Transcription
There was something that really struck me as I was listening to people talk this morning. Ten years ago, if you had a conference that was on capital and liquidity, two things would have been notable: one is everybody would have been talking about Deposit Insurance, and nobody this morning mentioned deposit insurance at all. So that's a dramatic change from what I think we would have experienced any years ago. And liquidity probably wouldn't even been on anybody's radar. So there really has been quite a dramatic change in how people are thinking about capital, liquidity, systemic risk, and so I thought from my comments you were going to get plenty of details about banking in Basel III, that I talked more about financial stability more broadly, which includes a lot of the topics that we're talking about today, but it's not solely related to what's going on in the banking sector or with the supervisory requirements that we have.
So let me start by talking a little bit about financial stability. And I was going to start when I was planning on coming out with my bound copy of Dodd-Frank, but it was too heavy to carry in my carry-on, and it was too big. And I think you're all quite aware of how voluminous that legislation is. But I would say that, well everybody agrees that financial stability is very important, there is no clear definition in the legislation. So it's notable that in all those pages and with so much focused on how do we think about financial stability, we have an oversight council that the legislators chose not to actually define it. And the reason that's important is because I think people define it differently, and how you define it does make a difference for how you set regulations, how you think about the kinds of problems, and how you think about the possible policy responses.
So a lot of the focus is on interconnected failures, and we had some discussion of that this morning. But what I want to emphasize in my comments is it's not just large institutions failing that can create financial stability issues. So I'm going to talk a little bit about other types of mechanisms, some of them related to the banking industry but not exclusively related to the banking industry. So let's start with the definition. So on the top half of this slide, I've just put three things that when you talk about financial stability in common usage, you see that there are different ways of interpreting what financial stability means. So sometimes when you hear people talk about financial stability, they seem to be talking about volatility; sometimes they seem to be talking about clustered failures of financial institutions; sometimes they are talking about asset bubbles and how do we prevent asset bubbles in the early stages of development. I'm going to argue, actually, that I wouldn't be focused on any of those three in my own definition. I'll give you my definition in a minute. And so the definition we're using at the Federal Reserve Bank of Boston, but I do think it is important because if I'm focused on any one of those, you might come up with different policy prescriptions about what supervisory policy should do, what regulatory policy should do, depending on which of these things you really want to be focused on. So I do think it's important for us to get a better sense when we're using these kinds of terms, what do we actually mean?
So my definition—I'll just focus on financial instability; you can all read it—but why don't I just quickly go over it? Financial instability occurs when problems or concerns about potential problems within institutions, markets, payment system, or the financial system in general significantly impair the supply of credit intermediation services, so as to substantially impact the expected path of real economic activity. So what does that mean? Well, I think it breaks down into three elements. The first element is that you need to have a problem in the financial system; so I think that's pretty intuitive. But the other two are critical as well: one, you need a disruption or impairment of the intermediation process, so the supply of credit is affected; and finally, you need a substantial impact on the real economy.
So why do we care so much about financial intermediation? This is something that actually, when I was going to graduate school, there was no money in banking at all the economics then in many of the business schools. And actually, the institutional knowledge that's tied to money and banking and financial markets virtually evaporated over the last 20 years, and it's actually hard to get people outside of the financial sector itself that really understand a lot of the things that we've been discussing today so frequently. If you're talking to an academic and you're talking about Basel III, you get a pretty blank stare, unless it's somebody—and it's an unusual academic—who actually really understands the mechanics of this. They're in relatively short supply. I think that's unfortunate. This crisis has told us that we actually need a better supply of people coming out of our graduate programs that have an appreciation for this. And so what normally would we talk about for intermediating and why it's so critical is that it plays a critical transformation of assets, and there are two aspects to that: one is the maturity transformation, and the other is the risk transformation. So people that are providing deposits to a financial intermediary frequently want to have immediate access or close to immediate access to their funds, and when they want to pull out their funds they don't want their principal to be at risk. But those people who actually want the funds tend to want the funds for a longer period of time and tend to want to put it into riskier assets. So we have a natural matching problem, and that's why we have financial intermediaries in the economy to try to bridge this gap between borrowing and lending. And what we found is that when that matching process gets disrupted, it has a really substantial impact potentially on the economy and can have significant macroeconomic consequences. Reinhart and Rogoff have a very good book that kind of documents some of the experiences over a very long period of time with the kind of disruptions that occur when we have financial intermediation that becomes broken. And so I think it's much more intuitive after going through the financial crisis, and I think that book certainly documents some of the issues about why that matching process is so critical for a well-functioning economy and why when it gets seriously disrupted you usually have a very long recovery, you usually have a halting recovery. And in fact, that's what we've been exhibiting so far is that it's been a very slow and halting recovery that we've experienced today.
Now I want to highlight what's not in my definition, and that is that you don't necessarily have to have failures of institutions in order to have significant disruptions. That's certainly a way that you can have it, but it's not a requirement. So even if no individual institution fails, even if no group of institutions fail, you could still have financial stability being seriously compromised. And one of the things that I want to highlight is you care about the substitutability of this intermediary, and I'm going to talk a little bit about that in a couple of examples in a minute. But the intermediary services, if somebody can immediately step in and do the same thing costlessly, then I wouldn't have to be all that worried about intermediation getting disrupted. But if they're very poor substitutes, so that we can't easily transfer that matching to another organization, then that can be a real problem for the economy. And so I think it's important to think about what is the substitutability of the intermediary services. I would also highlight what's not in my definition is just asset bubbles. That's not to say that asset bubbles can't create a financial stability problem; they can. But not all asset bubbles are sources of financial instability. A critical component of whether or not it becomes financial instability is what is the role of leveraged financial intermediaries. So if leveraged financial intermediaries have a very large exposure to that asset, then potentially if the value of that asset declines quite rapidly, it has a broader impact on the intermediating process. So when we think about asset bubbles and whether it's having an impact on financial instability, we have to ask what are the role of the intermediaries in the economy, how is it being financed, what is the likelihood that similar organizations have the same exposure. But when I think about financial instability, it's not synonymous with talking about asset bubbles in the economy.
Now I'm going to talk about three things first of all that I think are not examples of financial instability, and then I'm going to talk about three examples that I do think are better examples of financial instability, just to give a sense of what I'm trying to convey by this definition, and then I'll have some concluding remarks. So the three that I'm going to argue are not instances of financial instability are some recent experience with silver prices, the failures—the grouped failures—of Savings and Loan institutions, and the experience that we had with the dot-com problem earlier last decade. So let's start with silver prices. Its silver prices are actually kind of interesting. If you look back to the late 70s, it was a period when there was an attempt to try to corner the market in silver, and as you can see there was a very rapid increase in silver prices. It was not a completely successful attempt to corner the market in silver, so there was also a very rapid decrease in silver prices that happened shortly thereafter. You can then see that we had a long period where people were—where silver prices didn't really move very much at all, and then very recently we've had a very big run-up in a number of commodity prices, both precious metals but also other types of commodities. But more recently there was a relatively rapid decline in the silver prices. So this is an example of something that has a lot of price volatility, but I would argue that there's nothing in the silver market that would make me worried about financial instability. So why would I say that this price volatility is not synonymous with financial instability? Well, one, it's a relatively small market. So while some people took a big risk in the commodity market, it wasn't widely held, and particularly it wasn't widely held in financial intermediaries that were highly levered. So it wasn't really financed by highly levered institutions. It's a small enough market that even if the price goes down doesn't have a huge wealth effect in the economy overall and doesn't really affect GDP, unemployment, the kinds of macroeconomic variables that we would care about. So even though it is very volatile, this would not be something that I'd be particularly concerned about with my financial instability definition.
Let's turn next to what happened with the Savings & Loan industry. So this was a situation where there were clustered failures, and so it's not that the problem of the Savings & Loan industry was not a problem; it certainly was. We had a Deposit Insurance Scheme that was extremely costly and wasn't structured appropriately, and we had regulatory and supervisory oversight of savings and loans that was inappropriate, particularly for the monetary policy that we were running at the time. So if you recall, the savings and loans were in a situation where they were borrowing funds and investing in long-term mortgages at a time when interest rates went up very rapidly. That was partly a result of regulation and supervision, partly a result of choice. So there's a combined culpability there, but they—as a result of, for the most part, supervisory forbearance—the full impact of those losses weren't realized immediately. We had net worth certificates and a variety of other different mechanisms to avoid closing down savings and loans particularly rapidly, and you can see it took quite a while before we really saw widespread failures of savings and loans. So while there were a lot of failures of financial institutions and while it was very costly for the taxpayer, why would I say this isn't necessarily an instance of financial instability? Well, this shows you the cost relative to the 10-year Treasury yield of getting a 30-year fixed mortgage, which is what the savings and loans were primarily focused on, and you can see shadowed is the—the failures. And what's notable is that you're not seeing much of a disruption in the funding of mortgages at a time when we had an awful lot of failures. So why is that? Well, this is a case where there was a lot of substitutability between the Savings & Loan industry and other types of financial intermediaries: commercial banks, savings banks; the securitization market was picking up. So in terms of a borrower, if I wanted to buy a house and borrow money in order to buy that house, the fact that the Savings & Loan industry was having a lot of difficulties didn't really impair the cost of credit and didn't really impair the ability to get credit. And so again, while there was a widespread failure of savings and loans and it was very unfortunate, we could have done a much better job with the regulation and supervision, I would not call this period of financial instability because it didn't have a big impact on the real economy because there was a lot of substitutability between financial intermediaries for this type of financial product.
Next, let's turn to the internet and the bubble that we had, or the presumed bubble that we had at the end of the 90s and beginning of the last decade. And as you can see, there was a very rapid increase in dot-com stock prices that shortly thereafter also fell. Something I don't have to say to this audience, but I would highlight that I—again, this actually was more widespread; it did have an impact on the economy; it was mostly through a wealth effect. But wealth effects are something that we are actually pretty capable of addressing through more traditional monetary policy, and I would argue in this case that despite the—the rapid increase and decrease in prices here, this was not a situation where most—the financing was financed by highly levered institutions; it was mostly equity finance. So it did impact wealth, but it did not impact the financial intermediation process. As a result, we had a very different issue coming out of this recession than we did over the Great Recession. Okay, so let—let me next go to examples where there is financial instability. And again, I think a lot of focus on Dodd-Frank has been on large, too-big-to-fail type of organizations, but I want to highlight in this talk is there are lots of other ways that you can get financial instability. And so the first example I'm going to talk about is the experience of the money market mutual funds during the crisis, because I think there's been an under-appreciation of how important that has been for the disruptions of short-term credit during the crisis. And that really was the smallest institution is the biggest risk rather than the largest institution. So it kind of turns Dodd-Frank on its head in terms of it's not too big to fail; it's too small to be able to address the problem. And so I want to talk a little bit about that. Second, I want to talk about the risk of widely held exposures that may not involve failure at all, and in that I'm going to talk about, in effect, credit crunches where a financial institution doesn't have to fail in order to have a problem in the intermediation process that could impact the real economy. And then finally, I'll talk a little bit about the interconnectedness.
This shows you the daily change in money market mutual fund assets at prime money market funds. So there are three types of money market funds: you have the government-only kind of money market fund, you have ones that are tax-exempt, and then you have the prime money market funds which are able to invest in a wider group of assets, which tend to include commercial paper, asset-backed commercial paper, bankers' acceptances. And you can see that after Lehman failed, one of the consequences of the Lehman failure was that the Reserve Primary Fund didn't have the capacity to meet the obligations of not breaking the buck—in other words, maintaining a stable net asset value, which is what the money market funds promised to their investors. And as a result, there became concern not only at the Reserve fund but also at a wide variety of other money market funds. So there was a rapid movement out of money market funds, and then—and I'm going to show you another slide that shows you one of the places that it moved—but I would just highlight that those are pretty big numbers, and these are days. And so day by day, having hundreds of billions of dollars leaving a financial intermediary, even if you have a lot of liquidity and even if the duration of these are fairly short, those kind of very rapid movements are going to be a problem for any kind of financial intermediary as long as there's any transformation of the asset at all. So it's not surprising that they had a very significant problem and in part had to be addressed by government intervention. So there was insurance provided by the Treasury during the crisis for an organization that everybody had assumed was not going to get insurance, and there was also a liquidity facility actually run by the Federal Reserve Bank of Boston which was focused on two components: how do we make sure that the asset-backed commercial paper market doesn't get completely disrupted—and I'll show you a slide on that in a minute—but also though to try to staunch the flow out of the money market fund industry. And you can see that with the Fed announcement of the AMLF as well as the start of the insurance program, within a couple of weeks we started to have a more stable situation in the money market fund industry. I would also highlight that at the same time the New York Fed was very active in a whole variety of other programs to try to deal with the disruption that was happening to short-term credit markets. So one of the concerns was the people that were pulling their funds out, but from a systemic standpoint you care about the asset side as well as the liability side of the balance sheet. So on the asset side, the concern was asset-backed commercial paper, commercial paper, bankers' acceptances—all of a sudden there was a very large investor in those type of assets that was not willing to buy those assets because they were having to quickly get funds available to take care of redemptions. So it was a situation that immediately required attention, and you can see that one of the things that occurred—so the purple, which is the bottom, is the prime money market funds that I was just illustrating; you can see the green, which were the taxable governments, was where an awful lot of the organizations moved. So you got out of the prime money market fund, you could easily transfer—usually within the same fund family—into government-only. During a crisis, governments actually tend to do pretty well; the risk was actually quite small, and the default risk was negligible. And you can see that there was a very rapid increase in that green bar as we got into the fall of 2008. So there was a natural shift that was occurring in the types of money markets, but it also highlighted why it's so easy to run because these are funds that you can pull out immediately, and you could easily pull it into another account quite easily that didn't have the same attributes as the one that you were in.
So what kind of disruptions did that occur? Well, this focuses on the asset-backed commercial paper market, but I think you can see a similar type of pattern if you look at the other kinds of financial assets the money market funds held. And so the first thing to note is that I'm showing the spread relative to the effective Fed Funds rate, and there was a very dramatic increase in the cost of funding for asset-backed commercial paper. But perhaps even more importantly is the maturity structure. So the share of asset-backed commercial paper maturing in one to four days went up to 90 percent. So basically, if you wanted any kind of longer-term financing, it wasn't available in the market, and that's not only for this market; it was true for a wide variety of other markets. So there was a real serious disruption in short-term credit markets that I think clearly made it systemic. So in that case, we had a very small money market fund that was not able to live up to the expectations of a stable asset value, and as a result of its inability to make those payments caused a broader run that disrupted the cost of financing and the availability of financing, and I would argue had a big impact on the real economy as well. So money market funds I do think are something that are critically important for financial stability, and actually not addressing that is a financial instability that we actually want to worry about. And so I want to just talk about that because while there were very substantial things done by the SEC, and I think those are welcomed in terms of thinking about the duration of the assets, thinking about the liquidity of the money market funds, the basic problem with the Reserve fund was there was a credit risk. So there was at least one money market fund, and actually quite a few money market funds, that had Lehman as the credit risk of potential Lehman failure. And so that credit risk is what actually caused the problem. And so I think the risk is that that has not been addressed in any of the regulatory changes. This is two and a half years after the crisis, and the money market fund situation is basically, in terms of the credit risk, very similar to what we were encountering during the fall of 2008. And I would highlight that there is a vulnerability here. So money market funds are looking to have a higher yield, and one of the ways of getting higher yield right now is to invest in short-term debt instruments in Europe. So they do have an exposure to European banks, for example, and as we're all aware there are some issues going on in Europe right now. So there is a potential risk—I'm not expecting necessarily to have a bad outcome—but nonetheless it is a vulnerability that I think that we have to be aware of. Similarly, taking it from the perspective of the Europeans, a critical source of short-term dollar financing for the European banks is the money market fund industry. So we have a pairing here that is a systemic vulnerability: the money market funds have some risk to unexpected credit loss in Europe, and the European financial institutions have a risk that could even just be a regulatory change that causes the money market funds not to want to hold short-term debt instruments in European institutions. So I think this is a situation that deserves a fair bit of attention.
So what are the possible solutions? Well, there are a number of solutions that have been recommended; some of them are on this slide: allowing the asset value of funds to float, so instead of promising net asset value that's fixed, allowing there to be some variation; a second is that you require the money market funds to have capital, so that if there is a loss they have the capacity to actually meet the cost; the third is potentially requiring a source of strength, so a money market fund that has no capacity to finance some bad assets in the portfolio maybe that is not acceptable; maybe you want a source of strength, or maybe you want to require some kind of insurance bond, or there are a number of different financial mechanisms that you could actually adopt. So there have been a number of solutions that have been suggested. I would highlight that none of the solutions are very far down the path right now, and I think that's unfortunate. What does a solution need? Well, I think there are three components: one is it has to address the impact of an unexpected credit loss; we at least need to be able to handle the last crisis, and if we can't even handle the last crisis that is a potential problem; second, we have to deal with the incentive for investors to move funds quite quickly, particularly when there frequently is an alternative that may be very easy to do, which is moving into a government-only money market fund; and the third is there is an operational convenience to money market funds, which is because there's a stable net asset value; every time I write a check on my money market fund I don't have to have a capital gain or a capital loss. So if you used it as a transaction account, you wrote hundreds of checks—though most people aren't writing checks anymore—but if you were one of those people writing hundreds of checks, then if it wasn't a stable net asset value, you'd have all these capital gains and capital losses. So thinking about how we address the problem in a way that continues to have the transaction account vehicle is a component that I think that the money market fund industry is very concerned about. But I think despite the various challenges with getting a solution, this is a vulnerability that continues; it is a vulnerability that I think is very important.
For us to address, and I'm hoping that we'll have more of a discussion over time to think about what is the best way forward. The second financial instability I want to talk about is a banking example. And in this example, I'm going to focus not on banks that have failed, but just banks that have become troubled.
So in this slide, I'm looking at the first year of the recession and asking what is the difference between institutions with high CAMELS ratings? So the CAMELS ratings gets at capital, assets, management, earnings, liquidity, and sensitivity to market risk. And so if you have a 1 or 2, it means you have a low probability of failure; if you have a 5, that means that failure is more than likely imminent. So 3, 4, and 5 are the more troubled institutions. Not a perfect comparison for capital constrained, but actually some of the research which I've done in the past has indicated that it isn't a bad indicator of capital constraints by and large. And you can see that while none of these banks were failed banks, there was a difference in behavior between those banks that got troubled and started to reduce lending and those banks that were not troubled.
Now that's very logical. If you have a binding capital asset ratio, there are a number of ways to address that. One is you can increase capital. We'll have a discussion of how costly it is to raise capital later this afternoon, but assuming that at least if the banks think it's costly to raise capital, so they're not particularly willing to do that, then the other way that you meet a capital asset ratio is you shrink assets. While shrinking assets is reducing the intermediating process for the rest of the economy, so problems that the banks have an impact on the borrowers of those banks doesn't require a bank failure, doesn't require a Lehman Brothers in order for this to happen.
And so in the early 1990s, we had a situation like that. Arguably, right now we still have a situation where there are many capital-constrained banks, particularly in some parts of the country that have a disproportionate amount of real estate losses, both residential and commercial real estate. So this just shows you what the pattern of real C&I loans have been, and I would highlight one thing—all right, highlight a couple of things actually. One is that the decline in real C&I loans has been pretty appreciable during the most—the Great Recession and the recovery from the Great Recession. But the other is each of these recessions have been very slow recoveries. So the previous two were called jobless recoveries; we didn't get much of a recovery. And interestingly enough, it's coincident with there not being a very rapid increase in C&I lending. And I do think in the most recent crisis there has been a very substantial decrease in credit, at least in part by organizations trying to meet their capital ratios by shrinking some of their assets. And this just provides another way of looking at it, which is from the borrowers: what is the cost, what are the standards that are occurring? And you can see in this Great Recession we have had a period where there was a very substantial increase in the tightening standards of the banking industry as they ran into some difficulties.
So the third example is one I think is well understood, which is the Lehman Brothers—the very large institution that fails and gets at interconnectedness. There is a number of ways to think about interconnectedness. The first is kind of the immediate exposure of the firm. So while we may not know who the creditors to Lehman were, we at least know one of them was the Reserve Fund, and through the bankruptcy proceedings we're learning about what the exposures of lots of other organizations were. But normally we don't know who the immediate—what the immediate impact of the organizations that have a credit exposure that isn't public information. So that's one source of interconnectedness. A second is—and it got to one of the questions that was asked earlier—yes, the counterparty exposures are not well known, and so there is no Paik miss, and that you don't really know what the counterparties are to a particular organization that you think is getting deeply into trouble. And that was one of the factors that caused the spread between the one-month LIBOR rate and the overnight index swap rate to really grow very substantially during the crisis. Banks were even concerned about evaluating some of their largest peers. So when this first appeared, it was really quite dramatic that some of the largest banks in the world were unwilling to lend for one month, or sometimes even one day, to some of their peers that they just didn't have a degree of comfort that they could really ascertain what the credit risk was and what the potential impact on them would be of an unsecured position.
The two other components of interconnectedness: one is the criticality of the firm. So many of these firms service market makers and particularly critical markets, and so we don't want that kind of intermediary disrupted. Auction rate securities and a variety of other markets were disrupted when some large financial institutions that were very critical to those markets got into trouble. And then the increasing global nature of the large financial intermediaries. We haven't had a lot discussing kind of resolution and the difficulties with resolution, but the international component remains a very big challenge. How do we address a very large organization? What does that mean for interconnectedness when it's not only a domestic and/or connectiveness but a global interconnectedness? And that even makes it more difficult to understand what the patterns of a potential failure would be.
Now I think there are a lot of questions to explore. They're a bunch of academics in this room. This is an area that I think actually is a fairly unexplored area, and I think we have to do an awful lot more work here. So one is just how do we measure interconnectedness? I don't think there right now is a good way, even if you have access to confidential information, how do you actually come up with a good measure of interconnectedness? If you can come up with a good measure of interconnectedness, how should that affect capital requirements? What does that mean for how much you want to reduce the probability that a highly interconnected firm becomes insolvent? And then we get to the disclosure. Banks are monitoring their counterparty risk; how much should be revealed to supervisors as a matter of course? How much should be revealed more generally to the market? If interconnectedness is an important component of systemic concern, then maybe there's a reason for the public at large to have some sense of how interconnected a firm is. And finally, what kind of role can stress tests play? So the Federal Reserve's gone through two stress tests over the last couple of years. One of the things that we've been asking is about counterparty exposures, and I do think that there's probably room to think about how do we use stress tests more effectively to think about this interconnectedness problem that is more likely to be something that we saw. I don't know if we solve that; we can at least better understand through stress testing than maybe through other mechanisms.
So just some very brief concluding remarks, and then I'll be able to answer some questions. I think the concept of financial stability means different things to different people. I actually think that we have to settle what a definition of financial stability is so that we understand what are the problems that we're trying to address. I've suggested my own definition, which is the definition we're using at the Federal Reserve Bank of Boston as we talk about financial stability and what we should be doing, but I think that we need to get a more widespread agreement on what financial stability is, what financial stability isn't. And then I think it really highlights there's a lot more work to be done understanding what really causes impairment of financial intermediaries and how that impairment of financial intermediaries can affect a broader economy is something that I think is not particularly well understood. It doesn't fit into our macro model. So during the financial crisis, as we were trying to think about what is the effect of financial intermediaries, virtually all our macro models don't model this particularly well, and so it is a real problem from any of the central banks; it's a problem for private regulators. And so I think it's a real challenge to really get a better understanding of how this kind of financial stability issue plays out more generally. So that concludes my remarks, and I'd be glad to answer some questions.
I want to ask you about the money market fund. So you talked about the question is with respect to liquidity demanded by individuals and check-writing and all of that. We have deposit insurance, and there's a—goes to what, 250,000—whatever goes through. The question is does—do we need a safety net system sort of for check-writing beyond that? And what's this sort of total cost-benefit of sort of having floating net asset value? Do we know for sure that, you know, people need the buck guaranteed? I'm asking for myself, but for example, I just read this week a proposal by—by Kansas Fed, which—which specifically says that the recommend is a way to handle, for example, shadow banking, specifically removing the any kind of guarantee. So the original concept of the money market funds was in effect the old narrow bank, which is there going to hold very short-term assets of very high quality, and that as a result they're going to be able to have a stable asset. And what wasn't taken into account were tail events and the fact that to the extent that you're taking any kind of credit risk, there is a potential that the net asset value is going to change. So one possible solution would be to think about extending the insurance to those organizations. That doesn't sound like what's likely to occur. My sense is that Congress is not anxious to extend the—the safety net to money market funds, and impart money market funds originally were created as a capital arbitrage. So it was a structure that didn't require capital; banks did have capital, and so there was an arbitrage in effect from the banking industry to the money market fund industry to avoid capital requirements. So when we talk about what are some of the distortionary effects of capital requirements, this is one of the examples. This is an industry created out of a capital distortion. So I'm not sure the solution I would choose, at least, would be to extend the safety net to these organizations, but I so—and I don't think there's much public policy support for extending this to money market funds. What would I personally do? What I would personally do would probably be a combination of both a capital requirement and floating NAV. And a floating NAV by itself doesn't solve the problem, and the reason for that is you still have an incentive to run. You don't know what the exposures are, and so just imagine if you were in a money market fund on the day that the Reserve Fund announced that they had an exposure to Lehman. You don't have any idea whether your money market fund has exposure to Lehman; you're not going to wait to find out; you're going to transfer the funds as quickly as you can. So even with a floating rate NAV, you don't eliminate the risk of very rapid movement out of that fund. So I think you actually need a combination of things; that would be my own personal view.
Eric, I wanted to go back to one of your earlier slides when you talked about some of the key roles the banks play in society, one of which being the transformation. This morning we talked about a number of firms—Mipham—own firms like this. We learned a lot from this crisis how precious liquidity was and that that was one of the key things to her own solvency and her own safety. And now we're also building a lot of regulations around to institutionalize that among a number of other institutions. I wonder how you balance the issue of if all firms are racing to kind of keep their own rainy-day liquidity funds and lessen the amount of maturity transformation, what that does to the supply of this to the overall economy, and how you balance that macro issue of supplying key service to the economy with a prudential issue of all firms now trying to take the lesson of this crisis and be—be more safe.
So I think that's a really good question, and I don't think we have a good way of quantifying actually exactly what the costs are if we're wiring too much or potentially too little liquidity. Our models aren't particularly good at calibrating that. I do think we have had an example of what can happen when we have too little liquidity, and we found those costs are actually pretty large, or at least potentially pretty large—the cost to society. I mean, if everybody was a money market fund and somebody wants to borrow for a home or for other types of business investments, you know, money market funds—not a vehicle that's going to give you a long-term financing for a business. So there is a natural problem with any kind of financial intermediary, and part of what makes them valuable is this transformation process. And it's a public policy balancing act to say what is the appropriate amount of capital, what's the appropriate amount of liquidity, and we can push too—too much in either direction. I think we're moving in the direction of getting that balance more right than we had during the process, but I think it remains to be seen whether we go too far or whether we don't go far enough, and hopefully over time we'll get a better idea of how well to calibrate our models so that we have a better sense. So there was an argument earlier that we needed—Rafael highlighted—what are the costs and benefits of this regulatory policy. I agree with that, but it's not that easy to do. And so one of the challenges is if you require everybody to be extremely liquid, it means a lot of the transformation doesn't occur. So I think that is something we have to work on. I don't think we have a good way of modeling.
So you talked about interconnectedness across financial institutions. I want to talk about a slightly different interconnectedness that's come up in the press, and that is when the government is virtual and debt that then the financial institutions need to hold to meet their reserve requirements, and more and more of the government debt is not going to necessarily into the real economy so to speak; it's going into debt service. So how do we—I mean, you've thought about this issue, or how do you think about the interconnectedness with that kind of government debt funding through—some called it repression—but through the—the ability to require reserves held in government paper, in fact, and what does that do to these financial institutions as they're trying to sort out their ability to get what we didn't stay with—would—and the price of that? So certainly government securities are one of the assets that banks are going to hold, and both the liquidity requirements and the risk-weighted assets are providing an encouragement to purchase what are—at least from a default sense—or at least we presume from a default sense for the US government—low, if not zero default risk. And so our banking regulations do have an encouragement to hold that asset, but it's more because of the characteristic of the asset—but because it doesn't have a default risk—that that is the one type of asset that during the crisis it clearly was able to be converted into cash for institutions that needed to raise cash. So I think there's a logic to why we have in some sense encouraged government securities holdings. We wouldn't want to over-encourage that. So we certainly want the transformation of assets; you want institutions to be taking credit risk; you want institutions to be offering financing for longer-term maturities, so that is an important part of getting the private sector financed. So I don't know if that addresses your question.
Quieter, I'm just wondering what flexibility there is within the reserve requirement to—to diversify somehow from—so you talk about the reserve requirement specifically. Yeah, this probably gets more technical than I'm going to talk about right now, but so the Treasury securities are different than the reserve requirement, and the quantity of reserves in the system is more tied to the Federal Reserve's balance sheet than the composition of assets being held by the banks. But rather than going into a torturous description, maybe we could talk offline and I can talk more about that.