Fixing the financial system

<p><em>Romesh Vaitilingam interviews Raghuram Rajan for Vox<br />
</em></p>
<p><em>July 2010<br />
</em></p>
<p><em>Transcription of an VoxEU audio interview [http://www.voxeu.org/index.php?q=node/5282]</em></p>
<p><strong>Romesh Vaitilingam:</strong> Welcome to Vox Talks, a series of audio interviews with leading economists from around the world. My name is Romesh Vaitilingam, and today's interview is with Professor Raghuram Rajan, of the University of Chicago's Booth School of Business. Raghu and I met in London, in July 2010, where we spoke about the Squam Lake Report, of which he was one of 15 contributors . The report aims to map out a long term plan for financial regulation reform. And I began by asking Raghu what brought the group, informally known as the Lakers, together?</p>
<p><strong>Raghuram Rajan:</strong> It was Ken French, I think, who had the idea of getting a group together. He's been the driving force right through. It's hard to get 15 academics together to agree on anything. And so getting this out and getting these papers out, and eventually the book out, is really a tour de force that he should be complimented on.</p>
<p><strong>Romesh:</strong> The fundamental question you're trying to answer in the book is how to prevent a replay of the world financial crisis. Tell me how you came together collectively to think about addressing that big question.</p>
<p><strong>Raghuram: </strong>Let me start off by saying, I don't want to represent 15 people's views. So think of these as my personal views on what we're trying to do. I mean, a bunch of things that we're trying to do. One: we're trying to get the incentives right within the financial firms. Two: We are trying to get the incentives right within the investors in these financial firms. One of the problems is, investors didn't fully appreciate off pricing the risks that were being taken.</p>
<p>Three: I think we're trying to structure the system so that, if in fact it gets into trouble, there are plans in place to make it easier to resolve the system. And fourth: I think we're trying to create a system which was in a sense, less likely to fail by creating redundancies, by creating transparency, by creating smaller exposures and so on. So those broadly summarize the kinds of things that we've been proposing here.</p>
<p>There are lots of small things. And no... again, to use a favorite phrase of mine, &quot;No silver bullet here.&quot; No one thing that's going to fix the system. But I think when you put all these together, they could lead to a much stabler system. And one which may well, be much less prone to crisis.</p>
<p><strong>Romesh: </strong>But why do pick one or two of your favorites, perhaps. You have a number of chapters going through specific recommendations, one of which is called, &quot;Reforming Capital Requirements.&quot;</p>
<p><strong>Raghuram:</strong> There is a general sense that we have to raise capital requirements, to increase the buffers, so to speak. But I believe that if you just raise capital requirements, double them, triple them--pick a number--certainly with financial funds, it seems as if equity is much costlier than debt-like finance. And this is going to raise the cost of capital to the financial firms. They're going to pass them on. So this is going to raise the cost of intermediation. There are other ways we can do this, which are cleverer, but will give us similar levels of stability.</p>
<p>So that's what prompted the search. And the idea that we came up with which relies on fast work by Mark Flannery, for one. And some work that Anil Kahsyap, Jeremy Stein and I have done. But then developed from that, I think is more for our proposal, is what we call, contingent capital.</p>
<p>And the idea is to get some kind of convertible debt, which will convert in bad times into equity, and provide a cushion for the bank at that time. Not only does it provide a cushion, it has good incentive properties, because it dilutes the existing equity holders and therefore gives management an incentive to avoid that conversion. So they don't have an incentive to trigger the conversion, because it'll upset the equity holders and make the equity holders very angry. Anticipating that, management has an incentive to raise equity well in advance.</p>
<p>So this contingent convertible, or reverse convertible has some good properties. There are some issues that are quite thorny that need to be dealt with here: One is, could you get a debt spiral with these contingent convertibles? In a sense that, anticipating that the convertibles will convert, will equity holders take fright? Depress the value of equity?</p>
<p>And if in fact one of the triggers for the convertible converting is equity values, then you could get this, what we call, multiple equilibrium, a bad equilibrium where it gets converted. Equity values plummet. And the bank gets into trouble just because it issued these, now toxic instruments.</p>
<p>And so, some of the thinking is how do you deal with that problem. And typically that kind of problem is dealt with by not having the conversion based on equity values, but something else which is more immune to equity values.</p>
<p>Second issue is, how do you prevent potential moral hazard? Firms themselves forcing the conversion, because they get equity relatively cheaply, compared to the alternatives. Again, can you get some sort of an outside trigger that helps make this happen? Rather than everything be an inside trigger.</p>
<p>So we came up with the idea of two triggers. And we can debate what those triggers might be. There's a lot of healthy debate within the group. But the whole point is, an external trigger, based on the systemic condition of the economy.</p>
<p>One way that trigger could happen is if the regulators say we have a systemic problem. (Of course, will regulators do that?) Another possibility is to base it on, say, the profits of the largest banks, and when those banks start making deep losses beyond a certain number, that trigger is invoked.</p>
<p>The second trigger could be something related to the condition of the firm itself. Barclay's recently issued a contingent convertible based on its book capital. You could also think about something based on moving average of profits or something like that.</p>
<p>So when both these triggers are invoked, the convertible converts, and provides a lot of equity. Why is it a good thing? Because otherwise the firm would have to issue equity. And that would be much costlier. With this, it issues debt. And the debt converts only in a small set of states of the world, states which upfront are highly unlikely. But when it does convert, it provides a whole lot of cushion to the fund. So in a sense, this is cheaper insurance than the insurance provided by equity.</p>
<p>Think of equity as being insurance that's available all the time, when you don't need it also, which gives perhaps too much flexibility to the bank. The convertible debt gives only conditional insurance. And as a result, first, it&rsquo;s not used all the time, second, it&rsquo;s cheaper, third, because it still limits--you have to make debt payments--it still limits the bank, to some extent, and it&rsquo;s part of its leverage ratio. It prevents greater freedom risk taking by the banks than would otherwise be desirable. That's one idea.</p>
<p><strong>Romesh: </strong>Well let's come on to another, perhaps one that's the general public care about the most, certainly in Europe. Its bankers&rsquo; bonuses. Tell me about the outcome of ideas you're thinking about in terms of remuneration.</p>
<p><strong>Raghuram: </strong>Well here the big question is why don't the boards themselves set remuneration and in appropriate levels? What is it that they're smoking and not doing this? One possibility that I think, is the boards themselves have an incentive in taking risks. Because of course, the nature of equity is, upside I get, downside somebody else absorbs. But you couple that with the potential that some firms are too big to fail, then the downside is not absorbed even by the debt holders. It's absorbed by the government. So there is a certain sense in which boards can be a little too willing to take big risks, and not discipline their management, or even further down, the traders.</p>
<p>So one idea was, can we create kinds of instruments that would actually suffer in a time of crisis? So following the same sort of logic as the contingent convertible, is it possible that management stand to lose something? Not so much tied to equity. The problem is, if you tie it to equity, governments have an incentive to come in and bail out equity.</p>
<p>So can you have a kind of junior equity? And this was the logic behind the argument that some fraction of bankers pay should be held back. It should be held back and exposed to potential losses that the bank might face.</p>
<p>But in particular, if the bank gets support from the government in one form or the other, then these claims would be wiped out. So, even if equity is still alive and flourishing, management tends to lose a portion of its stake, if in fact the bank is helped out.</p>
<p>The details of how this needs to be done have certainly got to be worked out. But, it's more than just telling, your management has to have a long term position, and just to say, that they have to have a long term position which is even junior to equity--and therefore, is the first loss, if in fact there are interventions--that, hopefully, will provide more incentives.</p>
<p>Now, will that be the solution? No. That's going to be part of the set of things that need to be done to make management price risk better, to make claim holders price risk better.</p>
<p><strong>Romesh: </strong>You conclude the book by saying, what would have happened during the financial crisis if the reforms you propose had been in place.</p>
<p><strong>Raghuram: </strong>This is where each one of us has a different view of how the crisis could have bailed out and what would have happened. But certainly, take a couple of other things, and this is giving me an opportunity to talk about a couple of other things, if in fact we had living wills in place. It's not that we'd have an easier time resolving these banks. But perhaps Lehman may not have had 600 subsidiaries, knowing that the regulators would object to such a complicated web of subsidiaries.</p>
<p>Lehman, when it filed would actually know, have a better sense of where the liabilities were, who the exposures were to, how much they were, so that you could can, sort of, in a sense, cauterize the problem much more quickly.</p>
<p>So, a couple of things, living wills: These are attempts to get banks to make plans for their eventual closure. But it would involve first, simplifying bank structure over time, in order to allow for prompter closure. Also, knowing what the exposures are, what the jurisdictions for these exposures are, et cetera, et cetera.</p>
<p>So, there's a much better sense of where you'd have to file, who you'd have to file under, who would have to be informed, et cetera, et cetera, when you actually have one. So, it is a contingency plan, and would you actually follow it when you get in trouble? No. Because the situation may have changed, when you get in trouble. But, it gives you much more information.</p>
<p>I think we spend too little time thinking about failure, because failure was unthinkable. This forces you to think about it, and allows the authorities to participate in that process, so they have a better sense of resolution. So perhaps, Lehman would have been actually fail-able, certainly far less messy.</p>
<p>The other part of the coin is, because the powers that the regulators have to close Lehman, in an appropriate way--that again is something that we talked about, but also is in the Financial Reform Bill--whether, in fact, the authorities have the appropriate legal bars to deal with a situation like Lehman in the crisis, it was highly uncertain. Hopefully, now we can do it.</p>
<p>So, how would this have changed things? Well hopefully, if we had this in place we wouldn't have had a disaster like Lehman. Yes, Lehman could have gotten into trouble. Hopefully, some of the measures to prevent them from taking the kinds of risks they took would have worked in limiting that risk taking.</p>
<p>But also, when it did get into trouble, perhaps, the exposures could have been more limited, would have been more on exchange. Some of them would have been easier to shield from the rest of the system, and would have had an orderly resolution.</p>
<p><strong>Romesh: </strong>How do you think we're doing, practically, now with financial regulation? How much is that kind of stuff is being talked about in Washington, following the principles that you would ideally like to see them follow?</p>
<p><strong>Raghuram: </strong>I actually think that the Financial Reform Bill that is being talked about has much of what we've suggested. And I'm happy that it doesn't seem that the sole objective of that Bill is to look at the past and to try and punish the perpetrators, but more look at the future and how do we prevent this from happening again, which is as much what we should be talking about. What is uncertain--I mean, certainly there have been political compromises, as many people have said. If we wanted a regulatory system in the U.S. we wouldn't have the kind that we have now. If we're starting from scratch, we&rsquo;d never come to what we have now--the divisions between the regulators, and so on. Those kinds of things haven't been fixed.</p>
<p>But that said, there are lots of good ideas that are in place, but like the Health Care Bill, there's a lot of good that can be done with what's in place, but all of it can be whittled down to become a meaningless nothing at the end.</p>
<p>So, we'll have to wait and see what happens, not so much in legislation, but so much is left to the regulators to develop. For example, capital requirements will have to be developed over time, and that means just not the U.S. regulators but also Basel and the international arena.</p>
<p><strong>Romesh: </strong>Well I was just about to ask that question, about international consideration. How much can financial regulation in this globalized world, particularly financially globalized world, be done in just one country, or one region?</p>
<p><strong>Raghuram: </strong>Well if it's a country as big as the U.S., and as central as the U.S., I think there is a certain amount that the U.S. can get away unilaterally with, and set the pattern for other countries to match. But of course, there are some elements that the U.S. will have to adapt, as it sees others move. You cannot, in a sense, go alone on some dimensions. For example, if U.S. capital requirements are much higher than anywhere else, you are going to have the arbitrage that takes place. So, there has to be some harmonization of these over time.</p>
<p>I do think there's a lot of for the international arena to do, which it's not doing. For example, cross border resolution. When a large bank which has branches in many countries fails, how do you resolve that? Who bares the costs? Those are things we're not talking about. We should be talking about more.</p>
<p><strong>Romesh: </strong>Is it right to say that a lot of the financial innovation of the past has been driven by financial institutions trying to get round regulations? And, do we see that as a problem going forward, and that the new regulations will come in to place, and then that will drive different kinds of innovation to try to bypass them?</p>
<p><strong>Raghuram: </strong>Well, for sure. And this is why one of the things I feel about regulation is, first it should be uniform. There shouldn't be shades of light and dark. We shouldn't say financial banks be regulated really strongly and we let the hedge funds go completely unregulated, because then a lot of activity would move to the hedge funds. So, we need more uniform regulation. But also, I think we need far more information as to what's going on across the system, so that we know when these exposures are building up elsewhere. That's where there is a little bit of a catchall kind of possibility in the Bill. And in general, that systemically large institutions will come under the purvey of...</p>
<p><strong>Romesh: </strong>So that could be anything.</p>
<p><strong>Raghuram: </strong>That could be anything. Of course, that leaves the possibility of abuse of power, but it also allows you to monitor the system and see where activities moving so that you can actually... We need a certain amount of dynamism in the regulators. You can't just set the rules and then think that it's going to work forever.</p>
<p>But also, I think we need to regulate the regulators, and this is where I think transparency about the regulatory process, as well as much more public information as to what's going on, will help keep them on their toes. I mean, things like real estate exposures, things like exposures to one another, there's no reason why over time we can't find a way to make this information public.</p>
<p>So far we've said, oh, it's going to create problems, et cetera. But, on a steady basis, as we start gathering all this information, and there is now a new Office for Financial Research, which is going to be set up, which is going to describe the information that has to be created. I think that could help quite a bit.</p>
<p><strong>Romesh: </strong>Final question, Raghu, there are lots of jokes about economist's propensity to disagree. Two economists in a room, you'll get three opinions. If you have Maynard Keynes in a room, he'll have three opinions. How did you manage to get 15 economists to agree on these kind of proposals, particularly when you've got such a range of different backgrounds. You've got some Chicago guys, you've got some guys that are much more interested in the behavioral stuff. How do you guys work out some of your differences, and come to some firm conclusions?</p>
<p><strong>Raghuram: </strong>Well I would say Ken French played an enormous role here. I think he was a good moderator, and sometimes, he knew when to shut off discussions. Given I think, he commanded tremendous respect in this group. But also, I do think that there was a common sense that one, this was an important thing we all had to do, but also, I think, that after vigorous debate, people realize at the end of it you have to produce something. So, we have to come to a conclusion. There were champions of different pieces, and eventually they put it together and people stopped saying, no, I disagree.</p>
<p><strong>Romesh: </strong>Raghu Rajan, thank you very much.</p>
<p><strong>Raghuram: </strong>Thank you.</p>