Toggle Menu

Ask the Expert Webinars

Ask the Expert (in a crisis): How to Forecast Economic Outcomes in a Crisis


April 30, 2020

Max Simkoff: Secretary Summers, I wanted to start with a few questions around advice to consider based on previous experience dealing with similar situations. It’s only natural that we look to recent economic crises to try and find relevant analogs. And clearly, comparisons to the 2008 financial crisis seem to be the ones showing up everywhere. Is there anything about what we’re seeing right now that makes that even a relevant comparison at all? What is similar about this crisis and what is different?

Larry Summers: What’s similar is that events we had not been factoring in as important risks suddenly come to define the situation. The degree of breakdown-in-meltdown after Lehman failed was not something that had been in the universe for most investors prior to the breakdown. The kind of events that took place with Covid were not in the universe of thinking for most investors. So broadly, what’s similar is that there’s a highly salient unknown, or you could call it Black Swan.

Or, you could go back to the famous economist, Frank Knight, who distinguished between risk and uncertainty. Risk is like the probabilities at a roulette table, where you know that you don’t know what’s going to happen, but you can calibrate the possibilities. Uncertainty is a broader concept of ignorance. So left tail events are what is in common. What is different is the sheer magnitude of this event. You can measure it in terms of unemployment. The number of people who apply weekly for new unemployment benefits has exceeded 3.5 million for five weeks now. And the worst previous week in the last 40 years was 800,000.

What is dissimilar is the way in which this is all leading us to live lives of restriction and confinement. That wouldn’t have been in the universe of possibility if we had been asked in January. What is dissimilar is the magnitude of the drop in GDP. So it’s dissimilar in scale and in particulars. But broadly, there are common features to crises, shocking left tail events, de-leveraging contagion as elements of financial crises.

Max Simkoff: You’ve pointed out the similarities and the dissimilarities from 2008. Are there other crises that we’ve seen even in the last hundred years? Certainly the Great Depression comes to mind — that would be more relevant for us to look to as comparisons to think about how to respond here.

Larry Summers: In some ways, the 1918 flu — the Spanish flu episode — is a very relevant historical episode. The Great Depression I suspect, I hope, is a less good precedent because it played out over a much, much longer time period than one hopes that this one will play out. We have seen in many emerging markets so-called sudden stops in economic activity when financing dried up.

In other respects, the kind of cooperation that’s necessary between central banks and governments, wars in the past against foreign adversaries, now against a microbe, are the right parallels. So I think what’s usually true is that history doesn’t repeat itself. But as somebody said, it rhymes, and there are elements of this — elements of a variety of events — that one has to look at to inform views and in judging outcomes.

Max Simkoff: Got it. A number of folks on the webinar here are senior executives at financial institutions, having to shift dynamically between being reactive on a daily basis and also putting new plans together. Based on what you’ve seen in previous situations, where the conditions and the resulting outlook for improvement shifted significantly over mere weeks of time, how would you advise organizations to be dividing their time between planning mode right now versus just purely reacting to real time movement in the market? Is there any general rule of thumb on how planning time should be allocated during a crisis like the one that we’re in?

Larry Summers: There’s a famous quote, that’s attributable to Eisenhower, that plans are absolutely useless when the battle comes, but it’s nonetheless essential to have them and to make them.

I guess my general views and general experience would be whatever you think you’re going to have to do in two months, you should probably do today. And that if your idea is that there’s something that you don’t have to do yet, but you’re very likely to have to do in the future, it’s probably better if you face it and do it today. The corollary of that is you can’t bounce till you hit the floor.

And those who hit bottom quickest start to climb fast. So I think there are many, many more examples in history of entities that underreacted and reacted too slowly to gathering storms. Then there are, too, entities that overreacted to gathering storms would be kind of a principle of reaction.

I remember when I was overseeing the team that was working full time and that really deserves the credit for having managed the automobile bailout during the 2008/9 crisis for the Obama administration. I remember having a conversation with Jack Welch to get his advice. We talked about February or March of 2009, and he said, “What’s it look like car sales are going to be?” and I said it had been 16 million and it looked like they were gonna be 11 million. He said the company should be figuring out how to be viable at 9 million. And I said, do you really think it’s going to go that low. He said, no. He said pretty unlikely that it will go that low. But they need to be ready. If it does, and if it’s clear that it’s going to go to 9 million, it’ll be too late to get ready for 9 million — and a bunch of stuff they’ll do planning for 9 million will turn out to be good to have done if it ends up being that car sales were 11 million. So I think I would overreact. I would generally counsel overreacting on the downside in the time of crisis.

And I would say, if your plan doesn’t involve something that sees the crisis as a strategic opportunity, then either you’re in bad shape because there’s nothing you can do in a crisis that is a strategic opportunity, in which case you should sort of soul search about the place of your institution in the cosmos, or you’ve got the bad plan and you’re too focused on minimizing opportunity.

So, for example, people have asked my advice at Harvard and I’ve said, no, Harvard’s strategy should not be to try to figure out how to absolutely minimize layoffs. Harvard’s strategy should be to use this moment to engage in layoffs that are probably the right thing to do, but would be too painful to contemplate and too jarring in normal times. And that Harvard should, among universities, have deep pockets, and therefore, there should be opportunities to recruit superstars who would otherwise be unavailable, but aren’t getting the kind of support they think they need from their existing institutions. Well, that’s special to one particular context, but I think the principal — look for the creation of strategic options and react very aggressively in the short run — would be the general advice I would give relative to what are most institutions instincts.

Max Simkoff: And that’s actually a great transition to a question or two about predictions on how we might capitalize on opportunities as we’re waiting to return to a new normal. And this is more specific now to the mortgage market: We’ve seen an unprecedented amount of dislocation in the secondary mortgage market right now, where despite significant government purchasing of mortgage backed securities, many large lenders are actually tightening mortgage lending standards. For example, number of large lenders, mortgage lenders have announced in recent weeks the requirement of a 700-plus minimum credit rating floor or not accepting home equity lines of credit (HELOCs) at all. And so what is your opinion on the effectiveness of these policies at the mortgage originator level? And are there more effective ways for mortgage lenders to digest the volatility of the secondary mortgage market right now, while not having to constrain origination via tighter borrowing at a time when it’s probably needed more than ever?

Larry Summers: One of the things that I think is important for people like me who hold themselves out as experts on certain things that people care about, is that when the honest answer is “I don’t know,” that we’d be willing to say, “I don’t know” — and I’m not enmeshed enough in the mortgage market to provide any answer that’s intelligent. 

I guess what I would say is this: If I look at the 5 to 10-year Treasury yields that are normal benchmarks for looking at mortgage yields. They’ve declined 125, 140 basis points since some date early in the year. And if I look at the rates at which I could get a mortgage in order to buy a new house or in order to refinance my existing mortgage, they’ve declined, first approximation, zero.

And if I looked at people in less fortunate circumstances than I, what their availability of getting that mortgage is, it’s down not up. And so it can’t be that we have a well-functioning economy. If at a moment when consumers are more strapped than ever before, at a moment when it would be more important to stop a declining house price dynamic that could feed into a recession at a moment when supporting consumer incomes through reduced debt burdens has been especially important that we’re having. Max, we’re having minimum pass-through of monetary policy in the form of lower rates to the mortgage market. And if I was in government, I would regard that as a pressing problem to solve.

Now to what extent? That is because the Fed has liquidity facilities for everything except mortgages — or in greater scale for everything except a variety of things outside of mortgages. To what extent? That is because of mortgage servicer problems with liability in the context of a period when we’re having a rising level of at least temporary nonpayment. And to what extent that is due to the fact that a very clumsy industry has been made much clumsier by the fact that people can’t go into offices and are working from homes.

I don’t know, but I guess the fact that extra clunkiness is leading to larger spreads and therefore more profits for a variety of financial intermediaries rather than benefits for a variety of middle class homeowners — if I was in government — I would be very, very heavily focused on those issues.

Max Simkoff: Got it. And maybe staying on a similar theme of if you were in government and you were the sole decision maker on whether we needed a new round of stimulus. What specific pieces of information would shape your view each time you had to make a decision, and would they be the same factors as if you were, say, setting interest rates?

Larry Summers: Well, I’m not sure that I relate well to the concept of stimulus in a context like this. I think that the way in which I would be thinking about things is if people are too scared to go to the store, and they’re too scared to go to their job if there are other people who sit near them at their job, then there’s probably not going to be much economic activity and it doesn’t matter what the interest rate is. It doesn’t matter what the budget deficit is if people don’t leave their homes. There’s not gonna be much production of GDP.

So I would think of that task of policy as less being stimulus and more being support — supporting the capacity of the economy to resume as rapidly as possible. So the first lesson that I would draw from that is the need for, Max, what I would call macro money for micro health policy.

If we were running this right — it looks like the administration is actually moving to get right — we would be taking the five best vaccine candidates or the seven best vaccine candidates and we’d be producing all of them. And we would know that we would have produced 40 million doses of some vaccine and it turns out not to work, and that we would have wasted $5 billion. But when this is costing the economy $80 billion a week, more than $10 billion a day, if we end up wasting several billion dollars on some kind of vaccine that doesn’t work,. But as a consequence of that, we’re able to reopen the economy a week earlier and we’re able to accelerate the path back to normal by a week — that is a very good investment. And something similar can be said about testing. And the first level should be any money that can be spent effectively on treatment, on vaccines, on testing, on contact tracing. That money should be being spent, and I don’t think we’re having that kind of attitude. That’s the first thing that I’d say.

Beyond that, I’d be tracking the available information — as best one could create economic forecasts. In general, I think we need to make sure that people are well-supported through this. I think we need to be very careful to find ways to target money well, and that we should be targeting places where there’s real distress because of covid. But I’m concerned about programs for which, for example, I’m eligible to get money from the government to help pay the people who helped me with my consulting. There’s no reason why I should be able to get that. There’s no reason why a variety of institutions should be getting large amounts of essentially free money from the government, because I think we’re going to need that money for a long time to support the recovery of the economy.

Max Simkoff: Got it. We’re going to move to the last part here for taking some live Q&A. I’ve got a few questions come in. And just as a reminder, if you have a question, you can use the Q&A button down at the bottom of the screen. Secretary Summers, this is one that is certainly on the minds of a number of folks on this webinar. And it relates to when we might return to the levels of normal mortgage lending activity in 2019. The question reads: Six million people bought a home that required a mortgage and 3.7 million people completed a refinance. When do you think we’ll get back to that level of activity and will it be a linear return?

Larry Summers: I spend my days thinking about a very large number of issues, and there are many of you on this call who are much more focused on the mortgage market. So I would caution you not to give my opinion too much weight. But since you asked, I would say, first I do not think it will be linear. I think that things will feel kind of closed down. And then at some point things will open up and there will be big backlogs of people who’d been looking to buy homes. People have been looking to put homes on the market.

And so I would expect that — if you think about economic activity in the context of a power failure — the power’s off and there isn’t any economic activity, and then the power turns back on and there’s more economic activity than there was before that. That’s the right kind of way to think about this rather than the linear model.

I don’t think you’re going to get back to normal until we have a universal vaccine or cure, and I don’t think we’re going to have a universal vaccine or cure for a while. If we’re there on July 1st, 2021, that will be a good outcome from my point of view. If I could make a commitment right now and we would be there on July 1st, 2021 and it would be no sooner and it would be no later, I would take that relative to my senses, the uncertainties.

Max Simkoff: Great. Another question that just came in, and again, just for timing purposes we’ll probably take one or two more of these before we close this one. Secretary Summers, what do you see as the future of cities versus suburbs versus rural areas? Will we see a shift of economic activity out of large coastal cities towards less dense, lower tax, lower cost areas?

Larry Summers: If you look at what has happened to the relative price of a house in Scarsdale outside of New York versus an apartment in the middle of Manhattan, or the relative price of a condo in Back Bay in Boston versus the price of a house in a suburb like Lexington, there has been a dramatic relative price change over the last 25 years. I think there’s a possibility that that pattern is going to change quite substantially. And certainly I’ve been exposed to, anecdotally, evidence of people who had been thinking of urban townhouses and had changed their plans towards suburbs with better public schools and lawns.

I am hesitant to make a confident prediction because there was a fair amount of that kind of talk two months after 9/11. Sort of for the same reasons, safety, comfort, all of that. And it turned out to have no real enduring kind of impact. So I am unsure how much it will lead to de-urbanization.

My guess is that whatever the right predictions were about urbanization before, the combination of crowded cities feeling more dangerous and a growing sense that you could do more of your work from home and so you may not have to commute in every day, so traffic congestion is a less serious issue and commuting time is a less serious issue, and perhaps a sense that people are going to be more willing, more desirous of having their kids spend more time at home rather than in crowded daycare and after school type settings, that all of that may drive some kind of re-suburbanization, if I can call it that.

I think that it has to be that the substantially worse performance of New York City relative to other American cities through this has to be sobering with respect to mass transit as a strategy, and that reinforces the same kinds of tendencies. So I think my guess would be a meaningful re-suburbanization impact. But I’m very much in mind of the conversations people had in November of 2001 when they overstated the impact of September 11th [on residential housing trends].

Max Simkoff: Got it. And maybe just as a last question before we close here, I’ve got one that’s asking: Many have criticized the reluctance to shut down entire parts of the economy fast enough by some leaders, both nationally and internationally, in response to the leading indicators and the virus rapidly spreading in late February and early March. What do you think we learned from that more recent experience that can be applied if we see further flare ups in the next three-to-six months as we try and reopen the economy? And what advice would you give leaders to act such that they minimize further extended closure of critical parts of the economy in these situations?

Larry Summers: There are many contexts where a stronger, faster response ends up being a cheaper response. A stronger military deterrent can mean you have to fight less war. I’m associated with some set of doctrine around financial crises that if you put more money up more quickly and more strongly as a guarantee, you end up having to spend less because you instill confidence more quickly. You know, I think in general, we’re seeing that the places that moved more aggressively to stronger shutdowns are the places that are able to come out of this in better ways. So my main advice I think would be: Test fast and respond very vigorously with very strong kinds of quarantining policies.

And I have to say, I’m really quite concerned that we’re having a bit of a strategic letting up. If I felt that our letting up was very carefully designed, with careful testing so we would find hot spots early and special, extensive efforts to protect the vulnerable and all that, I would be more comfortable with where we are as a country. But it feels more like we’ve done our time, we’re kind of sick of this, we’re going to let it rip a bit — and I’m worried that that will be an expensive set of choices.

Max Simkoff: Thank you, Secretary Summers, for taking the time to talk with us today, your insights on the current crisis and how to navigate uncharted territory, as well as where we can find relevant historical analogs, has been extremely valuable and I really appreciate the time, so thanks for joining us.

Larry Summers: Thank you very much, Max. And it’s good to be with States Title in the States Title team and their many partners. Thank you very much.

Max Simkoff: Thank you.