Statement by Richard Spillenkothen, Director, Division of Banking Supervision and Regulation, Board of Governors of the Federal Reserve System, and Donald H. Wilson, Financial Markets Officer, Federal Reserve Bank of Chicago, before the Subcommittee on Policy, Research, and Insurance of the Committee on Banking, Finance and Urban Affairs, U.S. House of Representatives, May 6, 1992.
I am happy to be here today to discuss the securitization of commercial real estate loans and the related questions you raised in your letter of invitation, and I am most pleased that Mr. Donald Wilson of the Federal Reserve Bank of Chicago is with me. Mr. Wilson has worked extensively in the area of securitization and stands ready to help answer any questions you may have after my presentation.
To provide perspective, I will begin by reviewing the evolution and current state of the general asset-backed securities market. This review will be kept relatively brief because our colleague, Mr. Franklin Dreyer, of the Federal Reserve Bank of Chicago, presented an extended discussion on securitization when he appeared before this committee last summer.
In the beginning, government-guaranteed mortgages were placed in pools, and securities were issued that entitled the holders to the proceeds of the principal and interest payments that flowed from these mortgages. These securities were also guaranteed by the government. Nongovernment-guaranteed residential mortgages were also securitized, and most of the these securities were backed by government'sponsored agencies. Today more than $1 trillion worth of government-related mortgaged-backed securities are outstanding. Moreover, in the past decade or so the process of securitization has been extended to nonmortgage consumer loans such as automobile loans and credit card receivables. Securities that are backed by these assets are not guaranteed by the government but generally carry a credit enhancement provided by the private sector, and those securities account for a significant portion of the entire market. Furthermore, as I will discuss shortly, a limited volume of commercial real estate loans have been securitized in recent years without the benefit of government enhancement.
The range of terms available on securitized mortgage assets has also been greatly expanded over time. In particular, the innovations of collateralized mortgage obligations (CMOs) and real estate mortgage investment conduits (REMICs) have resulted in securities that provide different claims and priorities on the principal and interest payments made on underlying loans, thus accommodating the various needs and preferences of investors.
This very impressive growth and development of securitized loans has occurred because of the benefits they provide to both issuers and investors. Originators of loans that sponsor assetbacked securities benefit from improved liquidity, enhanced fee income, and, to the extent that assets are removed from their balance sheets, less need for capital. Investors, on the other hand, acquire securities that require no management of the underlying loans on their part and yet provide an attractive return for instruments that pose, depending upon the nature of the credit enhancement, little or no credit risk. Furthermore, as I have noted, these securities increasingly have been structured to meet varying investor preferences for safety and predictability of cash flow and variability of values in relation to changes in interest rates. The importance of the safety and predictability of cash flow deserves to be given special emphasis because, to date at least, these qualities have been primarily responsible for the wide and deep attraction that assetbacked securities have had for various groups of investors.
Given the benefits provided by asset-backed securities, the Federal Reserve and other agencies that supervise insured depository institutions have viewed their development as very salutary. Moreover, the participation of depository institutions in this market as both issuers of, and investors in, these securities has been considered altogether appropriate, provided that their participation is conducted in a safe and sound manner. To foster that end, we have issued guidance on how banking organizations should carry out securitization activities and given instructions to our examiners on how to review these activities for purposes of safety and soundness. We have also issued guidance on, and instructed our examiners to review, the investments of banking organizations in assetbacked securities. A principal concern in this regard is the interest rate risk that is inherent in some of the more complex tranches of securities issued under CMO and REMIC arrangements.
Finally, a key concern of our supervisory efforts has been to insure that the exposure to credit risk is properly reflected in capital positions of institutions that sell loans into pools with recourse or that provide guarantees of assetbacked securities. The supervisory agencies are currently reviewing their common policies pertaining to these capital requirements under the auspices of the Federal Financial Institutions Examination Council, and I shall say more about this effort toward the end of my remarks.
With that background now presented, it is appropriate to turn to the focus of this hearing, which is the securitization of commercial real estate loans. As I previously noted, to date there has been very little securitization of commercial real estate loans and it is instructive to ask why that has been the case. Our basic answer is that, because of the nature and quality of these loans, the market generally has not been able to design a securitized product that meets the requirements of investors without imposing unacceptable costs on issuers.
Commercial real estate loans are relatively heterogeneous in nature. Such loans are made to finance a wide range of building projects and properties, including strip shopping centers, office buildings, hotels, and various industrial plants and parks. Because of this diversity in property use and because of the circumstances affecting different borrowers, the loans are made with considerably different maturities and repayment terms. This heterogeneity greatly complicates the process of predicting the future cash flows that even pools of the highest quality loans will produce. Without that predictability it is doubtful that demand for securities based on these pools will be broad. Given the innovations in the market in recent years, this aspect of the heterogeneity problem--uncertainty in cash flow--may possibly be amenable to resolution by firms that are expert in designing securitized products.
The lack of homogeneity among commercial real estate loans in terms of the credit quality of borrowers and of other characteristics that can affect the certainty of repayment is a much more formidable problem. This problem greatly complicates the task of performing due diligence on a loan pool and reaching judgments on the overall quality of a pool. A typical pool of securitized consumer loans, for example, consists of a large number of small loans, made on very similar terms to borrowers who have all met or exceeded pre-established and time-tested credit standards. Thus, these pools readily lend themselves to statistical assessments of loss expectation with a high degree of predictability. Such refined actuarial methods of loss estimation simply cannot be applied to pools composed of a much smaller number of large and diverse commercial real estate loans. This characteristic and the market's perception (based on the dramatic experience of recent years) of the riskiness of commercial real estate loans appear to be primarily working to impede the securitization of these loans and the development of a secondary market for securities that are backed by them. Indeed, most of the relatively small number of securitizations of commercial real estate loans that have been done so far, until the recent offerings of the Resolution Trust Corporation (RTC), have been privately placed, and there has been no active secondary market for any securities that are backed by commercial real estate loans.
We understand that the recent offerings of the RTC were generally well received by the market, and dealers are reported to be prepared to make an active market for them. The RTC has indicated, moreover, that it intends to make similar offerings in the future, and this securitization activity may serve as a basis for the development of a broader market for securitized commercial real estate loans.
Whether that proves to be the case, however, remains to be seen, for the key to the success of the recently completed offering was the very sizable credit enhancements provided by the RTC. These enhancements included setting aside liquid assets to cover possible losses of approximately 30 percent of the value of the loans in the pool, as well as structuring each securitization into three tranches, with two of them subordinated so that the third, senior tranche was provided additional protection against credit risk. Altogether these enhancements provide holders of the senior tranches of securities, which received AAA ratings, with protection against loss on the pool of approximately 45 percent of its value. Market concern over the quality of commercial real estate loans held by the RTC may account for some important part of the exceptional size of this enhancement, but it seems clear that even better-grade commercial real estate loans will require considerable credit enhancement, given the general characteristics of these loans and the experience of recent years. For example, a recent proposal, reviewed by Board staff, to pool a combination of performing and nonperforming commercial real estate loans would have had the seller provide credit enhancement by retaining a subordinated interest in excess of 40 percent.
Thus, an ever-present tradeoff exists between the nature and quality of loans to be securitized and the level, and hence the cost, of enhancement that is necessary to make a securitized product marketable. Obviously, this tradeoff is one of the most important determinants of the viability of a securitized product, and the substantial credit enhancement demanded for commercial real estate loan securitization has, to date, been a principal obstacle to the issuance of such securities and the development of a secondary market for them.
With respect to the question of whether supervisors would have concerns about insured depositories participating in such securitization activity, either as investors or issuers, our view is that it is acceptable if it is done in a safe and sound manner. We would, of course, expect any such securitized loans purchased by insured depository institutions to be sufficiently protected from the riskiness of the underlying pool of loans so that they are of investment grade. As for participation in the issuance of these securities, we would expect institutions to structure these transactions properly to ensure satisfactory performance of contract terms and at the same time to achieve that structuring without incurring excessive costs for credit enhancement and other expenses. If these conditions are met, the participation of depository institutions in securitized commercial real estate loans, either as investors or issuers, will provide the same benefits to depository institutions that other securitized products have provided to them.
Let me now turn to our policy pertaining to the maintenance of capital when assets are sold with recourse (that is, the seller agrees, under one of several possible arrangements, to cover some or all losses on the assets). Our policy is predicated upon the principle that a government-insured depository institution should hold capital that is commensurate with its risk exposure. That principle translated into our current policy on recourse and other credit enhancements is that generally banking organizations selling assets with any recourse must maintain capital as if they owned the assets.(1) This reflects the long-held view of the agencies that when an institution's obligation under a recourse agreement exceeds the normal loss expectation on the underlying assets, the institution is exposed to most, if not all, of the risks normally associated with owning those assets.
We emphasize that this policy has not precluded banking organizations from engaging in the securitization process. For example, in the case of consumer loans, banking organizations have taken the option of selling their loans without recourse, in which case there is no capital requirement to the selling banking organization. This has been accomplished by paying a third party to take the credit risk or by using a spread account, which is an escrow account whose funds are derived from a portion of the spread between the interest earned on the assets in the underlying pool and the lower interest paid on securities issued. These arrangements provide protection to investors without placing a bank's capital at risk.
Finally, we would note that the rapid growth in size of the securitization market and the extension of the securitization process to various kinds of consumer loans that has taken place in recent years also suggest that our policy on recourse has not been a major impediment to securitization activities nor to bank participation in these activities.
All this being said, however, the agencies have recognized that there is a need to review carefully and reassess our policies on recourse and other guarantee arrangements. In particular, we believe that the policy as it pertains to cases in which there is only very limited recourse exposure deserves scrutiny, and, more broadly, we are also considering whether we need to make our capital treatment of recourse arrangements more consistent with that for other types of guarantees when the same level of risk is involved.
Although we will not presume to predict the outcome in any detail at this time, we do believe that it is appropriate to consider the need to refine our capital charges for transactions when the bank's risk is limited to an amount less than the normal capital charge associated with holding these assets on the balance sheet. Furthermore, we are also considering the advisability of refining our capital charge in those cases in which a depository institution is in a junior loss position-that is, when its own risk position is protected by credit enhancements provided by others. Our study of these matters has been under way for some time, and we hope to request public comments on concrete proposals in the not-too-distant future.
To sum up, it is clear that the securitization process as it has evolved to date has been beneficial to both the originators of asset-backed securities and to investors. And, moreover, similar benefits can be derived from the securitization of commercial real estate loans if the problems stemming from the heterogeneity and the uncertainty of the risk in these assets can be overcome. Such a development appears to require sufficient credit enhancement for such securities in a manner that is not prohibitively costly to issuers. As for the participation of insured depository institutions in the securitization of commercial real estate loans, we see no regulatory problems, so long as their activities are carried out in a safe and sound manner and a level of capital support is maintained against credit risk exposure that is commensurate with the degree of that exposure.
1. In the Federal Reserve Board's recent clarifications to its risk-based capital guidelines, the Board approved a limited recourse exception with respect to the sale of residential mortgages with recourse. Banking organizations using this exception would not be required to malntaln capital against residential mortgages that they have sold with recourse if, at the time of the transfer, the maximum possible recourse obligation is less than the expected loss on the transferred assets and a liability or specifically identified, noncapital reserve is established and maintained for an amount equal to the maximum loss possible under the recourse provision.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Statements to Congress|
|Publication:||Federal Reserve Bulletin|
|Date:||Jul 1, 1992|
|Previous Article:||Record of policy actions of the Federal Open Market Committee.|
|Next Article:||Deregulation and competition in Japanese banking.|