Current Legal Issues Affecting Central Banks, Volume IV.

Chapter 24 Securitization: Has It Matured?

Robert Effros
Published Date:
April 1997
  • ShareShare
Show Summary Details


Securitization has become an endeavor dependent upon acronyms and technology. There are MBSs, ABSs, CMOs, REMICs, IOs, POs, PACs,1 and a host of others—all acronyms for various kinds of securities, many of which would be difficult if not impossible to create, describe, evaluate, and understand without the increasingly sophisticated modeling capabilities made possible by computers. Accordingly, securitization is a financial and securities product that the computer age has made possible.

The focus of this chapter is on the general concepts underlying securitization, the factors that have impacted on its growth and development and may impact on its future growth and development, and the benefits and risks to financial institutions and other parties that participate in various parts of the securitization process (as well as the marketplace generally). This chapter will not dwell on the more exotic products (some of which create a variety of significant benefits, and some of which create significant risks) that the market wizards have created. It also avoids overconcentration on the technicalities of the U.S. legal and regulatory framework affecting securitizations. However, that framework, particularly insofar as banking regulation and tax and securities laws are concerned, has had a significant impact on the pace of development of the securitization markets in the United States and the direction that they have taken. Moreover, the U.S. legal and regulatory scheme has analogies in other countries. Accordingly, securitization—in the United States and elsewhere—cannot be understood without some reference to that framework.

This chapter has five basic themes: (i) securitization is a significant activity in the United States with important consequences for the health of the financial system, as well as for the liquidity of banks; (ii) securitization generally has been a positive development; (iii) commercial banks, because they are the major originators of securitized assets, have been and will probably continue to be at the center of developments in the area; (iv) as an activity, securitization has for the most part matured in the United States; and (v) in the short term, there is likely to be some retrenchment in securitization, as the instruments offered to investors are likely to become more rather than less standard.

What is securitization? The basic concept is simple. Securitization fundamentally involves the process of taking relatively illiquid assets originated by financial institutions and repackaging them into securities that can be sold to capital markets investors; hence the term “securitization.” The objective customarily is to create a security that capital markets investors can analyze and invest in by focusing only on the risk characteristics of the underlying assets and the structural implications, both legal—taxes, for example—and purely financial—prepayment risk, for example—of the transaction. Generally, there must be no material risk that, if the seller of the underlying assets has financial difficulties, the securities purchased by investors will be adversely affected. Accordingly, it is important to be able to conclude that, if the seller of the underlying assets ultimately does become bankrupt or get into financial trouble, the securities sold to investors and the investors’ indirect interest in the underlying assets will not be affected under applicable bankruptcy and insolvency laws.

Mortgage-backed securities (MBSs) are securities supported by mortgage loans (generally residential mortgage loans). The term “assetbacked securities” (ABSs) is sometimes used by practitioners to mean securities supported by assets other than mortgages, and it is sometimes used to encompass the entire world of securitization—both securities backed by mortgages and securities backed by other assets.

Although precursors to securitization date back to the nineteenth century, the modern day market in MBSs and ABSs in the United States is not really that old. MBSs began in the early 1970s when the Government National Mortgage Association, a government-sponsored entity, issued the first pool of residential mortgage loans, in which the pass-through certificates purchased by investors benefited from a U.S. government guarantee. The first private MBS offering (in the form of a single class of pass-through certificates) backed by residential mortgage loans involved the Bank of America as seller-sponsor in 1978. The world of non-MBS ABSs began in 1984 with the first two public offerings of ABSs backed by automobile loan receivables, one by Marine Midland Bank as seller-sponsor (with Salomon Brothers as underwriter), and the other by Valley National Bank as seller-sponsor (with First Boston as underwriter). Both offerings took the form of pass-through certificates, identical in structure in all material respects to the mortgage pass-through certificates issued to date.

Since 1984, securitizations have been accomplished using a variety of other types of underlying assets and receivables, including credit card receivables, first securitized by Republic Bank Delaware in 1987; boat loans, first securitized by Chemical Bank in 1988; recreational vehicle loans, first securitized by Fleetwood Enterprises in 1988; equipment leases, first securitized by Sperry Lease Finance Corporation (using leases on data processing equipment) in 1985; and nonperforming loans and other real estate loans, first securitized by Mellon National Bank through Grant Street National Bank in 1988. In addition, junk bonds, highly leveraged transaction or HLT loans, and tax-exempt bonds have been used in securitizations.

As the scope of securitization expanded, the development ultimately having the greatest impact was the first securitization of credit card receivables. This development was important because credit card receivables, given their revolving nature, could not simply be securitized in the normal form, namely, by placing a pool of identified receivables in a trust and passing through payments on the receivables to the investors when received. If that form were used, the trust would liquidate in very short order (six months or less).

MBSs and ABSs, taken together and based on new issuances, are much larger than the corporate equity market. MBSs and ABSs are closing in on the overall corporate market (corporate bond and stock offerings lumped together) as the second-largest securities market in the United States after government securities. Approximately $418 billion of MBSs (by principal amount) were issued in 1993 alone, and nearly $1.3 trillion in the five years ended December 31, 1993. At the end of 1993, well more than $1.5 trillion in MBSs were outstanding. With respect to ABSs, approximately $59.5 billion were issued in 1993, and more than $225 billion in the five years ended December 31, 1993. By comparison, corporate equity and bond offerings in 1992 totaled approximately $560 billion ($472 billion of bonds and $88 billion of stocks).

The securitization market in the United States is more developed than elsewhere, although the markets in a number of other jurisdictions are developing quickly. The U.S. head start is largely the result of two factors: U.S. government policies favoring the development of a secondary market for residential mortgages; and the deregulation of interest rates payable by U.S. banks and savings associations in the early 1980s. However, securitization as a financial activity is already quite significant in the Euromarkets and in several other countries’ domestic markets. Standard & Poor’s reported in April 1991 that, since 1987 (when it rated its first non-U.S. structured financing), it had rated over 100 non-U.S. issues, totaling approximately $28 billion and involving assets originated in seven countries. Anecdotal evidence indicates that securitization has grown exponentially in Europe over the past several years. The Bank of England’s 1989 notice on “Loan Transfers and Securitization” removed some of the uncertainties surrounding the treatment of securitized assets for U.K. banking regulatory purposes. France has promulgated a legal and regulatory framework targeted specifically to securitizations.2 Sweden, given the important role of its Government in supporting and assisting the construction of new housing, seems to be the birthplace for mortgage securitizations in continental Europe, and Japan has devoted a number of research committees, both in and out of its Government, to studying securitizations for a number of years.

Banks and Their Objectives

Commercial banks and thrifts (including their mortgage banking subsidiaries) tend to be the largest sponsors of securitizations, followed by the large, captive finance subsidiaries of the automobile manufacturers. The reason is self-evident: these entities are the originators of the assets, and, particularly in the case of banks, they are under intense competitive pressure to generate predictable earnings streams while limiting risks. Their interests and needs have been and will continue to be the principal determinants of developments in the U.S. securitization markets.

Securitization is essentially a large-scale disintermediation, in which traditional bank lending funded by deposits is replaced by the funding of the same assets directly by the capital markets. There are six principal incentives and objectives that have pushed seller-sponsors to securitize.

  • Credit risk with respect to the underlying assets can be transferred to third parties (whether the MBS or ABS investors or credit enhancers).
  • Capital markets can be accessed as a source of liquidity.
  • As part of its task of managing assets and liabilities, a seller-sponsor can reduce maturity risk, the risk that a financial institution will not be able to match perfectly the funding of assets and liabilities on its balance sheet. In a securitization, there is no maturity “gap.” Additionally, prepayment risk is transferred to investors (except to the extent of the servicing fee income that the seller-sponsor, in its separate capacity as servicer, may expect to receive).
  • Like maturity risk, any interest rate basis risk for the seller-sponsor can be eliminated and transferred (collectively) to the investors in the securitization.
  • Cost savings can be generated by a securitization, as the securities can be structured (including through “overcollateralization,” extrinsic credit enhancement, or both) to obtain a higher credit rating for the MBSs or ABSs than the originating entity itself would obtain, effectively resulting in a lower cost of funds.
  • Seller-sponsors can produce fee income without having to fund the related assets and bear the risks of ownership and funding described above. The seller-sponsor of the securitization customarily will act also as servicer and will receive, as servicing compensation, a designated percentage of the balance of the underlying assets from time to time outstanding. The servicing fee is funded from the excess of the interest rate received on the underlying assets over the interest rate paid on the securities being issued.

Critical Developmental Factors

Undoubtedly, a variety of direct and secondary pressures, including the factors referenced above, have borne on the development and growth of securitizations, particularly the dramatic growth that began in the late 1970s. Banks were under pressure to reduce assets; accordingly, they required capital that generally (even before risk-based capital requirements were introduced) was determined in connection with some measure of assets. Banks were also under pressure to generate fee income and manage risk generally.

As indicated above, two factors were particularly important in the growth of securitizations in the United States: (i) U.S. government policies favoring the development of a secondary market for residential mortgages, reflecting to some extent the importance traditionally assigned to home ownership in the United States; and (ii) pressures resulting from the deregulation of interest rates payable by U.S. banks and savings and loan associations in the early 1980s (as mandated by the Depository Institutions Deregulation Act of 1980).3

U.S. Policies Supporting the Residential Mortgage Market

Government support for the residential mortgage market is manifest in a host of ways. Perhaps the most important and fundamental was the creation by statute of the Government National Mortgage Association (a wholly owned corporate instrumentality of the U.S. Department of Housing and Urban Development), the Federal National Mortgage Association (a federally chartered and privately owned corporation first established in 1938), and the Federal Home Loan Mortgage Corporation (a federally chartered corporation created in 1970, the stock of which is owned by the Federal Home Loan banks). These agencies were charged with developing a secondary market for mortgage loans and given the power to issue their own MBSs. It has been estimated that today approximately 40 percent of new originations of residential mortgage loans are eventually packaged into MBSs issued or guaranteed by these three agencies.4

Beyond the creation of these agencies, government policies have been supportive of mortgage credit in many other ways. First, in 1984, Congress passed the Secondary Mortgage Market Enhancement Act,5 which relaxed, among other things, usury limits, state “blue sky” and securities regulations, and legal investment limitations on mortgage-related securities applicable to a variety of financial institutions (including banks, savings and loan associations, and insurance companies). Mortgage-related securities, for this purpose, are essentially MBSs supported by residential mortgage loans and issued by banks, savings and loan associations, insurance companies, and other institutions that (i) are rated in one of the two highest categories by a nationally recognized rating agency and (ii) are supported by first liens on the underlying real property.

Second, the real estate mortgage investment conduit (REMIC) provisions6 were added to the Internal Revenue Code in 1986 to permit the issuance of multiclass mortgage-backed securities in qualifying transactions without treating the issuer (whether a trust or some other vehicle) as an association taxable as a corporation (that is, without imposing federal income tax on the issuer).7 As will be addressed later, this action removed a major impediment to improving the economic efficiency of MBSs; the seller-sponsors had been unable under applicable U.S. federal tax law prior to 1986 to issue in securitizations multiple classes or tranches of securities. Although comparable relief has been discussed for some years for non-mortgage-backed ABSs, legislation has not yet been enacted and does not appear imminent.

Third, Federal Reserve regulations support the development of a secondary market for residential mortgages. The Federal Reserve Board’s Regulation D requires that reserves be maintained against certain kinds of “deposits” (defined broadly under this regulation to include most liabilities) of depository institutions.8 Prior to December 1990, Regulation D included a 3 percent reserve requirement on time deposits having maturities of less than 18 months. Accordingly, MBSs issued by a trust sponsored by a bank that retained a subordinate interest or otherwise provided credit enhancement generally would have been treated as a “deposit” of the bank, subject to reserve requirements. However, the current (1995) version of Section 204.2(a)(2)(ix) of Regulation D includes an express exception for a retained interest not exceeding 10 percent in pools of conventional residential mortgage loans.9

Finally, the risk-based capital requirements of the U.S. banking regulatory agencies favor mortgage loans and MBSs in a variety of respects. Most important is the 50 percent risk weighting available for high-quality residential and multifamily mortgage loans (as well as for MBSs backed by such loans).

Deregulation of Interest Rates

Depository institutions in the United States had been subject prior to the 1980s to limits on the interest rates that they could pay on various kinds of deposits. These limits were implemented through the Federal Reserve Board’s Regulation Q.10 In the early 1980s, the interest rate ceilings were effectively eliminated, introducing for the first time real competition for deposits and creating the specter of significant mismatches between interest earned on long-term, fixed-rate assets (the largest single component of which were residential mortgage loans) and interest paid on deposits at competitive rates. The incentive to manage interest rate risk by intermediating between capital markets as a long-term funding source and mortgage borrowers was a very important component in the push by banks, commencing with Bank of America in 1978, to develop the private (that is, nonagency) MBS market.

Characteristics of Assets Affecting Securitizations

Although any kind of loan, receivable, or other asset that generates a cash flow stream can, in theory, be repackaged into a security, as a practical matter it is much easier to securitize assets that are more (rather than less) homogeneous, and to securitize when each asset is only a small part of the entire pool of assets that supports an MBS or ABS. Legal and regulatory factors, which will be addressed below, are likely (at least in the near term) to constrain the expansion of ABSs into new types of assets. However, equally important is the practical difficulty of the marketplace in coming to grips with less homogeneous assets (commercial mortgage loans, for example, and, to an even greater degree, commercial and industrial loans) that, because of their lack of homogeneity (and often this less predictable loss and delinquency experience), produce less predictable cash flows. The less predictable are the cash flows generated by the underlying assets, the more protection against loss is required by investors in the securitizations. For example, the Resolution Trust Corporation, the U.S. government agency created to administer the resolution of failed savings and loan associations, was by far the largest seller-sponsor for securitizations of performing commercial mortgage loans during 1993. The transactions succeeded because the Resolution Trust Corporation provided recourse, in the form of liquidity reserves, in amounts of between 30 percent and 40 percent of the securities sold. The need for substantial credit enhancement that cannot be satisfied by excess servicing (the difference between the interest rate demanded by investors and the interest rate borne by the underlying assets) precludes as a practical matter commercial banks from participating as seller-sponsors in similar transactions, at least if the selling bank proposes to get sale treatment for regulatory accounting and capital purposes.

To summarize, a number of characteristics facilitate securitizations:

  • Predictable cash flows, which are partly a function of the nature of the underlying assets and partly a function of their contractual terms, facilitate securitizations. For example, residential mortgage loans in the United States generally must be prepayable at the borrower’s election (subject to limited exceptions). A large amount of data is available with respect to prepayment experience on fixed-rate residential mortgage loans. An extraordinary amount of effort and expense (and technology) is devoted to trying to predict the likely prepayment experience on mortgage pools and related MBSs in different interest rate scenarios. There is significantly more basis for analyzing and extrapolating likely prepayment experience on fixed-rate residential mortgage loans than on many other kinds of assets that might be securitized (commercial loans, for example). As to the contractual terms, fully amortizing, level-payment obligations are the most easily securitized. “Balloon payments” (that is, large, unamortized balances due at maturity) and other nonlevel payments lead to nonuniform cash flows to investors. Furthermore, obligations with balloon payments are thought to be more likely to default (because of the uncertainty as to whether the borrower will be able to refinance at maturity). Similarly, assets bearing a fixed interest rate are generally easier to securitize than variable-rate instruments.
  • A history of consistently low delinquency, default, and loss levels facilitates securitization. The level of credit enhancement required is generally based on an estimate of the performance of the asset pool under “worst-case” scenarios or stressful economic conditions. Although the performance of a pool will be impacted by a variety of factors that may have nothing to do with the historical performance of similar pools originated or serviced by the same seller-sponsor, the level of uncertainty is reduced if the seller-sponsor of the assets has originated and serviced similar obligations in a consistent manner over a long period of time and has maintained accurate records or payment histories and loss experience on such obligations.
  • Demographic and geographic diversity of obligors on the assets facilitates securitization. Demographic diversity is particularly important in the case of consumer receivables, in order to mitigate the effects of regional economic downturns; geographic diversity is important in the case of mortgage loan securitizations, in order to mitigate the risk of natural disasters (for example, earthquake risk in California or hurricane risk in Florida).
  • A greater seasoning of assets facilitates securitization. Experience has shown that delinquency and loss levels are generally highest during the early years, particularly for installment obligations. The greater the “seasoning,” or average age, of the obligations, the lower is the expected delinquency rate.
  • High liquidation value and utility to the obligor of the underlying collateral facilitate securitization. This is a particularly important positive factor for MBSs, for which ultimate losses on foreclosure have historically been relatively small, and on which obligors have a great incentive to avoid defaults (and keep their homes). Credit card securitizations are at the opposite end of the spectrum; they are generally unsecured. The risks for credit card securitizations are offset by the small proportion of the total pool accounted for by each receivable and by the well-developed databases for delinquency and loss experience of the major originators.
  • Standardized, high-quality underwriting and collection policies by the seller-sponsor and servicer facilitate securitization.

Conversely, a number of asset characteristics make securitizations difficult:

  • A small number of assets in the collateral pool and a high ratio of the largest asset to the average asset make securitization difficult. The more “large” assets in the pool, the more concentrated is the risk to investors from particular assets, and perhaps the greater the need for asset-specific disclosure.
  • A lack of standardized documentation for the assets makes securitization difficult. Less standardized documentation is likely to be accompanied by less standard (and predictable payment) terms, as well as by less certainty as to the servicer’s rights and remedies upon default.
  • Infrequent payment dates increase the difficulty of securitization. The capital markets generally require payments on a quarterly or monthly basis.
  • The ability of asset obligors to modify payment terms (interest rate formula elections, for example) makes securitization difficult. This contributes to greater uncertainty about the cash flow available to service securities.
  • The absence of sufficient historical loss and delinquency experience for the asset type or for the seller-sponsor increases the difficulty of securitization.
  • An inexperienced or undercapitalized servicer makes securitization difficult.

These characteristics make it much more difficult to analyze the expected cash flow stream from the asset pool on an actuarial basis and are likely to engender a need for substantially greater credit enhancement or recourse on the part of the seller-sponsor.

The Securitization Process

Elaborating on the definition proposed at the outset of this chapter, securitization can be described as a process whereby a pool of assets with similar characteristics is “packaged” into securities that either pass through or pay principal and interest to security holders, based on and derived from the principal and interest (and sometimes disposition proceeds) from and on the underlying assets. Generally, the monies received from payments on the assets in the pool are the source of funds to pay interest on (and retire) the securities. The product of the securitization might be a pass-through interest in a pool of assets, which is a beneficial ownership interest—not necessarily pro rata with every other beneficial ownership interest—in the pool entitling the holder to periodic distributions of available funds up to specified amounts. Traditional mortgage pass-throughs, for example, are in this category. Alternatively, the product of the securitization might be an actual debt security issued by a legal entity (a special-purpose corporation or an owner trust, for example).

Notwithstanding its young age, the market for MBSs and ABSs in the United States is relatively mature. Most assets that are material assets for financial institutions and that, given their characteristics, can be securitized are in fact being securitized in significant volume in a market-efficient way. These assets include mainly residential mortgage loans, home equity loans (including both term loans and revolving credits), credit card receivables, and automobile receivables. These assets generally have characteristics that facilitate securitization, as explained above: predictable cash flows, standardized documentation, low weights in the collective assets pool, and valuable collateral (except for credit card loans). Additionally, except for residential mortgage loans, these assets are essentially consumer loans, on which the interest paid by consumers exceeds significantly the interest rate that will be payable on ABSs resulting from a repackaging of the assets into a security.

Basic Structure

Because securitization is simply the repackaging of loans and other receivables into securities, the scope of what falls under this term—the types of issuing vehicles and the types of securities issued, for example—is quite broad. It is not particularly useful to spend much time exploring the intricacies of alternative structures of securitization.11 Instead, this chapter focuses specifically on the most common structure used in the United States (a “grantor” or “pass-through” trust) as a way of illustrating the impact of the U.S. legal scheme on the development of securitization and the risks and benefits to the parties involved.

In a typical pass-through trust12 transaction, the relevant parties are (i) the seller-sponsor, the customary originator of the assets; (ii) the servicer, (often but not necessarily the same party as the seller-sponsor); (iii) the trustee (generally a bank) that holds the assets and issues pass-through certificates evidencing interests in them; (iv) one or more underwriters or placement agents distributing the ABSs; (v) in many cases, a “credit enhancer,” charged with issuing a letter of credit or surety bond or funding a cash collateral account to protect investors against some multiple of the expected risks; and (vi) the investors themselves.

Typically, the following basic steps occur at about the same time in the securitization process: the seller-sponsor will convey the assets to the trustee in return for pass-through certificates (which may or may not be in multiple classes), evidencing 100 percent of the ownership interests in the trust that holds the assets; the credit enhancer (if any) will issue its letter of credit, surety bond, or other similar instrument to the trustee (or fund a cash collateral account held by the trustee) for the benefit of certificate holders; the seller-sponsor will sell some or all of the certificates to the underwriters, pursuant to an underwriting or purchase agreement; and the underwriters will resell certificates purchased by them to ultimate investors.

Responsibilities of Parties to the Securitization

The operative agreement under which the assets are transferred to the trustee, the trust is created, the servicer is retained to service the assets, and the pass-through certificates are issued is customarily called a pooling and servicing agreement. Under the agreement, the seller-sponsor, servicer, and trustee take on a number of obligations. The seller-sponsor, in connection with its conveyance of the assets to the trust under the pooling and servicing agreement, typically makes a variety of representations and warranties as to the characteristics of the assets, their enforce-ability, and their compliance with law, but not as to their collectibility or value. If it turns out that those representations and warranties are breached with respect to one or more of the assets, the seller-sponsor customarily will be obligated to repurchase the assets to which a representation or warranty was breached.

The pooling and servicing agreement will generally spell out the duties of the servicer with respect to collecting payments of principal and interest on the assets, the standard of care that it should apply, and the flexibility given to it to deal with assets that ultimately become troubled or go into default. Moreover, in order to “smooth out” the cash flow going to investors in the pass-through certificates, the servicer will normally be obligated to make out of its own funds “servicer advances” on payment dates for the pass-through certificates to cover delinquencies in payments of principal or interest scheduled on the underlying assets, but only to the extent that the servicer determines that the delinquent payments will ultimately be recovered. The servicer undertakes this obligation because, for a variety of innocuous reasons (borrowers may be on vacation or forget to pay their bills for a week, for example), payments by the obligors (generally individuals) on the underlying assets may not come in exactly when they are supposed to; in these circumstances, the late payments do not reflect credit concerns. If, however, the servicer ultimately concludes that a delinquent payment is not recoverable, the servicer will be entitled to reimburse itself for servicer advances out of the next available cash flow from all the assets.

The pass-through certificates issued by the trust evidence beneficial ownership interests in the underlying assets and an entitlement to specified cash flows from those assets. The terms of the pass-through certificates can be simple or complicated. At the simple end of the spectrum is a transaction in which a single class of pass-through certificates is issued, with each certificate representing an undivided pro rata interest in, and entitlement to, principal and interest payments received on the underlying assets in each period. At the complicated end of the spectrum, generally represented by multiclass MBSs, a large number of separate certificate classes (sometimes as many as 15 or 20) may be issued by a single trust, with each class having different payment entitlements. These might include, for example, sequential pay classes (in which principal payments on the underlying mortgages are applied in sequence to classes until one class after another is retired, with the consequence that the later classes are riskier), as well as interest-only classes (IOs), principal-only classes (POs), planned amortization classes (PACs), and a host of others. Clearly, computer technology plays a critical role in orchestrating these payments.

The trustee’s role in pooling and servicing agreements is customarily fairly limited. The trustee (i) holds record title to the assets, (ii) receives from the servicer collections of principal and interest on the assets, and (iii) pays to the appropriate classes of investors their respective entitlements to the cash flow received from the assets. If the servicer defaults in performing its obligations (for example, fails to remit to the trustee on behalf of certificate holders in a timely manner payments on the underlying assets), the trustee customarily would have the right (and under some circumstances the obligation) to dismiss the servicer and itself act as a backup servicer until a replacement is found.

Role of Law and Regulation

Securitization in the United States is a legally intensive exercise. The principal legal and regulatory areas involved are tax, banking, securities, and bankruptcy and insolvency. The legal and regulatory framework in the United States has had a critical impact on the pace with which securitizations have developed because of certain tax, banking regulation, and securities matters. The legal and regulatory environment in the United States today is quite different from the circumstances as recent as the mid-1980s, which was the major developmental period for securitizations. Specifically, in each of the three critical areas, legal and regulatory changes have been or are being made to accommodate securitizations, which underlies the view that securitization as a financing activity is relatively mature in the United States.

It is instructive to touch upon the three most important legal and regulatory changes: (i) the 1986 REMIC tax legislation permitting multitranche MBSs;13 (ii) the treatment by the federal banking regulatory agencies for capital and regulatory accounting purposes of asset sales with recourse;14 and (iii) the adoption by the Securities and Exchange Commission in 1992 of Rule 3a-7, providing relief for certain qualifying transactions from the Investment Company Act of 1940.15 The legal and regulatory changes have been largely reactive: laws and regulations have changed after market participants have managed to do most (but not all) of what they wanted to do under the existing legal and regulatory framework, generally with some inefficiencies because the framework did not take account of securitizations.


Generally, for a securitization to make sense it is essential that the issuer (the trust, for example) not itself be subject to income tax (technically, under the U.S. Internal Revenue Code, that it not be treated as an association taxable as a corporation). If the issuing trust were taxed as a corporation, the trust would be deemed to have substantial taxable income. In the worst-case scenario, the income would be equivalent to the difference between the interest received on the assets, without any offsetting interest expense, as the pass-through certificates would likely be treated as equity interests instead of debt obligations on which interest is paid. Accordingly, a substantial portion of the cash flow generated by the trust’s assets would be paid as taxes to the Federal Government instead of being available for distribution to investors.

In the early days of securitizations (before the 1986 REMIC legislation), a trust could be “tax passive” (that is, not give rise to income tax at the trust level) only if it issued a single class of pass-through certificates. As an exception, certificates could be issued in two classes if one class were subordinated to provide default protection to holders of the senior class and, absent defaults, received payments pro rata with the senior class. Accordingly, tax passivity and the complicated tranching and cash flow allocations that are characteristic of today’s MBSs generally were not achievable.

Through the mid-1980s, the early MBSs, as well as ABSs backed by nonrevolving consumer loans (automobile receivables, for example), were grantor trusts, as described above. In terms of market efficiency, grantor trusts issuing a single class of certificates were not optimum, particularly when the underlying assets had long final maturities and would generate cash flow over an extended period. For example, if the underlying assets were 30-year mortgages, investors would receive some cash flow throughout the 30-year term of the instruments, even though the average life of the mortgages might be 12 years. Seller-sponsors, underwriters, and other market participants realized that it would be much more efficient to tranche the cash flow from the underlying assets to create multiple classes of securities when such tranching might generate more predictable cash flows (for example, by creating sequential pay classes and PACs, in which the cash flow for a particular class is paid within a narrower time band or pursuant to a specific schedule). Alternatively, the market participants could create “hedge” instruments, the value of which would respond in predictable ways to movements in interest rates (IOs and POs, for example).

The 1986 restatement of the U.S. Internal Revenue Code provided real estate mortgage securitizations (both residential and commercial) with substantially more flexibility, including, among other things, the ability to issue multiple classes of securities. The REMIC provisions added to the Code in 1986 provide tax passivity for a pool of assets that meets the requirements for a REMIC, irrespective of the form of ownership—whether partnership, corporation, trust, or some other entity—of the pool.16 Although most REMICs are, in fact, trusts—largely because the marketplace’s familiarity with trusts dates back to when MBSs were issued without REMICs—there is no requirement that the issuer be a trust. The flexibility afforded to REMICs is not available in other contexts (that is, the REMIC provisions of the Internal Revenue Code apply only to mortgage assets). The REMIC provisions allow for the issuance of multiple classes of “regular interests” (each having defined payment entitlements) and a single class of “residual interests” (which entitles investors to whatever residual cash flow is left). These provisions permit very limited substitution of assets and are available only where the underlying assets are mortgages. They give the flexibility, at least in connection with MBSs, to model the cash flow from the underlying mortgages in increasingly complex ways, thus creating increasingly complex and esoteric securities.

The major deficiency in the REMIC legislation is that it covers only mortgage securitizations. If there is a tax frontier yet to be crossed in the United States, it is to provide similar relief for nonmortgage securitizations. An industry group has been working to develop REMIC-type legislation that would cover such securitizations. Because the nonmortgage assets that have supported ABS securitizations to date have largely been shorter-term consumer receivables, there has been less pressure to resolve the market inefficiency of creating from a nonmortgage asset pool different tranches of securities for which prices and yields can be calculated based on more precisely identified segments of the yield curve.

Banking Regulations and Related Capital Treatment

Commercial banks are the largest securitizers in the United States. The regulatory accounting and related capital treatment of securitizations is critical for banks. Although a number of factors create incentives for banks to intermediate between the capital markets and borrowers by securitizing assets, one of the most important is to reduce the assets against which capital must be maintained.

The appropriate treatment for capital and regulatory reporting purposes of asset sales with recourse admittedly presents a number of problematic issues. Bankers generally would say that the current treatment is too restrictive. However, the industry recognizes that there are no easy and definitive solutions to the banking regulatory agencies’ concerns.


An understanding of the basics of treating securitizations for capital purposes by U.S. banks is useful. As a first step, whether a purported sale will be recognized as a sale for capital purposes depends upon its treatment under so-called regulatory accounting principles (RAP), which are the accounting principles that apply to the statements of condition (call reports) filed by banks with their regulators. If a bank’s purported sale of a pool of assets in a securitization is not recognized as a sale under RAP, the underlying assets will remain on the bank’s balance sheet for RAP purposes (irrespective of the treatment of the transaction for purposes of generally accepted accounting principles (GAAP)). If these assets remain on the selling bank’s balance sheet for purposes of its call reports, capital must be maintained against them. Consistent with the Basle Committee on Banking Supervision’s 1988 Capital Accord,17 each of the federal banking regulatory agencies’ risk-based capital regulations impose an 8 percent capital requirement against risk-adjusted assets.

As to what is a sale under RAP, these principles essentially divide the world into two parts: pools of residential mortgage loans and other assets. With respect to residential mortgage loans, RAP will permit banks to treat a transaction as a sale provided that the selling bank does not retain “significant risk of loss.” Significant risk of loss means, essentially, retained risk of loss (either through direct support, such as a guarantee or a letter of credit, or indirect support, through retention of a subordinate interest) that exceeds historically expected losses on the pool. (These principles are embodied in the definition of “participations in pools of residential mortgages” in the glossary to the call reports.)

The RAP treatment of sales of assets other than residential mortgages is embodied in the definition of “sales of assets” in the glossary to the call reports. For receivables other than residential mortgage loans, retention of any risk of loss from the transferred assets resulting from any cause essentially precludes sale treatment. This definition can lead to some untoward results. For example, if a bank transferred a $100A; (where x is a fixed number—for example, x = $1,000,000) pool of receivables to a pass-through trust in a typical securitization and provided $1x in recourse (through a $1x guarantee or a retention of a $1x subordinate interest), the transaction would be treated as a financing instead of a sale for RAP purposes, and the selling bank would have to maintain 8 percent capital (or $8x) against the sold pool of receivables, even though the maximum exposure was only $1x.

Regulatory accounting principles, obviously, do not follow GAAP. The basic tests for a sale under U.S. GAAP are set forth in the Financial Accounting Standards Board’s Statement of Financial Accounting Standards No. 77.18 This statement specifies three conditions needed for a transfer of receivables to be recognized as a sale:

  • The transferor surrenders control of the future economic benefits embodied in the receivables. . . .
  • The transferor’s obligation under any recourse provisions can be reasonably estimated. . . . [and]
  • The transferee cannot require the transferor to repurchase the receivables except pursuant to the recourse provisions.19

Although there has been a great deal of discussion in the United States as to whether RAP should simply follow GAAP, the U.S. banking regulators have declined to do so, principally based on the argument—which may not be entirely correct—that GAAP emphasizes the degree to which benefits have been transferred while RAP is more properly concerned with the degree to which risks have been retained.

Excess Spread

In order to create a security attractive to investors, the process of packaging bank loans and receivables into marketable securities normally requires that the seller directly or indirectly provide some sort of credit enhancement.20 How then can securitizations be accomplished at all with banks as sellers?

In theory, a bank could simply sell the assets underlying a securitization at a discount so that investors are effectively overcollateralized. This approach would be highly inefficient for the bank, however, because (i) the selling bank’s own estimate of the amount of losses expected and, accordingly, the size of the discount are likely to be substantially less than what investors (or rating agencies, as their proxy) would require and (ii) selling at a steep discount has the consequence of creating a windfall for investors if the selling bank’s expectations of anticipated losses are borne out.

The solution to date has largely been to take advantage of an off-balance-sheet contingent asset that reverts to the seller, as servicer, over time, namely, “excess spread.” Excess spread is the excess of the interest rate borne by the receivables being securitized over the interest rate demanded by investors on the securities issued in the securitization, plus the customary servicing fees. Suppose, for example, that each receivable in a pool of automobile loans bears interest at 10 percent but that, in the current interest rate environment, the marketplace for the desired rating level and the expected weighted-average maturity require that the securities issued in the securitization bear interest only at 6 percent. If the customary servicing fee to be paid to the seller-servicer is 0.25 percent, a pool of $100x in receivables can be sold for $100x in cash, and the excess spread of 3.75 percent received over time can first be applied to cover any losses on underlying receivables. To the extent that it is not necessary to cover losses (because they do not occur), the excess spread can be remitted to the servicer as “excess servicing.” Provided that the seller-servicer does not treat the excess servicing as an asset (or take it into income) until it has irrevocably reverted to the seller-servicer and is no longer available to cover losses, that contingent asset may be used to enhance the credit of the securities. From a regulatory perspective, the theory is that the RAP balance sheet at all times reflects the worst-case scenario because the contingent asset inherent in excess servicing has not yet been recognized as an asset on the balance sheet. Accordingly, no asset on the call report balance sheet, which, if its value goes to zero could reduce the equity of the bank’s shareholders, is at risk.

One problem with the use of excess spread to facilitate securitizations is that many assets on a bank’s balance sheet do not bear interest at rates sufficiently above the interest rates demanded by the capital markets to create an excess spread sufficient to enhance the credit of a securitization. Arguably, this constraint has hampered some securitizations, for example, of commercial and industrial loans. However, the major constraint is the lack of homogeneity—as well as the predictability—of losses and payment streams.

Notice of Proposed Rulemaking

In these circumstances, the federal banking regulatory agencies, acting through an interagency council known as the Federal Financial Institutions Examination Council, have proposed far-reaching changes in the current rules (some good and some bad, from the industry’s perspective).21 One set of proposals, included in a Notice of Proposed Rulemaking that has been made public by the Federal Reserve and the Federal Deposit Insurance Corporation, would require credit enhancers (in the same manner as sellers providing recourse) to maintain capital against the entire asset pool undergoing credit enhancement instead of against only the amount of the credit enhancement.22

The regulations described above, under which a purported sale would be recharacterized as a financing, apply only to sellers. They do not apply to third-party credit enhancers. Instead, a credit enhancer writing a letter of credit or funding a cash collateral account to support some other bank’s securitization is required under the existing rules to maintain capital only against the amount of its credit enhancement. The proposed change would implement the regulator’s principal of “equal capital for equal risk.” That is, if credit could be enhanced for a $100x pool of assets underlying a securitization by either a $6x subordinate interest retained by a seller or a $6A: letter of credit provided by another bank, the amount of capital required by the seller on the transaction (if the first alternative were chosen) or the unaffiliated credit enhancer (if the second alternative were chosen) should be the same. Under the existing system, the credit enhancer has to maintain 8 percent capital against $6.%, or $.48 in capital; under the Notice of Proposed Rulemaking, the credit enhancer (like a seller) would be required to maintain 8 percent capital against the entire $100.% pool, or $8 in capital. Both sellers and credit enhancers would be subject to a capital charge capped at the amount of the recourse.

Advance Notice of Proposed Rulemaking

An Advance Notice of Proposed Rulemaking was published on this matter in 1994.23 It implements a ratings-based, multilevel approach, under which a bank’s risk-based capital charge for certain asset securitizations would depend on its relative risk of loss.24 Although credit enhancers view the equal capital for equal risk principle embodied in the Notice of Proposed Rulemaking as anathema, the Advance Notice of Proposed Rulemaking would provide some relief. Industry participants commenting on the Advance Notice of Proposed Rulemaking strongly argue that the portion of the Advance Notice of Proposed Rulemaking addressing credit enhancers should be implemented only if relief for qualifying transactions is simultaneously provided for in the Advance Notice of Proposed Rulemaking.

The Advance Notice of Proposed Rulemaking is based upon a hypothetical, simplified transaction that assumes (i) a senior, highly rated security; (ii) a “middle-level,” subordinated security or interest that is in a second-dollar loss position and enhances the credit of the senior security; and (iii) a subordinated security in a first-dollar loss position that enhances the credit of the two more senior tranches.

The Advance Notice of Proposed Rulemaking would make three provisions. First, if a middle-level, subordinated security or a second-dollar loss enhancement (whether in the form of recourse or a direct credit substitute) is rated at least investment grade by a nationally recognized statistical rating agency, the seller providing recourse or the third-party providing the direct credit substitute by writing or holding such position would be assessed capital only against the face amount of the recourse or credit enhancement. Second-dollar loss positions that do not qualify for this approach (including purchased subordinated securities and letters or credit) would be charged capital, based on the bank’s exposure and the more senior portions of the pool.

Second, the Advance Notice of Proposed Rulemaking would provide that the most senior class of a qualifying securitization transaction would be risk weighted at 20 percent (regardless of the underlying asset type) if this class received the highest credit rating from the same rating agency that rated the second-dollar loss position. However, senior positions that are supported by prior credit enhancements carrying third-party performance risk (for example, standby letters of credit) would not qualify for this treatment.

Third, with respect to recourse arrangements or direct credit substitutes in a first-dollar loss position, the low-level recourse rule (capping the capital charge at the amount of the recourse) included in the Notice of Proposed Rulemaking would apply. That is, capital would be charged dollar-for-dollar against the bank’s exposure up to the full risk-based capital charge for the assets.

Rule 3a-7 Under the Investment Company Act

The Investment Company Act25 provides a comprehensive regulatory scheme for managed investment funds. Although the perceived need to regulate managed investment funds gave rise to the enactment of the statute in 1940, its actual reach is much broader. The Investment Company Act by its terms imposes its regulatory scheme on any entity whose assets are primarily “investment securities.”26 Accordingly, for purposes of the statute, almost any obligation to pay (including all kinds of bank loans) is an investment security, with limited exceptions made for government securities and stock of operating subsidiaries. Therefore, most trusts and other securitization vehicles, absent an exemption, are by definition “investment companies.” The Investment Company Act imposes a series of substantive restrictions on investment companies (including with respect to capital structure and dealings with affiliates) that would make it impossible to conduct the normal securitization of assets as a registered investment company.

The Investment Company Act has included since its adoption exceptions from investment company status for companies that are primarily engaged, among other things, in (i) purchasing or otherwise acquiring notes, drafts, acceptances, open accounts receivable, and other obligations representing part or all of the sales price of merchandise, insurance, and services; and (ii) purchasing or otherwise acquiring mortgages and other liens on, and interest in, real estate.27 Subject to sometimes arcane interpretations by the staff of the Securities and Exchange Commission as to the availability of these exemptions, most securitization vehicles for the assets traditionally securitized in great volume (residential mortgage loans and various kinds of consumer receivables) have been able to rely on one or another of these exemptions.

In 1992, however, the Securities and Exchange Commission adopted a new rule, Rule 3a-7,28 intended to permit, subject to the rule’s specific requirements, securitizations free of regulation under the Investment Company Act, irrespective of the type of loan or receivable being securitized (including any collateral for, or the use of proceeds of, the loan or receivable). Rule 3a-7 exempts from status as an investment company a vehicle in which the assets are (i) “eligible” financial assets that by their terms convert into cash within finite time periods and (ii) “fixed-income” securities issued to public investors that have stated interest or principal amounts, or both, and are rated in one of the four highest categories by at least one nationally recognized statistical rating organization.29 (The foregoing describes only the essence of the rule, which includes a number of other detailed requirements.)

Rule 3a-7 reflects a major regulatory step by the Securities and Exchange Commission to remove an artificial limitation on securitizations. At the margin, Rule 3a-7 will make a difference because it will permit securitization of a variety of receivables that would not otherwise have fit under the existing exemptions. However, it is not likely to contribute to a fundamental growth or change in direction in the U.S. securitization market (particularly where banks are the seller-sponsors) because the types of assets that most lend themselves to securitizations, given their characteristics, could be fit within the existing exemptions, as described above.


The track record for securitizations in the United States is remarkably good for investors and seller-sponsors. Apparently, a rated public securitization with a commercial bank as seller-sponsor has never gotten into trouble or incurred a loss for its public investors.

The analysis of risk for commercial bank seller-sponsors must be a bit more subtle because, as described above, they are constrained in providing credit enhancement but allowed to treat transactions as sales for regulatory accounting and capital purposes. The federal banking regulatory agencies, through the examination process, look very closely at the exposure of banks that are major securitizers. These agencies focus primarily on three areas. First, in light of the overall securitization activities of a particular bank, has that bank sold its better assets and retained its weaker assets (“adverse selection,” insofar as the bank is concerned)? Second, does a bank have an incentive to lower its loan underwriting criteria for asset types that it securitizes in order to generate assets to support its securitizations (and, in turn, generate fee income)? Third, does a bank that acts as a seller-sponsor, irrespective of its contractual obligations, feel compelled to provide “moral recourse” (that is, to reimburse investors for losses even when it is not contractually obligated to do so), in order to support the market for its future securitizations?

The concerns cited above are appropriate focal points for bank examiners. Each of the federal banking regulatory agencies is well attuned to these risks and has developed specific guidelines to assist its examiners in their evaluation. However, there has not apparently been a case in which a bank has been deemed to raise supervisory concerns as a result of its securitization activities.


Securitization, as a process of intermediating between capital markets and borrowers (largely individuals), has been a great success in the United States. It has contributed significantly to the availability of consumer credit at attractive rates while enabling banks and other financial institutions (which, absent securitization, would themselves be funding these loans) to pass a variety of risks on to the marketplace.

As is almost inevitable when new frontiers are being probed, changes in the U.S. legal and regulatory framework concerning securitizations have largely been reactive. This is certainly true of (i) the REMIC tax legislation,30 which permits substantially greater flexibility for multitranche MBSs than the old-style, sequential-pay collateralized mortgage obligations (CMOs) issued by owner trusts; (ii) the “with recourse” bank capital proposals currently being considered by the federal banking regulatory agencies;31 and (iii) the adoption by the Securities and Exchange Commission of a structured finance rule (Rule 3a-7), which provides an exemption from the Investment Company Act for qualifying transactions.32

Securitization in the United States is a relatively mature activity that has reached a plateau. In other words, those kinds of assets that represent a material portion of financial institutions’ balance sheets and are likely candidates for securitization (given their homogeneity and other terms) are already being securitized in a significant way.

The principal developments in securitization in the United States over the next couple of years will principally be around the edges, that is, the changes will increase the efficiency of transactions that are already occurring. These developments may include tax legislation permitting REMIC-like flexibility for non-MBS ABSs, as well as more favorable capital treatment for bank investors in MBSs, along the lines proposed in the Advance Notice of Proposed Rulemaking under consideration by the federal banking regulatory agencies.



Securitization, of course, predates computer technology. One could say that nonrecourse financing techniques—such as the sale of participations in loans and the factoring of accounts—are early forms of “securitization.” These types of transactions, which admittedly do not always achieve the objective of protecting purchasers from sellers’ insolvencies, have been conducted for many years. One of the burdens faced by participants in the more modern U.S. securitization process and one that will be a theme in nearly every other jurisdiction in which this type of securitization is implemented is that the analysis of these modern transactions is based on precedent, if any, that addressed the more primitive type. In the system based on common law, a sedimentary buildup of case law addressing financing techniques of the adding machine age is being used for guidance in the computer age. The cases do not address anything nearly as sophisticated as what the new securitization transactions have created. (In civil law jurisdictions, the problem may even be worse as, in some countries with such a tradition, it is unlikely that statutes addressing these concerns have been written.)

Insolvency Risks

Some understanding of U.S. law issues is helpful in thinking about international securitization. This thinking by analogy becomes very obvious when it comes to one of the underlying premises of securitization: “insulation.” One of the primary objectives of participants in this area is to insulate the assets supporting the securitization from any adverse consequences that might flow from the seller’s insolvency. Rating agencies, referred to by some as “gatekeepers,” and sophisticated investors alike are well versed in the risks associated with insolvency proceedings under U.S. law and have developed structures and credit-enhancement features to mitigate these risks and ensure that the investors’ interest in the assets will not be affected adversely by a deterioration in the seller’s financial condition.

The question that may come to mind at this point is, How does one know whether the assets need insulation? In the United States, several aspects of insolvency law can create delay, lead to the termination of contractual relationships, or even result in disregarding the investors’ interest and the recapturing of assets (including cash flow) by the bankruptcy trustee, all of which would present barriers to a successful securitization. With respect to international securitizations, the gatekeepers typically ask whether these risks are present; they expect to be told not only about the risks of which they have become aware under U.S. law, but also about anything else that should concern them. The following is a brief description of the major U.S. insolvency risks.

Automatic Stay

The commencement of a bankruptcy proceeding under the U.S. Bankruptcy Code operates as an automatic prohibition against the taking of action against the debtor or against property owned or controlled by it.1 Thus, if the issuer of the securities became subject to a bankruptcy proceeding, investors could not gain access to the underlying assets, and the cash flow would be held up until bankruptcy court approval for continued payments could be obtained. Termination of the seller’s role as servicer would also be prevented. The need for such approval would, at the very least, entail temporary delay and under certain circumstances might not be obtainable until the bankruptcy proceeding was concluded.

Avoidance Power

A bankruptcy trustee has the power to “avoid”—that is, to set aside—unperfected interests in the debtor’s property (often referred to as the “strong-arm” power) and to avoid perfected interests that are considered to constitute “preferential transfers” or “fraudulent conveyances.”2

Strong-Arm Power

Under U.S. law, most interests of investors in securitizations can be perfected without undue effort or expense. Interests in receivables, for example, can be perfected by a fairly simple filing in one jurisdiction. Interests in securities and other instruments can be perfected by delivery or another method of transfer to the trustee acting on behalf of security holders. Interests in mortgage loans typically require delivery of the mortgage notes and recordation of assignments of mortgages to the trustee (recordation does, in a few jurisdictions, require payment of a recording tax).

One type of securitization in which all steps necessary to perfect the investors’ interests are sometimes not taken is the securitization of automobile receivables. The need in some cases to make a notation of the trustee’s interest on the certificate of title has been thought to be too cumbersome, given the number of automobiles involved and the effects of local (state) variation in these requirements. Moreover, the payment obligation of the automobile purchaser or lessee is considered the primary source of payment to investor; foreclosure on the automobile itself would be troublesome and possibly costly.

In the vast majority of cases, however, U.S. law provides a relatively painless and inexpensive method of ensuring that the investors’ interests will have priority over competing claimants and will be respected in the event of an insolvency proceeding. This legal situation may constitute another factor that has given the United States a head start in the securitization business.

Preferential Transfers

Under the U.S. Bankruptcy Code, the trustee can avoid interests created in favor of lenders to, and purchasers from, the debtor if (i) the interest conveyed were “on account of an antecedent debt”; (ii) the transfer occurred within the “preference period” (generally 90 days before the bankruptcy proceeding, but one year before if the transfer were to an “insider,” including an affiliate); and (iii) it enabled the transferee to receive more than would have been received if the debtor were liquidated.3 This description of preferential transfers avoids many technical issues and other nuances that may or may not be significant in any particular deal. Several exceptions to this avoidance power exist, most notably for transfers in which the debtor contemporaneously receives “new value” for the interest conveyed.4 This would occur if the assets were purchased for cash but not if the assets were conveyed to meet collateralization requirements. (Thus, additions of collateral would not be available until they were “seasoned,” that is, until the expiration of the preference period following the transfer.)

Fraudulent Conveyances

A related concern under the U.S. Bankruptcy Code is the trustee’s power to avoid transfers of property (and incurrence of debt) when the debtor is insolvent and receives less than “reasonably equivalent value” in exchange.5 In contrast to the preference provisions, transfers of collateral to secure existing indebtedness are not avoidable because antecedent debt constitutes “value” under the fraudulent conveyance provisions. Sales of assets would also be immune from avoidance as a fraudulent conveyance as long as the purchaser paid “reasonably equivalent value” for them.


The result of any avoidance of preferential transfers or fraudulent conveyances is an ability on the part of the bankruptcy trustee to recapture the assets or their value from the immediate or subsequent transferees (with certain limited exceptions). Thus, the value of cash flow received or additional collateral posted within the preference period or the applicable limitations period for fraudulent conveyances (one year under the Bankruptcy Code6 and up to six years under relevant state law) could be required to be returned. The presence of serious risk on this front would obviously make the securitization unworkable.

Not surprisingly, investors worry—and the rating agencies assume—that the originator-seller of the assets will become insolvent and that one or more of the foregoing principles will be applied to disrupt payments to investors (temporarily or permanently). Consequently, a fairly common structure has been developed in the United States involving a “true sale” to a separate, “bankruptcy-remote” entity (which investors and the rating agencies assume will not become insolvent), in order to insulate investors from these risks.

Establishment of Bankruptcy-Remote Issuer

Use of such special-purpose vehicles as bankruptcy-remote issuers has been a part of securitization almost since the beginning. Typically, a new corporation, partnership, or business trust is created under organizational documents that restrict the activities of the entity to those necessary to the securitization, most notably acquiring and caring for the assets and dealing with the cash flows. Other indebtedness is typically prohibited.

In addition to limiting the activities of the entity to prevent a direct bankruptcy filing, other steps are taken to ensure that the entity will be respected as a separate entity, notwithstanding the bankruptcy of its affiliates. U.S. bankruptcy trustees have the equitable power to order the “substantive consolidation” of the assets and liabilities of more than one entity. This power has typically been exercised when affiliates conduct themselves in a manner that shows disregard for their own separateness and is misleading to creditors.

Accordingly, in order to make it unlikely that the new entity will be substantively consolidated with its affiliates (whose bankruptcy is not “remote”), a number of actions are frequently taken. First, a separate office can be established for the new entity. Second, the new entity can be required to observe the necessary corporate formalities and to have at least one officer and director that is “independent” from all affiliates. Third, the entity can be prohibited from dealing with affiliates on other than arm’s-length terms, and generally prohibited from giving or receiving guarantees to or from affiliates. Fourth, the entity can be prohibited from having its assets commingled with those of, or otherwise held by, affiliates. Fifth, separate (typically audited) financial statements can be produced. Finally, the new entity can be “held out” as separate from the affiliates and its assets made unavailable to creditors of those affiliates.

When partnerships, rather than corporations, are used as the new entity, the general partner must ordinarily qualify as a bankruptcy-remote entity itself (meeting the criteria just mentioned). Obviously, when a business trust is used, a certain amount of adjustment must be made to satisfy these criteria—in spirit, if not in letter.

Absolute Conveyance of Assets to Special-Purpose Vehicle

The establishment of a separate, bankruptcy-remote entity does not of itself result in the requisite insulation. Once the appropriate type of entity is established, the assets to be securitized must be conveyed to that entity in such a way that they will not—for reasons other than “substantive consolidation”—be considered or become part of the seller’s property in the event of the seller’s insolvency. The assets must, therefore, have been “truly sold,” rather than pledged, to the new entity. If they have merely been pledged, they will remain property of the seller’s estate, and the automatic stay provision would prevent the special-purpose vehicle or anyone acting on behalf of investors from exercising control over the assets without court approval, as discussed above. There is some difference of opinion among U.S. law firms on the factors that would provide a basis for “recharacterizing” a sale as a secured borrowing. It is fair to say, in general, that the presence of direct or indirect recourse against the seller is often at the center of this debate.

In addition, the initial and any subsequent conveyances must not be avoidable as preferential transfers or fraudulent conveyances. For this reason, the seller cannot be obligated to convey additional assets except for “new” and “reasonably equivalent” value. In many securitizations, the initial conveyance of assets involves both a purchase by the new entity—using the proceeds of the securitization—and a capital contribution by the seller (assuming, as is usually the case, that the seller is the parent). Normally, capital is contributed to give the new entity sufficient “excess coverage” to make the securitization viable from the point of view of investors and rating agencies.

Financial guaranty insurance is also used fairly commonly in U.S. securitizations. Such insurance can be used to protect investors with respect to debt and pass-throughs, as the highly rated insurer takes the risk that the seller’s insolvency will prevent full and timely access to the underlying assets. These insurers are typically risk averse (and have their own ratings to keep) and, therefore, seek comfort on the very issues that would otherwise concern rating agencies and investors.

The law remains unsettled concerning the claims that holders of securitization may bring against the institution responsible for the securitization. Assuming that the holders of the securitized claims retain some residual claim against the bank that managed the securitization, the issue arises whether, in the event of the bank’s insolvency, its receiver has the power to stay and avoid claims that the holders of the securitization may have against the bank. State banks governed by state law in this regard may be subject to different rules. National bank insolvencies are governed by the National Bank Act7 in the first instance but become subject to the provisions of the Federal Deposit Insurance Corporation Act8 (insofar as the law requires the Federal Deposit Insurance Corporation to be appointed as the receiver of such banks if they are declared insolvent by the Office of the Comptroller of the Currency; the Federal Deposit Insurance Corporation may also become the receiver for state-chartered banks). Insurance companies are governed by the laws of the states in which they are chartered, and it is to these laws that reference must be made on these issues if an insurance company is responsible for the securitization.

International Structured Finance

International structured finance has been examined elsewhere.9 One of the earliest securitizations originated in Latin America (namely, the securitization of the receivables of Mexico’s national telephone company from AT&T). Bancomer has securitized certain MasterCard and Visa credit card receivables, and Mexico’s national petroleum company has securitized its oil receivables. In addition, the securitizations of various Mexican toll roads have been successfully marketed.

A certain impetus toward securitization in the international arena, as well as in the United States, has been generated by the Basle Capital Accord.10 In fact, some commentators expected even more of a surge in international securitization after it came into effect, but the path to securitization outside the United States has not been without stumbling blocks. For instance, to the extent that hedging against interest rate and currency risk is involved, the counterparties must be sufficiently creditworthy for investor and rating agency confidence. Sovereign risk may also come into play. More fundamental, however, is the need to analyze asset and seller risks under the laws of the relevant jurisdictions. The U.S. law analysis set forth above is a starting point in many cases, and it certainly informs the thinking of the rating agencies worldwide. The initial questions posed by rating agencies and investors often focus on whether the U.S. legal risks are present in the country in question. Sometimes, even the disclosure included in international securitizations reads very much like that included in U.S.-based transactions.

As briefly noted, the steps required to insure that the investors’ interests are insulated from claims of other creditors of the seller are often more cumbersome in other jurisdictions. In some countries, for example, the account debtor must apparently be notified for receivables to be sold. Failure to do so renders the investors’ interest at most one of an “equitable” nature, easily displaced by subsequent claimants. This has been the case in Japan, although the Ministry of International Trade and Industry has been instrumental in passing a new statute to create a public notice system for perfection without notification of account debtors.11 Wholly aside from the practical burden imposed by a notification system, an issue of psychology has historically generated resistance to the securitization of receivables in Japan, where the selling of a customer relationship has been viewed as an act of desperation by a failing company. Such a situation would, of course, not bode well for collecting on existing receivables, let alone generating new ones.

Worse, perhaps, than the burdensomeness of the steps required to be taken in some jurisdictions is the lack of certainty as to how the conveyance of the assets can be legally accomplished at all. However, the combination of pressures favoring securitization and the investment appetite for the products that securitization produces will continue to create an incentive for finding or creating greater certainty in an ever increasing number of jurisdictions.

    Other Resources Citing This Publication