Prof Venky Venkateswaran - Faculty Insights: Covid-19 and NYC Series

Prof Venky Venkateswaran - Faculty Insights: Covid-19 and NYC Series

Show Video

Thank. You everyone welcome. Back to one. More of the faculty. Inside, series. Today. We have with us professor Venky. Venkateswara. From, the. Economics, department at, Eastern, though, you did not see Venky, earlier, in. April. He, was a co-author also, of the presentation. By Tama Philippa, on, optimal. Mitigation, policies, during a pandemic. So, think he has clearly been one of the true real-time. Researchers. During. The last few months he's. Going to present today his work with Julian, Kozlovsky, and Nora well camp, on. An important, issue which, is I'm sure at the top of the minds of everyone. Which. Is weather when as and when the dust, settles, on the pandemic, will. It be back to life as normal or will. There be some long lasting effects, of, the pandemic. Over. To you Weinke and. We'll do Q&A once. You are finished with the presentation, thank you. Thank. You so much well and, thank you all and, thanks for including. Us in this series, it's. A pleasure to present this and this forum and, thank you all for attending I, look forward to you, know hearing your thoughts and, comments, as. You can imagine in a spirit of mention this is real-time research, so you know I think it's, still very much being, cooked as, we speak. So. I think. I. Think the the the goal, we try other the the question, we are trying to grapple with here. Is what are the long-term longer-term. Consequences of. The kovat, 19 and then it can be. Associated economic. Crisis, that it is then, it is coming on a spot. So. The, hypothesis. We're going to develop or at least trying to kind of quantify and, measure, is. That an event, of this magnitude. Reshapes. Our thinking, about risk, when I say our thinking I just don't mean us as researchers I mean businesses. Consumers, what have you so everybody, is as, about risks going forward, has. Changed, just a little bit but it has changed and that is going to induce a change in behavior which. Would in turn have implications, for economic. Activity going. Forth and, so that's kind of our hypothesis. And. What we're trying to try and do is to quantify, that, how big is this is this small relative, to what we are currently going through or is, this, roughly. Of the same order. And. The, way early on our approach in this paper is. We're. Gonna start. With a very simple model a pretty. Standard, model and our may be it with a couple of new elements here and there now point out where they are where what those are as we go along and. Then kind of simulate, a kovat, style. Event, in that in. That, particular. Episode and. In. That particular model and then try and measure what. The longer-term Casas and just. A headline the result when, we do that we find that these long-term cost the the negative consequences. Of this belief, change, are. Many, times larger than, the, last, GDP. In 2020. 2021. All right so that's kind of our main, result. And I'll try and show you where. Alright, so, here's, the idea in a nutshell. We're. Gonna pass it that, make. A very realistic premise, that nobody knows the true distribution of, aggregate sharks in particular, sharks, to the return to capital, everybody. We, estimates, they're these. These. Distributions. As new data arrives we're. Going to kind of put a particular we're, going to put some structure on it we're, going to kind of keep it as flexible, as possible we don't want to assume the result we don't want to kind of take in, something. That makes the magnitudes. Much bigger than they are so, we're going to let the data speak, so. To speak. And. That's going to be the core of our of the paper core of the belief the, core of the, belief formation, process, and I will show you more about that in a few slides and then, we're going to kind of simulate code 19 by throwing in like an an extreme, adverse. Into that model and. As I'm sure you that's going to induce these large fairly. Long, live changes, in beliefs and in particular, assessments.

About These disaster. Risks assessment, about something, really bad happening, in. In. The future, and. Then we're going to kind of build that into a macroeconomic. Model and then show how. What. That means for for. The trajectory, of the US economy going. Forward and. The macroeconomic, framework itself, is based on something we're used, to. Explaining or to think about the, post financial crisis. Experience. Of the US economy if, I have time at the end or a human a I'll show you what, the model says about that and. The fact that that framework turned, out to be somewhat successful in. Understanding. This. Episode, is. What kind of led us to use this as our as, our kind of benchmark for thinking about the. Cope with nineteen situation, as well all right so that's kind of our that's. The paper. So. Let me start you know maybe that we don't have a lot of time let me kind of walk you through the main ingredients, of the model there's, going to be a fair amount of math here on the slides but I'm gonna try and given that we don't have a lot of time I'm not going to walk you through all the equations, but I'm going to hopefully give you the highlights of what's really in there so, the model has kind of two blocks one, block is. The. Epidemiological, stuff. And there's a hundreds. Of papers written you. Know about, how. Epidemiological. Considerations. Shape. Policy. Shape. Behavior. Of agents, and so on and so forth we're, going to take a slightly, different tack here and given our given. Our codes so, the basic model is something that you know you, may have seen already. It's. What's called the susceptible. Exposed infected. Recovered. Framework, or the sei our framework, and. The basic, idea here is, that the, number of new, infections of. The disease are a function of the, compact, rate between members, of the population right. So, that's kind of that's the basic idea of the so. Called R naught. That you. Often hear that meteorologists. Talk in terms of what, we're going to and, so what we are going to do is basically. Map. A realized. History of infections. As. Denoted by this Delta IT term to, behavior. Which, in turn shapes this contact, right so, that's the idea of it and have this economy, it's going to see a series of infections. A rise, in number of cases and so on and so forth observing. That number of cases some. Combination. Of government mandated. Shutdowns, and private, mitigating, behavior, is going, to reduce the contact rate, when. It reduces the contact, Rea it's going to idle, productive, capital so, buildings. May not be used machines, may not be used, airplanes. Might be lying idle, so this going to, associated. With this mitigating, behavior, is an, idling of capital, and. The. Novel, but all of this is pretty standard the, novel part is to kind of pass it that ass capital, idles. There's. A greater chance that, it becomes obsolete that. Something. Changes, in the structure of that activity which, makes the existing capital. Essentially. Obsolete. For for. Productive, purposes so. We're going to kind of think about that is like a faster, so the longer, you idle the, faster, you depreciate, and. That's going to be the novel, kind of ingredient, here all the other stuff you know we're going to kind of try and just build, on what other people, and. This, of solitons of capital. Is going to be the shock, that agents, are the. Economic, agents face and, learn. About so. That's how that's, kind of the basic idea from. The epidemiological, model. In. Terms of the macro model, again. We're not doing anything fancy there, we're not really innovating, we're. Going to have the shock as denoted, by this Greek letter fee, that's, going to be this of Solace and shock that's, going to enter the production, function and, that's going to reduce the. Amount of output. That a given unit of care given set of inputs can. So. In particular it's going to affect the return the capital it's going to show a, realization. Of fee less than one means. That you're going to get lower returns, than.

We're. Also going to have default. Default. Is going to play a couple, of different roles and I'll explain that in, a few slides, so, forms are going to be borrowing. Debt. Markets, and. So. A bad. Realization. Of fee is going to induce more firms to default so. That's something that some the, model will have implications for. But. It's also going to have an additional ingredient, which I think captures something, quite. Realistic. Obsolescence. Itself, is going to be determined, or influenced, by the extent, of default and. So this is the idea that if a whole bunch of firms go bankrupt, supply. Chains are going to be disrupted. Even. Surviving, firms have to make more kind, of costly adjustments. You kind of have to we. You. Know you reconfigure. Your business, in some way or your business your, operations, in some way. Also. Capture, things like faces. Kind of straightforward externalities. If. Half the stores, in a mall. Are. Bankrupt. Then even the sole even the surviving. Stores in that in. That mall are in that property I am going to experience a lower value, of operating, because fewer customers are going to come it's, going to be lower footfall, and, so on, so all of those things are going to be folded, into a, mathematical. Representation by, which default. Itself, is going to feed back into the shock fee so. That's what this functional, form here is the. Exact functional, form is not that important, what's important, is that default, and absurd, lessons have this kind of two-way, feedback. So. That's the kind, of the macro sign of the of the model, and. So. All of this in some sense is pretty standard maybe a couple of new elements but for the most part it's pretty sad. What's the belief part of the model that's kind of the core new. Or. Perhaps the the thing that's unique about our analysis, we're, gonna kind of have agents, in the model behave like statisticians. And econometrics so. They're, going, to, acknowledge. That they. Don't know the true distribution or, the true uncertainty. Of. A true distribution of sharks hitting the economy so, at every day they're going to have a finite, sample they're gonna have some realizations. Based on history and then they're going to apply a. Particular, statistical. Procedure and. That's what this Gaussian kernel, density estimators. And. So. This is let, me give you the intuition of this in two sentences, it's, basically it takes the histogram, of realized. Time, series so, if you just plot in a standard frequency. Standard. Histogram, and then, try and draw a smooth, probability. Distribution. Function. Around that that's, what the kernel density estimator. Does and. So this is a very standard procedure in non parametric estimation. And. We're just going to literally take it out take, it out of the textbook without any, almost. No modification. We're just going to literally feed in the data into. This into. A canned procedure, for. Estimating the distribution, precisely. Because it's kind of flexible, it you know we're not imposing that the distribution is normal or, has fat tails. None of that stuff there is no you know this length this allows the shape of the distribution to. Be determined. By the observed, data that, you know that. We measure from from. From historical, time series so that's going to be the key, advantage. Of doing this it, allows. Us to kind of conduct this analysis, without, taking a stand on what exactly, is the shape of the distribution that, much like agents in our model we don't know either. Right. So that's the belief part of it and, so the main mechanism, is basically going to act as follows you're going to see a an extreme, realization, of feet when. You see an extreme realization, of fee that's. Going to make the beliefs a little bit more pessimistic and show you some pictures in a second that's, going to lower the return from investing, going forward, and. Not. Only is it going to lower the return. Investing. It's also going to increase the risk of bankruptcy, so you. Know to the extent that you have to borrow to invest that, whole enterprise, becomes, less attractive, than it was before so both of those things are going to you know there's going to be a double whammy on your incentives, to invest and. So what that means is you're going to have lower capital, and lower employment and lower GDP and, so on so, that's kind of the the, the way the.

Way In which a shock, is going to get a transmitted, itself, or. Propagate, itself. And. So where does persistence, come from the persistence, comes from the idea that these believes. You. Know are. On, average going, to remain pessimistic. Now. Depending, on you can come you can modify this so that the beliefs eventually, leave the system that's. Not quite so crucial, however, you do it these, beliefs are going to be with the aged are going to stay that way for a long time so they become slightly more pessimistic and they stay slightly, more pessimistic for. In. Arc you know by, most. Of our calculations. For potentially, decades, and, so these effects I'm going, to be with us for. 401. And that's where that's the source of the long-lived effects so, child coming in some sense from the fact that these beliefs are. Persistent. Are not something, I'm not transitory, even. After the shock itself. Even. After, vaccine. Comes and. And. Makes the immediate, health, crisis. You. Know pass. So. With that kind of introduction let me kind of I want to kind of fair amount of time talking about this also let me kind of switch gears and talk about. Talk. About the, results. Right. So. Here's kind of the game plan here's, what we do so, first step we're going to do is to try and discipline, what, we're beliefs of agents, before the Kovach crisis, and. That's where we're going to build very heavily or not earlier work we're, going to use historical data. From. US economy, kind. Of public macroeconomic. Data you. To construct a time series of this shot fee and apply. The non parametric estimation, procedure, from, from. A confirmed two slides ago so. That's going to give us a, measure. Of what risk. Perceptions, we're in the. Pre-code and. Then. We're going to construct you, know two. Scenarios, for how code 19, is, going to play out. This. Is partly because there's, a lot of uncertainty, about him even the immediate, effects, of. 19. So. The. Point we want to get across here is not that we have a model, that's going to tell us what's going to happen in the next few months or the next few, quarters that's, nothing that's not the the inside I want to communicate what, I want to you what I want to pay the communicate, is the idea that whatever. Happens, in the next few months is, going to have you, know. An. Amplified. Effect because, it's going to affect beliefs, going forward, and. So that's the, the. Main idea I want to communicate and. So what we're going to do is to show you two scenarios one, scenario. Is one where is is, going to be one with aggressive mitigation of, the disease so there's. Going to be a. Fairly. Drastic. Response. By, policy. And by. Agents. Behavior. In. Response, to infections, and that's, going to have you, know that's going to work in terms of mitigating the disease or the health crisis, immediate, with. A high degree of success but. It's also going to have more drastic, economic, consequences. And I'll show you pictures on all of this. The second scenario is one, where, the. Way we respond, to the health crisis, is going to be relatively, lacs that's. Going to have bad outcomes or. Horrible outcomes as far as infections. And fatalities, and sickness and other things go but. It's going to mean less. Economic. Disruption, so these are the two kind of scenarios and in both of these scenarios we're, going to calculate what beliefs, are going to be like and. What future, time paths for the US economy. And. I'm going to show you a bunch of a series of wraps which are going to basically show the. Average. Expected. Future path for GDP. For, investment, for employment, defaults. Etc so. That's what that's going to be the main output, of the model and. I'm also going to show you a, simple. Calculation, that we take these out. Is. Losses, in terms of lost GDP, and kind. Of turn them into a present, value number. So we can kind of directly compare, what, the, immediate effects. All. Right so that's going to be the game plan I'm going to show you data on all of that I'm going to walk you through. So. Let me show you first, sorry. Let me start with the with, the disease. Scenarios. Again. The, disease scenarios, per se are not on focus, there's a lot of people looking at that as. You know mentioned I've looked at that another one here, the focus is really not on the disease side the focus is really on the economic, consequences of that so the disease side is only going to give us a way to think about economic. Consequences, they're not directly our interest I won't really want to emphasize them for, example we are not really thinking about what a policymaker. Should do if. He's trying to trade off health and economic outcomes that's really not what, I'm after today.

So, Here is the here. Is what happens on the disease side under, the aggressive, scenario. And under the lacks mitigation. Scenario. So. May. You know this, is the graph of the number of susceptible infected, and, recovered, including. Fatalities. In. The US so. This is millions, of residents. So. Here is what I, want. To maybe pay attention to the red line which, is the number of infected people in the US under. This policy so. The policy, start and on the right panel here, is the policy, is this odd, not the. Policy expressed in terms about now so, we're going to start off with an R nought of about, 3.5, that's. Probably at the higher end of estimates, the some work suggesting, maybe it's a little bit lower but it's a very contentious, parameter, at this point so we're going to start off you, know. High level of 3.5. As. Infractions. Committee. Policy. Here aggressively. Is going to step in and. Reduce are, not pretty aggressively, from three point five to abort one which. Is often viewed as a critical, threshold. But. Then once, that kicks in after a while you, know things are going to lighten up and, so the policy, kind of net stakes is foot off the pedal and, then, we're going to see the odd not go back up again and, then we're going to get a what, is sometimes referred to as a hammer and the dance type of battle and to, kind of take our foot off the pedal but then slam it back down, when. When, things go south again, so you're going to get a little bit of this wave-like pattern. Here. The, last scenario, does. Not respond, nearly, as aggressively, so, as a result the infection, great Peaks this is going to be it significantly, more infected, people and under, the, last scenario then there are the aggressive scenario but. You know associated, with that is a much more, modest. You, know disruption. To. Contact, something economic activity. Alright, so like I said these, are disease scenarios, this is not you. Know what we are really trying to innovate or we're really trying to push here what, matters is one end us to our. Estimates, for the economic, disruption, in particular what it matter what it does to our estimates, for for. This. Capital, up sources, so, this first scenario, under. Our calculations. You, know leads, to about 10% up sauce so it reduces the economic, value of capital, in the United States by, about, ten percent the, second scenario reduces. It by about five cents, so that's what that's what these two scenarios were loosely speak. So. Just to give you a benchmark, so if, you look at commercial. Real estate prices, in the u.s. in. You, know start you know relative to whether we're in February, they, are about ten, eleven. Percent lower so you know that's kind of gives, you an idea about what this number. Looks like so. What is the heart of the beliefs work so, the blue line here the blue solid, line is the is the, estimated, distribution, pre : that's what we call the price then. You're going to get a new observation and then you're going to draw a new updated, distribution. Which is this red dashed line as you, can see for the most part that distribution looks exactly the same as it did before all we've.

Done Is basically add, a small. Bump, all the, way on the left where we got this unit so, that's how this kernel, density estimation, works we're just going to add you, know a little bit of a bump whenever. We get in, the, under scenario to that. Bump of course a little bit to, the right because, you, know the the observation, itself was much, was. Perhaps closer. To one, so. These are the two distributions and we're going to contrast, what happens for economic activity. Right. And that's basically the second. Part so. Here is what happens, when you when you kind of feed this into the model, so. These graphs show, the change in aggregate. GDP. Relative. To the, initial, point relative to what we have everywhere, before the. Crisis, so. Let. Me focus on the blue line which is the model with learning where. In other words with beliefs, are changing. So. There's, an immediate shock, and. That just reflects I just made a bunch of your capital obsolete. That forced you to ID your capital, all of, that put together basically, means you don't you're not going to have much GDP, in 2020. Right and that's you're. Going to have lower GDP in 2020, and that's what this immediate, impact. Represents. So, that in scenario one who is about nine percent right. So GDP. In 2020, on the scenario one is forecast. To be nine percent, below. Normal, and, that's in the ballpark of what most, small street for grass and most what, effect forecasts but several other. Professional. Economists are forecasting. The, interesting thing is what happens after, that the. Shock goes away. For. A while because, I've just rendered, a bunch of your capital obsolete, and it takes time to replenish it output. Is GDP is going to be lower anyway. Right so that's what I'm going to call the obsolescence, of I'm just I've just taken away some, resources, from you and so, you are going to have depressed, activity, for, a while, the. Innovative. Part is the second one which is what we are really interested in your, beliefs have changed, and so because your beliefs have changed, you're, not you don't want to go back to what think how things where you're not you don't want to go back to the normal, level of capital from, the fourth advice you're. Going to kind of in a permanent, fashion invest. Less. And that's what the beliefs carding mechanism. So. The next slide I'll show you how much of this comes from the first and second, there's, also the possibility that these pandemics, occur again, in, the future, that's. Kind of mechanical, so I'm not really I'm, not going to focus on that and that's part of this but it's a relatively small particles, Sun, so, here is this calculation. Here is that decomposition. Between. GDP, and between. Capital obsolescence, and behaves calm so. Just to kind of go back here our baseline, prediction, is that JP, is, going to be 5% lower. Kind of almost permanently. Out. Into the future, right. And if you look at where that comes from this orange area is the component, that's coming from the belief scarring and this gray shaded, area is the, component, that's the direct effect of the observer since we have kind of fed it so, you can imagine the absorbance, is kind of important, in the short run but, as your horizons, extends, out you're, going to see you, know less, and less of this obsolescence, effect because the economy kind of figured that out or sorted that out but, the belief scarring, effect really kicks in as we go, forward.

And That's what you see this. This orange, area basically, accounts, for almost all of the long-term. And. If you cannot calculate a net present value, the risk adjustment and. All the fancy stuff you want do if, you did that you're going to end up with you, know a number where. The net present value of this orange region in scenario, 1 is going, to be about 50 percent of of, 2019. Gee whereas. The contemporaneous, drop, the crop in 2020, GDP, is about f9. Scenario. Two all these numbers are basically. You know about. Two-thirds of these numbers the some non-linearity so, the worse the crisis gets the worse the belief scarring so there's a little bit of that perhaps, in that the scenario to is just a milder, version of this even, it's the important thing to note is that even in scenario two the, belief scarring, component, is several, times the. 20/20. GDP drop and that's the sense in which you, know these, longer-term. Effects, are much larger than. Anything. In, a world where you think people are going to learn from this experience and modify, their behavior going, forward, you. Know these come these longer-term effects, can. Overwhelm, the, immediate, consequences. Well. So that's kind of our my, headline this out here, I'm. Going to show you a couple more slides and then I want to kind of conclude and the mean and then we can get into Q&A so what's really underlying, this as I said earlier is this idea that when you are pessimistic you, know yes I've rendered some of your capital obsolete. Normally. If I didn't have to worry about your beliefs changing, if, I render some of your capital obsolete, you're going to jump in and replenish, that capital, put in new buildings new factories, new machines what have you and that's what happens when, technological. Changes, make you know capital. Obsolete, here, there's, an additional negative effect, which is that you, know you're a little bit more pessimistic going. Forward, so you're not that excited about replenishing. That cap and that, is the effect that's going to you know so it essentially acts through investment, it's going to keep investment.

Low And throw, investment. Low it's going to keep capital, low and therefore lower GDP. And. So that's basically, the core of the. Mechanism, so. I'm running a little bit low on time so let me just show you two, other pictures, and then I will the. Rest we can get to our Q&A. Right. So this is what happens to two, defaults, so, in the, short run in 2020, are the morning you, know predicts, that, defaults. Are going to spike from a normal, of about 2% to a high, of about 6%. So. This is pretty dramatic, but, it's still quite a bit lower than what we saw in 2008, 2009, for example during the financial crisis, but interestingly the. Spike, in defaults, that's not really last and. The reason it does not really last is because the pessimism, that is introduced by the beliefs cutting mechanism also. Makes firms less, willing to borrow so, debt also goes down so firms are kind of be leavening, in a sense to respond. To. This new into living to operate in this new world and so the actual change in defaults, going forward, is not. That. Hot so that's something kind. Of interesting. So. Here's my last slide and then I will hand. Over to brown for for. Q&A, the, model has a bunch, of implications, for our asset markets, I do, want to make one kind, of overarching. Comment, here. The. Police you, know most of the implications. For asset prices in this model I'm going to be relatively. Modest, so, you, know credit spreads are going to change that I said about my amount half basis, point or a basis point I mean there's going to be a lot of default, this year but then going forward defaults, are not going to change that much so it's not a lot of change in credit spreads there, isn't that much of a change in equity or a casebook the market type of ratios, those that the PD type of ratios don't change a whole lot either, risk-free. Rates change, so. That's one place you can see that a safe asset, like US Treasuries, becomes much more attractive, so you can see a big change in the interest, rate on those claims what. Is interesting is, on, what is perhaps these are all changes, which are might be kind of rather imprecise.

Signals, Of a change in perceived. Tail. Risk where. You see a much more kind, of clear effect, is in. Options. Data so these are two kind of statistics, we're going to talk about from options data so, if you look at out of the money put options, on the S&P 500. The model predicts a fairly substantial increase. In the, implied, probability. Of a negative and. That's. What these two numbers. Here are. All. Right so let me kind of end. With and, with that I'm happy to talk about policy, so one of the things that the model implies is. If. You you know to because there is this feedback between policy, between defaults, and absolutions, by, mitigating, default. You're, also mitigating. Of so lessons and therefore the scarring effect so, by. Kind, of intervening, aggressively. Through financial, policy, which, reduces, defaults, let's say if you go back to my picture but defaults went up by about 6 to about 6 percent if, you could bring that down to something much more normal, the, payoff, from that in terms of longer-term. Benefits. Dwarfs. The immediate. Immediate. Benefit, is obviously there you have less you, have more economic activity in 2020, and maybe in 2021. But these longer-term, benefits about, how you manage, the economic, fallout of the crisis, are. Substantially. Bigger. What. You're going to get in the next year or so so, that's something that's kind of the main message of the paper so, now I want to leave you with that kind of punchline, now, what. What the this is an unprecedented, event, in. Several, dimensions and so the after effects are going to be felt for. A while, and. You. Know at least the qualitative, sense those. After, effects are much much bigger than the. Question of whether we're going to be in a recession in 2020, or at the end of 2020, or a funny one and now kind of overarching message is that those, numbers those estimates, of longer-term effects, should. Form part of any policy, debate whether on the health side or. On. The economic slash, financial. I think. That, no matter what you think what scenario, you think how this is all going to play out those. Longer-term effects, must be part of the conversation, and I think that's what we want to emphasize. Is the main necessarily. All. Right so, let me kind of stop there and hand, it over to we're all for. For, Q&A, Thanks. Thank you when key those, really terrific I think you presented, it in a, very transparent, fashion. In spite of all the technical. Details let. Me start off with, a couple, of questions we were actually, casually, bantering, around, these questions before. The. Forum started, which. Is so one, view of bankruptcies. Is. That, they are socially. Costly. Another. View is of course sort of more, of champey Tyrion which is that you. Know there's some creative. Destruction involved. In bankruptcies. And, usually. They enable, a shift in the allocation of resources to, the more productive use, perhaps. The world is as of. Now somewhere, in between in, the sense that many. Of us, would. Agree, that, if. The. Revised. Beliefs, about risks of pandemics, are right, then. Perhaps there should be some alteration. In. The way production, is organized, in, the economy, clearly. Given. The irreversibility. Of investments. This. May require some, painful adjustments. Right. Nevertheless. You. Know it's as, you said you know defaults, probably, do have externalities. Especially, if they happen in a very large number in a very short period of time. And. You know there's there's this talk, about, trying. To ensure that the bankruptcy, is don't rise beyond, a certain point as you mentioned to limit these externalities. So. That's. A non-polluting, to the question I'm going to ask the, question I wanted to ask you was. Should. One think in the model and. In general about potential. Beneficiaries, of. An. Episode, such as the pandemic where, even though the beliefs revised, the way in the manner you have discussed. There. Is some production. Alterations. That take place, so. That actually, the loss in the output is not entirely, due to the capital, that becomes obsolete but. Some new capital, is one created, in these productive, frontiers, of the economy, for facility, yeah so I think that's, a great observation and. In, the way I've presented, of them the way the model is written right now it kind of abstracts, from that in the following sense it says oh you, know yes, we built a huge mall in, 2018. And come, 2020. We took a big hit to that all so. The way we were and we are kind of thinking about agents, in the model is that they're thinking about the next unit of capital they're going to invest it's also subject to a, similar, risk well, one possibility is that the next unit of capital is kind of qualitatively.

Different, From, this and might, be less vulnerable to. You, know to the same risk so, even if everybody says yeah pandemics, aren't going to keep coming up they might say but no I'm kind of investing, in a different, form of capital, that might be less. You. Know susceptible. To the type of obsolescence. That that, you highlight and I think that's a first-order question, so in some sense that the question of magnitudes, about how much how. Much that substitution, away, from the traditional, types of capital, to the newer forms, of capital, can happen and. I think you, know if. You kind of think about some, of the more most capital intensive, sectors of the economy it's, things like healthcare it's, actually incredibly, carefully it's, also very labor intensive, but it accounts for a big chunk of the of investment. As well so some of those sectors you. Know a banking now we haven't seen you, know at. Least. I. Want to be careful here we, haven't seen a dramatic kind, of technological, transformation. That has happened yet but, to the extent that happens, that's going to mitigate, the negative consequences, I highlight so. I think you know at, the end of the day we didn't really thought, about that but we didn't really have a, quantitative. Way to discipline, it so we kind of stayed away from it all together the, way you would think about it in the context of this model is that you can imagine there are two forms of capital, you, could have always thought, remotely, but we chose not to because the cost of doing that relative, to the benefits, led, us to teach you know in classrooms. Now. That, calculus, would have shifted a little bit away and we're going to use that, alternative. Form of capital, which, is somewhat. Imperfect substitute, for what we were doing earlier but, that's how. Going, forward and I think that is something we're trying to figure out. My. Kind of maybe somewhat pessimistic view. Is that you know for. The economy, as a whole. Those. Substituting. Technologies. Are not going to be big enough to. Kind, of do, to kind of make these effects, go away but. It is but, I but, I do agree that they are going to kind of attenuate, the. Numbers. That I showed you. Let, me raise, a question that come from Tom fugle Venki, which is. Does. Your model have implications, on, the effects, on national, saving. On. The equilibrium, parts his question is bettors the national saving that is not being put into real investment. Though if. There are if. Is because of the scarring of the leaves not enough real investment, is taking place so, that's a great point. So, in, the in the way the model is set up. So. There are two two things happen at the same time I've just, come in and whacked a bunch of your capital, and made it obsolete so, you have less stuff anyway. So. You are you know as a fraction, of that less stuff you're still saving you. Know you're, saving a certain fraction of that and investing, in it but, it's not like I started, you out with the same pool of resources and, you decided to invest less so, let me maybe kind of state that a little more clearly I could, have posited, the following, world where there was no solace, in shock, but. Agents, beliefs went forward, then, they would have extra, resources that they would have something to do here. The, you know the fact that we had a big negative shot means that the extra for the amount of resources, on the table to be split between consumption. And saving itself, is a lot lower so, there isn't a lot of extra, resources so if I looked at the time path of consumption, in the model try to show you that also, drops and stays low so you know in this system. The. The main. Thing maybe. The point I want to emphasize is the opposite, you, know if, we, lived in a world where beliefs, did not change, as a result of this then. You. Know what the economy would have done in the next few years is cut, back much more drastically. On consumption, and invested. Much more aggressively, to replenish, the capital, that's. What the beliefs cutting mechanism and. It turns off that's what the beliefs cutting mechanism and us so, in. Some sense the the. Excess, sailing, here, is going to be you, know relative, to that benchmark, rather than I. Hope. That you know I know it's, hard, to do that without the equations, but I hopefully, Tom. No, that's a great great, point Winky, just, coming back a little bit again to the issue of belief, scarring so. I, think, your last discussion, on the policy, implication, was particularly intriguing, which, is that. That. The use of a.

Pool, Of social, resources in, order to mitigate. The, incidence of bankruptcies. Could. Actually bring, the levels. Of investments. Back up, to. Higher levels, at, a faster, faster, speed so. I have two questions related to that which is, is. The is. The belief updating. Inefficient. In some manner. What. I mean by this is that, what. The reason. Why the limiting. The number of, bankruptcies. Improves, outcomes is it because you are improving, the. Belief updating, process because. Somehow that's. Sort. Of not. Not. Getting to the truth at the right speed or at the right level, or. Is it that it's. Directly, actually, by reducing. The pace of also on the sense making. It more attractive to, invest in capital, that's. Great question so. Let me try I mean so let me clarify the efficiency, part and then maybe that'll help me answering your question so, the you, know the belief updating, per se is. There's. Nothing irrational about that there's nothing behavioral, we have an ad anything any, other. But. The economy itself is inefficient, for various reasons one, which is of particular relevance to, the, question you raise is that default, itself, is inefficient because private, agents are going to decide whether default, or not based on whether, they have enough resources. They. Don't internalize the, fact that by defaulting, they, are exerting this negative, effect on on everybody, else so, you. Know even though people whose beliefs are right you know it's this mapping, between, defaults. And this thing that leads to an inefficiency, of sorts so what the policy, since in a sense doing, is, trying. To kind of correct, at least in this extreme, event you, know correct, that extra, that negative, externality, by correcting that component, so. So, let me kind of maybe use, a numerical example so, the, say. The primitive, shock is at 10 percent of services. Default. Terms that 10% of solitons, and to say a 15%, of services so, what policy, is doing is saying well we can't do anything about the 10% of solutions and people learning from the 10% of sources we're, going to bring the 15 back to 10 and then. So people are going to go forward rather than think we lived in a 15%. Of Solace and small they're going to be thinking about the 10% one so, that's what that that. Does I mean it it raises a broader question about you, know in a world where default I mean so the the crucial, thing is this default amplification. Really only kicks in in these extreme events you. Know when things are normal, the default related, inefficiency, is kind of small so, this tells you that one way to make it one way to manage, these extreme events is through a concrete, policy, which says I'm going to address what is obviously, an inefficiency, because, people are making private decisions. To to, close down shop without. Internalizing. The effect on the. So. I think that's how I want to kind of frame freedom, if, on top of this if you thought that beliefs. Were. Inefficient. And there, are lots of evidence for how this recency, biases, all kinds of biases. You could throw in those, things would only strengthen, the, case to kind of and. Try and not cheer that way here, all of that's happening, only because the Falls are you know so in a world where the default of solitons, feedback, was absent. It's. You, know we'd be hard-pressed to recommend, policy. Any you know any useful, policy, in this one again. You could posit, that the policymaker, knows, better than the economy rather than the agents, about the true nature. Of risks, facing the economy but short of that paternalistic. View there'd, be no reason to intervene before, I think gives us a concrete reason to intervene even, without the paternalism, of, knowing. The truth. There. Seem to be two questions, from Yaakov. Imijo but I'm not seeing the questions, unfortunately so, let me jump to Larry White's question. In the meantime. So. Laddies question is on the differential, between your, two scenarios Venki. He's, saying are. The, differences. Potentially. Being overstated he's, saying after all let's. Take the example of the airline sector, the domestic airlines were fact idly. Even. Though they were never forced, to shut down correct.

And. Maybe with a little bit of a lag consumers. Would have similarly, learned to avoid the restaurants. Movies etc even. If they were not formally, shut off and so, I think he's really asking the question, of whether. Shutting. Down by itself is. Really the mechanism through which. My. Father should have been much clearer what. We call you, know what. I call, policy. I wasn't. Only thinking, about you. Know a mandated. Lockdown. From, governments. Or policy. Makers I was, we, were trying to capture all, mitigating. Behavior. Whether it's the result of agents, by themselves, kind of scaling back activity, or by. The. Result of. Mattad and. As Larry pointed out, there's a lot of evidence both. From, around. The world and even within the US that. Even, when an official lockdown, is lifted it's, not like things just run back. So. That's, you. Know that's one thing I want to emphasize the, same thing so the way we kind of are the way I like to think about those two policies you. Know you, can kind of think about them is in terms of the disease of the epidemiological kind, of parts, you, know my preferred, ways to think about them as representing. To, you, know, economic. Scenarios, you, know one. Represents, a fairly dramatic you. Know crisis, where, 10%. Of the capital in the u.s. suffers, from this of solace in stock and the, other were only 5% does so I think you, know the. Exact. Constellation. Of, ingredients. That delivers, those economic, outcomes, I think there's a lot of uncertainty, about that, not. In the least about you, know what. Behavior, would be going forward like, in the take they take the example of Airlines saying you, know are we going to return you know let's say a vaccine, comes back are we going to jump back to travel, to the same degree as we did before, so. Those there's a lot of uncertainty, surrounding that and I think that you cannot the, scenarios, kind of love. All of that into, one number saying one is a 10% of the other. And. I think at this stage you spend a lot of time trying to think of you know how, we might discipline, them further I, think, you, know this is one of the pitfalls of real time stuff, we really don't have data yet to, start, you know piecing that together so most, of the forecasts. We have today, you. Know confound, what's going to happen in 2020. What's going to happen perhaps going forward, and I think we haven't found an effective way to disentangle, the. Thing we are really interested in which is the longer term stuff from. From. The immediate effects but. I think that's something that certainly. Okay. So I do have the questions. From. Code, now, his first question is is. On the, nature of default. Or bankruptcy. This. Is something that Larry White has also alluded, to in, the forum, and he was talking about public policy which is that default. Or bankruptcy. Is not a shutdown. In. Fact the. US Bankruptcy, Code, allows the, assets to continue, operation, and. One view of bankruptcy. Is really that it's a financial restructuring. The. Most part rather than an operational. Restructuring. Of the assets. So. There may be operational, restructuring, of assets as well and. So, he's, saying how. Crucial then. Is. The fact that. You. Are implicitly, assuming, some. Significant. Real. Investment. Or operational, costs, and the. Example, is the ACOs giving is that even. An all equity, firm in a mall would shut down if the future were bleak and, so that's not really an. Inefficiency. No. That's fair, so okay two comments let me give you an a quick answer on the magnitudes, so suppose we turned off this obsolescence. Default, feedback. That. Knocks, off, loosely. Speaking a quarter, of the results so it's economically, meaningful, and it's non-trivial. But. It's not the. Basics even if as, you know as exactly as you pointed, out we could have said you know imagine you lived in a world where there, is the possibility of default there, is no real consequence. Of default so one thing I want yeah so in other words you know default is a purely private affair, the maybe the debt holders lose at the expense of somebody, else who makes money on this we can do all of that stuff even in that world almost. 75, percent of the numbers, are the effects I showed you are still.

That's. One second, I, don't disagree that you, know chapter, level is different from chapter seven in terms of thinking about going. Operations. What we want to capture is that even. In even when a fall is in Chapter eleven if, it's going to be in Chapter eleven for a while, you. Know it, creates an, certainties and vulnerability. Is all along the supply chain so if you thought you know this was something that came up in 2009, when they were trying to adjudicate, the GM. Situation. And really worried that an extended. Long winded, restructuring. Process would, have ripple effects on you. Know on on an entire kind of you, know ecosystem. Of suppliers, and. Other. People, and so those are the type of effects we're trying to capture, it. Should also be something much more mechanical. That if, United is under Chapter 11 and I'm going to imagine that they're going to maintain their flights a little bit less or, maintain, their air fleet a, little. Bit less effectively, than they did before. There. Might be fewer investments, in upgrading, so there could be things like things. Like that as well but I think the first kind of things I talked about we're a. Major. Part, of the economy mired. In chapter 11 proceedings, for an extended, period of time, you, know has. Negative, consequences. It's not literally, business as usual it has negative consequences, for. A whole bunch of people that's. Something we, don't have a you. Know so the default, absolves and feedback is something we don't have like. Direct estimates, for is trying to kind of do that but nothing and that we can use for the aggregate economy we have some more micro estimates, but we don't have something much. More so. We kind of calibrated. Inside the model, but. I think that's something, we. Should probably refine. With. Better micro, estimates. Thanks. Randy let me jump to one, of your. Sort. Of very interesting, insight. From the model is that because. The occurrence. Of these tail events leads to a revision, in the expectations. For future pandemics.

Type, Events being more likely. That. Belief updating, leads to an aversion towards, taking on debt. Now. While. We have seen a bit. Of that on, the one hand there. Are many anecdotal. Examples. Of companies. That have actually, tried. Very hard to actually generate forms, of cash. So you know there were these examples, of PNC, Bank getting. Rid of its stake. In one of the large asset management company, because you know they were reliant, on wholesale finance and they wanted to have more cash at hand, on. The other hand we've also seen dramatic. Issuances. Of debt over, the last two. Three months, and. While you know clearly, one may attribute a, big chunk of that to. The backstop that's been offered by the central, bank and the government. I, I just wanted to see if you could link a little bit on, the. Scarring, of beliefs that's happening, in the model with. The, dramatic. Reversal. In the stock market's, credit. Spreads and and the sheer quantum, of bond market issuance is that we have seen over the last to two months. So. How do you make how do you make. It all sort of, that's. A great point so in, the you, know the model kind of you, know leaves those things out partly, because we have like an annual frequency this, is not sufficiently. High, frequency, enough so the way I kind of try to think about that is, you. Know if you wrote down a more you. Know more, granular, model, reforms by making decisions, day by day and, trying. To kind of mitigate this disaster, at a higher, frequency I'm trying, to deal with the immediate. Consequences. Of disaster, I bet. We would see you, know my. Conjecture is that we would see an uptick, in that immediately, as well so I think you. Know in the model, in some sense is a little bit artificial. You know firms, walk in with some bet they, see that their economic fortunes have gone south they. Just have to decide whether to continue operating, given, the debt obligations, they have or, to shut up shop so. Some, fraction, of them choose to just shut up shop and walk away if on the other hand a more realistic description. Is you walk in you. Know yes this this there's, a shutdown or some kind of there's. A disease you don't know how long it's going to last you don't let the states even know whether you should fold up shop or not you, want to certainly stick around and see how all this plays out before, you do that and, I think that, behave, like, that that initial, stage that, that richness, we've, kind of completely left out and I'm, almost positive that will that, will lead. To at least a significant. Chunk of firm saying okay. Obviously. I don't have any revenues I have to kind of continue to operate I have some expenses, I need to mean how do I know that I'm going to borrow draw down on lines of credit use whatever I can to kind of survive this and then, maybe at the end of this let's, try and decide whether, we want to you. Know continue operating or not like for example you, know everybody's, been talking about this wave of bankruptcies. That's coming the, bankruptcy, filings, haven't, really, yet reflected, that and in part it's because everybody's, playing this option value game of saying let me see how this thing plays out before. I rush out the door and start. Finally. Start. What, is really, a very costly, default, restructuring. Process. That's. One comment then the second which is maybe you, know which, I'm going to build on what you were saying maybe, go a little bit beyond. One. Thing we haven't done which we might see in going forward is other forms, of precautionary. Behavior, by firms. That. Might involve things, like lines, of credit, going forward, in other words they might say one, way I'm going to prepare. For this type of scenario is to have a contingency, source. Of capital available, if. I have the luxury of issuing, equity and sitting on a pile of cash great. If. I don't I may have to think about putting in place a back-up, plan I. Think that's, something we have definitely, likely, to see, we. May see a combination of all of these things where firms may be a less, levered, but.

Have Contingency. Financing, in place. Or. Have, a greater demand for contingency, financing, in place and. Perhaps. Whole. Greater cash as. And. I think some, so there's going to be a lot of richness in the way I think businesses, are going to adapt to, this, new world especially, the one little ones and. That's something you, know the model kind of leaves out but I certainly, acknowledge. What we see, it's. Great Weinke we've just, run, out of time I. Particularly. I. Want, to end with this one observation, which, i think is striking. Also in your last set of slides. Which, is that there. Was this phenomenon observed. In the options markets, that after, the, you. Know the October, 20th, 1987. Crash jump. Risks, which did not use to be priced in the tail in, the sort of the implied volatilities. In the options markets, actually, started. Getting priced and, of course a whole set of new wave. Of models, for option pricing was written thereafter, and, I think. It. Very much sort, of squares in with this fact that beliefs shift. Permanently. Post some of the realizations. Of these daily events thank. You very much Venki for joining us thank you to everyone in the audience as well enjoy, the summer and Phase two as. Of. The, gradual relaxations. And I hope you, stay well, and safe through the rest of the summer well thank you for all thanks everyone for participating, Allah, giving me the opportunity, and hope. You are safe and look. Forward to seeing you all in, person soon. Thank. You thanks Becky.

2020-09-06 06:19

Show Video

Other news