Crypto and Consumer Choice: Q&A with Giovanni Compiani
Professor Compiani shares insights from his research on cryptocurrency demand, investors’ beliefs, and consumer behavior.Crypto and Consumer Choice: Q&A with Giovanni Compiani
With lockdowns easing and COVID-19 infections rebounding in many countries, policymakers and corporate leaders face a profoundly uncertain future for which historical data and precedents may not be reliable guides.
Against that backdrop, Randall S. Kroszner, deputy dean for Executive Programs and the Norman R. Bobins Professor of Economics, led a discussion with Mervyn King, former governor of the Bank of England and coauthor with John Kay of Radical Uncertainty: Decision-Making beyond the Numbers. This virtual event was hosted by Booth’s campus in the heart of the City of London and marked the first in Booth’s Road to Economic Recovery series on COVID-19.
Kroszner, a former governor of the Federal Reserve System, told the audience one of his takeaways from King’s book was that making rational choices under uncertainty was not a “simple card game where you know what the probabilities are.”
King agreed, saying it was time to move away from the classical economic idea that everyone is assumed to know the subjective probabilities of any given event. COVID-19 was a case in point.
For example, said King, he and his coauthor wrote their book in summer 2019, and in it they warned that people should expect an epidemic resulting from a virus that did not yet exist.
“But that didn’t mean we had the power of foresight and could make statements that the probability of a virus coming out of Wuhan in December 2019 was 17 percent or 32 percent or any other number,” King said. “There’s no basis for making that statement. The essence of our book is that we say, ‘Don’t pretend that we know more than we do.’”
Avoid the Data Trap
King said decision-makers should focus on collecting data that would help them understand key aspects of the problem. However, he cautioned against what he called “bogus quantification.”
He cited the 1980s HIV/AIDS crisis in southern Africa. The World Health Organization built a complex computer demographic model. Robert May, the late renowned scientist, pointed out that it used the average number of sexual contacts per person per year as a key parameter without stipulating whether that was 100 contacts with the same person, in which case the epidemic would die out, or 100 contacts with 100 different people, which would mean it would spread like wildfire. “Homing in on the key numbers that matter is of fundamental importance,” King said.
Focus on Resilience
Kroszner said CEOs facing uncertainty right now wanted to know what lessons they should learn. King said they should focus on resilience and robustness rather than just cost minimization or profit maximization. “It’s all very well to minimize your costs in normal times. But if a sudden event comes along and you fail to survive it as a business, then you’re gone,” he said.
“It makes sense to ask, ‘How would we cope with a pandemic?’ and forget the probability. Just as in the financial crisis, we had allowed the banking system to go into 2007 with little resilience. When there was a shock, it couldn’t withstand it.”
Have a Clear Narrative
Kroszner said one lesson he took from his time at the Fed was that Chairman Bernanke was open to criticisms that certain solutions were not addressing the problem. “He made sure people understood there’s a framework, but there’s also the flexibility to pivot.”
King said that there is now a common belief that central banks are “the only game in town” because of their ability to cut interest rates and inject stimuli. But he said when people are worried about contracting a virus, lower borrowing costs are unlikely to encourage consumers to visit shops and restaurants.
The focus should also be on governments’ ability to use taxpayer resources to replace businesses’ lost revenues, contingent on how effectively the country is coping.
“Governments need to have a clear narrative to navigate a path between two very unattractive rocks. One is the rock of a large number of infections and deaths, and the other is the rock of a very large cost in terms of lost economic output. We have to navigate it, discovering through trial and error what works and what doesn’t work.”
This virtual event was the first in Booth’s Road to Economic Recovery series on COVID-19. Watch the full presentation below, and visit the Road to Economic Recovery website to stay updated on future events in this series.
Veronica Guerrieri: Good afternoon everyone. Welcome to the first virtual event in the Booth Road to Economic Recovery series. My name is Veronica Guerrieri and I am the Ronald E. Tarrson Professor of Economics, and Willard Graham faculty scholar at the University of Chicago’s Booth School of Business. I hope you’re all doing well in this difficult time, and I’m very pleased to see such an amazing turnout today. Thank you to Lord King for today’s event. Unfortunately, the current pandemic is an unprecedented event that is having dramatic consequences not only in terms of human lives but also in terms of economic impact all over the world.
Policymakers, corporate leaders, and nonprofits face uncertainty and daily challenges in their decisions to the road to recovery. Today is the first event in Booth’s Road to Economic Recovery series on COVID-19. All events will take place digitally and will include discussions with industry leaders, policymakers, and other experts from Booth, but not only. To open the series today, we have an excellent program. I have the honor to introduce our highly distinguished speaker, Mervyn King, who has served as a governor of the Bank of England between 2003 and 2013.
He was previously deputy governor from 1998 to 2003, chief economist and executive director from 1991, and nonexecutive director of the bank from 1990 to 1991. He was knighted in 2011, made a life peer in 2013, and appointed by the queen to be a knight of the Garter in 2014. Mervyn is the Alan Greenspan Professor of Economics and Professor of Law at the New York University and emeritus professor of economics at the London School of Economics. He’s also co-author joint with John Kay of the new book, Radical Uncertainty: Decision-Making Beyond the Numbers.
To moderate the conversation with Mervyn, we’re thrilled to have Randall Kroszner, who is deputy dean for executive programs and Norman R. Bobins Professor of economics at Chicago Booth. Randy served as governor of the Federal Reserve System from 2006 until 2009, and he chaired the committee on supervision and regulation of banking institutions, and the committee on consumer and community affairs. In these capacities, he took a leading role in developing responses to the financial crisis and in undertaking new initiatives to improve consumer protection and disclosure, including rules related to home mortgages and credit cards. Let me leave now the floor to Randy and enjoy the event.
Randall S. Kroszner: Great. Thank you so much, Veronica. I really appreciate that introduction and also, I really appreciate the incredible research that you with your co-authors have been doing on the pandemic. I mean, one of the things that’s really been amazing is how quickly research has come out, and I think particularly from my colleagues here at Booth, that are relevant for the current times and the current issues. Obviously these are things we’ve never seen before, and so that’s posing an enormous challenge for doing research as well as for doing policy, and that’s really the crux of the book that Mervyn and John Kay wrote on Radical Uncertainty, thinking about the challenges of making decisions and the whole decision making structure and process when it’s not a simple card game where you know what the probabilities are, you know what everything is. You don’t know exactly the outcome. You don’t know which card is going to be drawn, but you know how many cards are in the deck.
Here we don’t even know what deck we’re playing with, and that was, I think also true a decade ago when Mervyn and I were both in our policy positions at our respective central banks dealing with the financial crisis. With that brief intro, I now want to turn to Mervyn. Again, thank you very much for being here and being with us.
One of the standards that we typically teach in economics is that you have rational choice under uncertainty. As I said, no one says that we know what the outcomes are going to be, but we have a framework for trying to make those decisions. You just simply try to gather data, quantify risks, and then you make the trade-offs to make the better decision. Now we call that rational choice under uncertainty.
When I read your and John’s book I get the feeling that’s irrational to do that. Maybe you could give us a little bit of a framework for how you think about decision making in these challenging times.
Mervyn King: Well, thank you, Randall and let me say first, it’s a pleasure to join you and in particular, to see you in your splendid London campus with St. Paul’s. behind you. It looks really beautiful. Decision making under uncertainty. You’re right that economists tend to say that we know what is rational and that’s because of the very first line of the argument. They assume that we know all possible future events. We can write a long list of everything that could conceivably happen, and everyone is assumed to have some subjective probability which we attach to those events.
Once you know that, then you can go on to argue that the right thing to do is to maximize your expected happiness, utility, whatever you want to call it. Rational decision making becomes a rather straightforward thing, although the application to individual problems can be complex and difficult, and if you get it right, you can get a Nobel Prize for it if the problems are difficult enough, but no one goes back to challenge that very first step, which is that everyone is assumed to know what the subjective probabilities they attach to events are.
That really goes back to the 1920s when a young Cambridge mathematician, philosopher, economist, Frank Ramsey, who argued that you can ask people what odds they would bet on various outcomes and events. Provided the bets were bets that didn’t guarantee that they would lose money, that is, they behave in that sense rationally, you could infer from those bets that they were subject to probabilities that they attached to all events, which obeyed the normal laws of probability. The weakness in that argument is a very simple one, which is that most sensible people don’t bet on most things. Partly because they don’t believe they know enough about it. They believe that the person offering them the bet may well know more about the outcomes than they do and so they’re unwilling to bet on those outcomes and there’s just no basis for assuming that people can attach these probabilities.
The best illustration of this is indeed COVID-19. In the book, John and I wrote that . . . we wrote this last summer. That we must expect an epidemic of an infectious disease resulting from a virus that does not yet exist, but that didn’t mean that we had the power of foresight and could say, make statements of the kind that probability that there will be a virus coming out of Wuhan in China in December 2019 is 17% or 32% or any other number. There’s just no basis to making that statement. What we have to do is to be content with the statement. It’s likely that at some point there will be an epidemic coming out from a virus that doesn’t yet exist, and therefore we need to prepare for it just as in the financial crisis.
We knew the banking crises existed so that we knew something about them, but that didn’t enable us to say the probability that there will be a collapse of the US financial system in September 2008 is any particular number, and so the essence of decision making under uncertainty is recognizing that most of the big decisions we have to take are four events, which are one-offs. They’re not repetitions. Your card game, tossing a dice, they’re not repetitions of a stationary probability distribution. The world is non-stationary. That’s what differentiates the behavior of business, the economy, human behavior from the behavior of the planets or some scientific phenomenon. It isn’t just the playing out of a stationary distribution where we know the odds. The rules, the situation is changing. The model driving this all the time is changing.
The essence of our book is that we say, what we need to do is in any given situation, don’t pretend that we know more than we do. We know something, or we must use that information. We knew something about pandemics, but don’t pretend that we can attach numbers where we have to make them up in order to fill in the cells in a spreadsheet. Let’s just ask ourselves the question, what is going on here and try to work out, puzzle through the decision we have to take. That is what all good decision takers whether in central banks or businesses actually do. They don’t sit down and maximize some mathematical function.
Randall S. Kroszner: Yes. I certainly know that, and we both know that from a decade ago. You mentioned one very important, Frank Ramsey, but there’s another very important Frank, a University of Chicago economist, Frank Knight, who in 1921 published a book, Risk, Uncertainty, and Profit where he was coming at it more from the economic side rather than the statistics side, but came to the same kind of conclusions that there is a very important distinction between things that we’re . . . we know the probabilities, we don’t know the particular outcome versus things where we’re just not sure about what’s going on here. Getting at exactly what you had said.
He was a very important influence at University of Chicago. He actually trained Milton Friedman, George Stigler, others who went on to get Nobel Prizes. James Buchanan for example, and I think that’s an important part of the University of Chicago tradition, is drawing that important distinction. Not that every University of Chicago economist adopted that, but I think that is an important part of our tradition and that’s why I think it’s great to have you discussing these things.
One of the things that I think is super important for both policymakers and businesspeople is exactly what you were getting at. What is the right question to be asking? And this is something that comes up a lot in my classes and also now being an administrator at the business school about, well, what’s going to happen next and what’s going to happen next with the pandemic. It strikes me that that’s not the right question to be asking for exactly the reasons that you outlined, Mervyn. What basis can we draw from making any statements about that? But what we can do is perhaps ask other questions that are more specific to some of the risks that may be posed by that to try to get at it.
I want to move from . . . because I think it’s extraordinarily important that we shouldn’t have an excess of confidence in our ability to put down numbers that have very tight standard errors on them and know what’s happening, but then how do we take that next step to say, “Okay, well, how do we make decisions?” Because obviously we make decisions in this world. How do you ask the right questions, and then how do you make those decisions?
Mervyn King: I think the key is to ask the question, as I mentioned before, what is going on here to try to tease out the things that are really important to focus attempts to collect data on those aspects of the decision. John and I are not against models or quantification, but we are definitely against bogus quantification, of which there is a very great deal. One of my favorite examples is that when AIDS started to spread, particularly in Southern Africa, the World Health Organization started to put together a very large, very complex computer model of the demographics of the whole of Southern Africa, linking together demographic models of different countries, and they wanted to predict how quickly AIDS would spread.
Well, the one good decision they made was to invite Bob May Australian, went to Princeton, then Oxford in England. Became chief government scientist in the UK A mathematician, biologist. He and Roy Anderson of Imperial College literally wrote the textbook on the epidemiological models that people are now using. He went out and he looked at their work and he said, look, there’s a key parameter here in your model, and it’s the average number of sexual contacts per person per year in each country. This is a key parameter, but don’t you realize that if the . . . let’s suppose the answer that they’d collected from surveys was say, 100 contacts per person per year. Don’t you realize if it’s 100 contacts with the same person, that it’s utterly different from 100 contacts with a 100 different people. The former means the epidemic will die out. The latter means it will spread like wildfire.
What you need to do is forget the complexity of a model because that’s irrelevant. What you need to do is to go out and discover more about the sexual mores of the people who may be spreading AIDS through Southern Africa, and that’s what they did and they discovered that it was being spread quite by often by lorry drivers moving through Southern Africa. Homing in on the key numbers that matter is of fundamental importance, yet those violated in almost all attempts to apply cost benefit analysis to a well-defined problem in trade public policy where you’re given a spreadsheet and you have to fill in every cell in the spreadsheet, and you end up making up numbers and you don’t realize which ones are the really important ones that are driving the results.
What quantitative work should be doing is trying to tease out from the problem which other numbers that will really matter here to this decision and get the orders of magnitude right. It doesn’t matter if it’s to the third decimal place. What does matter is that you’ve got the right numbers that matter that will determine the outcome, and you have some feel for what the order of magnitude of those numbers are. That’s not something that fits naturally into a belief that what we’re observing is the playing out of a stationary distribution that’s playing out day after day, day after day, but all we need to do is just to feed in the numbers into the spreadsheet and out will come a number that tells us whether the decision is a good one or not.
It’s very rarely of that kind. Most of these decisions are one-offs and understanding the nature of it is far more important than pretending that there is a spreadsheet where we just fill in the numbers into each cell, press a button and out comes the answer.
Randall S. Kroszner: I think that’s incredibly important because it’s about, as you said, asking what’s going on here. What is important about this issue and what data might be able to get to actually help us to deal with it rather than sort of rely on the sophistication of a model or something like that. I recently had a conversation with one of our professors Eugene Fama. He’s one of the founders of, and really developers of using data in finance to try to think about risks and risk trade-offs.
He is incredibly sophisticated econometrically, but I asked him about his approach to the data and he said, you never want to use those complicated things. He basically said the same thing that you had said. You really want to understand what’s going on here. Try to understand the basics, the fundamentals. Sure, you can be fancy and put all these other things to dot the I’s and cross the T’s, but you really want to get the big picture and the understanding, and I think that’s something that is often lost both in policymaking and in business, especially when something like this comes up that’s so sort of scary. People want to sort of fall back on, quote, “science” or something that seems very sophisticated, but what you really want is something that’s very straightforward asking what’s the relevant issue. What’s the relevant question.
I want to move from this bigger picture to some of the challenges now. What would be the key questions that you think other businesspeople should be asking and thinking about what they should be doing next, or that the policymakers should be asking and what they do next for policy? What are the best questions, and where should we be focusing our data gathering efforts?
Mervyn King: Well, I’ll give you one example from COVID-19. The epidemiological models that are used are helpful in understanding the nature of an epidemic. That is, it tends to start slowly. You don’t really notice or have enough information about it until it suddenly accelerates, reaches a peak and comes down again. Early on, various modelers particularly at Imperial College were asked to make predictions about what the course of the epidemic would be and they turn out to be not very good predictions, which is hardly surprising because in these highly non linear phenomena the results are sensitive to the values of the parameters that you feed in, and there were many key parameter values that we just didn’t know.
It was a new virus. What was the nature of this virus? We didn’t know the fatality rate, for example. We didn’t know enough about the nature of the virus. We didn’t know how people would respond to various lockdown measures, and we certainly don’t know how they’re responding as we try to ease the restrictions that have been imposed on people. One of the messages from that was, well, we need to know far more about the fatality rate of this virus, but to do that what we need to do is to do a large random sample test of the population in order to find out how many people have got the virus and then to infer what the mortality rate is. Because if you . . . you may feel you can measure the number of deaths reasonably accurately, or that even that is actually mired in controversy now, at least in the United Kingdom where many people have been described as dying from COVID-19 when in fact they were tested months before and died from something else.
The realization that we needed to know more about this parameter and therefore the way to do it was to carry out a large random sample test was not followed for a very long time. The reason I think was that understandably all the people in the medical profession are basically focused on treating patients. They’re brought up to help a patient, and what you need to do in a public health emergency like this can be very different. Despite the fact that both the United States and the United Kingdom were described by various international bodies as the best prepared countries for a pandemic, it turned out we weren’t, and we weren’t in a number of different dimensions.
One of which critically was that we didn’t have in place a series of procedures to learn about the nature of the pandemic. We assumed we’ll know about the pandemic, how do we respond to it? But we didn’t know the nature of this pandemic and we therefore ought to have put in place measures to find out more about it. I think this is one of the key lessons that you mustn’t sort of pretend through bogus quantification that we know enough to make predictions. I was very struck in Britain that early on the chief medical officer was asked by a journalist in the daily press conferences, “So how many people are going to die?” And he said, “I don’t know.” And that was the best and most honest answer that anyone gave in this whole episode.
As the episode continued, what you could see was that many of the medical experts were sucked into giving very positive advice about exactly what you should do, and the trouble was that as we learned more about the virus, that advice changed. You lose credibility if you say I’m certain today that you should follow this course of action, the next week you say, but I’m equally certain that today you should do something different. You have to tell a narrative in which you explain to people we don’t know very much about this virus.
This is a sort of advice we think makes sense today, but we’re going to try and learn more about it and modify the advice we give you as time goes by. That way you build up credibility because people think you’re being honest with the public, but if you pretend to the public that you know more than in fact you do, they will come to realize that over time and then you lose credibility and you can have some very adverse consequences as a result of that as I think we’ve seen in other medical areas, such as the anti-vaccination campaign which had been very costly to society in terms of deaths, particularly of children.
I think credibility comes from being honest and being prepared to say in response to a number of questions, I don’t know, but that answer is one that very few politicians feel able to give.
Randall S. Kroszner: I agree. I think there’s so much pressure to say, we know what’s going on and we’re going to do this to solve it, but I think you’re exactly right that what . . . I think it’s difficult to just say, I don’t know. You have to have sort of the next step and the next step was obvious to do randomized controlled trials. I mean the most recent Nobel Prize winners in economics received the Nobel Prize for doing exactly this in emerging markets. People weren’t sure what was effective for education. They weren’t sure what was effective, actually did a lot of work on infectious diseases. One of the challenges is that people were estimating the frequency of a particular disease by looking at the data they have. It gets back to exactly what you were saying, Mervyn. What they were doing is collecting the data from the local health clinic.
Well, of course, a very high fraction of people who go to the health clinic have a health challenge and if you try to estimate the frequency of the disease from that data, you’re going to get a number that has nothing to do with reality. You have to go out and do randomized control testing, and you can do that in health. You can do that in so many other areas and they’ve actually done a lot in this. There’s a well worked-out field in this, just got a Nobel Prize, but what’s amazing is that we haven’t seen this anywhere. People should have said, I think exactly what you said. We’re not sure, but we’re going to find out, and this is how we’re going to find out.
Instead, as you said, they didn’t do that. I recently gave a talk on a joint seminar that we did with the Centre for Economic Policy Research that Booth did a joint workshop with them, and I titled my talk, “The Data Deficit.” A lot of people have been talking about the fiscal deficit, but not about the data deficit. If you look at the fiscal expenditures from the $3 trillion, so 15% of US GDP, that has been put forward, the most generous less than 8% could be said to go for gathering data for directly addressing the disease issues and finding out more about it. It’s just not sexy to do that and it’s also suggested that you don’t really know what’s going on, but it’s super, super important.
Is there a way to get over this in the public policy realm? And also, is there a way for CEO’s to do this? Because obviously they’re facing something very similar within their companies, either talking to their employees or talking to their investors.
Mervyn King: Well, you’re right. It does apply to businesses too. I hope that one thing that will come out of this episode is a greater focus on the importance of resilience and robustness rather than just cost minimization or profit maximization. It’s all very well to try and minimize your costs in normal times, but if a sudden event comes along like this episode and you fail to survive it as a business, then you’re gone and for most good companies, that’s not what they want to see happen. It makes sense to ask the question, how would we cope with a pandemic. Forget the probability of it. I mean, there’s . . . we know enough to say that it is likely that there will be another pandemic at some point in the future. We can’t be more precise than that, but that’s enough to know that we need to think about how we would cope with it.
I think that there will be an adjustment of business practices to focus more on resilience and robustness, just as in the financial crisis it became very clear that when we asked the question in 2008, early on in 2008, what is going on here? The answer was not a shortage of liquidity but deep concerns about the underlying solvency position of financial institutions, which then led to the symptom that people weren’t willing to lend to banks because they didn’t know whether they were short or long in many of these complex financial instruments and we had allowed the banking system going into 2007, ’08 to go into it with very little resilience.
The equity buffers were too small. There were no holdings of liquid assets, and the result was that when there was a shock, then the banking system couldn’t withstand it, and afterwards we then put in place regulatory measures to improve the resilience and robustness of the banking system. What we didn’t do was to ask the question, are there other parts of the economy that we also need to think about the resilience and robustness of, and there are many aspects of our economies ranging from potential weaknesses in the cyber area or are IT systems resilient. Are we absolutely confident that we have an electricity supply system that is resilient and robust to a number of different outcomes? The risks of terrorists with nuclear weapons in a suitcase in a major city.
These are things which threaten our survival and they’re worth spending time and effort thinking about how we can improve our resilience of our businesses and our economic and society systems to prevent the downside risk from these things occurring, and I think . . . to get back to something you mentioned earlier. One of the important conceptual distinctions we need to reintroduce into economic debate is the difference between risk and uncertainty. That was Frank Knight’s great contribution.
It’s rather unfortunate that the person who argued against it was another Chicago economist, Milton Friedman.
Randall S. Kroszner: And Frank Stevens actually. He was a student of Frank Knight.
Mervyn King: He was.
Randall S. Kroszner: ... so it’s interesting. Yes.
Mervyn King: And he argued against it, and I think he was wrong and I think there is a real distinction and it matters because risk is something we want to avoid or risk is a bad outcome relative to what John Kay and I call a reference narrative. That is, a story of how we think of business or our lives will evolve, and a risk is a downside outcome relative to that and we need to take steps to insure against that or to ameliorate the event if it were to occur.
Uncertainty on the other hand is something we, in many ways, we should welcome because as Frank Knight pointed out, uncertainty, that is where you can’t quantify the uncertainty is the essence of what entrepreneurs are all about and entrepreneurship is something that’s essentially been missing from a large part of conventional economic analysis. A lot of work is done on technology and patents, et cetera, et cetera, but the process by which people come up with new ideas which no one has had before, work with other people to put them into practice is the essence of entrepreneurship and it’s part of uncertainty.
Uncertainty is indeed the spice of life. If we could tell all our students on the day they graduate, that these are the three or four careers they could follow and these are the probabilities attached to each of these career paths. Here’s a list of the seven people who could be their life partners with probability attached to each of them, they would go home deeply depressed. When the students say to me, this very uncertain . . . there’s a lot of uncertainty about my future. I say, “That’s the most wonderful thing about it.” When you get to the end of your life and there isn’t a lot of uncertainty left about your life, that’s when you start to worry.
So uncertainty is the spice of life, but risk isn’t. Risk is something we should guard against and take actions now to minimize the chances or ameliorate the consequences of those risks and that’s why it is important to distinguish between risk on the one hand and uncertainty on the other.
Randall S. Kroszner: I very much agree. I tried to describe things in a similar way. My inspiration comes from a great entrepreneur in the US, Peter Thiel, who wrote a book called, Zero to One. He said, it’s the creation of something out of nothing that people hadn’t seen before. That’s what true entrepreneurship is. It’s sort of easy if someone has already done this for you to just maybe do it a little bit better, but to see something that other people hadn’t seen, to create something. Sometimes it gets back to these broader discussions of . . . deeper discussions of some free will and such as everything determined or not.
The kind of probabilistic world is where everything’s kind of determined, then you just have to figure out the probabilities. What I would consider, and I would agree with you in a sort of more optimistic world, people can create new things. You create the future. You’re not just given it and you figure out what the probabilities are and you’re just kind of stuck. You create that new future. You create the possibility of finding new partners, the possibility of new things that people might be interested in that they’ve never thought about.
I think that’s a positive message in all of this. Obviously we’re facing the challenges now, but that uncertainty is not something that is bad, but it’s actually something that can be . . . it can be inspiring. I want to turn to some of the questions, we’re getting questions in now. Please feel free to send more questions in, and it gets at exactly this. One of the questions was in both within the firm context as well as in the policy context, you certainly want to be honest that we’re not quite sure what’s going on here, but how do you still inspire hope and confidence rather than people saying, oh my goodness, he doesn’t know what he’s doing, or she doesn’t know what he’s doing. I’m either going to sell the shares or I’m going to move out of the UK or . . . How do you get that balance right?
Mervyn King: Well, I think, by telling a narrative which is as compelling as can be given our lack of knowledge. I’ll just give one example. I would sometimes go to the parliamentary committee in London and be asked a question, “So where will interest rates be in a year’s time?” And I’d say, “I don’t know.” “What do you mean you don’t know. It’s your job to know.” And I’d say, “No, it’s not. I don’t know where interest rates will be a year from now, because I don’t know how the economy will evolve. If you tell me how the economy will evolve, then I can tell you how we’ll respond to it.”
We have a stable response functional, reaction function in the jargon of economists, but what will actually happen to interest rates depends on what will happen in the world. I don’t know what will happen. We can tell a story about what we think is going on and we’ll do that and we’ll revise our narrative every three months and tell you what we think we’ve learned over the previous three months and whether our narrative has changed. And you should trust us because if over the previous 10 years, you have seen that at least we can give an intelligent description of what’s going on without pretending to have a sort of crystal ball that gives us a magic view of the future. We don’t have that and it doesn’t make sense to pretend that we do.
I think one of the mistakes that central banks have made in recent years is to be sucked in to this position of saying well we’ve done, what we can in terms of cutting interest rates and printing money, and let’s think of something new to do. Why don’t we tell people where interest rates are going to be in the next two or three years, but they can’t possibly know that because they don’t know what the economy is and what you saw therefore, is that both at the Fed and at the Bank of England, the statements they made under the heading forward guidance basically had to be retracted because the world changed and therefore they didn’t want to raise interest rates in the way they had previously said they would and they changed tack.
It was sensible to change tack. It wasn’t sensible to try and predict what you would do yourself in the next 12, 24 months. I think that it is a question of trying to explain to people why we don’t know certain things, because people are perfectly capable of understanding that. I think, I’ve been astonished in the last 10 years in politics in the UK at how politicians talk down to the population and assume that they really are dumb and they’re not. They may not be technical experts, but they can certainly tell when someone is spinning them a lie.
We saw that in several cases. In the Brexit referendum, both sides of the argument in the public debate engaged in what we call bogus quantification and they misused numbers, or they made up numbers to bolster their arguments. The Remain side said, every family in the UK will be £4,300 worse off a year if we were to leave the EU. I’ve met many people who said, no, I have no idea whether this number is right or not, but I’m sure that they can’t possibly know that with that degree of precision. They can’t know that. There’s too much uncertainty about it. They just inferred the politicians were making up the numbers, which they, in large part were and the economist advising them bogus assumptions, false comparisons between episodes in the past and today.
You do not gain credibility by doing that. I think that regular frequent communication with your staff, your team around you, the wider public is essential to build up a reputation that you’re open and honest and don’t pretend to know things that you can’t possibly know.
Randall S. Kroszner: I very much agree and I think it’s important to do that, not just in a crisis but before the crisis comes. As you did in your testimony before the parliamentary committees, it wasn’t just in a crisis saying, well, I don’t know where interest rates are going to be. You gave a foundation, you gave a structure for people being to understand that so that when the crisis then comes, you can say, well, remember I told you that before. That’s how I think about these problems and then this is how I’m going to go about trying to answer the problems.
That’s actually another one of the questions that’s come up. Are there ways in which we could accelerate gathering data, particularly in a challenging situation like this? A sort of getting to, one, asking the right questions about the data and two, accelerating getting that data both for CEOs in a business context as well as policy makers.
Mervyn King: I think that it depends very much on the situation and the question. I mean here with COVID-19, it would’ve made sense to have launched . . . This could have been done relatively quickly, I think, a program to do mass testing, several hundred thousand people being tested to find out whether or not they had the disease. There are other situations where you have to make decisions much more quickly. If you’re managing an emergency response to a situation that would be a fire or a riot. Or take the Cuban missile crisis, where I think what President Kennedy did having learned from the Bay of Pigs episode earlier in which he began to realize that the advice he was getting from the military then, or the intelligence services was misleading.
What he decided in the Cuban missile crisis was to create a small group of people, less than 20, and divided them into two groups when it became clear that there were two major courses of action. An air strike on Cuba or a naval blockade, and to get them to come up with the strongest possible argument for their particular preference. Then to get them to critique the other side so that he built into this process without much delay because this was taking place over days, not weeks. A process in which the narrative of each alternative course of action was being challenged and this is vitally important because you need a narrative, but you need to challenge the narrative all the time.
There are different ways of doing it, that the Fed and the Bank of England, we did it with committees making our decisions on interest rates where the debates among the committee were a way of challenging the narratives that different people held, and in the White House in that Cuban missile crisis episode, it was a very effective way of ensuring that people were challenging the narrative and interestingly, Kennedy deliberately didn’t go to many of those discussions because he didn’t want to be in a room where people were thinking, “Oh, what can I say that the president would approve of.”
That way you don’t get the right discussion and debate. You have to think about the mechanisms for doing it and then you get . . . you want alternative views. You make your decision. You don’t sort of criticize or dismiss the group whose advice you didn’t take. You thank them for it and you create an atmosphere in which people feel that their job is to come up with arguments and counter arguments, and then you will stick with the decision that’s been made.
Randall S. Kroszner: I know. I think that’s very important and that’s very similar to the kind of approach that Ben Bernanke took at the Fed. He was always open for people to raise issues and questions and he would give responses to that. Some people in crisis situations manage by raising their voice and say . . . shouting people down. He didn’t do that, but he would also say, well, we do need to move to a decision relatively quickly on things and understanding that we didn’t always know that this was going to be the solution to the problem and being open to revisiting something.
Many of the programs that we stood up during that time, particularly things related to the money markets, we had to do multiple programs to try to address the underlying issue and we were ready to do that. We were willing to be open to hear criticism and concerns both from market participants as well as from others in other central banks about, well, that’s not really working and that’s not really addressing the problem. I think having that of making sure that people understand that there’s a framework, but also having the flexibility to pivot when you’re getting more information and to show that you are learning and that, that is part of the process of what you’re doing, I think is very important in maintaining that credibility.
We’ve gotten a number of questions about policy and so I’m going to kind of put a few of them together in saying, okay, so now you’ve been talking about this. Given the data that we have, if you were back in a policy position not necessarily at the central bank, but let’s say in a broader policy position, what would you think would be the most important thing to either do in the US or in the UK or in Continental Europe right now to try to deal with what’s going on here today?
Mervyn King: Well, I’m not in a position where I have access to the information which governments have. Let me start with central banks. I think there has been a very unfortunate tendency to describe banks as the only game in town and to think that if economic growth falters for any reason, the answer must be for central banks to regenerate growth to ease monetary policies. There is no basis for that in theory whatsoever. There are many reasons why economic growth could falter. Only some of those are amenable to central bank action.
Monetary policy is essentially a short-term counter cyclical instrument and after 10 years of slow growth, we ought to have come to the realization that further cuts in interest rates and yet more money printing isn’t really the answer. There was of course, when the pandemic started in earnest in the West, a concern about financial instability in mid-March and then central bank–created liquidity to prevent disturbances in financial markets, but this is not a moment to argue for monetary stimulus.
The idea that we are either . . . I mean, first of all, when the lockdown came in and it’s still being implemented in many parts of the world, if the government is trying to shut down the economy with one hand, it makes no sense to try and stimulate it with the other. If you’re told you can’t go to work, you must stay indoors. You can’t go to restaurants or bars, no point giving people money to spend. When you come out of the restrictions, if people are nervous about going to restaurants and bars, and certainly in the upper age groups, I think people will be nervous about that until there’s a vaccine. Then the idea that slightly lower interest rate is going to make them willing to save a bit less, borrow a bit more in order to go to their favorite restaurant is not very plausible. You need to worry about their concerns about the health situation.
The focus should not be on central banks. It should be on governments. Governments I think have a responsibility to support business through a period in which business is being told, you can’t open. This isn’t a recession in the normal sense of the word, and I don’t think it’s helpful to use words like recession, because then people think of the normal tools to deal with a recession. We don’t want to do that. This is a situation in which the government is basically saying to business, we’re suspending a market economy for the time being therefore, I think government has a responsibility to replace the lost revenues of business in one way or another with transfers from taxpayers in general until such point as we can get the economy back to somewhere close to where it was before the restrictions were imposed, and we don’t know when that will be.
I think having arbitrary time limits to some of these schemes isn’t very sensible. They need to be conditional on how effectively we’re coping with the spread of the disease and how far we’ve pulled it down. I think the focus should be on governments, not on central banks. And on governments, they obviously need to have a clear narrative. I think you don’t want a situation in which the health experts say these are the health measures that have to be implemented because we don’t know enough to give a precise interpretation of that, and the cost to the economy of the lost incomes, output, jobs in GDP is massive.
Indeed, if you were to calculate how much money you would be willing to spend using the normal measurement of a value of a life, let’s say, we’ve lost in GDP far more than that relative to the number of deaths. I think that the measures that have to be taken are to navigate a path between two very, very unattractive sets of rock from either side, one of which is the rocks of large numbers of infections and deaths and the other is the rocks of a very large cost in terms of lost economic output. Both are very unattractive. There is no simple way to say, “This is the right path.” We have to navigate it discovering as we go through trial and error what works and what doesn’t work.
Therefore, I think absolutism in terms of a policy response is not very sensible at this stage. I think it is important to try and find ways to reopen the economy, but with testing and tracing so that if you see outbreaks of infections in certain areas pick up, then you have to reverse course and institute a localized shutdown. There’s no doubt that as we are learning more and more about this virus, we are probably seeing that the fatality rate from now on is a bit lower than we thought it was at the beginning, which is encouraging, but we’re also seeing how infectious this disease is. I mean, it’s extraordinarily infectious.
Look what’s happened in Australia. Australia had very good marks for the way they handled this episode. Much smaller numbers of infections and deaths than the US or UK. Really, very small numbers and everyone thought they’d done it. Slightly reopened the economy in Australia. What happens? Melbourne is now shut down. Five million people in Melbourne told to stay at home. This is a very, very difficult situation. We shouldn’t pretend there’s any easy way through it. It is trial and error, but finding ways of getting more information, the thing that we ought to be willing to do is to spend very large amounts of money not only on treatment because if you can minimize the fatality rate through better treatment, that’s a big plus, but obviously most important of all is a vaccine.
That won’t kill the disease. I mean, we’ll still be living with this disease for many, many years I suspect, but it will be . . . With a vaccine, if people take their vaccination seriously, it won’t be as big a threat as . . . well, not a big enough threat to justify shutting down the economy again. Spending more money on finding a vaccine, I think we are. I mean, I don’t criticize what governments are doing. I think they are putting a lot of money into it and there are many attempts around the world. Maybe this is one of the rare opportunities in which international cooperation will naturally come forward in which we’ll all support teams around the world and then when the first two or three come up with a successful vaccine, we’ll all share it around the world.
Randall S. Kroszner: Speaking of international cooperation, I mean the European Union has just done something unprecedented. They have gotten together and agreed on a three quarters of a trillion Euro package to provide budgetary support as well as undertake other actions European wide. This hadn’t been done before. I wanted to get your thought on, what do you think of that approach that they are taking and then what metrics would you use to judge whether they’re using that three quarters of a trillion euros wisely? Because this is really unprecedented, there is a very big move forward in cooperation within the EU.
Mervyn King: Well, up to a point. It is at one level but it certainly is not a ship to a fiscal union.
Randall S. Kroszner: For sure.
Mervyn King: It took them five days to agree on this. Half the money approximately is in grants and half in loans that will have to be repaid. That’s much less of a burden sharing. Clearly the scenes of distress in Italy and Spain were very important in influencing the attitude of countries in the north to be willing to make special payments. The future occasions on which there could be challenges to the monetary union are unlikely to be one in which people can see that the cause of the problem was some external events such as either a pandemic or some... an earthquake or whatever, where people are willing to make transfers
It’s more likely be in the context of a sovereign debt crisis where countries in the north will not want to take responsibility for the consequences of profligacy of countries in the south. I don’t think they’ve resolved their underlying problems. It’s obviously all being exaggerated, the benefits of this. People are talking up the whole scheme as a great new event. In Europe, I think that’s an exaggeration. After all, some of the countries in Europe put up frontier ban. I mean the whole spirit of free movement of people just went out of the window when the pandemic came along.
I think they are a long, long way from moving to a political or fiscal union. I wouldn’t want to exaggerate that. The scale of it looks very large on paper, but actually the amount of money which the UK treasury has been handing out to the private sector in the UK in terms of either direct payments or lost tax revenues is much greater relative to GDP than this particular package. I don’t want to underestimate it. It’s obviously taking them a long time to agree on this and it’s something which is better that they agree than if they fail to agree, but equally I think it’s . . . one shouldn’t just say all this is a big new chapter in European history. I think that will be an exaggeration.
Randall S. Kroszner: If we wanted to think then, because there’s another question about what metrics would use to evaluate the performance and management of this tragedy both on the economic side and the health side, what particular metrics would you look at if you wanted to assess the relative performance of the US, UK, countries in Asia, et cetera?
Mervyn King: I think it’s important not to do an exercise with the intent of apportioning blame-
Randall S. Kroszner: For sure.
Mervyn King: . . . nor is it a good idea to do it with a benefit of hindsight. I think we have to know at each stage what did people know at the time, what efforts did they make to try to discover more? How do they try to get more information? Metrics in terms of ability to acquire and deliver both protective personal equipment for people working in the health services and care homes. I think certainly in Europe, there’s going to be a postmortem into what happened in care homes for the elderly, where in countries with very different approaches to this episode, we have seen seriously large numbers of deaths in care homes often linked to, I think, to the fact that the attempt to sort of quarantine them overlook the fact that the people working in the care homes, a) often came from abroad and b) worked in more than one care home, so they were taking the virus themselves from one care home to another.
Understanding that I think is going to be very important, trying to get a proper estimate of what you might think of is the true mortality rate. I think this is very difficult because clearly the consequences of COVID-19 is that what leads to death is often a result of co-morbidity. That seems very obvious in terms of the nature of the virus. We still don’t really understand why the virus has apparently a differential effect not only by age, but by ethnic background, often geographical area. I think a postmortem needs to be not just a question of saying, well, this country gets eight out of 10, another country, four out of 10. I think it’s hard to do that.
There’s no doubt that in the UK and I suspect the US as well, we didn’t score very highly in terms of logistics so that the ability to acquire and distribute PPE quickly, the ability to implement a testing strategy and to ramp it up very quickly, that varies a lot between countries. Why did it vary between countries? What was it used for? Were decisions about the population who would be tested, the sample population, were they good or bad decisions based on what people knew at the time? I think these are the things to look at.
I don’t think it makes sense at this stage just to say, the mortality rate per 100,000 of the population was X1 in one country and a smaller X2 in another country, therefore second country did better. I think that’s too simple. I mean, just obvious points, there seems to be a correlation between mortality with COVID-19 and obesity and countries vary in the degree of obesity. When we were thinking about strategies and policies about obesity in years before, no one ever said, when that COVID-19 comes along, we will realize the cost of obesity. It never occurred to anyone. I do think that we have to do postmortem without using the benefit of hindsight.
Randall S. Kroszner: Sure, sure. Which is always a challenge and certainly in the press, as we well know from being policymakers the press never follows that and with 20/20 hindsight, they always know that you had done something wrong or why didn’t you do this at that time? Then I think it’s very important as you said, to think about what data are available. One of the criteria that I would use is very much consistent with what you’re saying is, what were they doing at the time to try to get data, to try to understand what was going on? Were they just saying, well, we’ll do some testing, we’ll do this or that. Were they really trying to be systematic about it and saying, well, let’s try to set up randomized controlled trials.
Actually Germany, I think has done reasonably well on this. In Munich a few months ago, or just a month ago, they started doing some randomized trials. They’re doing this in the UK. They’re starting to do some of that so I think that’s something that if we could somehow get people to realize that that’s super important, that the learning part, it’s not the fatality rates per million people that eventually will be affected by that but in the short run, there are a lot of other factors that are coming into that and also we can see that as these things evolve, somewhere that had a very low death rate earlier may have a high death rate now and vice versa.
Really understanding that, as you say, the comorbidities, but also I think the process that they undertake of taking seriously that we’re not so sure and we want to find out what’s going on here is super, super important, which I think is the fundamental of your book-
Mervyn King: And I think there’s very [crosstalk 01:00:18]-
Randall S. Kroszner: We’re just past our time, so just very quickly Mervyn.
Mervyn King: There’s a very important point, which is the countries that were thought to be well prepared for pandemics had prepared for pandemics on the assumption that it would be very similar to ordinary influenza. If you get such detailed preparations in place because you believe you know what the disease will be, you can go badly wrong. The first thing you need to do when there’s a pandemic is to say, well, what is this virus? What is going on here?
Randall S. Kroszner: Exactly. Sort of understanding that you maybe . . . you can try to draw on some history, but being fully straightforward we need to gather the data, understand whether it is like the past or not. I think that’s a crucial, crucial point. Well, Mervyn, thank you so much for participating in our kickoff. This is a great way to kick this off, and globally, I know we’ve got people from the US, throughout Europe, Middle East, Africa, and some people from Asia in. We will be taking a break from this in August but starting again in September.
One thing that I wanted to highlight is we will be doing a virtual global conference on September 10th that is going to be related to corporate social responsibility and we call it Corporate Social Responsibility Revisited. We’ve talked about Milton Friedman a number of times here. September is the 50th anniversary of Milton Friedman’s famous essay on corporate social responsibility, which has been interpreted as him saying, it’s all about profits, not about anything else. I think a more careful reading of the paper doesn’t say exactly that, but that’s exactly what we’re going to be debating and thinking about how things have evolved in the last 50 years in that conference.
We’ll be starting in Asia, then coming to Europe, Middle East, Africa, then finishing up in the US and then the Stigler Center will have a more academic conference that will flesh that out, and I think many of the things that we’re discussing today really get at these issues of what are the responsibilities of businesses for keeping people safe, for keeping their employees safe. What are their responsibilities? What data should they be gathering? How they should be acting? And those would be some of the questions that we’ll be looking at then.
I look forward to seeing you again in September as the Road to Economic Recovery continues and for the Corporate Social Responsibility Revisited conference on September 10th. Thank you so much. Bye bye.
Professor Compiani shares insights from his research on cryptocurrency demand, investors’ beliefs, and consumer behavior.Crypto and Consumer Choice: Q&A with Giovanni Compiani
As economists reflect on the lack of representation in their field, a Booth professor is collaborating with his peers to fight for change.Fostering Diversity and Inclusion in Economics
Amid the pandemic, students worried the New Venture Challenge would be too hard to move online. But with their professor’s help, they rose to the occasion.How One of Booth’s Most Intense Classes Had a Banner Year Online
Stay informed with Booth's newsletter, event notifications, and regular updates featuring faculty research and stories of leadership and impact.
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.