The Dodd-Frank Wall Street Reform and Consumer Protection Act (the “Dodd-Frank Act”) was signed into law on July 21, 2010. Among many other sweeping changes to the securities and banking laws, section 941 of the Dodd- Frank Act inserted a new section 15G into the Securities Exchange Act of 1934, which, among other things, requires securitizers to retain some portion of the credit risk of any asset-backed security (“ABS”) they bring to market. Risk retention is the notion, born of the lessons from the recent economic crisis, and especially the housing market collapse, that those who sponsor securitization transactions should be required to retain a financial stake, or “skin in the game,” as a means to encourage honest dealing and adherence to prudent underwriting standards in the extension of credit.
As enacted, the Dodd-Frank Act leaves much of the detail to be supplied by key federal regulatory agencies. Congress did, however, leave critical signposts in the law to direct the agencies in their difficult task. In the case of the risk retention rules, among other instructions, Congress directed the agencies to differentiate among asset classes and provide appropriate exceptions to, and exemptions from, their rules to “(A) help ensure high quality underwriting standards for the securitizers and originators of assets that are securitized or available for securitization; and (B) encourage appropriate risk management practices by the securitizers and originators of assets, improve the access of consumers and businesses to credit on reasonable terms, or otherwise be in the public interest and for the protection of investors.”1
Section 941(c) of the Dodd-Frank Act also contains a requirement that the Federal Reserve Board (the “FRB”) study the impact of the new risk retention requirements on securitization, and issue a report detailing its findings and including “statutory and regulatory recommendations for eliminating any negative impacts on the continued viability of the ABS markets and on the availability of credit for new lending.” The FRB released its “Report to the Congress on Risk Retention” in October 2010 (the “FRB Report”).2 The FRB Report made several recommendations for the agencies charged with drafting rules and regulations to implement the risk retention rules, including the following: (1) consider the economics of different asset classes and structures in designing credit risk retention requirements; (2) consider the potential effect of credit risk retention requirements on the capacity of smaller market participants to comply and remain active in securitization markets; (3) consider the potential for other incentive alignment mechanisms to function as either an alternative or a complement to mandated credit risk retention; (4) consider that investors may appropriately demand that originators and securitizers hold alternate forms of risk retention beyond that required by the credit risk retention regulations; (5) the new rules may need to adjust over time to reflect the dynamic nature of the markets; and (6) care should be taken not to push these financial products into other venues with less stringent or nonexistent regulation.
On March 29, 2011, the FRB and the Federal Deposit Insurance Corporation (“FDIC”) approved a joint notice of proposed rulemaking (we refer to the notice and the proposed rules contained in it collectively as the “NPR”), which was subsequently approved by the Office of the Comptroller of the Currency, the Federal Housing Finance Agency, the Securities and Exchange Commission and the Department of Housing and Urban Development (collectively, the “Agencies”).
In this first iteration of their proposed rules, the Agencies, while acknowledging the existence of different asset classes, have conspicuously failed to follow through on Congress’ directive. The NPR provides for only one approach to risk retention: retention by securitizers of at least 5% of the ABS interests issued by the issuing entities in their securitization transactions. Although the NPR provides some flexibility for how this risk is to be retained, as we discuss below at least one corner of the industry—collateralized loan obligations (“CLOs”)—is going to find it highly impractical and costly to comply with the requirements of the NPR. Although the NPR provides an exception to the risk retention rules that seems to be crafted expressly for CLOs, it is largely unworkable, and thus virtually useless.
The NPR’s Risk Retention Scheme
As drafted, the NPR would generally require securitization sponsors to retain not less than 5% of the “ABS interests” issued by the issuing entity in any “securitization transaction.” The definition of ABS interest includes all but a limited category of securities and interests issued by the issuing entity. A securitization transaction is defined broadly to apply to all asset-backed securities, whether registered publicly or issued privately.
Pursuant to the NPR, a sponsor may retain its mandated risk in any one of the following five ways (there is also a sixth method, applicable only to revolving asset master trusts and not pertinent to this discussion):
- a 5% “vertical” piece of the ABS interests, retaining a pro rata portion of each class issued, akin to a slice of layer cake;
- a 5% “horizontal” first-loss position, i.e., a portion of the equity or subordinate debt tranche;
- an “L-shaped interest” whereby the sponsor holds at least half of the 5% retained interest in the form of a vertical slice and at least 2.564% of the par value of all ABS interests in the form of a horizontal first-loss position (excluding the assets constituting the vertical slice);
- an amount equal to the horizontal residual interest described in (2) above funded by the sponsor into a cash reserve fund account maintained by a trustee so that the sponsor experiences equivalent losses on the underlying assets as if it held a horizontal first-loss residual interest; or
- a representative sample, whereby the sponsor retains a randomly selected 5% representative sample of the assets that is equivalent, in all material respects, to the assets that are transferred to the issuing entity for securitization.
Exceptions and Safe Harbors
Several exceptions, exemptions and “safe harbors” have been carved out of the risk retention rules laid out in the NPR. Many of them are asset specific, and clearly not applicable to CLOs. As we will discuss below, however, there is an express exception from the risk-retention requirement for ABS backed by “commercial loans,” each of which has to meet a long list of very strict underwriting guidelines. There is also a foreign-related transactions safe harbor, which exempts from coverage certain non-U.S. issuers issuing ABS to a mostly non-U.S. investor base.
Commercial Loans Exception
The NPR’s exception for commercial loans excludes securitization transactions from its requirements if the ABS in question are collateralized solely by one or more commercial loans each of which meets rigorous underwriting standards. What follows is a summary of some of the more problematic requirements which, if followed, would hinder CLOs operationally, structurally and economically. Specifically, in order to qualify for the commercial loans exception from the 5% risk retention requirement under the NPR, a CLO manager must have determined that, for two years before inclusion in the CLO, the borrower under a commercial loan had (i) a total liabilities ratio of 50 percent or less; (ii) a leverage ratio of 3.0:1.0 or less; and (iii) a debt service coverage ratio of 1.5:1.0 or greater. The CLO manager must also be able to project that the borrower will continue to satisfy these very stringent tests following inclusion of the loan in its CLO. In addition, payments on the loan must come primarily from business revenues and be based on straight-line amortization over a term of no more than five years; the loan cannot have been made more than six months prior to inclusion in the CLO; the loan documentation must prohibit payments-in-kind; and, if secured, the loan must be backed by a first-lien security interest.
If it is determined after the CLO has closed that one or more loans did not satisfy the underwriting guidelines, the CLO will not lose the exception so long as the sponsor repurchases the ineligible loans from the CLO at par plus accrued but unpaid interest no later than 90 days after such determination is made. Lastly, the commercial loan exception is not available to securitization transactions that permit reinvestment periods.
Foreign-Related Transactions Safe Harbor
There is also an exception, or safe harbor, for foreign transactions that meet the following requirements:
- The securitization is neither registered nor required to be registered under the Securities Act of 1933 (the “1933 Act”);
- No more than 10% of the aggregate value of the proceeds of all ABS interests issued in the transaction are sold to or for the account of U.S. persons;
- Neither the sponsor nor the issuer itself is a U.S. entity, an unincorporated branch or office of a U.S. entity or an unincorporated U.S. branch of a foreign entity; and
- If the sponsor or issuer itself is a foreign entity, no more than 25% of the par value of the assets underlying the securitization transaction were transferred to the sponsor or issuer by a consolidated affiliate of either entity that is a U.S. entity or an unincorporated U.S. branch.
The foreign-related transactions safe harbor has an express prohibition on transactions that seek to evade its restrictions while remaining in technical compliance.
Restriction on Hedging, Financing and Transfers
What makes risk retention under the Dodd-Frank Act especially onerous is that the entity that retains the risk faces restrictions on hedging or transferring it. The NPR prohibits a sponsor from transferring any interest or assets that it is required to retain under section 15G of the 1933 Act to any person other than a consolidated affiliate, and prohibits a sponsor or any consolidated affiliate from hedging the credit risk the sponsor is required to retain under that section. The prohibition on hedging is limited to the credit risk the sponsor is required to retain; therefore, hedge positions that are not materially related to the credit risk of a particular ABS interest or exposure required to be retained, such as general interest rate risk, currency exchange rate risk or overall market movement risk are permitted.
Why the NPR’s Approach Doesn’t Work for CLOs
The NPR ignores the recommendations of the FRB Report (as well as the plain text of the Dodd-Frank Act) to distinguish among the different types of ABS and the vast universe of assets which underlie them. This leaves CLO managers, placement agents and other market participants wondering how their unique product will fit into the one-size-fits-all model envisioned by the NPR. The Agencies have, whether intentionally or not, failed to craft any exception or exemption that CLOs can realistically use, and it will prove very difficult if not impossible operationally for most collateral managers to retain risk as set forth in the NPR. This is especially unfortunate because, as discussed below, CLOs are already structured with a retained risk element and are, in any case, by nature a relatively low-credit risk financial product.
CLO managers working to ensure compliance with the risk retention rules will first have to determine which party has the obligation to retain the risk. The NPR’s discussion of this topic assumes a structure that would be typical of a residential mortgage-backed securitization transaction, where a sponsor originates or purchases assets in contemplation of a securitization and sells them to the issuer via a depositor entity. As in so many other respects, CLOs are different from other asset-backed products. Typically, a CLO manager does not own the assets prior to their securitization. It purchases them in the secondary market on the CLO’s behalf. Interestingly, in a footnote, the NPR assumes that the collateral manager acts as the sponsor and should therefore be the one who retains the risk.3 The Agencies’ stated intention in the NPR is to place the risk retention burden on the entity that has the most active and direct role in arranging a securitization transaction and selecting the assets to be included. Accordingly, in transactions where the CLO manager does not select the assets, e.g., where an investor or a placement agent retains a CLO manager to structure a CLO according to its own guidelines, perhaps such investor or placement agent ought to be the sponsor for risk retention purposes, or the CLO manager and such investor could agree to each retain some portion of the risk required to be retained. Another possibility was suggested in the FRB Report: since, as discussed below, the placement agent in a CLO may well have participated in the origination of syndicated loans included in such CLO, and may continue to hold a position in such loans, the placement agent’s position should be taken into account when calculating the risk retention for a CLO.4 One severe limitation of this approach is the fact that the placement agent and its consolidated affiliates would face restrictions on transferring or non-recourse borrowing against any loans incorporated into the risk retention calculation, or hedging the specific credit risk associated with such loans.
Impractical Risk Retention Options under the NPR
The horizontal first-loss piece is the only risk retention model permitted under the NPR that is workable for CLOs. Unfortunately, at 5% of ABS interests issued by the CLO, as specified in the NPR, it is unlikely that many U.S. CLO managers would be willing or able to commit so much of their scarce capital. This also applies to the reserve fund account model under the NPR.
The LSTA polled its larger CLO manager members to see how they would respond to the vertical slice approach. According to their survey, 87% of CLO managers polled lack the capacity and/or structure to retain a vertical slice as set forth in the NPR. Many U.S. CLO managers do not have large enough balance sheets to hold 5% of the ABS interests of their CLOs. Those that do, e.g., certain private equity funds or investment management units of broader financial institutions, have investment directives that would not be met by vertical slices of ABS, which may contain a distribution of as much as 70% highly rated bonds, which are likely to have a very low yield. Of the 13% who responded that they could comply, several managers indicated that they would be unwilling to burden so much of their limited capital.
The L-shaped risk retention approach suffers from its partial derivation from each of the horizontal and vertical slice models, neither of which is particularly workable for CLO managers.
The representative sample model outlined in the NPR is also unavailable for CLO managers, because it requires that the sponsor designate a pool of at least 1,000 separate assets for securitization. Most CLOs only package and securitize between 100-200 separate loans and, in fact, it is doubtful whether there are 1,000 eligible loans in the marketplace.
Limited Applicability of Exceptions and Safe Harbors
The exception apparently intended to cover CLOs—the “commercial loans” exception—is unfortunately so restrictive that very few CLOs could avail themselves of its provisions. The exception applies only to ABS that are backed solely (excluding cash and cash equivalents) by commercial loans that satisfy enumerated underwriting guidelines. This appears to exclude CLOs whose investment criteria permit investments in other classes of assets, such as high-yield bonds, ABS and synthetic securities. It also appears to exclude CLOs that enter into derivatives to hedge some or all of the credit risk in the portfolio, since a swap is clearly not a commercial loan.
The underwriting standards enumerated in the commercial loan exception are so conservative that they exclude most commercial loans in the market from its scope, including the loans of even the most creditworthy U.S. companies, such as AT&T, Wal-Mart, Johnson & Johnson, Verizon Communications, Time Warner Inc., Hewlett- Packard, Kraft Foods, PepsiCo and Deere & Co.5 It is also unclear whether the requirement that payments on the loan come primarily from business revenues will also serve to exclude loans whose credit documents permit prepayments sourced from asset sales or refinancings. As in other areas, the market would benefit from some additional clarity in the final rules. In addition, the prohibition on reinvestment periods conflicts directly with what is probably a CLO’s greatest strength—the retention of a dedicated asset manager whose job it is to diligently manage the portfolio in order to maximize returns, which will include attempting to capitalize on profitable arbitrage moves, as well as replacing nonperforming assets.
Possibly the most problematic aspect of the commercial loans exception is the requirement that the CLO manager repurchase ineligible loans from the CLO at par in order to retain the benefits of the exception. No rational asset manager purchases at par a distressed and possibly deteriorating asset. It is doubtful whether a CLO manager would even be permitted under applicable law to purchase for its own account assets owned by a fund under its management. This is certainly not compatible with historic CLO practice, in which nonperforming assets are sold by the CLO manager from the portfolio back into the secondary market at market price, and replaced with performing assets, purchased also at market price.
It appears that the foreign-related transactions safe harbor will only be available in certain extremely limited circumstances. Most CLO transactions are private issuances by offshore issuers. As drafted, this safe harbor or exception clearly excludes U.S.-based CLO managers and issuers. In addition, many CLOs are co-issued by offshore and Delaware co-issuers. Any CLO manager attempting to avail itself of the safe harbor will certainly have to abandon the co-issuer approach, and in the process possibly lose certain classes of investors as a result. Lastly, the 10% restriction on sales to U.S. persons is likely to prove restrictive and initially difficult to administer. The operational aspect of this limitation should, however, prove surmountable in the long run in cooperation with DTC and other clearing houses. In this regard also additional Agency clarification will be needed to determine whether the 10% limitation applies only at time of sale, or throughout the life of the CLO.
“ABS interests” as defined in the NPR (“any type of interest or obligation issued by an issuing entity”) could, without further clarification by the Agencies, be construed to pick up the fees which are payable by the CLO, including the indenture trustee’s and CLO manager’s fees, thus inadvertently increasing the risk retention piece.
Why Does It Matter?
The CLO market has shown signs of stirring since the first quarter of 2011. CLOs have historically provided a large source of credit to U.S. industry and commerce, in turn helping to maintain liquidity in the markets and keep down borrowing costs for corporate borrowers.6 If CLOs are unable to return to their historic role in the credit markets, it may take some time until other investors in the secondary loan markets are willing and able to pick up the slack, which portends a period of higher debt service costs for companies in a very fragile economic recovery. Alternatively, borrowers may be pushed into the high-yield bond markets to raise capital. To the extent that high-yield bonds are unsecured, investors who buy such bonds may be more at risk than investors who make secured loans, or who invest in CLOs backed by such secured loans. In any case, if CLOs go into permanent retreat, borrowers and capital markets will likely see resulting reduced efficiencies, as CLO managers, with their broad market knowledge, pull back, and are replaced by individual investors, each of whom has to analyze every individual investment they make.
Other reasons that CLOs deserve to be looked at differently from other securitization structures include the fact that they do not present the types of credit risk issues raised by “originate-to-distribute” classes of ABS products (e.g., loans that are underwritten often by brokers who retain no interest in the loans and who sell them off quickly into the securitization aggregation pipeline, collecting fees on the way). Commercial loans are thoroughly underwritten at issuance by a syndicate of lenders, and the administrative agent and other agents who led the underwriting of the deal often retain large positions in such loans. It is also quite likely that the placement agent for any CLO holds some for its own account. This means that those who underwrite the loans and package the ABS already have a huge incentive to make sure the extension of credit is done according to prudent standards. In addition, the CLO manager, who as discussed below has a duty to its CLO investors, also thoroughly reviews the loans it purchases for the CLO in the secondary markets. Pricing in the secondary markets for commercial loans is transparent (daily price quotes are often available from independent services). Often the borrowers of these loans have credit ratings from rating agencies, and are generally well-known entities that are often SEC-registered companies.7
A CLO manager is contractually required to purchase loans for the CLO using selection guidelines laid out in the CLO’s governing document. These guidelines are familiar to investors prior to the closing of the CLO and the issuance of its securities. In standard CLO deal documentation, a CLO manager who fails to perform its duties with the appropriate level of care may be replaced by the requisite number of investors.
It is also important to note that the fee structure of CLOs means that CLO managers already have significant “skin in the game.” In a typical CLO, the CLO manager receives the bulk of its fee only if the CLO performs.8 According to the LSTA, CLO managers typically receive a nominal annual base fee of between 10-20 basis points, which is senior to principal and interest on the notes issued by the CLO. A further portion (30-40 basis points) of the CLO manager’s fee is payable junior to the noteholders, but senior to the equity tranche. A final fee portion is payable after the equity tranche has recognized a targeted rate of return. Thus, the CLO manager clearly has a material incentive to ensure that the portfolio of loans performs well enough for all investors to be repaid in full, because only then will it receive the largest portion of its fee.
In addition, CLO managers are required, with few exceptions, to be registered as investment advisers under the Investment Advisers Act of 1940 (the “Advisers Act”).9 Accordingly, they already owe a duty to their investors under the Advisers Act and applicable state law. In practice, as the CLO manager observes that loans in the portfolio are starting to fail, it will look for opportunities to liquidate them, and will use the proceeds to purchase new performing assets. This active portfolio management aspect of CLOs is unique among classes of ABS.
At this juncture, it is worth mentioning the performance of CLOs during and after the recent financial crisis, in particular because the Dodd-Frank Act and the NPR are largely regulatory responses to the real and perceived recent failures of securitization. According to the LSTA, CLOs performed remarkably well during the recent crash: of a sample of more than 630 cash flow CLOs outstanding in the first quarter of 2011, there have been only two payment defaults at the CLO level, despite the much higher default rate on the underlying collateral (as high as 6.5% in June 2009).10 While the sector experienced some credit rating downgrades, many of these were attributable to changes in the rating agencies’ rating methodologies and, since early 2010, the rating agencies have been upgrading many tranches of CLOs. Interestingly, the FRB Report attributes the relatively sound performance of CLOs during the financial crisis vis-à-vis other securitized products to the transparency of CLO portfolios, relative simplicity of CLO structures, credit enhancements and incentive alignment.11 Other special attributes of CLOs identified in the FRB Report—such as conditional cash flow tests, overcollateralization and excess spread12—certainly also played a role in preventing a broader meltdown in this asset class. Whether this late vindication of CLOs arrives in time to influence regulators remains to be seen.
Interestingly, the Dodd-Frank Act’s definition of “asset-backed security” does not appear to capture synthetic CLO securities. Instead of incorporating Regulation AB’s broader definition of “asset-backed security,” the Dodd- Frank Act creates the following new definition: “a fixed-income or other security collateralized by any type of selfliquidating financial asset (including a loan, a lease, a mortgage, or a secured or unsecured receivable) that allows the holder of the security to receive payments that depend primarily on cash flow from the asset … ”.13 Synthetic CLOs generally allow investors to take long or short positions with respect to an underlying pool of loans via derivative instruments, but are not collateralized by them. Accordingly, if sufficient market appetite for synthetic exposure to pools of corporate loans returns, this investment product could benefit from exclusion from the risk retention rules of the Dodd-Frank Act. Of course, counterparties to the total return swaps that are at the heart of synthetic CLOs need to be wary of other aspects of the Dodd-Frank Act, and any CLO marketed to European banks may be impacted by Article 122a of the European Union’s Capital Requirements Directive.
It should be readily apparent that what is good for the goose, may not be good for the gander. A regulatory response intended to avert a recurrence of the ills wrought by the “originate-to-distribute”14 practices of certain parts of the securitization market may well end up restricting or even destroying a corner of the market that has been a powerful engine for economic growth and did not experience or cause the problems this legislation seeks to address. And, the loss of CLOs could have long-lasting negative effects on the availability and price of credit for American industry.
The LSTA, the American Securitization Forum and other trade groups are working with Congress and the Agencies to revise and clarify the rules. The comment period for the NPR ends June 10, 2011, after which it is hoped a clearer set of guidelines will emerge, and sufficient attention will be paid to the status of CLOs as a unique and successful standalone class of structured products lacking some of the riskier characteristics of other higher profile asset classes, while already possessing their own organic forms of risk retention. It is also to be hoped that there will be some recognition of the useful role CLOs play in the U.S. economy and that, accordingly, the final rules will either soften the risk retention rules outlined in the NPR where they apply to CLOs, or else carve out broader, more workable exceptions and safe harbors.