central banking community will think about what it can do to make sure that there is adequate finance available for more rapid and more decisive responses the next time the world faces a pandemic threat.
Altig: Is it your position that the climate focus is an inappropriate place for a central bank to put its energies, or it's just a matter of priorities and what are the bigger threats?
Summers: I think that central banks should be focused on making sure that financial stability is maintained and using the tools of financial policy that are at their disposal to assure that economies are able to function well. It's been my view that as grave a problem as global climate change is, I have not seen stranded assets associated with climate change policies that are not appreciated by the market as being a terribly important source of systemic financial risk. I think they are dwarfed by a whole set of issues in the shadow banking system, for example. I think central banks, in order to be relevant to something that's on political leaders and citizenry's mind, have rather stretched things in the degree of emphasis they placed on those kinds of stranded assets issues.
On the other hand, I think that assuring that credit is readily available to get masks to people, to enable relief payments to be made...I think that is a more pressing issue of systemic stability, and one that central banks, perhaps over time, need to be more focused on. I certainly would welcome greater initiative in Basel [Basel Committee on Bank Supervision], or in other central banking forums, around the risks associated with pandemic.
There's always a question as to how broadly versus narrowly focused you want central banks to be, and I can appreciate a variety of perspectives on that question. What I'm more confident of is that they have reached much further beyond their traditional mandates with respect to the climate change issue than they have with respect to the pandemic issue, and that the payoff at the margin to more focus on pandemic is likely to be greater.
Altig: Thanks. I do want to remind everyone that you can use Pigeonhole to submit any questions you would like considered. I want to get back to that; I want to get to that shadow banking comment in a second, but before I do, you also mentioned that one of the areas of insufficient forward-looking focus is emerging market debt. How do you think we ought to be positioning ourselves, either through fiscal policy or through monetary policy, in light of that concern of yours?
Summers: Some of it's about monetary and fiscal policy, and I think that we have underestimated the risks—very substantially—both to financial stability as well as to conventional inflation of protracted, extremely low interest rates, even in economies that are performing strongly as the U.S. economy is right now. I think with respect to emerging market debts, the principal issue is that we have substantial amounts of debt...that if you look at the size of the credit spread, you have to think there's a meaningful chance that that debt is going to need to be rescheduled or restructured.
We have a set of institutions that are well-suited to traditional creditors and traditional borrowers, but are not well-suited to a world like the present in which there is very substantial lending through bonds to private sector entities in emerging market countries, and in which there's very substantial lending by Chinese institutions that are not party to the Paris Club and not party to other standard institutional arrangements. There's an understandable reluctance to plan excessively for default, just like there's a reluctance to plan excessively for paying ransom in the event of kidnapping. You don't want to plan for failure, that will make failure more likely.
On the other hand, I think we are left with an architecture that is really quite suspect. I, along with Guillermo Ortiz, Jean-Claude Trichet, and others, have put out a report under the auspices of the G30 that looks in more detail at these issues. The substantial part of it involves the establishment of appropriate institutional architectures that would prevent a variety of problematic debt practices that will make restructurings that are likely to be necessary very, very difficult if that need arises in the future.
Altig: We have a question in that's related to this. The direct question is: "Are you concerned about whether a tightening phase in the U.S. is going to serve as a sort of tipping point for these problems?" And I guess a related notion would be is there time, in your view, to position ourselves to avoid any serious sorts of disruptions associated with pursuing what may be viewed as appropriate monetary policy in the U.S.?
Summers: I believe that the serenity that is being projected, with respect to inflation, the policy projections suggesting that rates may not be raised for close to three years, are creating a dangerous complacency that make it more likely that when, as I think is quite likely, there is a strong need to adjust policy, those adjustments will come as a surprise and jolt market participants in ways that will do real damage to financial stability, and may do real damage to the economy.
So I believe the idea that the balance of risks is anything like equal between deflation and inflation, between financial bubbles and credit problems, between acceleration and slowdown, I believe the reading that suggests that we are near equipoise is very far off of an accurate reading of the U.S. economy right now. And that to the extent that it is widely internalized, as I think it has been by market participants, we are setting the stage for, and indeed partway through, a vicious cycle in which rising inflation, coupled with constant nominal rates, leads to decreasing real rates, which leads to expansionary pressure, which leads to more inflation, which leads to lower real rates, and so forth. And so, I think we are underappreciating the extent to which we have put a very substantial, procyclical bias into financial conditions by anchoring nominal rates.
Altig: So, how do we get off of the stage that we are setting?
Summers: I think we need, and it needs to be done with care and gradualism, but it needs to be done clearly, to signal that we are in a very different place in our assessments of the economy than we were six months ago. That the primary risks today involve overheating, asset price inflation, and subsequent financial excessive leverage, and subsequent financial instability, not a downturn in the economy, excessive unemployment, and excessive sluggishness.
The rhetoric regarding future policy needs to adjust in a substantial way. It is not tenable to assert today that in the contemporary American economy, labor market slack is a dominant problem. Walk outside; labor shortage is the pervasive phenomenon, and the failure to recognize that, and the failure to begin an adjustment to that reality, puts at risk the kinds of mistakes that we have not seen made in the United States for a long time.
The new Fed framework put in place in Jackson Hole last year was a reasonable response to secular stagnation coupled with COVID. It is not a reasonable place for policy to be in a world where the budget deficit has been expanded by 15 percent of GDP by stimulative policy, in my view.
Altig: You referenced specifically the employment part of the mandate and framework. I presume, maybe I shouldn't presume, you feel the same about the average inflation targeting mechanism, and the implicit sort of backward look rather than forward look.
Summers: I believe the old-fashioned way, "we strive for price stability, and we'll know price stability when we see it," is the appropriate way to address that part of the Fed's mandate. I believe the focus on a numerical target has never been a helpful contribution to policy. I never saw why obsessing about 1.7 [percent] versus 2 was valuable, nor do I see the virtue of obsessing about 2.3 versus 2. I think the operative challenge is to recognize, what are the primary risks in any given situation?
We adopted a framework that was appropriate to a world where the primary risk was sluggishness, the primary risk was credit problems,; the primary risk was deflation, and we signaled in a broad way that we were seeing those things as the primary risks, and we were prepared to take other chances in order to avoid those risks. That was an appropriate general direction to set. It should have been set, in my view, as I said at the time, with more awareness that the world changes, and more awareness that you can't know where things are going to be, and so in a less detailed and prescriptive way.
The general orientation of it was correct; but no one could have imagined that when that framework was set, that eight months later we would have 15 percent of GDP in additional stimulus committed, with another $4 trillion headed forward in stimulus. No one could have imagined that the vacancy rate, the job openings rate, would be approaching record levels. No one could have imagined that survey expectations would be on fire, both for future sales and for future inflation, to the extent that they are. In that context, we need to recognize that we're in a very different place than we imagined we would be, rather than pledging fealty to the judgments we made in an entirely different context.
I think there's both a lesson about now in this, and there's a broader lesson about the dangers of trying to commit future policy for advantage today, and the dangers of relying on forward guidance or descriptions of reaction functions in a world where what Don Rumsfeld called "unknown unknowns" are pervasive. I think that both domestically and globally, we need to adjust ourselves for what Mervyn King has written about, "a world of radical uncertainty," and that includes being willing to adjust in the face of developments.
Altig: So just to knock off a question that came in: there was a conversation yesterday about whether a target should be 2 percent or 4 percent. I gather you think we ought to not be focusing on that question at all.
Summers: I don't think that. I certainly would not support a 4 percent target; I think a 4 percent target will become a 6 percent target. I think that 2 percent can be understood as reflecting price stability, but I think the efforts to have numerical targets...it's far from clear to me that the benefits exceed the cost slope. In the Fed community, we focus very intensely on certain price indices, and for the people who attend a session like this the distinction between the PCE [Personal Consumption Expenditures] and the CPI [Consumer Price Index] is a salient and meaningful one.
Both those indices have the property that housing is acting as a drag and an anchor on measured inflation. Well, out there in America, where for the first time in American history that we can measure a majority of houses are selling for more than their asking prices...out there in America, where house prices by some measures are up 18 percent in the last 15 months, they think we've got a major phenomenon of house price inflation, and that bears on the question of price stability. However, some owners' equivalent rent measure is calculated within the CPI. I think we've got to bring back the exercise of judgment, rather than mechanical modeling in the treatment of inflation.
Altig: We've got a question in that is pushing back about your judgment on the labor market. I'll read it directly. "The payroll employment in May was still more than 8 million below where it stood before the pandemic. How are you squaring that with your observation that the labor market is tight everywhere you look?"
Summers: My judgment that the labor market is tight everywhere I look comes from everywhere I look there are vacancies, people eager to fill the vacancies, and unable to fill the vacancies without very rapid wage increases, and that even with very rapid wage increases having started, the vacancies remain. Why is that happening at a time when employment is still lower than previous peaks? I think the logic of supply and demand tells you that when you have price up and quantity down that you've had a supply shock.
Looking at that data, you would be inclined to think there was a supply shock. And then, if the government is paying large numbers of people more to not work than they're able to earn by working, it would stand to reason that that would create a supply shock. And if there were substantial fears and inhibitions coming from COVID, around the fact that people are not able to work, that would also create a supply shock.
I find there to be a reasonably consistent interpretation of the data that there's been an adverse supply shock to labor at the same time that there's been a very substantial demand shock, and that those two things together point towards increased prices, and not such great quantities. That's exactly what we're seeing, and the appropriate monetary policy response to that is surely not to push your foot even harder down on the accelerator.
Let me repeat: constant nominal rates in an environment of increasing inflation expectations represent a kind of policy easing. I think the important debate is maybe this supply shock is all going to solve itself when unemployment insurance benefits revert to normal, and when kids go back to school in September. That's certainly possible, and certainly that's what I hope we see.
It's odd for the people who are most enthusiastic about expansionary policy to be most committed to the belief that unemployment insurance isn't having any effect on labor supply. I'm surprised every time I hear my friends in the administration assert that unemployment insurance is has having no effect, since I would have thought the best argument for their kind of fiscal stimulus was to suggest that we're going to get a big surge in labor supply in September.
I'm very impressed by the work that just came out from Melissa Kearney and Jason Furman. It basically shows that the labor supply for adults with young children has not gone down more than the labor supply for adults without young children. And so, the idea that we can think of this supply shock as somehow related to schools and all of that, and childcare responsibility, is just not in the data, so it's just not right.
Now it may be that, as everybody feels like they can take their mask off, large numbers of people are going to go back to work. I think there will be some of that, but it would surprise me, as the stimulus checks are spent over the next months, as the overhang of $2 trillion of savings is worked off, as everyone goes back to the stores, and onto airplanes, and starts to take trips, if that's a world where the number of job openings goes down. It could happen, but that would not be what I would expect without some significant upwards pressure on wages.
Altig: Let me flip this conversation to the financial price side of things. You're drawing a picture of supply and demand imbalances that are not in the right direction, or maybe they're in the right direction depending on your point of view, on inflation concerns that are front of your mind. Why have we not seen a sharper run-up in market yields?
Summers: First of all, it depends on how you look at these things, I guess. I haven't looked at this in the last several weeks, but I think that of all the first quarters in the last century, the run-up in the 10-year yield was greater in the first quarter of 2021 than any other quarter except for 1980. I would argue we've seen a fair-sized move in markets.
Second, the Federal Reserve has been doing its best to suppress any tendency towards an increase in yields by remaining committed to QE policies, and not even beginning a process of considering the end of QE policy. To some extent, we've been suppressing the natural market kind of response.
Third, if you look at what I think is even more relevant, if you look at what's happened to break evens as inferred from the spread between index bonds and nominal bonds, those have continued upwards and reached their peak for the last decade or more this month. I think it's not right to say that markets have a sharply different view.
I think the other part of it, frankly, is that market prices are to some extent set globally. Policy in the United States has been substantially different. The kind of massive fiscal stimulus that we've had is not something that's taken place elsewhere. I think the lesser inflationary tendency, lesser labor shortage tendency elsewhere, has been somewhat inhibiting of increases in interest rates.
Altig: This time has gone lightning fast. I'd be hard pressed to say that your views aren't pretty clear on these things, but let me just one more time tie this up. The Federal Open Market Committee is committed to reviewing how it's thinking about things annually in its long-run strategy document. I'm presuming you think that they ought not wait until next January to do that, and the time for reversal in policy is now.
Summers: I've got enormous respect for Jay [Powell, Federal Reserve Chair], for your President [Raphael] Bostic, for Vice Chair [Richard] Clarida, for New York Fed President [John] Williams, for all the members of the FOMC. I know from my time as an insider in government that people on the outside often say things that they might not say if they knew all the things and all the considerations that I had. I'm reluctant to make tactical prescription, or to criticize specific comments and specific actions.
I would say that I think the prospects for avoiding turbulence over the next several years, both in the real economy and in financial markets, would be substantially greater if there was a sense that monetary policy authorities in the United States were focused on the need to avoid overheating rather than focused on the need to reassure people that they won't focus on overheating. I would rather see us go back to a Fed that is concerned about preempting inflation, rather than a Fed that is concerned about preempting fears that it will be concerned about inflation.
I think that points towards the desirability of some change in what's being projected from the Federal Reserve System, how that integrates with policy reviews, how that integrates with specific meetings, is something that I would not presume to judge. I would caution that I'm offering my best judgments, that I offer them in the spirit of recognizing that all policy choices are judgments under uncertainty that involve balancing risks, and that anyone who is certain of almost anything is making a mistake.
Also, with an awareness that from within the system, one has access to information and sees things that are hard to see from the outside. I think outsiders like me can make a positive contribution by raising concerns and being a bit of a pressure point against inertia, given that it's always a tendency for policymakers to want to stick with the beliefs that they've had. But equally, it's important to respectfully recognize that these are very difficult judgements, and that the Federal Reserve is sitting on top of a vast body of information.
Altig: Professor Summers, thank you so much. You didn't disappoint on any dimension there. We will now take a break, and the program will continue at the top of the hour at 11 o'clock. Thanks, everyone, for joining us.