Securitisation Core Features and Early Developments

The basic tenet of "securitisation"1 is simple: the creation and issuance of debt securities for which the payment of principal and interest primarily depends on cash flows generated by an underlying pool of assets. The earliest securitisations came about in the 1970s when the U.S. federal housing authority, Government National Mortgage Association ("GNMA"), sought to increase the supply of housing loans by increasing their liquidity by means of providing guarantees of pass-through securities of pools of whole loans backed by HUD and Veterans Administration guaranteed mortgages. Fannie Mae and Freddie Mac, which are federally chartered organisations aimed at promoting home ownership, soon followed with similar guarantees that further increased the securitisation of mortgage loans.

The ease of trading securities issued by such pools represented a significant improvement over the trading of whole loans and other illiquid assets. Securitisations soon became repositories of innovation and additional features aimed at addressing identified economic and regulatory issues. For example, fixed-rate mortgages carry a significant prepayment risk (i.e., when interest rates decrease, borrowers refinance), making traditional pass-through securities unattractive for many investors. To address this problem, GNMA introduced securities in 1983 with different tranches tied to varying rates of principal prepayment. As the regulatory landscape changed, additional innovation resolved emerging issues and captured a broader investment base.

Securitisations were given a significant further boost by the Tax Reform Act of 1986, which introduced the concept of a real estate mortgage investment conduit ("REMIC"), which was intended to facilitate the issuance of mortgage-backed securitisations by extending beneficial tax treatment to entities that issue securities with different levels of seniority, maturities and interest payments. The introduction of the REMIC paved the way for the nowubiquitous waterfall and associated tranching of risk.

The ability to create tranches of seniority, maturity and return characteristics with credit ratings as high as "AAA" meant that securitisations could be tailored to the specific needs of various investor classes. For example, beginning in the late 1980s, money market funds experienced explosive growth, and tranches with high credit ratings and specific liquidity rights (created through repurchase mechanisms) were developed to satisfy the investment mandates for such funds imposed by the U.S. Investment Company Act, as amended, and rules promulgated thereunder (the "ICA").

Securitisations began to emerge for non-mortgage asset classes, beginning in 1985 with the $192.5 million securitisation of computer equipment lease payments by the Sperry Corporation. As the demand expanded, a liquid financing market for all types of consumer loans, mortgage loans and other traditionally illiquid assets emerged, and the supply of assets underlying securitisations increased. Essentially any asset having a predictable cash flow can be securitised; today, there are securitisations covering asset classes from the mundane to the esoteric, in industries ranging from airlines and automobiles to retail, trucks and heavy equipment. In 1997, we saw the advent of synthetic securitisations, where the underlying asset being securitised was an instrument with a pay-out that depended on the performance of assets that had not been transferred to the securitisation but that continued to reside elsewhere (initially as a tool to lay off risk of a pool of loans that continued to reside on banks' balance sheets). By de-coupling the physical asset from the securitisation, it became at least theoretically possible to securitise a single asset an infinite number of times.

Developments Between 1999 and 2007

Most early securitisations were funding transactions, where the originator selected the assets to be securitised, and, once transferred from the balance sheet of the originator to the securitisation special purposes entity ("SPE"), the assets were purely passive. Any subsequent transactions in the securitisation vehicle not dictated by the sponsor were pre-programmed so as to remove investment discretion that otherwise would have to be exercised by an investment manager.

Over time, those fundamentals began to change. The SPE progressed from a "brain-dead" entity with a role "limited to the collections of principal and interest on passive assets and distribution of the cash flows to the beneficial interest holders in the entity based on a predefined formula"2 to one of active management of both the assets and liabilities, in an effort to achieve maximum return on equity. No longer were most securitisations funding vehicles; rather, they were designed to arbitrage the risk-weighted regulatory capital requirements imposed by Basel I.3 Under Basel I, banks were required to hold capital equivalent to 8% of their risk-weighted assets. Securitisation allowed them to take assets with high capital charges and effectively exchange them for senior tranches of the securitisations, which had relatively higher ratings and thus lower capital charges. The lower tranches would of course carry high capital charges, but it was often possible to structure a transaction with lower total capital requirements than the underlying assets being securitised. Basel II replaced the fixed-percentage capital requirements of Basel I with a "value at risk" ("VaR") methodology, which (at least in theory) allowed for the risk weight of each particular asset to be tailored to the particular characteristics of that asset, but it did not alter the fundamental benefit that securitisation offered to banks attempting to reduce their capital charges. By the mid-2000s, the securitisation market was a multi-trillion dollar industry, with issuance approaching $4 trillion annually, and was a leading tool of corporate finance. Securitisations provided liquidity to markets for assets that could otherwise not be sold. Securitisations also provided new sources of funding, essentially increasing the supply of credit available to the market as a whole while dispersing the associated credit risk. Auto loan securitisations, for example, enable purchasers of cars in the U.S. with subprime credit to obtain loans that would otherwise either not be provided at all or come at an exorbitant cost. Indeed, the capital markets rely on securitisation as a vehicle for providing credit to, and dispersing the credit risk of, many of the corporate debtors throughout the world. According to the Federal Reserve, by the late 1990s, traditional banks no longer provided the majority of credit in the U.S.; today, the nonbank financial sector (which is largely composed of securitisation vehicles) makes up more than two-thirds of the total credit markets.4

The explosion of securitisation issuance helped drive some of the innovations in the industry that have been much criticised as a significant contributor to the economic crisis. In addition to those mentioned above, securitisations – particularly MBS5 and CDOs6 – ceased being vehicles simply for banks to transfer the credit risk of originated assets off their balance sheets, and became key sources of revenue for underwriters. As asset classes dried up (because they were being sourced by so many securitisations), those banks had strong incentives to develop alternative asset classes and/or techniques to protect their underwriting revenues. This led to the securitisation of less traditional and more risky asset classes (e.g., subprime residential mortgage loans) and ultimately to securitisations of securitisation exposures, sometimes multiple times. Additional demand was driven by an active repo market, which required highly rated securities for use as collateral. CDOs of securitisations came to be used as a means to "slice and dice" normal securitisations to create additional highly-rated paper.7 Through CDOs and synthetics, the default of a single asset could impact multiple securitisations, causing unprecedented correlation and a complete breakdown of the benefits of diversification. The overheated demand for assets also drove deteriorating underwriting standards and, as outlined below, a key goal of post-crisis securitisation-related legislation is to align investors' and sponsors' interests in maintaining adequate underwriting standards.

The Securitisation Market 2008 and beyond

In the aftermath of the global financial crisis that began in late 2007, securitisations have been heavily scrutinised for their role in causing or augmenting the crisis. The collapse in the home mortgage market was a key cause, and the securitisation market was identified as key enabler. The criticism extended beyond mortgagelinked securitisations (for the purposes hereof, considered to include RMBS, CDOs, CDO squared and synthetic mortgage-linked CDOs) to include even securitisations of well-performing asset classes such as automobile loans, credit cards and corporate loans.

The scrutiny, suspicion and post-crisis regulatory environment was a full reset button for the securitisation industry. Some products, such as CLOs, are back to pre-crisis levels; however, the industry as a whole is continuing to innovate, this time to address regulatory and market demand for more transparency and less complexity. Below we highlight some key drivers of innovations and adjustments within the securitisation industry.

i. Basel III and Risk Retention Requirements as Drivers of Securitisation Features

Post-crisis, the entire capital regime for banks and other financial institutions has been subject to significant change and scrutiny, with the overall effect of correcting for (or in some instances, over-correcting for) the perceived shortcomings of the pre-crisis capital regime. This has resulted in the disappearance of certain securitisation features and the emergence of others and is likely continue to drive the innovation of additional features in the near future.

a. Re-securitisations and Re-tranching

One obvious change driven by the revised capital rules has been to end synthetic securitisations. The demand for CDOs has largely evaporated as a result of the current market aversion to excess complexity and securitisations that are difficult to value. In addition, given the large risk weights assigned to re-securitisation exposures and the inability to use the internal ratings based approach ("IRBA") for such exposures, the demand for such securities will likely remain infinitesimal, even should investors otherwise be willing to overlook their inherent risk concentration and the high embedded correlation. On the other hand, re-tranching of individual securitisations may be increasingly used to adjust the credit risk of a securitisation exposure to optimise an investor's capital purposes. Bank regulators have determined that the re-tranching of a single securitisation does not result in the increased correlation risk and complexity seen in the re-securitisation of multiple underlying securitisations, and they are thus not subject to the same punitive regime.

b. Disappearance of Monoline Wraps and Ratings as Proxies

In some pre-crisis securitisations, a monoline insurance company would write an insurance policy on the performance of the senior (and sometimes mezzanine) tranche of the securitisation; this "wrap" would guarantee the ultimate or scheduled payments of principal and scheduled payments of interest. A wrap by a "AAA" rated insurance company would increase the rating of the senior tranche to "AAA", which meant that a lower spread would be required to sell that tranche of notes. Transactions were structured such that the premium paid to the monoline for providing a wrap was less than the spread payable on the equivalent unwrapped tranche (which would have a corresponding lower rating), thus (theoretically) increasing the return on equity achievable by the securitisation. The revised Basel rules do not permit securitisations to take into account any credit enhancements provided by insurance companies predominantly engaged in the business of providing credit protection (such as monolines).

Furthermore, the possibility that ratings ascribed to such enhancements may push the relevant exposure out of being able to rely on the external ratings-based approach ("ERBA") in countries that otherwise permit the use of such external ratings for purposes of determining capital requirements, will likely further negatively impact demand for such guarantees or credit enhancements. Credit enhancements that effectively come from the bank itself also will be discounted and may not be taken into account as part of the ERBA. For example, whereas a rating agency may ascribe a ratings-enhancing effect to various forms of support for a securitisation, a bank is not permitted to rely on the credit-enhancing effect of such support in determining its own risk weight.

Following the default of many highly rated entities and instruments during the financial crisis, regulatory and investor confidence in the rating agencies' ability to accurately assess and monitor the creditworthiness of an entity or instrument suffered. Part of this failure resulted from an oversimplification, or outright misunderstanding, of the assumptions being used in the Gaussian copula, which was the theoretical underpinning of the securitisation market in the 2000s.8 Essentially, the correlation model used by the rating agencies vastly underestimated the level of correlation in a declining market. Initially, small downward changes in the market began to have a bigger effect than the model indicated they should have, a phenomenon that only accelerated as the markets deteriorated. In the end, the entirety of the financial markets proved to be vastly more correlated than had been assumed. The U.S. congress responded to the perceived failure by the rating agencies to accurately forecast this result by eliminating the various ratings requirements in the regulations governing various participants in, and products of, the financial markets (importantly, including by the elimination of rating-based approaches to determining regulatory capital requirements under Basel III).

Going forward, there will undoubtedly be a demand for guaranteetype credit enhancements, especially if such enhancement cannot be eroded other than for reasons relating to losses in the underlying assets. It is possible that other qualifying guarantee providers will step into the space left behind by the monolines to provide certain credit enhancements. Such guarantee providers could either be insurance companies (or reinsurance pools) not deemed to be primarily engaged in writing financial insurance or pools of assets capable of providing the necessary enhancement. This is likely an area where there is room for significant innovation, not only as part of securitisation in the narrow sense but also as part of creating an optimal risk allocation structure under the revised Basel rules. Furthermore, without the assistance of the monolines to help drive underwriting standards, and lacking trust in the ratings, investors have been much more involved in the deal-making process, demanding both transparency and relative simplicity. Even in managed transactions, many investors have sought to limit the discretion of the manager. A new entrant into the market has also emerged – a "deal agent", whose role, while still evolving, is essentially to look over the shoulders of the principals to the transaction to ensure that the rules of the transaction are being followed.

c. Increased Asset Level Disclosure

One of the core post-crisis initiatives is to increase the amount of asset-level disclosure and emphasise transparency and simplicity, which is driven by a number of different regulations as well as investor demand. In the U.S., registered offerings of many securitisations have to be made in compliance with Regulation AB II ("RAB2"), which became effective in November 2015 (although the biggest change introduced by RAB2 – requiring up-front and ongoing asset-level data points about the pool assets backing the securities – does not become effective until November of 2016). RAB2 is similar to the prior asset-backed securitisation rules in that the disclosure requirements do not apply to securitisations that are offered in reliance on Rule 144A or otherwise pursuant to an exemption from registration. RAB2 eliminates the common market practice of using a two-part disclosure document (consisting of a base prospectus and a prospectus supplement) in favour of a single, integrated prospectus, and introduces a certification requirement by the chief executive officer of the entity "depositing" assets in the transaction.

While it remains unclear how the changes introduced by RAB2 will impact the overall volume of issuance,9 certain themes have emerged. First, the asset-level disclosure requirements have elicited privacy concerns, both from the perspective of the consumer (for example, under Graham-Leach-Bliley) and the enterprise (which often views the ability to identify clients as a competitive advantage). The regulators attempted to stem consumer privacy concerns by requiring disclosure of only the first two digits of the zip code of a mortgaged property. Ironically, some issuers have been willing to provide fivedigit zip code information in private transactions (where investors are subject to confidentiality restrictions) where they are not willing to provide the leading two digits in public transactions. Although these asset level disclosure requirements cut, to some extent, against the longstanding view that they are below the materiality threshold when looking at the securitised pool as a whole, there is a long tradition for Rule 144A offerings to produce disclosures that are as close to the requirements of a registered transaction as practicable.

We therefore expect investor demand for asset-level disclosure to increase as investors build models to analyse registered deals subject to RAB2 and seek to apply those models to other securitisations. We are seeing the emergence of third party due diligence providers filling a function that until the crisis was occupied by monolines and rating agencies. The growth of these types of service providers will also likely drive a demand for more asset-level disclosure, in part as a means of producing the information required for a thorough due diligence examination and in part because more information tends to justify the hiring of an outside due diligence consultant. Second, we are seeing some issuers publicly offer only the "AAA" tranche of securities, with the lower tranches being privately offered pursuant to Rule 144A, where they are not subject to the CEO certification requirement. In other words, the efforts of the regulators appear to have the effect of pushing at least some of the public market into the private market, where the regulations have no direct impact. This might be another area where we see investors demand a third party service provider to provide additional diligence.

The Basel III capital rules are also expected to be significant drivers for increased asset-level disclosure. In order for banks to be permitted to calculate their capital requirements using IRBA, the banks are required to collect sufficient information for these models to be deemed adequate. In addition, the Basel III capital rules have a fall-back capital requirement of 1,250% if the bank is viewed as not having done sufficient diligence on the relevant securitisation exposure. The information required for a bank to use the IRBA, coupled with this due diligence requirement, incentivise increased disclosure as well as simplicity both in the nature of the underlying assets as well as in the securitisation structures, such as the waterfall and various triggers. As pointed out in the latest Basel Committee consultation: "A bank must have a thorough understanding of all structural features of a securitisation transaction that would materially impact the performance of the bank's exposures to the transaction, such as the contractual waterfall and waterfall-related triggers, credit enhancements, liquidity enhancements, market value triggers, and deal-specific definitions of default."10 Similarly, in the U.S. the due diligence requirement dictates that "the banking organisation's analysis would have to be commensurate with the complexity of the exposure and the materiality of the exposure in relation to capital of the banking organisation".11

As detailed as the disclosure requirements under various securities regulations may be, it is possible that the granularity required by the banks may be higher still, driven by the inputs required for use of the IRBA. These disclosure requirements coincide with various simplification and transparency initiatives, such as the BCBSIOSCO work on developing criteria for simple and transparent securitisation structures, "to assist investors with their due diligence on securitisations . . .".12

d. Risk Retention

In what is viewed by many as the single most important change to the securitisation market, the risk retention rules in both the U.S. and Europe impose an obligation to retain a certain amount of risk on the various participants in a securitisation. From a policy perspective, the rationale is straightforward: the more "skin in the game" a sponsor has, the more prudent it will be in structuring the transaction and sourcing and diligencing the assets. For some securitisation transactions, particularly those using the originator model, this rationale may have some merit; however, for other asset classes, and particularly for non-funding CLO transactions, the rationale disintegrates somewhat. In the context of a CLO, a manager selects a portfolio of loans to be acquired in the secondary market. In order to obtain the ratings necessary to place the CLO liabilities, the manager must work closely with the rating agencies to develop portfolio characteristics (loan quality, diversification, average coupon, and so on) that the acquired loans must satisfy. The manager is paid based (largely) on the performance of the transaction – it does not need (nor does it necessarily have the capital) to hold the regulatorily-required risk in the transaction in order to be properly incentivised, by virtue of its interest in maximising fees (not to mention its reputation), to do its best to ensure the success of the transaction.

The risk retention requirement is imposed in Europe indirectly through additional capital requirements for covered institutions that hold non-compliant securitisations, and in the U.S., directly through the risk retention requirements that prohibit securitisations that fail the retention requirements. Both jurisdictions effectively require the retention of risk in the form of either a 5% "vertical slice" (i.e., 5% of each tranche of securitisations offered), a 5% horizontal slice (i.e., an investment in the most junior of at least 5% of the market value of the entire securitisation) or a combination of the two. In addition, risk retention rules in the U.S. provide for some alternative risk retention structures for certain specific types of securitisations, and exempt some securitisations from the risk retention rules altogether if they either (i) consist solely of U.S. government, or agency, securities, or (ii) include only assets meeting specific origination criteria.

Risk retention rules have been in place in Europe for a few years, although the rules and their application are still evolving. In the U.S., risk retention rules have become effective only with respect to RMBS, and will become effective for all securitisation beginning December 24, 2016. However, one thing has become clear: risk retention is costly. A fair amount of innovation therefore has focused on financing risk retention. Securitisation managers in the U.S. have discovered that investors are less willing to invest in securitisations even before the rules become effective, unless the manager can demonstrate an ability to effectuate a refinancing that complies with the risk retention rules. Large, well-known managers with independent ability to access the capital markets have an easy time demonstrating this capacity. Smaller or less established managers have to demonstrate, at least in broad strokes, the ability to establish a feasible risk retention structure. We expect that this will ultimately result in fewer managers coming to market, and may result in more "sticky" money as investors develop confidence in one or few managers, and invest in multiple transactions with them (in these ways, a return to the CLO 1.0 model).

The risk retention rules in Europe were originally viewed as focusing more formalistically on risk retention by the entity that sells the securitised assets to the securitisation vehicle (the "originator"), which resulted in attempts to establish special purpose risk retention entities that would be capitalised by third parties and would then be inserted into the origination chain as the entity that acquires assets from various other originators and sells them into the securitisation. Subsequent guidance from the European Central Bank has criticised these structures, and this type of special purpose risk retention entity is therefore less likely to be acceptable, as Europe has moved more towards a risk retention model where the risk retaining originator also must act as "sponsor". The sponsor in this context must be an entity empowered with real decision-making authority (in terms ofasset selection) and is the risk retention model we have in the U.S..

It is possible for a securitisation to have multiple sponsors; but the rules in Europe and the U.S. require that only one sponsor retain the risk. In the U.S., risk may be retained by a majority-owned affiliate of the sponsor, whereas it is not clear whether a majority-owned affiliate is permitted to retain the risk in Europe. There are a number of possible compliant structures in each of the U.S. and Europe, but when looking at the intersection of the key goals, the field of optimal solutions is much smaller – the key goals being using an entity (a) that can obtain the maximum amount of financing for the least cost in the market, (b) with minimal permissible recourse to the sponsor, (c) that is compliant in Europe as well as in the U.S., and (d) that ideally can be used for multiple sponsors so as to quickly achieve scale. The jury on the optimal structure is still out, but our prediction is that it will involve establishing a financing vehicle with real decision power over the asset selection of underlying securitisations. We expect the underlying securitisations to have a sub-manager that selects various investments subject to the financing vehicle's approval, ensuring that the financing vehicle maintains its status as sponsor. In European transactions, the financing vehicle would be the entity that, in consultation with the sub-manager, aggregates the assets to be securitised and sells these assets to the securitisation vehicle; in the U.S., that formal step would likely not be viewed as necessary and the assets could be purchased through traditional channels.

The risk retention rules have also become an important driver for innovation through legislative action. The various origination criteria that, if complied with, reward securitisations with not having to retain additional risk, are largely the result of significant industry input into the rule-making process. Currently, there are proposals for establishing criteria for qualifying collateralised loan obligations that will, if complied with, permit sponsors of compliant securitisations to be deemed compliant with the risk retention requirements by means of retaining only 5% of the equity, which is a far less costly proposition than the current requirement of retaining equity equal to 5% of the market value of the entire securitisation.

e. Securitisation Innovations to Effectuate Balance Sheet Optimisation

As outlined above, under Basel I securitisations became an important tool to maximise capital relief in response to the insensitive Basel I risk-weightings, by allowing banks to tailor tranches with variable risk/return characteristics. The evolution of securitisations, particularly synthetic securitisations, was influenced by the need of a bank to retain ownership of underlying assets while transferring the credit risk, thereby reducing its capital charges.

Securitisations are likely to continue to play an important role as a balance sheet optimisation tool, and the revised capital regime and various post-crisis rules and restrictions will significantly shape securitisation structures going forward. However, it is worth noting that while senior securitisation tranches tend to have comparatively lower risk weights, junior tranches tend to have relatively higher risk weights, and the sum of the junior and senior risk weights exceed the risk weight that would have been assigned to a non-securitised exposure to the underlying asset pool.

Changes in accounting rules following the financial crisis have made it more difficult to deconsolidate securitisation SPEs where a sponsor retains too much control and residual exposure. Deconsolidation of underlying assets is one of the operative requirements under Basel III for transferring credit risk using a traditional securitisation, but the same is not required for transferring credit risk when using a synthetic securitisation structure. However, market scepticism to synthetic securitisations results in additional costs that often render such structures uneconomical. It is therefore likely that a new form of "synthetic" securitisation may be developed which on the one hand satisfies the applicable Basel III criteria but which on the other hand assures investors that the correlation risks and failures of precrisis synthetic securitisations will not be repeated.

The high cost increases that the banks face by virtue of the leverage ratio requirements (requiring capital to be retained based on a relatively crude measure of total leverage) and the costs imposed on banks making lending commitments through the liquidity coverage ratio (requiring banks to have investments sufficient to cover expected net funding commitments over a rolling thirty-day period multiplied by a factor that can be as high as 100%) will together likely also drive significant innovation in the securitisation industry. For example, in Europe, there are proposals for high quality securitisations that would be structured according to common, transparent and conservative criteria such that they would qualify as permissible investments to satisfy the liquidity coverage ratio. The demand for this type of structure, and the resulting improved lending conditions for the target groups benefiting from such legislation, is likely to be significant and could be an important driver to increase the demand and supply side of European securitisations.

ii. Volcker

In the U.S., the efforts to disentangle banks and their affiliates from the perceived risk of proprietary trading and knock-on effects of avoiding investments in investment vehicles that operate similarly to hedge funds, culminated in the Volcker Rule. The Volcker Rule restricts banks and their affiliates from engaging in most proprietary trading and also restricts banks and their affiliates from sponsoring or taking an "ownership interest" in any entity that is deemed to be a "covered fund" for the purposes of the Volcker Rule. A covered fund incudes any entity that relies on Rule 3(c)(1) or 3(c)(7) under the ICA (permitting up to 100 investors or requiring each investor to be a "qualified purchaser", respectively) or that relies on Rule 4.7 under the Commodity Exchange Act for an exception from registration as a commodity pool and that does not otherwise have an exemption available to it. The definition of ownership interest is very broad, and includes any interest that directly or synthetically has return characteristics or voting rights similar to equity, including the right to remove the investment manager outside an event of default.

The Volcker Rule has effectively driven securitisations to either comply with the restrictions of Rule 3a-7 under the ICA or rely on 3(c)(7) under the ICA and avail themselves of an alternative exception to the Volcker Rule. Actively managed CLOs and other funds that cannot rely on Rule 3a-7 tend to rely on the loan securitisation exclusion to the covered fund definition under the Volcker Rule, which permits securitisations backed solely by loans and a few other permissible assets (such as interest rate hedges). As a result, actively managed CLOs have eliminated the bond basket (traditionally around 5%) that permitted managers to engage in certain arbitrage trades between high yield bonds and leveraged loans.

Although Volcker Rule issues are relatively settled for CLOs that are able to qualify for the loan securitisation exclusion and for securitisations able to rely on Rule 3a-7, Volcker Rule restrictions will need to be revisited if synthetic securitisations or securitisations that seek to capture more advanced arbitrage opportunities resurface.

At that point, it is likely that the industry will focus on the "ownership interest" prong of the Volcker Rule (a bank can hold a covered fund so long as its interest does not consist of an "ownership interest"). We believe that the most likely outcome is that the ability of senior noteholders to vote to remove the manager, absent an event of default, will be neutered and, instead, those manager-related events about which senior noteholders are truly concerned will be folded in to the events of default for the securitisation as a whole. This would eliminate the ownership interest in those senior notes.

Conclusion

Early innovation in the securitisation industry was aimed at creating a liquid market for illiquid underlying assets. Immediately prior to the crisis, innovation was largely driven by excess demand for securitisation paper, which arguably led to many of the oft-criticised features and misaligned incentives. Much of the innovation postcrisis has been aimed at curtailing these perceived excesses and shortcomings, and we expect that, going forward, innovation in the securitisation industry will be focused on creating securitisation structures that are both regulatorily compliant and satisfy investor demand for simplicity, transparency and other quality-enhancing features, while being optimised for a wide variety of underlying assets and investment strategies.

The Basel III framework, including the leverage ratio and net stable funding ratio, pressures banks to shed long-term assets and reduce risk-weighted assets overall. Capital requirements also drive divestitures but can be more readily managed by changing the credit quality of the underlying assets. Traditional securitisations provide a means for both removing assets from the bank's balance sheet and transforming the credit quality of the retained securitisation exposures. Market pressures are therefore such that the banks will be incentivised to shift assets and risks to markets with less stringent capital rules.

Despite other legislative initiatives that may significantly impact securitisations, such as risk retention requirements, and extension of capital requirements and liquidity and leverage constraints beyond the traditional banks to the so-called "shadow banking" sector, securitisations would still provide capital efficiencies by allowing banks to originate various underlying exposures, transfer the bulk of their exposures to non- (or less) regulated parties wishing to take the credit risk on the underlying exposures. The consultation issued by the Basel Committee and the International Organisation of Securities Commissioners ("IOSCO") to identify criteria for "simple, transparent and comparable securitisations"13 highlights the need for building sustainable securitisation markets by increasing investor demand. One likely result will be more favourable regulatory capital treatment for standardised and "high quality" securitisations.

The shift towards non-bank lenders and less regulated participants is coupled with increased demands for high-quality collateral. As confidence in the securitisation market returns, it is reasonable to predict that demand for senior securitisation exposures in well performing, familiar and established asset classes, will rebound and complement the banks' need to sell assets to remain in compliance with their capital regime.

Endnotes

1 One prominent definition of the term "securitisation" is "the carefully structured process whereby loans and other receivables are packaged, underwritten, and sold in the form of securities (instruments commonly known as assetbacked securities). As such, it is a subset of a broader trend seen throughout the capital markets for many years, "securitization", that is, the general phenomenon whereby more and more fund raising is occurring through the agency of securities". J. Rosenthal & J. Ocampo, Securitization of Credit: Inside the New Technology of Finance 3 (1988).

2 Containing Systemic Risk: The Road to Reform, The Report of the CRMPG 45 (6 August 2008), available at: http://www. crmpolicygroup.org/docs/CRMPG-III.pdf.

3 See generally Basel Committee on Banking Supervision, Capital Requirements and Bank Behaviour: The Impact of the Basle Accord (April 1999), available at: http://www.bisorg/publ/(April 1999), available at: http://www.bis. org/publ/bcbs_wp1.pdf.

4 The Importance of the Nonbank Financial Sector, speech by Vice Chairman Stanley Fischer (27 March, 2015), available at: https://www.federalreserve.gov/newsevents/.

5 Mortgage backed securities ("MBS") are, as their name implies, securitisations of commercial and residential mortgages ("CMBS" and "RMBS", respectively).

6 Collateralised debt obligations ("CDO") is an umbrella term sometimes used synonymously with its components: collateralised bond obligations, or "CBOs", which are securitisations the underlying collateral of which is primarily corporate bonds; and collateralised loan obligations, or "CLOs", the underlying collateral of which is primarily corporate loans. Traditionally, a CDO is a combination of the two – a securitisation of corporate loans and corporate bonds. CDOs tend to be non-funding transactions, though certain funding variants exist, such as CLO transactions securitised from the balance sheet of the originator and securitisations of certain trust-preferred securities ("TruPS"), usually – but not always – issued by banks as a way of improving their tier one capital ratios.

7 See Gary Gorton & Andrew Metrick: Securitised Banking and the Run on Repo (9 November 2010), available at: http://ssrn.com/abstract=1440752.

8 See Sam Jones: The Formula that Felled Wall St., The Financial Times (24 April 2009), available at: http://www. ft.com/cms/s/0/912d85e8-2d75-11de-9eba-00144feabdc0. html.

9 We note that the market for RMBS post-crisis is a fraction of what it was pre-crisis. This is, of course, partially due to the fact that, pre-crisis, there were a lot of a mortgage loans that should not have been underwritten in the first place. Another important distinction in the post-crisis qualifying RMBS world is that the bank originator model is completely absent (with the notable exception of a recently announced US $1.9 billion transaction by JP Morgan); the entire qualifying RMBS space now consists of "aggregator" transactions, whereby an entity purchases loans from disparate originators for the purpose of securitising them. Finally, the due diligence that is being done in the RMBS market post-crisis is much more robust than it was pre-crisis, when (at best) only a statistical sampling of a loan portfolio would be reviewed. Now, investors are demanding that every loan be diligenced (including, with respect to the originator, its underwriting practices, its credit review process, the integrity of its data, the related property valuation, and its regulatory compliance practices), even at the jumbo prime level.

10 Basel Committee document: "Revisions to the Securitisation Framework" (December 2014), available at: http://www.bis. org/bcbs/publ/10. Basel Committee document: "Revisions to the Securitisation Framework" (December 2014), available at: http://www.bis. org/bcbs/publ/d303.pdf.

11 78 Fed. Reg. 62018, 62114 (11 October 2013).

12 Basel Committee document: "Criteria for identifying simple, transparent and comparable securitisations" issued on 11 December 2014 at page 8.

13 Ibid.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.