3.1 Policy options and consultations

The overarching EU policy objective is to restart the securitisation market in a sustainable way that could ensure financial stability and investor protection. To this end, the STS framework is aimed at making risks easier to assess and analyse for investors, and at the same time transactions that comply with the STS criteria would be subject to a more risk-sensitive regulatory approach (EU Commission Proposal, 2015, p. 7).

The above goals would be achieved by firstly, limiting the tendency to concentrate risks in systemic institutions. STS securitisation is conceived to realise an effective distribution of risks across the financial system, where asset-backed securities would be held by less leveraged institutions. Secondly, a number of features are designed to improve the predictability of the transaction especially with respect to its performance. This would enhance investor confidence and demand for securitised products, because the interests of issuers and investors would be aligned and the transaction’s simplicity would allow a more controlled relationship between SPV and issuer. The idea of simplicity in the transaction chain is also central to control the linkages that certain types of securitisation create with the shadow banking system (SBS). In this sense, the EU framework is cognisant of the monitoring framework advanced by the Financial Stability Board, to track developments in the SBS (FSB, 2013). The question as to whether its design is likely to be effective in limiting the undesired spill-overs of securitised products into shadow banking activities will be discussed in more details in Section 5.

The first stage of the consultation process revolved around a set of high-level principles devised in the BoE/ECB discussion paper. These stressed the importance of a simple and robust transaction where risks and returns, rather than being eliminated altogether, could be easily assessed, thereby allowing investors to conduct due diligence with more certainty. In order to achieve robustness and simplicity, the BoE/ECB identified three main steps, namely: a) standardised disclosure of information; b) transparency of credit ratings; c) design of ancillary facilities (BoE/ECB, 2014b). These mechanisms accompany a set of high-level principles that cover areas related to: 1) the nature of securitised assets, which should not reference complex formulae or derivatives; 2) the performance history of underlying assets, which should be verifiable and available to investors; 3) the ultimate obligors, to which the securitisation will have recourse, with no possibility to rely on derivative-like claims; 4) the expectations of payment, whereby originators will have to demonstrate that the receivables are a homogeneous asset type and consistently originated in the course of business; 5) the receivables transferred to the securitisation, which should be current in payment and not include delinquent obligations; 6) the security – if receivables are secured over specific assets – which should be first-ranking (BoE/ECB, 2014b).

The above principles, as explained in the introduction, mirrored to a great degree the IOSCO/Basel task force and the existing criteria under Solvency II and LCR. The EU Framework further advanced the same policy objectives highlighted above, by laying down a modular approach centred on the key criteria of Simplicity, Transparency and Standardisation, followed by additional risk features that address sector-specific risks (EU Commission Consultation Document, 2015). Briefly, simplicity required the underlying exposure to be homogeneous, it restricted the use of derivatives to hedging purpose only, and it excluded re-securitisations; transparency required compliance with transparency and disclosure requirements; standardisation required the transfer of assets to the SPV (see Kerr, 2011, and Thiemann, 2012; who discuss the problematic use of SPV finance to take advantage of accounting loopholes) to be a true-sale (EU Commission Consultation Document, 2015, p. 7).

In addition, the EU Framework laid emphasis on three more aspects of STS securitisation, namely the risk-retention requirements which, it was argued, should shift from an indirect (where investors have the burden of verifying compliance) to a direct approach and which are designed to ensure the quality of underwriting standards. Secondly, there has been widespread agreements across the different initiatives that, while the eligibility criteria should help investors to understand securitisation risks, investors should continue to perform careful due diligence. Monitoring the compliance with the eligibility criteria is therefore vital to maintain investor confidence. The Commission identified three possible strategies to this end, namely labelling, certification and licencing. Most importantly though, the question raised by the Commission was related to who should be responsible for monitoring and whether a private or a public system of certification should be in place. This point has been clarified in the Proposal and will be analysed in the next section. Thirdly, the EU Framework introduced the question of how much harmonisation would be necessary to increase simplicity and certainty. In particular the standardisation of the securitisation structure has been evaluated with respect to three key areas, namely, the legal form of the SPV, the modality for transferring assets and the rights and subordinations among noteholders (EU Commission Consultation Document, 2015, pp. 8–10).

Throughout the consultation process, the critical question of what type of authority should supervise compliance with the above principles and with the STS criteria has remained somewhat eluded. While the different initiatives have not clarified this crucial point, the more recent Commission Proposal (discussed below) has confirmed the broader policy objective stated within the CMU Green Paper, that market-driven solution are preferred and new regulation would be considered only if necessary (EU Commission Green Paper, 2015). This approach justifies a number of concerns, which are critically discussed in Section 5, in relation to the laissez-faire environment within which the securitisation market risks to restart.

3.2 Market responses

The somewhat vague approach on the possible regulation and supervision of the securitisation market can be partly understood by looking at some of the controversial responses that followed the consultation process. These highlighted deep-seated reservations that most market-players have, with respect to any form of statutory regulation of the securitisation process. It is worth stressing that the originate-and-distribute transaction channel developed in the pre-crisis years was the offspring of an innovation process spurred by the private self-interest of market-players, and that before 2008 this process remained largely unhindered by regulatory intervention (Bavoso, 2016; Biondi, 2011; Stout, 2011).

A clear illustration of these feelings is represented by disagreements as regards the envisaged function and definition of securitisation. Market responses for instance have stressed the potential benefit of securitising high-risk assets, and pointed out that the transfer of such risks could not be achieved under the idea of simplicity that is advocated as a central criterion. A large section of the industry in other words believes that, notwithstanding the poor performance in the pre-crisis years, more sophisticated transactional forms, such as synthetic CDOs, may still be needed. In particular, the European Banking Authority Banking Stakeholder Group (EBABSG) highlighted how synthetic transactions are in some instances the only means for banking institutions to transfer risks when there is no efficient way to complete a sale of the underlying assets. Despite acknowledging the risks associated with synthetic securitisation only aimed at creating arbitrage opportunities and short positions, the EBABSG proposed a further distinction between synthetic products in order to allow in certain cases the synthetic transfer of risk without transferring the ownership of assets (EBABSG, 2015, pp. 7,8). This line of criticism took note of the fact that risk retention requirements discourage the transfer of risk in more straightforward cash-flow securitisation (BoE & ECB, 2014c).

The above observation reflects a broader policy concern that most industry respondents shared with respect to the definition of simplicity. Generally, a simpler transactional pattern has been well accepted by respondents, insofar as it guarantees analysability on the part of investors who would conduct their due diligence with more certainty. However, it has also been argued that the tout court ban of complex transactions, regardless of how they performed pre-crisis, would be undesirable. Striking a balance on how much complexity is inevitable, remains difficult though, given that securitisation per se represents a complex transactional form and given that a number of responses clearly disagreed on the exclusion of synthetic securitisation from the STS framework (BoE & ECB, 2014c, p. 2).

Above and beyond the disagreements on the definition of securitisation, there is a natural friction on the need to regulate the transaction process and on its possible forms of supervision. While there is consensus on the need to establish a consistent and uniform regulatory approach to ensure the application of the STS criteria across jurisdictions, there has not been much clarity during the consultation process on how this certification function should be governed and what type of regulatory body should be in charge of it. What transpired from the consultation process is that most market-players manifested a preference for private bodies, instead of public authorities, acting as certifiers for STS securitisations (EU Commission Proposal, 2015, p. 7). The consultations have also failed to sufficiently clarify the extent to which the advocated transparency should be accompanied by a greater degree of transactional standardisation. Responses again have shown very diverging approaches on this very sensitive issue. On the one hand, responses were positive in supporting a single repository for the disclosure of key information, but on the other hand market players have reacted negatively to the prospect of publishing more macro-economic data to investors or to further standardise investor reports (BoE & ECB, 2014c, p. 4).

The question of transaction standardisation is even more topical. Greater harmonisation of some contractual aspects of securitisation has been welcomed by investors because it would increase transparency and facilitate due diligence. Some fundamental obstacles in this sense are however represented by the difficulty to reconcile different tax and insolvency regimes across the EU, which make some respondents favour optional regimes rather than mandatory ones (AMIC, 2015, pp. 6,7). Interesting responses related to harmonisation have been produced by the French regulator, the Autorite des Marches Financiers (AMF), with respect to the standardisation of the true sale of assets to the SPV. The AMF in particular advocated the introduction of a set of rules to regulate the transfer of assets in a robust way, and more importantly a European legal vehicle reserved for STS securitisation that would ensure the establishment of an EU passport for SPVs (AMF, 2015). Very interestingly, this proposal would mitigate the fragmentation that currently underscores the market along national boundaries and it would also facilitate the governance of securitisation vehicles and their supervision.

As will be discussed in Section 5, some respondents have gone as far as advocating the exclusion of the practice of tranching from the STS framework (Finance Watch, 2014, pp. 41,42). This proposal would entail both a tighter definition of simplicity and necessarily a greater contractual standardisation.

Finally, responses related to the possible supervisory structure of the securitisation market (and more broadly within the CMU) highlighted the same concerns signposted earlier in this article. The AMF for instance envisaged the central role that a public authority (EU ideally or national) should have in the certification of the STS criteria and in the supervision of the transaction (AMF, 2015). This was echoed by the World Future Council that stated in its response that the current supervisory infrastructure is insufficient because the European Supervisory Authorities (ESAs) do not have adequate powers to oversee the industry. They therefore advocated a mandatory approval process of all complex instruments, to be conducted by a public authority (World Future Council, 2015, p. 7). Against this trend, Prime Collateralised Securities (PCS) was among the respondents that favoured a private third party certification regime rather than one provided by a public regulator (PCS, 2015, p. 7).

As a final comment to this section, the natural friction between market forces on the one hand and regulation on the other, that emerged very clearly from the consultations’ responses, is central to the understanding of the Proposal that will be analysed in the next section. It is clear in fact that the Commission has been under substantial pressure from relevant lobbies to come up with a regulatory framework that would still allow market players to benefit from a laissez-faire environment. Weak supervisory mechanisms however, and an insufficiently robust definition of the STS criteria are likely to generate the same level of risk-taking and instability experienced before the crisis. The next section critically evaluates how the Commission’s Proposal has tackled these fundamental questions.

5.1 Definition of STS

The task of defining securitisation for the purpose of the STS criteria has been challenging and it still presents some unaddressed questions. The responses to the consultations highlighted in Section 3 showed that different stakeholders have not agreed on what should have been included within the concept of high-quality securitisation. As explained earlier in this article, much debating went into the exclusion of synthetic transactions from the STS definition. The proposal has now clarified that the simplicity criteria will require the true-sale of the underlying receivables, and that synthetic exposures will fall outside the scope of STS (Art.8). Questions of market efficiency were raised by market-players, based chiefly on the fact that not all synthetic transactions performed badly during the GFC, and on the costs that would be incurred from the true-sale of certain classes of more risky assets (EBABSG, 2015, p. 8). These concerns, albeit pertinent, fail to reflect the systemic importance that securitisation can have on the financial system as a whole. The Commission’s stance therefore has to be welcome, especially when considered in conjunction with other provisions that limit the complexity and riskiness of STS securitisation, namely: the exclusion of re-securitisations; the prohibition of active portfolio management (which was typical of dynamic transactions like pre-crisis CDOs; see Long, 2006, p. 124); the requirement that the securitised assets are homogeneous, originated in the ordinary course of business and not defaulted (Art.8).

The Commission however has failed to provide more clarity on the central question of tranching. As noted earlier, the practice of tranching seems to be accepted throughout the proposal (it is for instance suggested that non-senior tranches should also benefit from adapted capital charges), and the prohibition of active portfolio management per se cannot be conclusive of a willingness to restrict tranching. In light of the critical role played by this particular feature of securitisation, it is worth providing an explanation of its implications. Tranching represents a way to pool assets whereby the SPV does not issue one type of security against the asset pool, but several types, due to the slicing of the security into tranches of different credit quality and subordination (Lucas, Goodman, & Fabozzi, 2006). Each tranche carries a different risk and rate of return, appealing investors with different risk appetite and providing therefore a more customised product (Coval, Jurek, & Stafford, 2009). This practice allowed banks in particular to originate and securitise loans of poorer quality, which would anyway receive a high rating (relative to the low quality and high risk of the underlying assets) for two main reasons, namely the complex correlation formulas that underscored the way in which assets were bundled (Salmon, 2009, p. 79, explaining how the Gaussian Copula formula allowed the creation of triple-A securities where none of the underlying assets were worth triple-A), and the credit enhancement mechanisms (Acharya, Schnabl, & Suarez, 2013).

Credit enhancements can become a problematic feature of the transaction when they take the shape of guarantees provided by the originator. Even though this has been the case chiefly in the context of ABCP, originators or sponsors often provide guarantees to investors so that the investors’ claim is paid off even in the event of the SPV’s insufficient cash-flow. The result is that in most cases SPVs are screened from credit and liquidity problems and the relating risks are borne by the originator/sponsor, giving therefore rise to what has been defined as a securitisation without risk transfer. Instead of dispersing and diversifying risks (which is one of the functions of securitisation), these securitisations aggregate risks in large banks that do not hold sufficient capital against these risks (Acharya et al., 2013, p. 521).

The practice of tranching was also combined with the use of credit default swaps (CDS, entered into to either hedge certain exposures or to speculate on certain assets) which led large banks to have asset portfolios that were similar to one another. This is what is usually defined as correlation between banks’ balance sheets, which increased dramatically in the pre-crisis years because of the use of credit enhancement mechanisms (as explained in the previous paragraph) and the long transaction chains involving complex derivatives (such as CDS). Correlation in turn led to interconnectedness which aggravated problems of systemic risk, as well as risks of contagion between different tranches (Finance Watch, 2014, p. 35).

It has also become evident that tranching is the source of increased complexity in securitisation and more difficult assessment of risk-taking on the part of investors, something that is at odds with the Commission’s overarching aims. The subordination process achieved with tranching enables the creation of senior tranches rated triple-A. These, despite being insulated from most losses (absorbed by junior and equity tranches), are still affected in case of extreme events. Moreover, risks related to senior tranches’ correlation in the past were largely understated, that is the risk flowing from exposure to systematic or market risks (those inherent to the entire market), which cannot be eliminated through diversification (Coval et al., 2009). The way in which risks were repackaged in tranches in other words led investors and the market as a whole to believe that, “by slicing and dicing”, safe investment products could be created from risky underlying assets (Turner, 2009). As a result, herding behaviour abounded among investors who relied on CRAs and on the myth that senior tranches would be risk-free securities (Coval et al., 2009). Tranching in particular attracted less informed investors with little capacity to understand and manage complex risk (Finance Watch, 2014, p. 43).

Investors’ bounded rationality was anyway aggravated by the complexity of correlation models, and by long and opaque transaction chains. Tranching in other words added uncertainty, which contributed to the difficulty to estimate risks (Finance Watch, 2014). Interestingly, the Bank of International Settlements (BIS) recently stated that even when securitised assets are simple, transparent and of high quality, their risk assessment will remain uncertain, in particular with respect to certain tranches (Antoniades & Tarashev, 2014). Tranching also exacerbated the conflicts of interests between different tranche-holders, because senior tranches and equity tranches naturally attract different attitudes to risk-taking. Moreover, these conflicts were not mitigated by asset managers who have tended to manage asset portfolios in the interest of the sponsor and have therefore maintained an incentive to increase the riskiness of the asset portfolio (sponsors typically invest in the equity tranche; see Mahlmann, 2012, p. 13).

The pre-crisis boom years saw an increased appetite among investors for highly rated, low-risk securities, and in particular the demand for senior tranches to be used as collateral in the repo market. Spiralling levels of leverage and a widespread underestimation of correlation risks (due also to flawed credit ratings) were among the consequences of this trend, as will be seen in the next section. The critical question that emerges at the end of this section is whether the framework proposed by the Commission provides a definition of STS that will prevent the resurgence of risky market practices such as those described earlier. The idea of having simple and transparent transactions does not seem in principle consistent with the practice of tranching, and to this extent, the Commission could have provided a tighter definition of high-quality securitisation, more consistent with the highlighted functions of the transactions.

5.2 Spill-overs into shadow banking

In the aftermath of the GFC it became increasingly evident that the SBS and the regulated financial institutions were highly interlinked, mainly because large banks had become very active both as providers of collateral and as participants in the repo market (Gabor, 2013; Moe, 2015).

Senior-tranche securities were before 2008 highly demanded forms of collateral in repo transactions, and were used by financial institutions to obtain short-term funding in the wholesale market. Even though securitisation was always conceived as a way to achieve diversification, the balance sheet of banks became increasingly homogeneous in the pre-crisis years, because their liabilities side was over-reliant on wholesale funding, and their assets side on securitised forms of credit (Haldane & May, 2011, p. 355). Excessive reliance on wholesale funding moreover is what exposed banks to greater levels of leverage and funding risks and it increased the linkages between banks and shadow banking (Avgouleas, 2009; FSB, 2013). These concerns revolve around the structure of the credit intermediation system that has been recognised as “securitised banking” model. This refers to the practice of distributing originated assets to entities that, despite replicating some of the functions of traditional banking, are not in the regulated banking sector and are not constrained by capital and liquidity requirements. For this reason, interlinks with the SBS are said to contribute to increasing leverage at systemic level (Meeks, Nelson, & Alessandri, 2014, p.ii).

The linkages between the banking sector and shadow banks are attributed to some uses of securitisation and in particular the way it interplays with the repo market and money market funds (Claessens, Pozsar, Ratnovski, & Singh, 2013). In the securitised banking model, the best tranches, as said, are in high demands as highly rated collateral for the repo market, which over the last three decades has become a huge source of short-term finance for banks and other institutional investors (Gorton & Metrick, 2012). The contagion, or correlation as explained in the previous section, is caused by the fact that collateral in repo transactions can be reused (rehypothecated) by banks, thereby increasing the correlation between banks’ portfolios, the homogeneity of the financial system and the overall systemic risk (Claessens, Pozsar, Ratnovski, & Singh, 2012, p. 16; Haldane & May, 2011, p. 355; Moe, 2015, p. 12). This problem was augmented before the crisis by some credit enhancement mechanisms described in the previous section which, while ensuring triple-A ratings, increased the linkages between originators and SBS.

The complex transaction chains that had become a common feature in securitisations before 2008 have been identified as the cause of the uncontrolled level of leverage among large financial institutions (FSB, 2013, p. 4). The policy recommendations directed at improving transparency and disclosure – reflected also in the Commission’s proposal – are certainly a step in the right direction (IMF, 2013; IOSCO, 2012). However, the Commission has remained ambivalent as regards the degree of product and transaction standardisation that is necessary, because the prevailing belief is that too much standardisation would stifle innovation. This article concurs with those views that have expressed concerns on the limited regulation of certain aspects of securitisation and in particular it posits that plain non-tranched securitisation may be the only way to ensure certain functions of the transaction, without creating pre-crisis systemic risk (Finance Watch, 2014; Wallison, 2015).

Another critical aspect of the pre-crisis securitisation market that the Commission has not addressed is the question of the European banks’ over-reliance on the wholesale market. As explained earlier, securities created through the securitisation process were used as collateral for short-term finance obtained in the wholesale market. It was this funding base that primarily impaired European banks before the crisis, above the financial system’s over-reliance on banks as funding sources. It is worth highlighting in this section the problems that derive from banks’ reliance on wholesale funding instead of more traditional retail funding.

Resorting to wholesale funding enables banks to extend their lending capacity and grow their balance sheet beyond their core liabilities; in essence they take advantage of the rise in their equity and the apparent decline in risk and leverage. This is so because banks access a large source of funding which is only limited by the quantity and quality of collateral that can be posted, that is the assets created through the securitisation process (Finance Watch, 2014, pp. 38–39). It is well known that to increase lending, banks should increase their regulatory capital or alternatively reduce their existing risks. Both of these strategies however represent constraints on banks’ profitability, while there is clearly more and cheaper lending enabled through the wholesale market (the only limit on how much assets can be used as collateral being the level of haircut, namely the risk of holding those assets, Finance Watch, 2014, p. 39).

The extensive use of wholesale funding before 2008, combined with a general understatement of risks, enabled financial institutions to build up high levels of leverage that they could not withstand. Much of this funding channel through repo transactions became a critical factor in increasing systemic risk because banks were purchasing each other’s securities with borrowed money, which again increased leverage and the concentration of risks in the financial sector (Moe, 2015 p. 8). This eventually triggered liquidity crises, due to the reliance on short-term assets to fund long-term liabilities and the inability to renew short-term funding (Gorton & Metrick, 2012, p. 426; Wray, 2013, pp. 293,294). In essence, credit intermediation through the SBS replicated the same problems of the traditional banking system, namely maturity transformation and leverage, but they were magnified by the lack of regulation in shadow banks (particularly reserve and capital requirements; Blair, 2010, p. 17 and FSB 2011, p. 3) and also by the complexity of transaction chains in the SBS (Turner, 2014, p. 5).

Some of the risks highlighted above are likely to be mitigated by the new liquidity coverage ratio introduced with CRR/CRD-IV, which create buffers designed to equip banks with necessary liquidity in the case of a run on their short-term exposures in the repo market (Clifford Chance, 2012, p. 21). Admittedly though, there is not much in the Commission’s work that provides a clarification of the linkages between securitisation and the SBS. Regulating STS securitisation may not be sufficient in this sense, given that the repo market (referred to as securities financing transactions (SFT)) has received very limited regulatory scrutiny in the pre-crisis years, despite being much larger than the securitisation market, and used to finance securitisation products (ECB, 2012; Gabor & Vestergaard, 2015, p. 3)

In 2015 the Commission enacted the SFT Regulation (Regulation EU 2015/2365) with a view to reduce risks in the SBS, by mainly increasing transparency in repo-type transactions. This will occur through a twofold strategy; firstly, the Regulation set conditions on the reuse of financial instruments received as collateral (art.15, whereby both conditions revolve around counterparties’ understanding of the risk entailed with the reuse); secondly, counterparties will be required to report relevant transactions to a trade repository (art.4, and 13 as regards transparency towards investors from UCITS managers). As the main thrust of the Regulation is on increased disclosure of information provided to counterparties in repo transactions, this should undoubtedly contribute to tackle problems of counterparty credit risk and systemic risk. However, it is equally pertinent to cast doubts on the effectiveness of these provisions for two reasons. Firstly, they revolve around the assumption that counterparties will act rationally if provided with adequate information. This has proven to be rather illusory in the pre-crisis years (Avgouleas, 2009b). Secondly, as is discussed later in the section, the enforcement of the above provisions will depend on member states implementing relevant regimes through their national authorities, with ESMA playing largely a coordination role.

The SFT Regulation has not provided any guidelines on minimum requirements for haircuts, admitting in its preamble that these would prevent excessive leverage and concentration of risk among large institutions. This gap is likely to be filled by the FSB through a regulatory framework for SFT haircuts. While the FSB’s recommendations on minimum haircuts will rely on market participants establishing internal processes and procedures in accordance with the framework, monitoring on margin requirements will be coordinated with BSBS and IOSCO in order to reduce regulatory arbitrage (booking transactions in different jurisdictions) (FSB, 2015).

The Commission overall, while having acknowledged the risks of rehypothecation and having stated the need to appropriately regulate the flow of collateral in the financial system, has only partly addressed these problems, relying instead on an architecture still largely grounded on market-discipline mechanisms.

5.3 Reliance on external ratings

It was explained earlier in the article that the practice of tranching involves longer transaction chains which also represent a longer and more complex form of credit intermediation. In the years before the crisis reliance on CRAs had become indispensable in the context of these transactions because of the difficult assessment of the underlying assets’ risk, which investors could not supplement. The role of rating agencies went beyond the rating of structured securities. They provided a number of ancillary consulting services related to the products they would subsequently rate and in particular they advised on how to structure the transaction in order to receive a triple-A rating (Bavoso, 2013, p. 145; Partnoy, 2006, p. 61). Once again, the overarching regulatory goal to limit regulatory reliance on CRAs is at odds with the persisting complexity that generates from tranching. The question that still remains unanswered is how originators and investors will supplement the function performed (admittedly quite badly) by CRAs to mitigate the information asymmetry that derives from complex transactions.

Regulatory intervention in the EU in the area of CRAs has followed a twofold strategy after 2008. On the one hand, EU legislation has sought to increase transparency and disclosure of credit ratings through enhanced supervision of CRAs, conducted chiefly by ESMA (Regulation 513/2011, amending Regulation 1060/2009). On the other hand, as already signposted, there has been a move to limit regulatory reliance on external credit ratings, to be replaced instead by either internal models (BCBS, 2014, p. 1) or more standardised and simplified forms of disclosure to investors. The approach endorsed by the Commission in this proposal is clearly leaning towards increased disclosure of information in a more standardised way (consistent with Regulation 462/2013, amending Regulation 1060/2009, art.8) in order to reduce reliance on third-party ratings. At the same time though, CRAs will still play a relevant role in structured finance in the foreseeable future, given the necessity to have information intermediaries between originators and investors, and the relative complexity of all securitisations, even STS.

Hence, while reduced reliance on external ratings has to be welcome given the widespread failure of CRAs before the crisis, the reliance on internal models as a way to supplement third-party ratings, proposed by the Basel Committee, needs to be seen with caution. Despite introducing an improved framework and a hierarchy of approaches to assess risks in securitisations, the new approach will allow large banks to employ their internal models to establish how much capital has to be set aside against securitisation exposures (BCBS, 2014, pp. 15–16). The concern with this policy is that it fails to address two failures that were experienced in the pre-crisis years. The first problem is that internal models rely on assumptions of market discipline which have been largely misplaced. The peril in particular would be to allow large financial institutions to devise internal risk-assessment models that enable artificially low risk measurements and as a consequence arbitrary capital relief, followed by accumulation of leverage. Secondly, the question of addressing systemic risk in financial markets is not adequately tackled through a system that remains largely geared to a micro-prudential approach, focusing on the risk-taking of individual banks rather than on the accumulation of risk and leverage system-wide. There has been widespread agreement after the GFC that a lack of macro-prudential approach was at the root of the regulatory failures (Alexander, 2010, p. 437; Persaud, 2010, p. 445). This was particularly the case, again, with respect to the reliance on market discipline that was envisaged as a means to supplement regulation (BCBS, 2006). The undisputed trust in the policing function of market mechanisms has however proven misplaced (Avgouleas, 2012, ch.5; Biondi, 2013, p. 163; Boyer, 2013, pp. 130–132).

Admittedly, the Commission has not introduced a valid solution to the above regulatory problem. Moreover, the notification system in the proposed Regulation risks to re-propose questions of micro-prudential approach and supervisory coordination. These will be examined in the next section.

5.4 Regulation and supervision

The ambitious agenda set out with the Green Paper on CMU has left with some uncertainty as regards the institutional framework under which the project would be implemented. The Green Paper stressed that the Commission supports market-driven solutions and that regulatory changes would only be considered when necessary and this is premised on the fact that the regulatory framework for capital markets has already been harmonised in the past and that supervisory convergence has been promoted since the creation of the ESAs in 2010 (EU Commission Green Paper, 2015, p. 22). The existence of ESAs, and in particular the development of ESMA with its regulatory and supervisory powers (critically evaluated in Schammo, 2011, p. 1890), means that on paper the CMU does not need the establishment of a SEC-type of regulator to be implemented.

However, this is only part of the picture. Questions on the effective powers of ESMA have been raised, especially in relation to the budgetary constraints under which it operates and the relevant powers that have so far not been exercised, such as those related to financial products (Lannoo, 2015, p. 6). Concerns over ESMA’s capability to face the task of supervising a fully integrated capital markets and in particular an EU-wide securitisation market are therefore well grounded, and the various consultation documents have provided little clarification in this respect. It has also been observed that while the securitisation market is cross-border in nature, capital markets remain fragmented along national borders, which means that local knowledge will probably be required in the day-to-day supervision of market practices (Lannoo, 2015, p. 7). As anticipated in Section 4, it appears that the necessary level and type of supervision that should be necessary in a fully integrated European capital market is not yet been defined. Despite the creation of a standardised prudential treatment of securitisation across the EU (resulting from the application of the STS criteria and the Basel III framework), the risk here is that market-based channels of finance will be developed in a laissez-faire environment, reminiscent of the pre-crisis period.

At a higher policy level, the lack of a supra-national regulator makes it very difficult to envisage how the CMU project and STS securitisation more specifically would achieve some of the announced goals. Without the steering function typically associated with central authorities, the “real economy” dimension of the CMU risks being overshadowed by the profit-making priorities of financial institutions, leading in turn to more “financialisation” of the economy.

With respect to the supervision of STS securitisation more specifically, the Commission has clearly excluded the involvement of a public authority (presumably ESMA) in the certification process, and more importantly article 15 of the proposed Regulation allocates supervisory powers to national authorities, with ESMA playing a marginal coordination role. The notification system moreover, laid out in article 14, places responsibility upon originators and sponsors. This mechanism risks leaving too much power to market actors, especially if enforcement procedures are not sound and consistent across national borders. The risk with this system is that it confers a private quasi-regulatory power to market-players, which may not be adequately countered by the competent authorities’ enforcement capacity. If too much power is left in the industry, the boundaries of the STS certification could be easily crossed and this could lead to the same undesired practices that were common before the GFC.

Finally, post-crisis regulatory efforts at the EU level have attempted to bridge the gap between rule-making and supervision that traditionally characterised EU financial markets. The lack of synchrony between centralised regulation and local supervision and enforcement has been identified as one of the key flaws of the pre-crisis EU architecture (Ferran, 2011, p. 4). The allocation of supervisory powers to national authorities in the context of STS securitisation is clearly at odds with the overarching shift to centralise, and more importantly to harmonise, supervisory practices across the EU, especially in a market that is intrinsically trans-national.