Robert C. Merton (2014) - Measuring the Connectedness of the Financial System: Implications for Systemic Risk Measurement and Management

Thank you very much. It’s a great pleasure to be here. And since I have a clock in front of me and I have a 45 minute talk to do in 30 minutes, I’m not going to say too much except to move as fast as I can. But I wanted to sort of open up on the topic here. It probably doesn’t need too much motivation. But this is addressing the issue. There are very big macro financial systemic risk issues for both governments and large asset pools. And in essence you can think in the example of the 2008/9 financial crisis that, you know, at the centre of that crisis or certainly a very important part of that was credit risk. It was money market stuff and some of the plumbing which is also important. But it was credit risk that was at front and centre there. So I will be talking about the application of credit risk. And what we’re going to talk to you about is how we can use some financial economic tools to create tools to try to measure the degree of connectedness among institutions and sovereigns; so institution to institution, institution to sovereign, sovereign to sovereign. So you take that collective and you ask in what sense they are connected and the sense that the connectedness may be an indicator of a potential for a systemic event because of that connectedness. So what I want to talk to you about is creating a tool to measure connectedness, look how it changes and see if it could be used, if not to avoid or mitigate crisis, at least to be able to measure the connectedness and then perhaps to use it to manage or be prepared for a crisis. So the work here that’s being presented is work in progress with several of my colleagues at MIT. And so, you know, we don’t have all the final results. But at least I’m willing to say I think we’re on the right track. And I’m sure all of you will be able, if you find it interesting, to do a lot better job than we are. So with that let me get started. And I want to talk about, again as I say, the measure of credit risk. Now unfortunately... I’m mostly going to show you lots of pretty colourful pictures, graphs and stuff. So it’s, you know, not a lot of equations or anything. But like everything in life... no free lunch. So if I’m going to talk about the degree of connectedness with respect to credit risk only, not all the other things, transactions and so forth, but just credit risk, then you’ve got to know something about credit. You’ve got to become somewhat credit experts. So with the time I’ve got... I’ve got 10 minutes to give you a speed course on how to become a credit expert. And those who want after the lecture at some point... if you want me to sign a certificate that you’ve gone through at least level 1, I’m happy to do it. Ok, so starting out with the notion here. I have here a loan, a mortgage, a bond, pure real, sort of things. And forgive me but I’m going to do it in dollars, as you’ll see why, US dollars. Right, so I have here some kind of credit instrument, ok? And I have in this hand a full faith and credit US government guarantee. The US has issued a few of those during the crisis and since. What if I staple them together? Functionally what do I have? I have basically a risk free asset because if the issuer doesn’t pay me the US government will. Now there are some differences but let’s agree risk free, right? So therefore if I pull the guarantee off, go back to the bond mortgage, what do I have? I have a risk free bond minus a guarantee. You see the logic of this. There’s no special model. And I did it here for a corporation but you could do it for a household where the asset is a house, the debt is a mortgage and so forth. This is the simplest balance sheet I could come up with. And, you know, on the left side you have the operating assets, you have simple debt issue. The promise is, let’s say, a billion dollars, B dollars at some date in the future. And to make it simple we’ll assume its zero coupon so I don’t have to deal with that. And we say ok, you know, ask the question in default, you know, what’s going to happen when that bond comes due or that mortgage comes due. One of 2 things: Either the mortgage will be paid, the billion dollars, the B. End of story. Or it won’t. And what happens when it’s not paid? Now in the real world it’s a very complex process of chapter 11 and negotiations and so forth and so on with lots of costs. But in every case with creditors, ultimately, if the borrower doesn’t pay, the creditor gets the assets that are pledged for that loan, ok? So I’m cutting through all the institutional frictions and costs. At the end of the day if you don’t get paid as the creditor you get the asset. Now, what’s the value of a guarantee? Well, on the day that the bond comes due it’s pretty straightforward. If the bond is paid and, you get your billion dollars, the guarantee turned out to be worthless, zero. If on the other hand you’re not paid by the issuer, then the value of the guarantee is the difference between what was promised, the billion or the B, and what you would get as recovery from the assets as the creditor V. So compactly the payoff to the guarantee is the maximum of zero, B minus V where V is the value of the asset. Now I hope at least some of you and I hope maybe all of you might recognise that pay off function. There’s a financial security that’s very well known that has that pay off, it’s called a put option. And it’s essentially nothing more than value insurance. It guarantees that you can sell the asset for B dollars, the strike price, anytime up to the date which it expires. So it’s a simple insurance contract but a very important one. So what do we see? We see that a guarantee is equivalent to a put option on the assets of the borrower. So if it’s mortgage, it’s on real-estate assets. If it’s corporation, it’s on corporate assets etc. Ok? Now those of you, you heard what I got my prize for, you’d figure I’d get in options in here somehow. But in fact that’s very useful to know if you only have 10 minutes. One because most people who have studied financial economics, even at the most basic level, know what an option is. But more importantly we have 40 years of world wide experience in pricing options, understanding their risk, in huge scale, in just about every kind of thing under the sun. There’s a huge amount of cumulated knowledge, experience in them. So if you’ve never heard of credit before but you understood options, you already know the structure of credit fundamentally. So I remind you, any loan, any of these things is equivalent to a risk free asset. That’s just time value and money. Put your money now and get interest. That we all understand. Minus a put option on the assets. So I’m going to be focusing on that part. That’s where all the action is in credit. So I wanted you to see this as a structure. If you accept that... By the way you’ve heard of CDS credit default swaps which were a topic of conversation during the thing. Those are nothing more than essentially put options or guarantees. They’re put options on the underlying assets. So let’s get rid of that mystery in terms of what they are structurally. Now let me show you... I want pictures, I don’t know if I can do this. What I’ve drawn here is a picture of the value of the debt as a function of the assets. And since it’s a risk free asset which is straight lane minus the put, ok, because it’s a negative, it means that in effect that the guarantee part, the put part in anybody’s theory the value of that put option is going to be that convex decreasing function of the value of the asset as it’s drawn there. This is generic. And I put it as a bank liability in a sense the banks make loans. They are writing these guarantees. Now, what happens from this picture? Let’s say we start with the assets at AC here and then assets fall, whether its real estate or corporate assets, to here. What happens? Well the first thing you see is the value of the guarantee goes up in the upper picture because it’s minus that. The bond price comes down. Now there’s been no default, everything is green, it all satisfies, nothing has happened. But the value goes up with the thing and so the debt has to fall. By the way parenthetically in the real world the tendency is to treat loans and so forth in institutions. If they’re green means they paid their last interest payment. And they’re the same ones they had last time. They say nothing has changed even though the underlying assets have fallen or risen. And you can see from this what is true we don’t know. What we know is not true is that the price doesn’t change. It has to change when the assets change independently of whether they’re paying their things and likely to pay and all that sort of thing. It has to fall. So the first thing you see is the value of the guarantee goes up as the asset goes down. That’s straightforward although as I said that’s not even usually recorded when we do this. The part I want you to see that’s insidious about credit is the following. You see this point here. If you ask the question at that point how much risk to whoever wrote the guarantee is in there as a function of the movement of the underlying asset, corporate assets or whatever, for small changes it’s going to be the tangent to the curve. So let’s say... You see it’s a negative because of the direction. Let’s say it’s -0.1. That says at that point in time a small move in that assets value will have a 10 cent change in the value of the guarantee. So you would say the risk at this moment in that particular guarantee or mortgage or bond is 10 cents sensitivity to the underlying assets. When you get to the second point, because the guarantee has gone up the equity or the assets of the bank have gone down. So of course they are more leverage if they do nothing, right, if they keep the same assets as donated capital. That we’d all do. That’s what we see in capital ratios if in fact they were market to market which they typically are not. But what do you see from the picture? What’s not being taken account of is that same bond, that same asset, that same guarantee, nothing has changed in terms of the paper it is. What do you see about it when it gets to the new point for the next roll of the dice? Because of the convexity and non-linearity it’s steeper. It’s not 10 cents per dollar. It’s 15cents per dollar. And if you have another shock and move down, it gets steeper and steeper. So same asset, same bond, same guarantee, nothing changed, still being paid, nobody has defaulted, ok? But two things happen when the assets underlying go down. One: asset value falls therefore equity, capital falls. But the risk goes up and goes up in a nonlinear way. So if people were measuring for example the riskiness of these assets when things were at high asset values, they’re going to get with regression or any other measure, a pretty flat slope. So they’ll get 5 cents, 10 cents. If they don’t understand in the structure that has changed this way and they continue to think this slope is 5 or 10 cents when it’s really 15 or 20 cents, you see that they are underestimating the sensitivity of the risk. By the way you’ve heard of these 10 sigma events that happen all the time? Now who knows? They may be fat tails, surely the fat tails exist. But you don’t need fat tails to get that in credit. Because you can actually have these slopes go from whatever they are to 5 times steeper. Not 1 or 2 times, 5 times steeper. So if you don’t correct your estimate and you’re using the old one, historical, then what is in fact a 2 sigma event looks like a 10 sigma event because you’ve underestimated the degree of sensitivity by a factor of 5. And that’s the part of credit risk that makes it interesting to understand. And you can also see how it was possible, I don’t mean this is the explanation, but possible to have the kinds of events we saw with banks where they stopped lending, they didn’t increase their portfolio and they reported ever larger losses which could have come from cooking the books which you see come from the structure. Because each time you have the same shock to the same assets, the impact gets greater and greater. By the way so you don’t think this is Doctor Doom talking first thing in the morning to you. The good news is that if you’re at that lower point if assets values recover, the risk falls down very rapidly. And whereas we’ve seen corporate assets represented by the stock market have increased. Real estate in many places has started to come back. So it can very well be that the risk goes down very dramatically too. So it’s a 2 way street. Ok end of story on that. I can see I’ve got... it took me half the time to teach it, sorry. But if you just got this, if you just understood this structure, you’re well on your way to be able to be a trustee or sit on a board or even in senior management. And when they come in with the risk report on the mortgage or loan book, you know what questions to be asking. And if the answer that you get is same assets as last time, everybody is paying, therefore the risk and the price should be the same and you ask them the follow up question: What’s happened to real estate assets? What’s happened to corporate assets? If they’ve gone down, you would be confident from this picture to be able to say to them: I don’t know what the number is but I know that what you’ve just told me can’t be true. Okay, anyway. So end of that. Now let me tell you what we’re going to do with that. Oh, let me point out one other quick thing before we get to that. That’s for a guarantee. Now we all know the governments guaranteed our banks, either explicitly or implicitly or both, given. So now you know how to... If we go back here, we look at governments. They’re writing put options on the banks, assets. But what are the banks’ assets. Yeah they’re risk free minus they’re writing insurance. They’re in an insurance writing business. So what are the governments doing functionally? They’re writing puts on puts. So if you went to the real assets underlying these, not the bank assets but the corporate, what do you have? You have a put on a put. You have a convex function of a convex function, swish, that’s very convex. I don’t have the picture of that. Keep that in mind when you think of things like Thailand ’97, going along fine and then seemingly overnight ending up in a mess. It can happen very rapidly because of this high degree of convexity. I don’t mean that’s the only reason but that’s to point out. Now, so we know governments guarantee banks. But what happens sometimes when governments have trouble funding their own bonds, you know, the final lack of liquidity to buy their bonds. They often go to their banks and say: “You know, we have a very good investment for you: your sovereigns bonds. I mean, you know, we’re your sovereign. We’re even guaranteeing you so shouldn’t you buy us?” And they do. In the United States FDIC who guarantees, issues bonds which are bought by banks, by the same banks that they guarantee. So you have a situation when that happens where the banks are holding bonds of the government. But we know that if you hold a bond you’re writing a put option, you’re writing a guarantee. That means the banks are guaranteeing the government. And the government is guaranteeing the banks. So never starts with... well almost never starts with government. But let’s say the banks have a shock. The assets fall either from bad management or whatever. That makes the government weaker because they wrote these guarantees. You saw their assets go down, their risk went up. Well, what’s the feedback? That bank guaranteed the government. The government is now riskier, okay? So their guarantee goes up which means their assets fall, they become weaker and you can get a very nasty feedback effect. It gets more complicated, you know, and I’ve listed here... if you look at something like Euroland because it isn’t just one sovereign with their banks. It’s sovereigns issuing to banks of other sovereigns etc., etc. And this is just, you know, loops of what happens when you have the banks and the thing. So you have German banks holding Greek bonds that are guaranteed by Greek government. And so you can imagine. And then the banks dealing with each other. Don’t try to read this, just see this as frictions. So now you’re credit experts, you now see the potential of the feedbacks in systems, especially when you have this, the guarantee guarantees the guarantor etc., etc. Now let’s talk about how we would use this to measure connectedness. What we do is we estimate the value of the put option, the guarantee, and we scale it to the value of what’s being guaranteed. So you get a number like a percentage: 6%, 4%. That’s the cost of the insurance of guaranteeing the bond. So the higher the cost of the insurance the greater the credit risk. That makes sense. And we all saw there’s lots and lots of ways of doing that evaluation. One would be to use theories of put options and experience with put options. That’s only one way. So what we do is we compute that number and we take that ratio. And then if we want to understand the sensitivity of one institution to another or an institution to a sovereign or a sovereign to an institution, what we look at is that the credit level cost. You think of that as the cost of insuring for that entity, insuring a sovereign, insuring a bank, insuring an insurance company. And we look at that credit and relate it to the credit of something else; so a bank to a government, a bank to another insurance company or the other direction. And we did here with the data, simple first pass, we did simple one-period lag, essentially a standard, you know, the standard one, one lag. So we looked at credit last month and related to credit of the other entity this month. And if there’s a positive relationship between them, we would say this one, the one that the last month ones influence. Now I won’t say cause because we all understand it’s not a causal statement. It’s only causal in the sense of Granger Causality type meaning. But he causes that. So we can do institution J’s credit against institution K’s credit one period later. And we can run reverse. And we look at that coefficient, the simplest regression. This is the simplest version. If it’s statistically significant we say influence that... You know if it’s J influences K in credit because it’s... If we ran the reverse one and we found it was significant, we say K influences J, alright? Only in credit notice, this has nothing to do with anything else but credit. Now if it goes both ways which it can, then you have what we call feedback. This one influences this one. And this one influences... just as I gave you in the little story between governments and their banks in terms of the credit impacts. Okay? Now, how do we measure empirically for this study the cost of the insurance? You’re all aware I think that there’s a CDS market out there, a credit default swap market, which does sovereigns, some banks and so forth and so on. The natural place to look because this is a grown up market, this is not... A natural place to look to get the market price would be just use CDS prices. That’s the insurance prices. You can also get them implicitly from bonds and other ways. But the nice thing with the data now is you have the CDS market. We didn’t use CDS though for the institutions. We did insurance companies and banks. We couldn’t do pension funds. We didn’t have the data. So we’ve only done this with these 2 big groups. And we did not use that for them. We did use it for the sovereigns. So we had sovereign credit default swaps as an estimate of that cost. And what we used for the insurance companies and the banks was a model. It’s a version of a model I had something to do with 40 years ago that does the valuation along the lines you just saw using option type theory, derivative theory. But it’s a model that’s been used everywhere now. And it’s good enough. We’re not trading on it. It’s good enough to give you an indication of how they change. I mean this is used in the real world on zillions of dollars. So it doesn’t mean it’s right but it’s not a bad approximation. Now, why did we use that instead of CDS? Because we want to get the total credit risk of the entity. What does CDS give you? The cost of guaranteeing the private markets piece of the risk of the entity. So when Ireland guaranteed all the Irish banks, the CDS rate went way down on those banks. Why? Because they got a guarantee. It isn’t because those Irish banks got to be a better credit. Their assets were the same. Their liabilities were the same. It’s just that a portion of what was being borne in the private markets in credit risk got shifted to the government. If you want to see it because that’s what data shows you, the CDS on Ireland went way up. So banks went down and Ireland went way up because Ireland took on that risk. We want the total risk here. We don’t want to have the view that something has gotten safer not because it’s really safer but because it’s shifted the risk outside of the private market. That’s why we use the model there. Ok, so you understand where the data comes from. We did a simple... this simple regression. This is just to convince you. As you all know you could use all these lags and leads of each and own and everything and do that. And we looked for... We put this against data we had from 2001 up to 2014, this year’s January. We had monthly data, we had 17 sovereigns, 63 banks and 39 insurance companies, a 102 institutions, 17 sovereigns. And we computed all of these regressions’ numbers for different periods and put them together. And you have a pile of regression coefficients this high. Now what do you do with them? And that’s what I want to show you in my last 5 minutes. I always knew I was going to run out. And that is we use network theory. We use a lot of other ways how to display data in a fashion that your eyes can see things that a pile of regression coefficients is a nightmare. So it’s like looking at a painting in a museum. You see the whole painting. And you get out of it. So I’m just going to show you pictures of what happened when we did this with the data for different time periods and I think it’s... So what we did is here’s a picture. Blue is insurance companies, reds are banks and blacks are the sovereigns. And these are all only statistically significant connections drawn. And if it’s from A to B, if that’s the direction, remember with the lag, then we point an arrow in that direction. If it’s from B to A, it goes in the other direction. If it’s both ways there are arrows on each one. You can’t see this. This is like looking at Google map from outside of the world. You’re looking down at the whole world. Okay? This is a time lapse photograph, you know, it’s like you left the lens open for 3 years. And this is a composite picture, just so you can see, prior to what most people would say is the crisis. So do you understand what this is? These are only significant, only with credit. This has nothing to do with transactions of other sorts directly, only measuring credit. That’s before the crisis. Take a good look. That’s from 2009. Any difference? Before, you know, a little denser, huh? What is that telling you? The degree of connectivity at least measured this way has gone way up. Now, like a good painting if you sit in front of it for 2 hours, the more you look at this the more you see. You see the colours. The black remember is the sovereigns. The blue is insurance. The reds are banks. And you can do all of that. But I just want you see that this is like Google map from outer space. Alright? Now, by the way that’s since 2011. All I want you to see here is that it’s still dense but the insurance company, blue. I’ll just point out one thing: have become much more dense. So you’re getting information that something is happening that’s changing. But let me just go on. Now you just blow it up. I don’t have time to go through all of these. You zoom down the way you could on Google Street right down to a single building if you want. And you start to look at more of the structure. And what we’ve done is locate entities closer to each other the stronger the relationship. So we use visuals. If something is very... not only significant but big and significant, they’re close together. So by clustering them that way you can see from the picture who is sort of centre. We do Eigen functions and things like that to see who are the real players at a given point in time or over a period. And we can zoom down to these things. So that’s at one point in time 2008 before the Lehman crisis. And here’s Greece and others, the black ones. I’m not going to go into this. There’s Spain right before Spain really got into a bit of trouble. And what I wanted to show you here though... This is March 2012. I want to point out something. In there first you see Italy, the black... I can’t get over the screen. I don’t have enough time. That’s Italy right in the middle of everything. Look at the US. It’s very isolated. How could the US with the most influential bank, biggest economy certainly, how could it not be a big part of the system at that time? Well, you know one can only speculate because we don’t really know the reasons. We only have the data. But it could be argued that in March of 2012 the US in no way was going to do another 500 billion facility as it did for the EC in an election year for president. There was a lot of pushback. So one could make the argument: It’s unlikely that the US is going to step in to do a lot of credit enhancement around the world in election year. So the arrow going from the effect of the US on others credit. And despite all the nonsense that goes on sometimes in our congress and everything else, the truth is in March 2012 as today the safest place for people to put their money, at least what they do, is in the US. So there wasn’t a lot of worry that the US in 2012 was going to be hammered, its credit by the other. So you can sort of argue that in terms of credit not all the other activities because we put them all in there, you would never get it, that it would have a big influence. So the idea of this... At this point we’re not talking about doing anything more than saying: I think this asks good questions. It says: Why is that there and it’s different from the past? And we can use this. And I’ll show you another. This is a blow up. Now you’re really getting down to... You can see the lines and the arrows in there. Of course there’s Italy with many of them. But you’re down here you see the black one is Spain. The red one is Goldman Sachs. The 2 look married. And Goldman Sachs has a lot of arrows going into it. Now why? I don’t know. But if you were a regulator and you saw this nd this isn’t a pattern that it’s been in the past. Wouldn’t it make sense to call up Goldman Sachs and say: “Guys, we see an interesting thing here. You seem to be the centre of an awful of stuff coming in from insurance, lots of sovereigns and you look like you’ve just gotten married to Spain. So what’s going on?” You may get 2 answers. One is we don’t know. Or here yeah, here’s what’s going on. If you get the first one, don’t you think that’s an interesting question if you’re an overseer? And maybe you want to probe a little more. Not because there’s anything wrong. The fact you have high degree of connectivity doesn’t mean it’s bad. It just suggests that you’re vulnerable to systemic event. So that’s the idea. In my last, minus 10 seconds, alright, these are the time series where we measure arrows coming in and arrows going out. So from means the arrows of sovereign are impacting others. Ireland is being impacted by the banks and so forth. And we plot these here. And you can see the main thing to see here is it’s not constant through time. That’s good. That means these things are really changing. You don’t take one picture and it’s static. It’s not static. It’s a dynamic process. Similarly here’s... You know they have a pejorative term for those 5 countries. We tried to find the best we could. We came up with GIIPS for those 5 countries. And what we plotted is they’re being impacted, ok, minus their impact on others. Now most of those economies except Italy but certainly Ireland and Portugal are tiny economies. So normally you would expect them to be pushed around by everybody else. They’re the lifeboat sitting next to the ship. But as you will see here and that’s what it was prior to the crisis... But when you get into crisis mode it rolled up. So it’s showing you that even small things at times can actually be in the sense of Granger Causality having that kind of impact. So, you know this is a tool that we use in doing it. Well, I don’t have time to go through the rest. But let me just simply say as a summery. We’ve done a lot of this. We use market data. We don’t use accounting data. We don’t use position data. We don’t use central bank data. But this I think would be useful for Central Banks and others who are overseers because this uses market data. Is the market always right? Of course not. But what it is the market is telling you something. Secondly it’s really cheap because you get this data every day. So you could run this thing every freaking day and have updates. Accounting data comes when it comes, often highly lagged. Position data the same thing. And then shifting through it. So I don’t mean this is instead of. But it’s an added tool. And while as I indicated before we have high hopes that at some point as we refine this and do this that we’ll be able to get a lot of information of how we might actually be able to do policy or other things using this as a tool. But one thing I’m very sure of, it ask good questions. So you know stress testing right. You take 100 scenarios somebody dreams up. How many scenarios didn’t you test? Infinity minus 100. So what this is, this is a smart scenario. But it’s not smart scenario by you know dreaming it up, you can still do that, this is the market saying this relationship and this place, something is different, changed dynamically. You know you look at that and if you don’t get the right answers, that’s telling you maybe you ought to stress this because this is what the market seems to say. Cheap to do, fast to do, gives you indications you will not get from accounting data and you will not get from position data no matter how much you have. Thank you very much. And have a great morning. Applause.

Robert C. Merton (2014)

Measuring the Connectedness of the Financial System: Implications for Systemic Risk Measurement and Management

Robert C. Merton (2014)

Measuring the Connectedness of the Financial System: Implications for Systemic Risk Measurement and Management

Abstract

Macrofinancial systemic risk is an enormous issue for both governments and large asset pools. The increasing globalization of the financial system, while surely a positive for economic development and growth, does increase the potential impact of systemic risk propagation across geopolitical borders, making its control and repairing the damage caused a more complex and longer process. As we have seen, the impact of the realization of systemic risk can be devastating for entire economies. The Financial Crisis of 2008-2009 and the subsequent European Debt Crisis were centered around credit risk, particularly credit risk of financial institutions and sovereigns, and the interplay of the two. The propagation of credit risk among financial institutions and sovereigns is related to the degree of “connectedness” among them. The effective measurement of potential systemic risk exposures from credit risk may allow the realization of that risk to be avoided through policy actions. Even if it is not feasible to avoid the systemic effects, the impact of those effects on the economy may be reduced by dissemination of that information and subsequent actions to protect against those effects and to subsequently repair the damage more rapidly. This paper applies the structural credit models of finance to develop a model of systemic risk propagation among financial institutions and sovereigns. Tools for applying the model for measuring connectedness and its dynamic changes are presented using network theory and econometric techniques. Unlike other methods that require accounting or institutional positions data as inputs for determining connectedness, the approach taken here develops a reduced-form model applying only capital market data to implement it. Thus, this model can be refreshed almost continuously with “forward-looking” data at low cost and therefore, may be more effective in identifying dynamic changes in connectedness more rapidly than the traditional models. This new research is still in progress. The basic approach and the empirical findings are encouraging and it would seem that at a minimum, this approach will provide “good “questions, if not always their answers, so that overseers and policy makers know better where to look and devote resources to discovery among the myriad of places within the global financial system. In particular, it holds promise for creating endogenously specified stress test formulations. The talk closes with some discussion of the importance of a more integrated approach to monetary, fiscal and stability policies so as to better recognize the unintended consequences of policy actions in one of these on the others.

Background reading:

Merton, R.C., 1977, “An Analytical Derivation of the Cost of Deposit Insurance and Loan Guarantees,” Journal of Banking and Finance, vol. 1 (June); pp. 3-11.

Gray, D. F., Merton R. C. and Z. Bodie, 2007, “Contingent Claims Approach to Measuring and Managing Sovereign Credit Risk”, Journal of Investment Management, Vol. 5, No. 4, pp. 5-28.

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT

Content User Level

Beginner  Intermediate  Advanced 

Cite


Specify width: px

Share

COPYRIGHT


Related Content