Early Lessons from Recent Financial Turmoil
South Shore Chamber of Commerce Quincy, MA
It is a pleasure to be with you today.
My goal is to share with you what I see as some "early lessons" from the ongoing period of turmoil in real estate and credit markets.1 It is important, given the turbulence we have lately experienced, to begin to analyze and act on such lessons as early as possible.
First, some background. For the past two years, the housing market has gone from bad to worse. What began as a mild reduction in residential investment2 at the start of 2006 has accelerated, with residential investment declining by more than 20 percent during each of the last two quarters of 2007.
Beyond the obvious and intense pain for distressed borrowers, a noteworthy aspect of this downturn in residential investment has been its impact on financial markets. Previous housing downturns caused problems for savings institutions and banks that were generally smaller than the large financial institutions being affected today. This downturn has primarily affected financial markets and large financial institutions. Large financial institutions have lost billions of dollars, and in some cases tens of billions; and some are forecasting significant additional losses.3
But this is not just a Wall Street problem. Our real concern is the ways those problems in financial markets translate back to all of us.
After a brief background on the genesis of the current financial turmoil, I am going to discuss three issues that I am sure you are increasingly seeing reported on the business pages of your national and regional newspapers:
- First I will discuss why the accuracy of ratings on securitized mortgage products have been so much less reliable than ratings for corporate-debt securities, and the lessons we should draw from those differences. Here is a peek at the main message: uncertainty surrounding ratings has caused a variety of markets to become less liquid. Less liquid credit markets can hurt borrowers of every type. Problems with the accuracy of ratings had their roots in the determinants of the ratings – and so we need ways to differentiate ratings that have different “drivers” in terms of the credit risk on the underlying assets. We need for example, to differentiate ratings on assets like corporate securities from ratings on assets whose ratings histories and price-drivers may be quite different, and less well understood, like certain mortgages-related securities.
- Second, I will touch on the difficulty in pricing complex financial instruments, and actions that might help with transparency and pricing of these products in the future. Again, the main message: As I see it, the recent difficulty in pricing assets should make investors and financial intermediaries consider whether such complexity is necessary, and whether some of these instruments should be more standardized or possibly moved from dealer markets to exchange-traded instruments.
- Third, I will discuss the continuing problems in the housing market, and touch on the ongoing discussion of policies that might help mitigate those problems. Here I’ll suggest that, if housing prices continue to fall, we will need to increasingly consider programs for those with negative as well as positive equity in their houses.
Background On Recent Financial Turmoil
First, a bit of background. In early 2007 it became apparent that delinquencies on subprime4 mortgages issued after 2004 were experiencing problems at a more rapid rate than would be expected, given the rather benign economic environment. We began to see an elevated rate of subprime delinquencies [See Figure 1].
In July of 2007, rating agencies began to highlight the fact that subprime mortgages that had been securitized - a process I'll describe in a moment - were performing poorly, leading to downgrades for securities that had significant exposure to the subprime market. Just as investors became more uncertain about valuing these securities, it became clear that subprime mortgages were part of various financial instruments that had been viewed as generally low risk.
Securitization relied on the reasonable premise that subprime loans might be more risky than prime, but the majority would not default - and higher interest rates and fees would compensate for those that did. Subprime loans were bundled for investors and "riskiness" was tiered. Investors in the least risky tiers were thought to be well protected from losses.5 [See Figure 2] Unfortunately, underlying assumptions proved inaccurate.
The market for short-term asset-backed commercial paper (ABCP), short-term securities used to finance a variety of loans from student loans to home equity, has been particularly impacted [See Figure 3]. As problems with mortgage-related loans emerged, some investors became reluctant to continue lending in the ABCP market. This reduction in the availability of short-term funds caused the rates on ABCP to rise; and also forced some financial institutions to buy back ABCP that they could no longer refinance, bringing it onto their balance sheets. The combination of uncertainty over the appropriate rating of mortgage-related securities and the expansion of bank balance sheets caused significant pressure on the availability of short-term credit. In addition banks, as liquidity providers, were expanding their balance sheets in other areas, much of which was not anticipated prior to the financial turmoil. Some banks have had to take write-downs on some assets, and the losses in combination with involuntary growth in assets have made some banks more reticent to expand their balance sheets further.
An indication of the difficulties in short-term financing markets was the marked elevation of the London Interbank Offered Rate (LIBOR) relative to the comparable U.S. Fed Funds target rate. LIBOR is the rate charged in a key international market for short-term lending between banks. The elevated LIBOR rate not only made it difficult for banks to borrow short-term, but also raised the rates on loans tied to LIBOR -- notably including most subprime mortgages, corporate loans, and credit card debt.
Indeed, while the initial trigger for the financial turbulence was related to subprime mortgages, the uncertainty surrounding ratings of complicated financial instruments has caused disruptions in a variety of other assets and markets that depend on securitization - including state and municipal financing, student loans, and commercial real estate. The effects have been felt up and down Main Street, as well as in some markets overseas.6
I. The Role Of Credit Ratings
With that as background, I'd like to turn to my first subject today - the role played by the credit ratings used by investors to gauge the credit risk on securities.
The use of credit ratings for corporate bonds has been longstanding and generally free from substantial concerns about the ratings' accuracy, with a few notable exceptions.7 So I think it is useful to consider the similarities and differences between rating corporate securities and mortgage securities, where the accuracy of ratings has been called into question. This is particularly important because much of the current turmoil is driven by investor uncertainty in rating difficult-to-value financial assets.
Relative to mortgage securities, ratings on corporate securities have been time tested. Downgrades have been quite modest. Defaults have often been tied to recessions or problems specific to a given firm, such as excessive leverage or poor management [See Figure 4]. In addition, investors in corporate securities can rely on a wealth of external sources to verify the health of a firm.8
Corporate ratings have generally performed well. Despite the recent financial turmoil, corporate default probabilities in aggregate remain quite low [See Figure 5]. With relatively few downgrades, and delinquencies, the corporate balance sheet has held up better than that of the consumer - although the recent widening of spreads for lower-grade bonds gives one pause [See Figure 6].
The mortgage market is quite different. Most mortgage securities are based on a diversified pool of underlying mortgages. Many investors assumed that there would be significant benefits from diversification, that regional real estate shocks would be dominated by local factors, and that national home prices were very unlikely to decline. These assumptions proved to be wrong, resulting in widespread downgrades of mortgage securities.
The housing price assumption in particular has been critical, and many investors may have significantly underestimated a potential national housing price decline and its effect on defaults. The S,P Case-Shiller national home price index fell 10 percent from its peak in the second quarter of 2006 through the fourth quarter of 2007. Home prices are down in every one of the 20 large metro areas covered by the Case-Shiller national home price index.
In contrast to corporate securities, corroborating information on mortgage securities is not as readily available. There is no equivalent to equity analysts and equity prices to give investors updated market information. The information needed to analyze the individual mortgages in the pool can be expensive to obtain. So investors are more reliant on rating agencies than they are with corporate securities.
The problems in the mortgage market highlight the need for caution where there has been limited ratings history, where the underlying characteristics that drive the asset's price may not be fully understood or anticipated, and where evaluations cannot be easily corroborated by others such as equity analysts.9 Certainly one way to highlight these differences is to differentiate ratings on corporate securities from ratings on assets like mortgage-backed securities.
II. Transparency and Disclosure
Now I'd like to turn to a second area, transparency and disclosure. The heightened uncertainty surrounding ratings has been aggravated by the lack of transparency in the pricing of complex financial instruments. Because as many of the more complicated financial instruments have ceased active trading, determining a market price has become quite difficult.
While defaults in housing move rather slowly, the pricing of financial assets has moved much more dramatically, causing many financial institutions to significantly change their expectation of losses from these complex financial instruments. In fact, the highest-graded securities are selling at a very significant discount [See Figure 7], implying a significant risk premium for holding even the highest quality securities of some complex financial instruments.
The opaqueness in pricing has caused a variety of complications:
- First, many securities have become illiquid, and sales are only occurring at “distress” pricing.10
- Second, the uncertainty in pricing has caused a loss of confidence in firms that provided insurance on high-graded securities. This has been a problem for certain municipal securities, and instruments known as auction-rate securities11 and variable-rate demand notes12 where uncertainty in bond insurance has led to changes in pricing for securities previously thought to have little risk of default.
- Third, banks often provide liquidity backstops should securities trading become illiquid, and the current market turmoil is leading banks to be more cautious in lending because they are not sure if or when investor demand could evaporate, requiring significant extensions of bank credit.
- Fourth, certain markets have been disrupted, as investors avoid complex securities where liquidity could be a problem. The recent reports of problems in auction rate securities, suggest that financing for all sorts of entities – universities or hospitals or some municipalities – can get a lot more expensive when investors lose confidence in the pricing and liquidity of financial instruments.
So, what lesson should we derive from some of these complications?
First, some financial products were not well designed to withstand liquidity problems. To avoid paying banks fees to provide a liquidity backstop, many financial products of recent vintage included provisions to force liquidation when necessary to insure payment to the holders of the higher-graded securities (or slices of securities). This structure was used, for example, by structured investment vehicles (SIVs)13. However, due to the recent financial stress, assets of SIVs could not be liquidated at prices felt to be reasonable. Broadly speaking, products should be structured to better weather periods of illiquidity, and ratings models should take better account of liquidity risk.
A second way to improve price discovery would be to have greater uniformity in financial products. Standardization of products makes it much easier to price and trade securities. A case in point: many of the positive innovations in mortgage markets resulted from more uniform standards for conforming loans. Standardization helped insure minimum underwriting standards.14 Non-conforming mortgage markets, as well as other securitized assets, might well benefit from greater uniformity and standardization so that pricing is less idiosyncratic to the particular security an investor holds.15
A third possibility is to seek more trading of financial products in exchanges rather than through dealers. Securities that are consistent enough to trade on an exchange are more likely to have market prices that all participants can use.16
Finally, investors should give careful consideration to whether such complex financial products are necessary at all. With simpler and more understandable structures, the difficulties in obtaining market prices are likely to be significantly reduced, as are the consequent uncertainties like those we are currently facing.
III. The Housing Market
Now I'd like to turn to the continuing problems in the housing market, and a few thoughts on policies might help mitigate those problems.
The housing market was the genesis of current financial turbulence, and a key point is that significant further declines in home prices could greatly complicate efforts to resolve current problems. As housing prices fall, loan-to-value ratios will rise, in some cases exceeding 100 percent, reducing the number of borrowers that qualify for existing government programs like Federal Housing Administration (FHA) loans.17 Thus when considering ways to mitigate the current housing problem, it is useful to consider borrowers that still have positive equity in their house, as well as those that do not.
Fundamentally, I encourage worried borrowers who hold high-rate loans to approach a responsible lender about refinancing. I also encourage lenders to reach out to borrowers, and to take a fresh look at the state and federal programs that can be of assistance.
In recent months we estimated that a fair number of borrowers with subprime mortgages may be able to refinance into a more affordable loan, because they had good credit scores and some home equity when they got a fully documented loan on an owner-occupied property. As time goes by, however, declining home prices are eroding borrowers' equity, and some are experiencing financial difficulties or mounting debt as the economy slows. These forces complicate the picture, and narrow the pool of readily "refinanceable" subprime loans.
Given the worsening housing scene, Boston Fed researchers recently updated prior work in this area in an attempt to calculate the share of people with subprime loans who might qualify for FHA programs. They looked at subprime loans in Rhode Island, Connecticut, and Massachusetts that had full documentation, were owner occupied, had a loan-to-value (LTV) of no greater than 97 percent, met current FHA loan-size limits by county, were never 60 days delinquent, and had a maximum debt-to-income ratio (including other forms of debt) of 45 percent18. This approximates current FHA standards.
They estimate that about 16 percent of borrowers with subprime loans would meet those criteria. However, over time this pool is likely to shrink, as the number of delinquent borrowers has been rising and housing prices have been falling, likely reducing the number of qualified borrowers. For borrowers that could convert their subprime loan to an FHA-insured loan, saving significant money from converting from a subprime to a prime rate is an alternative that should be considered.
One refinance option is the Mortgage Relief Fund (www.MortgageReliefFund.com). Five large banks joined forces to set up this program, with the encouragement of the Federal Reserve Bank of Boston, to join forces in reaching out to borrowers with high-rate loans. The banks can help borrowers explore refinancing into a more-affordable loan - maybe an FHA loan, a state-guaranteed loan, or a conventional loan. We believe a number of community banks will join the effort in the coming months.
With two months completed, and realizing the lags from first contact on a mortgage to closing, the banks have logged over 1,000 inquiries, taken in more than 50 applications, made 115 referrals to nonprofit housing counseling services, and are now starting to close some of those loans from the first applicants - with the first dozen loans having recently closed. It is a modest start, but this month we are pursuing a second wave of outreach, and the banks are pursuing an additional advertising push, especially in areas with higher concentrations of subprime loans. We see the effort as a marathon, not a sprint, and we are refining and adding to it as we go.
As I noted a moment ago, as delinquencies and home prices shrink the pool of potential borrowers in existing Federal and state programs, an important consideration for lenders and policymakers involves the situations of borrowers whose loan now exceeds the value of their house. Of course, any remedies need to take into account the future risk to taxpayers, and the incentives created for borrowers and lenders. But we need to weigh that against the problems for communities that can occur with widespread foreclosures - including the negative effects on neighboring homeowners, the burden on tenants, and the costs borne by municipalities and communities in addressing blight and crime in areas of clustered foreclosures.
As Federal Reserve Board Chairman Ben Bernanke said on Tuesday,19 this situation calls for a vigorous response - but care must be taken in designing solutions, so they represent safety and soundness for lenders and are characterized by fairness and minimal "moral hazard." I would note that foreclosure costs are often substantial to lenders (and by extension, I would add investors) - of course it goes without saying that they are painfully costly on many levels for borrowers. So there is considerable scope for negotiating a mutually beneficial outcome.
Somewhat along those lines, some parties are proposing variants of a "shared appreciation" loan approach.20 Lenders could write down the loan amount to the current home value, cap losses, avoid the costs associated with foreclosure, and receive a share of any future home appreciation when the buyer sells. The borrower could avoid foreclosure and reduce monthly payments.21 The FHA could provide insurance, but defray the increased risk with a share of the gains when the homeowner sells. Approaches like this, and other worthy ideas that are being proposed, should be debated by policymakers and interested parties - but without delay.
As I noted at the outset, the current financial turmoil is ongoing but it is not too soon to consider lessons learned.
The uncertainty surrounding ratings has caused a variety of financial markets to become illiquid and caused very significant write-offs at major financial institutions. Considering ways to differentiate ratings on assets like corporate securities from ratings on assets whose ratings histories and price-drivers may be quite different, and less well understood, is probably a first important step.
The difficulty in pricing assets should make investors consider whether such complexity is necessary, and whether some of these instruments should be more standardized or possibly moved from dealer markets to exchange-traded instruments.
In the housing area, thought will likely be needed regarding programs for those with negative as well as positive equity in their houses. As long as housing prices continue to fall, the decline increases the risks to borrowers, lenders, markets and the economy.
I thank you for exploring with me today these problems that have roiled Wall Street, and are beginning to significantly affect Main Street. We are facing some unique and complex challenges in all of these areas. Let me leave you with the thought, however, that there may be a significant cost to delaying needed actions that could restore confidence in the ratings process, the pricing of financial assets, and the impact of declining house prices.