Transcript

Washington, D.C.
Wednesday, April 6, 2011

MS. LOTZE: Good morning, and welcome to this press conference on the analytical chapters of the Global Financial Stability Report: Chapter 2 on Systemic Liquidity Risks, and chapter 3 on Housing Finance and Financial Stability.

Let me introduce our speakers, who are all from the Monetary and Capital Markets Department. To my immediate right is Laura Kodres, Chief of the Global Stability Analysis Division. To her right is Jeanne Gobat, team leader on the chapter on liquidity risks and Senior Economist in the Department, followed by Ann-Margret Westin, team leader on the chapter on housing finance and Senior Economist in the Department. And at the very end we have Jan Brockmeijer, Deputy Director of the Department.

Jan will start out with some brief opening remarks, and then everybody will speak a little bit about their chapter, and Laura will give an overview.

So let me turn over to Jan, please.

MR. BROCKMEIJER: Thank you very much, Conny. And good morning all.

It's a great pleasure for me to introduce the analytical chapters of our new GFSR. They cover two very important central elements of the financial crisis: the issue of housing finance and its relationship to financial stability. That clearly has played a very prominent role in the buildup to the crisis. And secondly, on systemic liquidity risk, the risk that markets dry up simultaneously, causing funding difficulties for a large fraction of financial institutions. These are big issues, they have been recognized, but they have not been dealt with yet. And it's not surprising that they have not been dealt with adequately because they are very complex—politically complex in the sense of the housing markets, and technically complex when it comes to systemic liquidity.

So certainly, housing in advanced countries has been a central feature, in many cases, of the tensions that we have seen of the financial crisis. The epicenter, after all, was with regard to the sub-prime mortgages that were embedded in private label securitized products in the United States.

The chapter on housing finance tries to take a step back and take a broader view to see how experiences in various countries, through a cross-country analysis, can draw out best practices that can help the set up of housing finance arrangements in such a way that they help preserve financial stability, rather than, in some cases, damage it.

While movement towards best practices cannot happen overnight, given the fragile nature of some of these markets in a number of the affected countries, a long-range goal is clearly identified in the chapter to get back to basics. And, to help make sure that future booms and busts in housing cycles are attenuated.

Another realization coming out of the crisis was with regard to the dependency on continuing funding liquidity, which is not at all ensured, it is clear, under crisis circumstances. The entire global financial system at one point was penalized when wholesale funding markets dried up at the beginning of the crisis, due to uncertainty with regard to the underlying secured funding markets—the assets that were underlying those markets.

The system-wide fragility of liquidity risk is the subject of chapter 2 of the Global Financial Stability Report. It has been a key feature of the financial crisis. Unlike previous ones, the degree of dysfunction was very much at the center of the functioning of the liquidity markets.

Large banks had become far more reliable on funding sources from a number of institutions, rather than depositors, as before. And those institutions turned out to be very skittish when it became clear that their funds were supporting potentially poor assets.

These are two—in the first instance, apparently different topics altogether. But they have one very important common feature. And that is that they deal with the big issue of under-pricing risk. In the housing markets, the problem arose primarily from the elements of explicit and implicit subsidization, which resulted in the buildup of large amounts of debt and overvalued house prices, and in which build-up of risk had not been properly priced into the markets.

In the case of systemic liquidity, something similar happens. There it's more a problem of being able to measure that risk, which is an issue that is being dealt with in our chapters. And the problem of it being so difficult to measure is also that it's always regarded as somebody else's problem. One can take care of one's own liquidity requirements, but the arrangements for the overall availability of credit within the system are much more reliant on the assistance by public authorities. Or at least, that's the way it is perceived. And they are not priced inappropriately by individual institutions.

Enough said on this by me. Let me turn to Laura Kodres, who has overseen the production of these two chapters. And she will take it from here. Laura?

MS. KODRES: Thanks very much, Jan.

As Jan emphasized, this GFSR covers two important and timely issues; systemic liquidity risk and housing finance. Let me just say a few words about the context for our work, and some of the main takeaways before turning it over to my colleagues.

In our chapter entitled, How to Address the Systemic Part of Liquidity Risk, we extend the work that we did in the last October GFSR. You may recall in that earlier chapter we focused on markets for funding financial institutions, and how they malfunctioned during the crisis.

Our policy advice was how to improve the infrastructure of trading in these markets, particularly the repurchase or repo market, and how to make sure that those institutions that are active in funding markets and look like banks are overseen like banks—like the money market mutual funds in the United States. This time we take a slightly different tack, and we're looking more towards the interface between institutions and markets.

During the crisis, we discovered that financial institutions were using risky methods to fund themselves using short-term, wholesale funding markets, using repos based on risky collateral, and generally not paying enough attention to what might happen if they all needed funding from these less-stable sources at the same time.

We had a bathtub full of liquidity, but it only took a few institutions to pull the plug and let the liquidity run out, drying up markets for everyone else. Many implicitly assumed—rightly, as it turned out—that central banks would step in and turn on the spigots.

So in this chapter we try to construct ways to gauge how much an institution contributes to this systemic liquidity risk. As a practical next step, policymakers could consider a so-called macroprudential tool, such as a surcharge, tax, or insurance premium based on our methods to give institutions an incentive to contribute less. Or, at least pay more for the privilege of drawing on the central bank during emergencies. This should help rebalance the burden between the private sector on one side and the central banks and governments on the other, for the costs associated with systemic liquidity risk.

As my colleague Jeanne Gobat will explain, we developed three approaches to measuring systemic liquidity risk. Each associated with the way of charging institutions for their contribution. These methods are quite flexible, and they could be applied to global institutions with cross-border funding needs.

We don't take a stand on which macroprudential tool would be best, whether it be a surcharge, an insurance premium, or a tax. In part because systemic liquidity risk needs to be looked at from multiple vantage points. And we don't know which one will ultimately be the most reliable.

Second, we are hampered in that we do not have exactly the right data about the institutions to be able to refine these measures. In some cases, one would need data typically only available to supervisors. And thus, we don't know yet which method will be cost-effective, as our numbers are still very preliminary. Our purpose, really, is to provide a “proof of concept” to generate more discussion about the potential ways of fixing this particularly vexing issue.

For our chapter on housing finance and financial stability, as the title suggest, our overriding question and our ultimate advice is to go back to basics. The chapter is not a study of housing cycles and their link to economies of various countries, but instead we take a hard look at what types of financing methods are used in an array of advanced and emerging economies. And from this, we tried to discern which ones add to, or subtract from, financial stability.

Of course, the motivation for this study is the large number of economies that have suffered housing busts in this latest crisis. And in particular, the United States, where securitization is blamed for many of the problems that it still faces.

I'll let Ann-Margret Westin provide the basic conclusions, but two results are striking. First, while most existing studies have shown credit growth leads to house price growth, we also find the reverse. House price increases beget credit growth. That is the relationship between credit growth and house price increases—it goes in both directions.

Second—and this one is a bit more surprising. On average, the more government involvement there is in the housing finance system, the more the run-up in house prices and the deeper the bust. The old adage, “the higher you go, the harder you fall”, is accentuated by government involvement—despite the laudable goal of helping some disenfranchised households own their own homes.

Before turning it over, let me just note that the housing finance systems across countries are incredibly varied. But no matter which way housing is financed—covered bonds, securitized products, or just old-fashioned on balance sheet loans—the depth of the bust is ultimately related to the quality of the underlying loans. Hence, the back to basics includes as a first step good underwriting standards for loans, a robust legal system that underpins real estate transactions, and basic risk management systems for those that ultimately hold real estate risks.

I am going to stop here and thank you very much. And then turn to Jeanne for a few more comments about this very tough concept of systemic liquidity risk.

MS. GOBAT: Thank you, Laura. The recent financial crisis suggests that extreme deteriorations in liquidity conditions in critical funding markets went under-recognized by the private and public sectors.

Financial institutions did not factor in how their own actions to liquidity shortfalls would have knock-on effects, and could make the entire financial system shut down and less stable. They underestimated their contribution to systemic liquidity risk during the good times, and during the bad times they did not bear the cost that their actions imposed on others.

A number of studies have attempted to propose various fixes to deal with the systemic nature of liquidity risk. Why is this important? Because lowering the incentives for financial institutions to take on excessive liquidity risk during the good times would, in turn, lower the support that central banks need to provide in times of systemic liquidity crisis—that is, in bad times.

However, most of the proposals on the table do not provide concrete advice on how to calculate an institution's contribution to systemic liquidity risk, nor how to figure out an appropriate charge associated with it. This is our contribution to the ongoing discussion. In this year's April GFSR, we have come up with a way to measure how an individual financial institution contributes to system-wide liquidity risk in a way that can be used to develop a so-called macroprudential tool to capture the cost an institution imposes on the rest of the system. For instance, this can be achieved by requiring that all financial firms that contribute to systemic liquidity risk buy insurance proportionate to the expected support they will receive by central banks in times of stress.

In the chapter, we propose three different approaches to measuring systemic liquidity risk. First, we develop a systemic liquidity risk index that captures breakdowns of various arbitrage relationships across markets and securities. This can be used to capture tightening of the market liquidity, and funding liquidity conditions.

When arbitrage relationships break down, investors have a hard time obtaining funding or borrowing to execute the necessary trades and to take advantage of the arbitrage opportunities, this showing up in wider spreads. We use this measure to calibrate a surcharge or an insurance premium, based on an individual institution's exposure to this systemic liquidity risk, identified in the index. This should be equivalent to the support, over time, that the firm receives from a central bank.

The second approach is a systemic risk adjusted liquidity model. It has the advantage of using daily market data and standard risk management techniques to calculate a market-based measure of liquidity risk. Using this as our base, we then calculate the joint probability that all institutions will simultaneously experience shortfalls in liquidity. With this overall measure, we can then calculate an individual institution's contribution to system-wide shortfalls in liquidity, and a surcharge based on it.

And finally, our third approach uses a macro stress testing model, which assesses the affects of an adverse macroeconomic or financial environment on the solvency of institutions and, in turn, on systemic liquidity risk. Essentially, we take a regular standard credit market risk-oriented stress testing framework, and layer on it responses to liquidity shortfalls. We can calculate the additional capital that banks would require to lower the chances of experiencing a liquidity run as a result of another bank's failure.

One interesting finding of our analysis is that we can show that the joint expected liquidity shortfall across institutions—that is, all institutions experiencing at the same time liquidity strains—was during the recent crisis much higher than if we would just add all expected losses from individual bank liquidity shortfalls. Our research generally underscores the importance of accounting for the interconnectedness of asset and funding markets, and of financial institutions and contributing to liquidity risk, building up in the financial system over time.

Going forward, our suggested models and others should be tested to see how they perform and whether a capital surcharge or an insurance premium more cost-effectively captures the contribution of an institution to systemic liquidity risk.

Ideally, the model framework should be expanded to include all non-bank financial institutions that contribute to systemic liquidity risk. This can include special investment vehicles, money market mutual funds, hedge funds, finance companies, and others. Much of this will depend on the structure of a financial system, and will vary from country to country.

In sum, while we don't purport to know exactly how to fix all the issues surrounding systemic liquidity risk, we suggested some interesting and targeted tools in this year's GFSR.

MS. WESTIN: Thank you, Jeanne. The housing finance chapter analyzes housing finance systems in a number of representative advanced and emerging economies in order to identify factors that enhance the stability of the housing finance systems and financial stability, more generally. It's important to note that the financial stability impact of housing market downturns is greater in some countries than in others. This is partly because of important differences in countries' housing finance systems, including the role of government.

Although most governments think that their housing finance systems are helping their citizens, this chapter shows that some aspects of housing finance systems in some advanced countries actually contributed to financial instability during the recent crisis.

Empirical analysis in the chapter, cross-countries and over time, show a close correlation between rapid mortgage credit growth and sharp increases in house prices. In fact, past mortgage credit growth, together with the state of the economy, explain the bulk of the variation across countries in the house price slowdown and the increase in bank loan losses in the recent crisis. As noted by Laura, “the faster you grow, the harder you fall”.

The chapter also examines the impact of a number of housing finance characteristics on mortgage credit and house prices. For example, government participation exacerbated house price swings and amplified mortgage credit growth in the run-up to the recent crisis, particularly in some advanced countries. And on average, countries with more government involvement experienced deeper house price declines. This might be a reflection of the lower cost of credit, thanks to government subsidies, and relaxed lending standards as the private sector was trying to compete with the government.

Moreover, higher loan-to-value ratios are significantly correlated with higher house price and credit growth over time in advanced countries, in line with the findings of many other studies. This means that in countries where you don't have to make a very large down payment in percentage terms, the impact on house price appreciation is much larger.

This effect disappears when emerging economies are included in the example, covering the most recent period. This might reflect the fact that unregulated sectors still tend to play an important role in the lending process in some of these countries. Credit originated in the unregulated sector will not be captured in formal lending data.

Three broad areas of best practices for stable housing finance systems emerge from the chapter. First, regulation and supervision of mortgage lenders needs to be much better. In the run-up to the crisis, relaxed lending standards and abundant liquidity led to an increase in mortgage credit growth. This, in turn, spurred the house price boom and bust.

Mortgage lenders need to go back to basics and take into account not only the value of the property, but also the credit worthiness of the borrower. Banks and other lending organizations need to better manage risks and the standards they apply when approving mortgages.

Second, there is a need for more careful calibration of government participation in housing finance to avoid contributing to financial instability. While most countries pursue policies to promote affordable housing, an excessive emphasis on home ownership rather than renting might contribute to financial instability. Rather, treating the tax issues of home ownership in the same way as renting, such as a reassessment of the ability to deduct the interest on your mortgage, would reduce the current bias towards owner-occupied housing.

Third, there is a need for better alignment of incentives in the private label securitization business, including in the servicing business, with those of investors. In the run-up to the crisis, private label residential mortgage securitization in the United States was associated with a deterioration in underwriting standards and incentive problems. Furthermore, as loans have become delinquent, servicers currently have little incentive to renegotiate loans even when they have the contractual ability to do so and when such a loan modification would maximize expected loan net present values.

The chapter discusses additional aspects of best practices that need to be considered by policymakers in emerging-market countries as they set up their housing finance systems. In particular these countries should first focus on developing solid regulation and oversight for all organizations originating loans to help ensure good underwriting standards. Credit bureaus, as well as consumer-protection schemes that help educate consumers as to the nature and risks of mortgage products, are also important.

Finally, using best practices as a guide, the chapter makes specific recommendations on the housing finance system in the United States. This system remains unique in many ways and an overhaul is needed. In this context, the U.S. administration's recently released Housing Finance Reform Proposal is a welcome step in the right direction.

Reform of the U.S. housing finance system should address current gaps in the regulatory, supervisory and consumer-protection frameworks. It should aim for better defined and more transparent government involvement in the housing market, showing relevant items on the government's budget. It should also reconsider the role of housing-relating government-sponsored enterprises given the need to create a more level playing field in mortgage markets.

In the foreseeable future there seems to be continued need for government guarantees for securitized mortgages in the United States. However, any such guarantee should be explicit and fully accounted for on the government's balance sheet. Over the medium term and with appropriate forms to encourage safe securitization of residential mortgages by private entities, Fannie Mae and Freddie Mac should be wound down to make room for the private sector to reemerge. Over the long run, such reforms, phased in carefully, would have a significant positive impact on the U.S. financial system and would help bolster global financial stability.

QUESTIONER: I won't give you all my questions at once. On mortgage interest deductions, you indicate that the U.S. government should reexamine this deduction. Aren't you saying that you think they should cut it? It exists right now so the only other reexamination is whether you increase it, keep it the same or cut it and you're not talking about increasing it or keeping it the same.

MS. WESTIN: First just to note that the current proposal by the U.S. Treasury does not even mention the mortgage interest deductibility and I think that is our major concern with the proposal. The mortgage interest deductibility is expensive and it is regressive. Most countries have some sort of mortgage interest deductibility. They also have deductibility of capital gains taxation when you sell your own home while they don't tax imputed rent. So at least for a first step we would very much recommend that you at least cap the mortgage interest deductibility. There is a proposal by the U.S. Fiscal Commission to lower the limit from 1 million dollars to 500,000 and to let it only apply to primary residences. That would be a very welcome first step.

MS. KODRES: Let me add that in this country, particularly where this deduction is probably one of the largest deductions on most households' tax forms, we need to do this in a slow fashion. Many people have bought homes on the assumption that this tax deductibility is present and so it's not the kind of thing that's going to be here today and gone tomorrow. One would need to phase this in over a fairly long period of time. But I think the direction is clear that we would like to see it reduced and we'd like to see a more tax level playing field in all countries but in particular this one because of the way that the tax deductibility works.

QUESTIONER: I wonder if you could comment on the particular case of Spain. As you know, Spain is going through one of the biggest crises in its history and it's pretty much related to the crisis in the housing market. I wonder if you could tell us what do you think what are the characteristics of the Spanish housing market or if it was government intervention that made the situation so bad. And what do you think of the response of the Spanish government? Are they doing the right things or are there are some things that they could be better? Thank you.

MS. WESTIN: Let's just first note that the only country that we give specific recommendations to is the United States given that the crisis in part had its origins there and there is a real need for an overhaul. We do not go into detail for specific country recommendations.

Let me just note that in Spain there was quite an emphasis on floating-rate loans which could have contributed somehow to the volatility in the mortgage performance. At the same time it should be noted that the fact that there is an abundance of floating-rate loans has helped Spain right now in the current crisis with historically low interest rates. Other than that I would say that our main message of better underwriting practices, making sure that there are good-quality mortgages, no subprime, not too high loan-to-value ratios also apply to the Spanish situation.

MS. KODRES: Let me give your attention to a table in the chapter, Table 3.5, where we have a number of different ways in which the government participates in housing markets and you can do a cross-country comparison to look at specific countries. Spain is not the highest in the advanced countries by any means in terms of its government participation in the housing market. Therefore, one would not want to blame the government for the problems that the housing market has faced.

It's also probably worth mentioning that Spain is one of the few countries that maintain what we call dynamic provisioning on the banks' balance sheets which means that they have to put away money in advance of potential loan losses. So at the start of the crisis the buffers in their banking system were actually a bit higher than other countries. So despite the fact it has been fairly damaging, it would have been even worse had that not been the case.

MS. LOTZE: I have a question here online which is also about the housing chapter and I want to take that since we're talking about that chapter:

"The study mentions property derivatives as a hedging mechanism for investors or households. Do these pose dangers? Do you think that households could end up with expensive products that they don't understand that increase their risk?"

MS. KODRES: Let me take that one at least at the top level. In general, individuals who engage in derivatives activity should know exactly what they're doing. Derivatives can be complex. They need not be, but they can be complex and one needs to understand exactly what you're hedging and exactly how that works before you take on those kinds of transactions.

The message of the chapter is to go back to basics and have simpler products. Simpler products will be easier to hedge. They'll be easier to hedge from the point of view of investors and potentially households. I would suspect, though, that households would have difficulty using derivatives because of the nature of the products at the moment. It's difficult to find something that's small enough for an individual household.

Note that in the chapter we also talk about the extent of foreign-currency loans in some countries that caused or at least were an exacerbating factor for some of their housing run-ups--in particular in Eastern Europe where the use foreign-currency loans was widespread. Again this is the type of contract that because of that risk, one would want to try to hedge it. That's very difficult for households unless they are the recipient of the foreign currency say through remittances or other types of flows.

So overall we would suggest that the mortgage contact itself should be simple, basic and understood easily. Then to the extent that one can hedge that, also with a simple derivative, that can be helpful. But we wouldn't advise individuals to move in that direction unless they know exactly what they're doing.

MS. LOTZE: Very quickly, there was another question online that follows onto the Spain one for recommendations on Ireland, "Can you offer some country-specific recommendations?" but I think, Ann-Margret, you already answered that unless you want to add something to that.

MS. WESTIN: No. Again I don't think we want to go into the details for individual countries. We're trying to stay on the higher level of broad practices.

QUESTIONER: On the housing thing really quickly, you talked about moving away from more homeownership to renting. Doesn't that have a social consequence. I know from my own experience that landlords are less likely to take greater care of their property than perhaps homeowners and then there are consequences that spill over. That may be for another time.

On the liquidity section, are there existing dangers right now to the global financial system given that there is no adequate cushion for systemic liquidity? Two, you seem to indicate that once this measurement of systemic liquidity has been applied that you will cross a hurdle to reluctance to apply a financial tax. Am I reading that correctly? Are you saying that then the G-20 will come leaping and bounding toward the IMF to embrace a financial tax?

MS. GOBAT: Regarding the first question, typically when we look at contractual risks, those are dealt with more in Chapter 1 which will be discussed next week.

Then regarding whatever macroprudential tool we would suggest, we're not suggesting a tax. It could be a charge or an insurance premium. I think further work is needed. I think first of all we need to develop a robust methodology for capturing all financial institutions that contribute to systemic risk and we used publicly available data.

What we'd like to do is supplement that also with supervisory data and capture better the interconnectedness of the institutions so we don't have their exposure to other counterparties. I think first of all we would need to fine-tune the methodology to back-test it to see how it performed before thinking of introducing a macroprudential tool and then we would have to see which one of the tools would be most cost-effective in addressing the risks so that I think further work is needed. These are just proposed methodologies and we would suggest that regulators would supplement it with their own data and work and fine-tune the methodologies.

MS. KODRES: What has happened so far is that the discussion has focused on systemic solvency risk and not systemic liquidity risk so that most of the discussions are about charging for the prospect that one firm would have a deleterious effect on other firms in terms of their failures. That it is really focused on solvency. What we notice in our chapter and what all of our methods show is that the higher the capital base, the more solvent you are, the less likely it is that you're going to need a liquidity charge.

So they move in opposite directions in the sense that the higher the capital, the less likely you are to face a run, therefore the less the systemic liquidity charge would be. So in taking into account these two items one would have to make sure that one takes into account that they interrelate and capture both types of risks so that you're not overcharging, which might happen if you were charging for each type of risk individually.

MR. BROCKMEIJER: I would want to endorse what Laura said. She pretty much caught my point. You were wondering whether there was an insufficient buffer to be able to absorb systemic liquidity risk. I think our main point is that you have to look at this in a broader context as Laura pointed out and there are very welcome initiatives underway at the moment to raise buffers and capital cushions more generally for in particular the banking system although our concerns in pointing out systemic liquidity risk go broader than only the banks. But there are overall initiatives and Basel III is one of them to strengthen the capital cushions of institutions. That will also help from a systemic liquidity point of view.

I think the main advantage of trying to pinpoint the contribution of individual institutions to particular sources of risk, in this case the particularly difficult one to identify, systemic risk, is that the closer you can like an institution to its contribution to that risk, you can then charge that institution for the contribution to the risk and as a result fine-tune much better the incentives, that rather than putting an overall charge on everybody, you target it on those entities that are the most important contributors to the problem. Thank you.

QUESTIONER: Just one follow-up. You talked about that in order to have accurate measurement tools that you need greater data. You mentioned access to supervisory data. The IMF has been moving toward greater monitoring of financial markets and capital markets in general. Are you recommending that the IMF then be able to have access and pool that supervisory data or are you recommending that these tools be given at the national level and let them deal with it?

MS. KODRES: That's a very important question and it's a very sensitive one. I would first of all note that what we would want to make sure of, in the first instance, is that regulators and supervisors have the information that they need to do their jobs, and we found that even in the case dealing with systemic liquidity, they may not yet have all of the information that they need to appropriately evaluate some of these new measures and exactly how they might operate. So that would be the first approach is to make sure that the people doing the supervisory job have the right information to do their job.

In terms of what the IMF should receive, we are a global overseer, if you will, of systemic types of risks and we would prefer to have more data to do our job appropriately. Exactly what data and exactly what level of granularity is under discussion, and there are concerns about confidentiality particularly with bank information that has very institution-to-institution types of characteristics, so we are working closely with the BIS and the FSB in the context of the Data Gaps Working Group. Those negotiations are underway. There is no formal decision yet about what types of access the IMF will be entitled to. But I would note that our interest is not necessarily on individual institutions and their closure or their intervention or their recapitalization, but our interest is in looking at the bigger picture and seeing where risks reside and where they can be transmitted and where spillovers are likely to occur, so that kind of information is the information we need to informs ourselves about those kinds of risks.

In terms of information that might be publicly available, which is yet a different level, we would suggest that in order for individual investors, be they bond investors or shareholders, in order to evaluate the systemic or even the nonsystemic liquidity risk of an institution, they too would require more information.

The typical information that's disclosed in annual reports is not very granular. It doesn't allow investors to see the maturity mismatch of an institution. In many cases the liability side of the balance sheet is divided into “less than one year” and “greater than one year” which isn't sufficient to be able to see exactly the kinds of liquidity risks. And usually the discussion of liquidity risk management in these types of reports is pretty attenuated especially relative to the discussion of credit risk or market risk. So we would suggest that in fact more information on this would be valuable so that investors and even depositors would have a better idea of what types of liquidity risks their institutions are undertaking.

MS. LOTZE: We have no more questions online. Are there any more questions in the room? Do you want to conclude with a few remarks, Laura?

MS. KODRES: I would like to make sure that we understand a couple of things about Chapter 2 and a couple of things about Chapter 3.

On Chapter 2, the methods that we're proposing are in fact really “proof of concept”. We're not proposing that these are the right measures necessarily. The numbers that we have in there are illustrative. They're not numbers that one could sit down and make a charge for any individual countries or the vast majority of countries. So we want people to understand that what we were attempting to do in that chapter is put more on the table for discussion and what we found lacking in the discussion heretofore is the concept of linking the systemic liquidity (which people have discussed and people now have an understanding about) exactly to how one would go the next step of trying to make sure that that's priced properly in the markets. That is to assess an individual institution for their contribution to that risk. So that's where we have, I think, added some value at least in terms of he discussion.

On Chapter 3 we'd like to make sure we come across with a chapter has a very cross-country approach. Even though the U.S. features prominently in it, in part because its mortgage markets were at the heart of the crisis, I think it's useful to take a step back and observe that many countries have differing ways of financing housing. Some of those are quite stable. We have a nice box in there on the Danish mortgage market which has a very nice system in terms of making sure that risks are transferred appropriately and conscientiously. And we have a number of cross-country studies in there which I think give rise to a much more deep and broad understanding of how housing finance takes place and we're hopeful that many emerging markets look closely at that chapter as a way of using it as an example of what to do and what not to do in forming their own housing finance systems.

MS. LOTZE: Thank you very much. Let me very quickly announce a few things. First of all, for the broadcasters there are news clips available on these two chapters and on this event on News Market and Unifeed. Then the GFSR Chapter 1 will be presented next week on Wednesday, April 13, at 9:00 a.m. here at the Fund. The WEO analytical chapters will be launched tomorrow at 9:30, same time, same place. Thank you very much for coming.