Will the Securitization Revolution Spread?
While securitization has dramatically altered the way many lenders do business, small business lending has barely changed
Published September 1, 1995 | September 1995 issue
... Many hurdles need to be overcome before commercial loan securitization becomes commonplace ... [but] it is by no means difficult to envision that a couple of decades from now, markets for business loan securities will be a reality ... Only those bankers willing to embrace the technological change will share in the benefits.
Federal Reserve Board of Governors
Chairman Alan Greenspan
Speech before the American Bankers
Association, October 1994
In "Casablanca," Capt. Renault orders his men to "round up the usual suspects" in an attempt to solve a murder. Policymakers took a similar tact in trying to identify the causes of the "credit crunch" in the early 1990s. Bank regulators were excoriated for being overzealous and examining institutions so aggressively that they limited credit availability. At the same time, bankers were accused of using local deposits to purchase Treasury securities instead of making loans.
While searching for the impetus for lower levels of bank lending, many policymakers focused on the plight of small businesses. After all, the perception of many elected officials and the public was that a credit crunch was curtailing economic growth by threatening the vitality of entrepreneurs and small businesses. This focus led many policymakers, supported by some financial institutions and small businesses, to insert an unexpected cause into the lineup of the usual suspects: the absence of a market for securitized small business loans.
Securitization is the process of repackaging loans and selling certificates, or securities, which entitle the owner to some or all of the repayments on the loans. More precisely, the loans are pooled and the cash flows from the loans in the pool pay off the interest and principal on the securities (see Diagram 1). This financial innovation has caused a revolutionary change in lending over the last 20 years in mortgage and consumer lending by allowing lenders to sell loans that they had previously held to maturity.
One- to four-family mortgages were the first loans securitized, in 1970, and there are currently over $1.7 trillion in securitized one- to four-family mortgages outstanding. Securitization markets grew to include non-mortgage assets in the mid-1980s, including auto loans (1985) and credit card receivables (1986). In 1994, about $31 billion in securities backed by credit card debt and $18 billion in debt backed by auto loans was sold. An even wider variety of assets has now been securitized, including boat loans, student loans and even delinquent property tax receivables.
This profound change in banking, however, has not extended to small business loans where securitization activity is extremely limited, involving about one-half of 1 percent of small business loans.
Why a market in small business loan securitization has been slow to develop
The rapid increase in the securitization of non-mortgage loans has been driven by market forces. More specifically, the explosion in securitization is partly due to the many benefits it produces for borrowers, investors and lenders and partly due to technological innovations that lowered the costs of securitizing loans. However, despite all of the potential advantages, small business loans have been resistant to securitization. Although exact numbers are not available, it appears that less than $900 million in small business loans have ever been securitized, while an estimated $155 billion in such loans were outstanding at year-end 1994.
Why have small business loans been absent from the list of securitized assets when securitizations of non-mortgage loans have spread so rapidly? A favorable relationship between market benefits and costs is responsible for the growth of non-mortgage securitization. As such, the lack of small business loan securitizations means that either the benefits of small business securitizations are less significant than those of other non-mortgage loans, or the costs associated with securitization remain comparatively high for these types of loans, or both.
The benefits of small business loan securitizations are similar to those of other assets. Former Federal Reserve Governor John LaWare testified before Congress that the securitization of small business loans could increase credit to small businesses, provide liquidity, fee income and asset diversification for banks and reduce their capital requirements, and offer investors a low-risk security with attractive returns.
In fact, the primary retardant to growth seems to be cost. Determining the expected cash flows from a pool of small business loans remains comparatively costly. These costs are driven by three factors:
- The need for specific borrower and local condition information. To make a reasonable decision on repayment prospects, investors need to acquire detailed information on the management quality of a small firm and on the local economy.
- The lack of standardization of loan terms. A pool of small business loans will be very diverse in terms of: the borrowers' line of business, the terms of the loan (for example, maturities, pricing, covenants and collateral) the loan documentation, underwriting standards and credit quality.
- The lack of long-term performance data. Unlike the home mortgage market, long-term data on the loan performance of small business loans rarely is available.
This lack of data, along with a lack of loan standardization, prevents the use of low-cost statistical methods to determine the expected performance of the pool and substantially raises the costs of the transactions. As a result, investors cannot be confident about their estimates of cash flow from the pool and must expend considerable resources in the analysis of the pools. Given other investing options, investors have been reluctant to take on the risk and cost of investing in small business loan pools.
Because of their reluctance take on the cost and risk of the securities, investors would demand that the seller offer protection from losses. This protection could be structured so that the seller either compensates the investor directly for losses or purchases a guaranty from a third-party financial institution for this purpose. These forms of loss protection eliminate the need for the investor to employ the resources to determine the risks of the loans. In addition, investors want lenders to directly bear losses from the pool to provide the lender with incentives to put high-quality loans in the pool. A lender originating and selling loans, rather than holding the loans to maturity, would not have cause to be as diligent as normal in assessing the credit quality of the loans.
Although loss protection assuages investors, it negates benefits that could accrue to the seller of the loans. For example, loss protection purchased from a third party will be very expensive for the lender as the guarantor has to recoup the costs of determining the assumed risk. Under bank regulatory rules, when the lender continues to bear the risk of losses from the loans, it is deemed not to have sold the loans. As a result, the benefit of lower capital costs cannot be achieved by banks that sell loans with a high level of loss protection. Moreover, by retaining most of the credit risk through the loss protection, the bank cannot use securitization of small business loans to reduce its overall exposure to losses. High loss protection requirements, as a result, eliminate much of the incentive for lenders to securitize their loans.
Bankers sometimes suggest they do not securitize small business loans because they make more money by holding these loans than by selling them. Initially, this sounds like a different explanation for the lack of market activity. However, the reason why banks earn such high returns on these loans is because of the disadvantages to securitization, specifically these loans are risky, information intensive and less subject to non-local competition. The higher returns merely compensate the banks for holding these relatively risky illiquid assets.
Steps to increase the market for small business
The lack of small business loan securitizations is a rational outcome of market participants weighing high informational costs relative to transaction benefits. Nonetheless, policymakers have offered several plans to "jump-start" this market. One recently enacted law would reform current regulatory policy. Another plan would make better use of existing information that the government collects on small business loans. Neither plan is likely to lead to significant growth of this market. Bigger increases in small business loan securitization could be achieved by using public resources to absorb the costs of these transactions. However, such government intervention would not improve the use of society's resources. In contrast, market actors are developing initiatives that could build the infrastructure and lay the groundwork for efficient and sustainable growth of this market. Two such efforts, described later, are taking place in the Ninth District.
Public-sector efforts to increase the number of small business loan securitizations
Because small businesses are considered an important source of economic output, the Congress has historically taken actions to increase the flow of credit to these entities. In this vein, President Clinton signed legislation in September 1994, which, in part, sought to increase the amount of credit flowing to small businesses by encouraging the growth of small business loan securitization. The Riegle Community Development and Regulatory Improvement Act of 1994 took two steps to achieve this goal: It granted favorable regulatory treatment to securitized small business loans (for example, it waived some restrictions on investing securitized loans), and it made changes in capital requirements to reduce the capital that banks selling small business loans have to hold.
This legislation recognized that high transaction costs were the primary cause of limited small business loan securitization. Indeed, the legislation could lead to more small business loan securitizations by reducing regulatory costs. However, the legislation does not reduce the far more significant informational costs associated with purchasing securitized loans. Thus, it is unlikely to lead to major volume increases in this market.
The government has other means of reducing the costs of these transactions. One proposed means of achieving this goal is the creation of a federally chartered, but privately owned corporation that would guarantee payment on securities backed by small business loans. Like other government-sponsored enterprises, such as Federal National Mortgage Association (known as Fannie Mae) or Federal Home Loan Mortgage Corp. (Freddie Mac), the corporation would have the implied financial support of the federal government. As a result, its guarantee would almost eliminate default risk on small business loan securities. Investors would no longer need to use resources to analyze the loan pools, and loan sellers would no longer need to provide loss protection.
The risks associated with these transactions would not have been reduced, however, but would be transferred to federal taxpayers who support the corporation's guarantee. Such a shift could benefit society if the corporation were able to more efficiently judge and bear the risks posed by the loan pools. But, the corporation would have no competitive advantage in these regards and moreover, as a pseudo-public entity, would have less incentive than private parties to diligently monitor potential costs. In addition, the creation of this corporation, with its federal support, could greatly curtail private innovation in this market.
To reduce information costs, federal agencies that already interact with small business loans or lenders could allow market participants to better exploit their contacts or data. For example, the Small Business Administration (SBA) could make more widely available its historical records on the loan repayment experience of SBA-guaranteed loans. In a slightly different tact, the Office of the Comptroller of the Currency (OCC) has proposed a program in which banks could request an OCC summary of how a bank's evaluation of credit correlates with the OCC's credit grading of the same loans. The OCC would then produce an "agreement table." The requesting bank can then pass out the agreement tables to investors to help them translate issuing bank credit terms into more familiar terms.
These programs have the advantage of adding few new costs as the agencies would already perform these functions. Both programs would also directly address informational concerns. Yet, if enacted, these plans are unlikely to greatly increase the number of small business loan securitizations. The SBA data may not be relevant for investors in evaluating loans that are not underwritten to SBA standards. The OCC would not be able to provide historical loss information for its loan classification system. Thus, investors will not be able to use these classifications to determine expected payouts from a pool of loans.
Ninth District activity related to small business loan securitization
No banks in the Ninth District have securitized small business loans nor have disclosed plans to do so. However, the largest banking organizations in the district, Minneapolis-based Norwest and First Bank System (FBS), are taking steps that could make such securitizations more likely in the future. These types of activities governed by private measurements of costs and benefits will ultimately lead to the expansion of the small business loan securitization market. The Ninth District is also home to an unusual program for purchasing and pooling commercial loans made by non-profits and local governments.
As noted above, lack of loan standardization is a major stumbling block for small business loan securitization. Internal loan standardization has become increasingly common among the largest banking organizations over the last five to 10 years. FBS has established three regional loan centers that are responsible for almost all aspects of their small business lending, including underwriting, documenting and servicing the loans. Because these functions have been centralized, FBS small business loan products are more standardized in terms of underwriting, terms and documentation. Moreover, this system allows FBS to better monitor the performance of their loan portfolio.
While the process undertaken by FBS is primarily driven by credit quality and cost concerns, the loan centers could have important implications for small business loan securitization. The more standard the small business loans of a banking organization and the better information available on their performance, the lower the cost of securitizing the loans. While FBS has not announced plans to securitize small business loans, and may never do so, it now has greater ability to undertake such a transaction.
A slightly different aspect of building the skill base for future securitization involves reducing the cost of analyzing loan purchases and making it more simple to purchase a diversified pool of loans. Here, Norwest is implementing a program called "Norwest Loan Partners," which uses existing skills to buy and pool loans.
In the Loan Partners program, Norwest purchases loans between $1 million and $10 million from a pre-selected group of banks. Norwest believes banks will sell the pool loans that would otherwise leave the banks exposed to the risk of loan concentration with one borrower or in a geographic region. Norwest will pool these loans with the intent of selling banks that originated the loans the right to receive the cash flows generated by the pool. The pool will be diversified by borrower type and geography.
If this program is successful it will allow a bank to sell off loans that pose concentration risks and to replace their income with cash flows from well-diversified loans, thus achieving a major benefit of small business loan securitization. The program, however, has grown slower than anticipated with fewer banks selling loans to be pooled. Although the Norwest program would not constitute full-fledged small business loan securitization, the experience gained from the program could make the sales of securities backed by these pools a more viable option in the future.
Outcomes from increased small business loan securitization
Given a long enough time horizon, small business securitization will be a much more common transaction. As Fed Chairman Greenspan noted in his speech to the American Bankers Association in 1994, the substantial benefits of securitization provide ample incentive for market participants to develop means to reduce the information costs of assessing a pool of these loans. In particular, the advantage of holding a security backed by a well-diversified pool of loans instead of a single loan should spur cost reductions. Moreover, competitors to banks that already fund their activities by issuing securities have a natural incentive to find new sources of funding and are likely to lead the development of this market. Market experience with other assets that were once considered hard to securitize (for example, credit card receivables) suggests that these hurdles can be overcome.
It seems certain that small business loan securitization will grow in the long run. But it is less clear that this form of financing will be as dominant in the small business loan sector as it is or will be in other loan sectors. Furthermore, while the growth of small business loan securitization should indicate the reduction of transaction costs, a higher volume of such securitizations will not produce net benefits for some banks or borrowers. For example, the increase in these transactions may not lead to additional credit to small businesses and may actually increase costs for some small businesses. The effect of this trend on bank profitability will also vary a great deal.
Small business loan securitization may never be dominant
The key to increased small business loan securitization is increasing standardization, which would bring down the costs of analyzing these loan pools. But, there are some attributes of small business operations and banking relationships that may inhibit standardization or low-cost analysis over the long term. Many small businesses and banks will resist standardization because the credit needs of small businesses vary more than most other borrowers and bank underwriting criteria for small business are, as a result, quite heterogeneous. Some small business borrowers may prefer, and in some circumstances need, lenders that are willing to maintain ownership of the loan and engage in frequent loan restructuring.
The loans of small business borrowers who are larger and have established predictable performance or that require less personalized structures will be the first to be securitized with other borrowers seeking out lenders who will not sell off their loans. By definition, many small businesses will fall into the second camp.
Small business loan securitization and the cost and availability of credit
Standardization of all business loans could limit small business access to credit. Former Fed Governor LaWare noted, "Standardization ... would introduce an element of inflexibility into small business lending and could preclude many small business firms from obtaining the credit accommodation they need because they do not fit the 'mold' ..." He also testified that securitization increases the need for documentation that could increase costs to small firm borrowers. As small business loan securitization becomes more prevalent, borrowers that do not want their loans sold off may also have to pay banks a premium to continue to hold illiquid assets.
Small business securitization and bank profitability
Securitization will allow lenders that are particularly efficient at one or more aspects of lending to generate fee income from securitization that matches or exceeds the revenue they generate from holding small business loans. For example, community banks may continue to have an advantage in originating and servicing loans to local, smaller firms because of their knowledge of local conditions, and of firms' specific characteristics, like quality of management. These banks could compete with non-bank lenders and they could continue to profit even if they sell their loans. Securitization could also allow community banks to originate loans, such as longer- term loans, that they would not make if they were not able to sell.
However, other banks' profitability could be hurt by the rise of small business loan securitization. For example, banks would find new competition to originate loans from non-bank competitors. Not only could banks originate and service fewer loans due to competition, but with information easier to process, the loans in which they are involved may produce less revenue. With a more standardized loan product, some banks could no longer have comparative advantages in making loans to small firms.
In the end, the market for securitizing small business loans will certainly grow over the next few decades without additional government assistance, as market participants lower transaction costs while exploiting benefits. The effect of this growth, though, will not be uniform or be viewed as an "improvement" for all small business borrowers or lenders.
How some small business loans have been securitized
Almost all the small business loans that have been securitized to date have been the unguaranteed part of loans partially guaranteed by the Small Business Administration. The biggest issuer of such securities is a subsidiary of The Money Store Inc., which has securitized about $380 billion in unguaranteed portions of SBA 7(a) loans. Much smaller issuers of such securities backed by the unguaranteed portion of SBA loans include First Western of Dallas; Emergent Business Capital of Greenville, S.C.; Zions First National Bank, Salt Lake City; and the Colorado Housing and Finance Agency. In contrast, Fremont Financial Corp., an asset-backed lender headquartered in Santa Monica, Calif., has securitized $330 million of its revolving commercial loans secured by assets such as account receivables, inventory and machinery.
These transactions prove that small business loan securitization is possible. But, the lenders and loans involved in these transactions have characteristics that eased the securitization process and may set them apart from other small business lenders and loans. For example, these lenders are perceived to have particularly clear and rigorous underwriting standards that are consistently applied. More generally, the SBA nature of most of the loans leads to fairly high loan standardization. In the same vein, the loans in these transactions were usually supported by similar types of collateral and documentation. In addition, the lenders were able to provide data on their loss experience with these types of loans. Finally, many of these issuers were very high volume lenders ensuring that they have enough loans to make securitization, which has high fixed expenses, cost effective. As such, these lenders were able to overcome many of the factors described in this article that limit the growth of this market. Even with these favorable characteristics, the lenders had to provide loss protection ranging from 20 percent to 11 percent of the loan pools, or purchase private bond insurance, to make the transactions acceptable to buyers.
Why asset-backed securitization has grown
Benefits for borrowers. Securitization allows lenders to routinely sell loans for cash. As a result, it allows lenders to make new loans and permits borrowers to indirectly tap new sources of funds. Thus, investors, like insurance companies and pension plans, which have only limited direct access to home and car buyers can provide additional funds to these classes of borrowers. The competition provided by these new buyers to purchase loans can also lower the cost of borrowing.
Benefits for investors. Securitization produces a new low risk debt instrument that can better match investors' maturity or risk/return demands without investors needing to acquire the skills necessary to make the loans themselves. Purchasing securities backed by loans also allows investors to diversify their investments.
Benefits for Lenders. Securitization can reduce various risks of the direct lenders. Lenders can sell their loans in securitized form and buy the loans of others also packaged as securities. This can provide geographic diversity and reduce business-line concentrations in loan portfolios and eliminate the need for lenders to hold loans to a single borrower. In addition, selling loans can reduce the risk that banks bear when interest rates change the value of the loans they are holding.
Securitization can also lead to cost reductions. For example, depository institutions that sell loans no longer have to meet "capital requirements" on the loans sold. This capital absorbs losses the bank incurs when assets fail to perform in accordance with their terms. Regulatory guidelines call for depository institutions to have capital protection of up to 8 percent of the amount of small business loans they hold. Lower capital holdings reduce regulatory expenses for depository institutions.
Finally, lenders can generate income from securitization by charging fees for originating loans, documenting the transaction or collecting payments. By dividing the lending process into its discrete parts, securitization allows lenders to focus on the aspect of lending they do best, making the lending process more efficient.
Lower costs. The costs of securitization have fallen with improvements in technology for producing and analyzing information. Investors who want to purchase loans, either as securities or as whole loans, need to understand the cash flows generated by the loans. They must estimate the probability that loans in the pool will default and understand the prepayment patterns of loans. When each investor had to analyze each loan in the pool and then manually aggregate the results, the costs of investing in securitized loans was prohibitive. But, the combination of computing power and statistical techniques have automated the analysis of pool cash flows, driving down the costs of securitization and making these transactions more feasible.
Pooling Commercial Loans Made by Non-profits and Local Governments
Since 1989, the non-profit Community Reinvestment Fund (CRF) has purchased "development loans" from non-profits and governmental agencies, often at a sizable discount. According to CRF, these loans support activities such as job creation or housing improvement. After purchasing the loans, CRF issues bonds whose interest and principal are serviced by the loan repayments, loan collateral and if need be a "credit reserve."
The credit reserve, along with overcollateralization, acts as the primary loss protection for the investors and generally equals 20 percent of the loans. The funds to finance the credit reserve have come from foundations and the state of Minnesota, among other sources. Banks have been a major purchaser of CRF-issued bonds in order to demonstrate community lending for purposes of the Community Reinvestment Act. In total, CRF has purchased $11 million in loans.