Part Two: Basel II – Bottomline Impact on Securitization Markets

IRB Charges Generally Lower on Rated Securitized Notes than Unsecuritized Pool

By fundamentally changing the allocation of regulatory capital, the IRB framework creates new incentives for how banks manage their investment and portfolio strategy.

One critical area explored by Fitch is how Basel II might affect a bank’s decision to invest in securitized notes. More specifically, a bank that seeks exposure to a certain asset market, e.g. corporate debt, can either purchase a portfolio of corporate bonds or, alternatively, invest in a securitization of these bonds. A question Fitch seeks to answer is how Basel II might influence this investment decision in terms of the regulatory capital charges that the bank would have to hold under each alternative.

In performing this analysis, the Basel II charges on an IRB bank’s investment in a portfolio of unsecuritized assets are compared to an investment in the securitized notes on the same pool. In this analysis, Fitch assumes that the bank purchases all notes and positions created through the securitization process. While one bank purchasing all positions in a securitization is certainly not typical of market practice, this construct helps contrast the capital charges on securitized versus unsecuritized investments of equivalent economic risk.

What Fitch has discovered is that, in many instances, IRB banks will face lower Basel II capital charges by investing in a rated securitization structure of a pool of assets rather than directly holding the underlying pool of these same assets. This finding is somewhat surprising given much-voiced industry concerns that Basel II will inflict excessive capital charges on securitizations and potentially hamper the market. In short, the results for the portfolios Fitch analyzes in this study suggest that this concern might not be so grave.

Certain Securitization Products More Attractive for IRB Banks

For IRB banks, certain securitization products become more attractive than others from a capital standpoint. Under Basel II, credit card ABS and CMBS become particularly attractive investments for IRB banks. As illustrated in the chart below, the unsecuritized charges are several multiples higher than the securitized charges (using the RBA) on both the credit card ABS and CMBS structures evaluated in this study. In other words, an IRB bank seeking exposure to the credit card and commercial mortgage markets will be better off from a regulatory capital perspective if it chooses to invest in a rated securitization structure rather than holding or purchasing an equivalent pool of unsecuritized assets. For RMBS, there is also a moderate Basel II benefit to holding the securitized notes of a rated structure rather than the underlying pool of unsecuritized assets, although the disparity between the unsecuritized and securitized IRB charges is not nearly as pronounced as in the case of credit card ABS and CMBS.

In contrast to the other asset classes, for CDOs, there appears to be little, if any, difference between the securitized and unsecuritized capital charges. Hence, an IRB bank has the potential to be indifferent about investing in rated CDO structures versus holding the underlying pool of corporate debt instruments. Therefore, in the CDO world, the Basel II framework appears to achieve a very close alignment between the IRB formula for corporate assets and the RBA charges on the securitized notes of these assets.

Under Basel II, IRB banks appear to be indifferent about investing in corporate debt versus rated CDO products. Thus, the demand by IRB banks for CDO products relative to corporate debt appears to be unaffected by Basel II, meaning that market dynamics, not regulatory requirements, are likely to continue to drive banks investing in these markets.

Mild Incentive for IRB Banks to Hold Unsecuritized Assets Rather than Unrated Securitization Structure

Fitch’s finding that banks are generally better off holding the securitized notes rather than the underlying pool of assets is premised on the notes having an external credit rating and, hence, being subject to the RBA securitization charges; this is an appropriate assumption given that the structures evaluated in this study are comparable to those in Fitch rated deals.

However, if these same structures were unrated, IRB banks would instead have to use the supervisory formula, which, for most securitization positions, appears to generate higher regulatory capital charges than the RBA. When evaluating these deals using the supervisory formula instead of the RBA, it was discovered that IRB banks no longer face a Basel II capital incentive to hold the securitized notes over the unsecuritized assets.

Fitch’s analysis for the unrated positions is very similar to that of the rated positions. Looking at the same sample of deals, the Basel II charges on a bank’s investment in a portfolio of unsecuritized assets are compared to an investment in the securitized positions on the same pool, with the difference in this case being that none of the securitized positions are assumed to be externally rated. To help make the unsecuritized versus securitized investments economically equivalent in Fitch’s analysis, it is again assumed that one bank purchases all notes and positions created through the securitization process.

Interestingly, for all products (except RMBS), IRB banks using the supervisory formula on unrated securitization positions generally would face slightly lower charges if holding the pool of unsecuritized assets instead of the securitized notes. Therefore, it appears that the supervisory formula will provide IRB banks with a different set of incentives from the RBA, namely for banks to hold the unsecuritized assets on balance sheet rather than investing in the unrated securitized structure.

The capital incentives for holding unsecuritized assets versus holding the securitized notes (for both rated and unrated structures) are illustrated for each product type in the chart below.

This chart extracts the two main findings of Fitch’s research, comparing the Basel II charges for unsecuritized versus securitized assets; namely that, in general:

• Basel II appears to provide IRB banks with regulatory capital incentives to invest in rated securitization structures rather than holding the underlying portfolio of unsecuritized assets.
• Conversely, for unrated structures, IRB banks generally will have to hold slightly less Basel II capital against the underlying pool of on balance sheet assets than on an investment in a securitization of these assets.

Credit Card ABS and Role of Excess Spread

The Basel II incentives – namely for banks generally to want to invest in rated, but not unrated, securitization structures – are especially pronounced in the case of credit card ABS.

Part of the marked differences between the securitization charges on rated versus unrated credit card ABS positions can be explained by the role that excess spread plays in providing enhancement on the most junior tranche in these transactions. Excess spread is essentially the income remaining from a pool of receivables after paying out investor coupon and servicing fees, as well as any chargeoffs.

Basel II does not explicitly recognize excess spread as a form of enhancement under the supervisory formula, meaning that the enhancement levels used as an input to the supervisory formula charges on unrated positions only reflect securitization positions or reserve accounts that are junior to the tranche in question. Thus, the capital charges generated by the supervisory formula would presumably be somewhat lower if excess spread was actually treated as a form of credit enhancement under Basel II.

There is also no explicit recognition of excess spread as a form of enhancement for securitization positions under the RBA. However, since the RBA charges are linked to credit ratings and to the extent that the credit ratings process accords some benefit to excess spread, in practice, the RBA provides some limited, indirect recognition of excess spread.

In other words, the RBA capital charges reflect the benefits of excess spread to the extent that excess spread is incorporated into the enhancement level and, in turn, the credit rating of a particular tranche. For example, when rating structured transactions, Fitch gives no credit for excess spread for tranches rated A or higher but provides limited recognition for tranches below this level, as long as the asset pool is of sufficient diversification and asset quality. In the credit card ABS deals evaluated in this study, excess spread trapping is assumed to provide some enhancement to the most junior tranche, helping meet the expectations for a BBB rating.

The design of the RBA charges means that a securitization deal whose most junior tranche is rated BBB would require relatively low levels of regulatory capital, since no part of the structure would face the high Basel II charges on tranches rated below investment grade. Thus, for credit card ABS, even the limited recognition of excess spread within the credit ratings process can contribute to rather important differences in capital between the RBA and supervisory formula and serve as a particularly strong incentive for banks using the RBA to hold securitized notes rather than the underlying pool of receivables.

Minimal Impact of Basel II on Bank Decision to Securitize Corporate Debt and Residential Mortgage Assets

A critical issue under Basel II is how the new capital framework might affect a bank’s decision to securitize assets.

Ideally, Basel II should have little if any impact on a bank’s decision to securitize a pool of on balance sheet exposures. That is, the Basel II capital charges on a pool of assets should closely mirror the total charges on a securitization of these same assets, as the structuring process does not alter the overall economic risk profile of the pool. To the extent that securitization results in a net increase in a bank’s Basel II charges, there could potentially be new regulatory capital disincentives for banks to securitize their assets (though any potential disincentives would, in practice, be limited by the Basel II cap that limits a bank’s securitization charges to the KIRB amount on the underlying pool). However, if securitization leads to lower capital charges on a given pool of assets, Basel II would create incentives for banks to securitize these assets.

To illustrate the capital impact of Basel II and its potential influence on securitization market activity, the IRB charges on $100 of unsecuritized assets are compared to the total capital requirements that both originators and investors must hold across all exposures arising from a securitization of these assets (including capital against reserve funds and, in the case of credit card ABS, the Basel II capital charges that address potential early amortization of a structure). In other words, the amount of capital an originator would have to hold on an unsecuritized pool is compared to the capital it would have to hold if it were to securitize that same pool and buy back all the securitized tranches. Within this analysis, Fitch assumes that the securitization notes are rated and, hence, subject to the RBA.

For CDOs, Fitch’s analysis suggests that the securitization process will not result in either a net increase or a decrease in Basel II capital charges on these products. This finding is consistent with expectations, as it is generally understood that Basel regulators looked extensively at CDOs when designing the securitization framework and calibrating the charges. Notably, the results for the RMBS portfolios also suggest that securitizing residential mortgages will not lead to much change in capital under Basel II. Based on Fitch’s analysis of these portfolios, it appears that the advent of Basel II may have little, if any, impact on a bank’s decision to securitize corporate debt (through CDO structures) and residential mortgage assets (through RMBS structures).

Basel II Offers Capital Incentives for IRB Banks to Securitize Commercial Mortgages and Credit Card Receivables

However, for credit cards and CMBS, Fitch found that the securitization process generates a net reduction in regulatory capital charges for IRB banks. These results suggest that the new Basel II framework may provide banks with capital incentives for securitizing credit card and commercial mortgage assets. While a number of factors influence a bank’s decision to securitize assets, the capital benefits under Basel II on these structured products could essentially spur future growth of the credit card ABS and CMBS markets.

Basel II May Influence Deal Structuring

Basel II may influence deal structuring, notably to minimize the size of subinvestment-grade tranches where possible.

Regulatory capital is one of many factors that influences an originator’s decision when structuring a securitization, including pricing, market demand for different tranches, and rating agency expectations. At the same time, one unique structure, in most cases, will minimize the overall Basel II charges on the transaction and, hence, be optimal from a regulatory capital perspective.

Fitch’s research illustrates that this optimal Basel II capital strategy is for originating banks to maximize the size of tranches rated BBB- and above and minimize the size of subinvestment-grade tranches, particularly equity positions. Additionally, the high Basel II capital charges on subinvestment-grade securitization tranches may pressure banks to sell these positions to market participants not bound by the Basel II rules, such as hedge funds.

Basel II Introduces ‘Cliff’ in Capital Charges

Basel II introduces a steep ‘cliff’ in capital charges between investment- and subinvestment-grade securitization positions. Thus, Basel II generates significantly higher capital charges on securitized structures with larger proportions falling below investment grade. This finding is broadly consistent with expectations given the higher risk of subinvestment-grade tranches. However, the difference between investment- and subinvestment-grade tranches is quite large under Basel II and may, in some cases, lead to potential inconsistencies in the capital charges across different securitization deals.

An important driver of the capital charges on a given securitization structure is the steep cliff that Basel II will introduce between investment- and subinvestment-grade securitization positions. For example, the RBA will require 2.5 times more capital for a BB+ tranche (20 per cent capital charge) than a BBB- tranche (8 per cent). Thus, deals with larger portions of the securitization rated below BBB- generally face considerably higher Basel II capital charges.

The impact of this cliff effect is particularly evident in the case of CDOs, which is not surprising given that the pool of corporate debt instruments underlying CDOs tends to be riskier than, for example, the retail credits that typically underlie RMBS deals. The higher risk of the assets underlying CDOs means that larger enhancements are generally needed to support risk on the more senior tranches in the structure.

Indeed, the higher risk, medium risk, and lower risk CDO deals evaluated have, on average, the largest proportion of the structure rated below investment grade (i.e. over 5 per cent of the deal). Not surprisingly, CDOs, on average, face the highest Basel II charges compared to other types of structured products examined in this study. For CDOs, the sizable equity piece needed to provide enhancement to the rest of the structure is a dominant driver of the overall Basel II securitization charge on these products.

For example, compare the contribution of each tranche in the medium risk CDO deal to both the overall notional exposure and the total Basel II securitization charges on the structure. Most of the transaction’s total exposure (over 90 per cent) consists of tranches rated A or above. However, in terms of capital requirements, the equity tranche contributes over 85% of the total Basel II charges on the deal. Clearly, from a capital perspective, the dollar-for-dollar Basel II charge on the equity tranche drives the overall capital requirements on the deal.

By creating a cliff in capital charges between the investment- and subinvestment-grade tranches (particularly equity tranches), Basel II gives banks a regulatory capital incentive to try to minimize the size of the equity tranche or reserve funds used to absorb first losses on the pool. Of course, a number of market factors constrain the ability of banks to minimize the equity piece or reserve fund, particularly the expectations of rating agencies that the mezzanine and senior tranches in a structure need to be supported by sufficient credit enhancement to cover the risks associated with each tranche’s rating level.

Given the high Basel II charges applied to subinvestment-grade and equity tranches, banks may seek to sell these positions to non-bank investors not subject to regulatory capital rules, such as hedge funds. Another possible strategy is for originating banks to try to use credit derivatives to mitigate their exposure on positions that consume large amounts of capital, such as reserve funds. Indeed, given the dollar-for-dollar Basel II charge against reserve funds, banks will have a capital incentive to try to lay off this risk exposure synthetically.

Potential Inconsistencies in Basel II Charges Across Some Deals

The wide gap in Basel II charges between investment-grade and equity tranches means that some structures may potentially face higher overall charges than a securitization of riskier assets.

In this study, Fitch discovered that the higher Basel II charges on subinvestment-grade tranches – particularly on equity or first loss pieces (such as reserve funds) – in some cases, can result in inconsistencies in capital charges across different securitization deals. Under Basel II, a given securitization structure could possibly face lower capital charges than a comparable securitization of a higher quality or less risky pool of assets.

Given the calibration of the RBA charges, IRB banks must hold about 175 times more capital on an equity tranche or reserve fund (dollar-for-dollar charge) than on an AAA rated tranche (0.56 per cent charge). The disparity in charges reflects the much higher risk of a first loss exposure compared with an AAA rated exposure. However, this disparity also means that, when comparing two broadly comparable securitizations, relatively subtle differences in the sizing of the more junior tranches could potentially result in riskier structures facing lower relative charges.

A hypothetical example of where a structure backed by a riskier portfolio faces relatively lower Basel II charges is the near prime versus subprime RMBS deals evaluated in this study. The pool underlying the near prime deal comprises relatively stronger credits and, hence, less risky assets than the subprime deal. However, the overall RBA charges on the near prime structure are slightly higher than on the subprime deal, an unexpected finding given that the near prime transaction is backed by a stronger pool of mortgage assets and, for example, has a larger AAA tranche than the subprime deal.

This counterintuitive outcome is a direct result of how the Basel II securitization charges are calibrated. The hypothetical near prime structure has a slightly larger reserve fund, in part because it is backed by relatively less excess spread than the subprime structure. The reserve fund is in a first loss position and, hence, subject to the stringent dollar-for-dollar Basel II charge. Thus, even though the near prime portfolio is less risky than the subprime and has a larger AAA tranche, the size of the reserve fund is the primary driver of the overall charges on the respective deals and explains a potential inconsistency in the securitization charges generated by Basel II.

Thus, given the manner in which Basel II has calibrated the securitization framework, the structuring decision of the originator has an important impact on the overall capital charges on the structure.

Basel II May Affect Competition Between Standardized and IRB Banks

Basel II could affect competition between standardized and IRB banks within securitization markets.

Under Basel II, IRB banks generally will face different charges on the same position or asset than a bank that uses the standardized approach. Of note, banks not adopting the IRB in the US will be subject to the US recourse rules for determining capital requirements on securitized positions, which are generally comparable to the Basel II standardized securitization charges. More specifically, for investments in securitized exposures, it generally appears that:

• IRB banks will face lower charges for higher quality tranches and, hence, have a relative capital advantage over standardized banks in this segment of the securitization markets.
• Standardized banks will face lower Basel II charges on investments in lower quality or weaker tranches and, thus, have a capital advantage over IRB banks on these exposures.

More specifically, standardized banks will be required to hold more capital than IRB banks on all tranches rated BBB- or above. However, standardized banks generally will face lower charges than IRB banks on investments in subinvestment-grade tranches, namely those rated BB or BB-.

Thus, Basel II appears to give standardized banks a capital incentive to purchase lower quality tranches, since they will generally benefit from lower charges on subinvestment-grade positions than IRB banks. At the same time, standardized banks will avoid investing in higher quality securitized notes given the higher charges they will face relative to IRB banks on investment-grade positions.

To the extent that standardized banks make investment decisions based on these new capital incentives, Basel II could inadvertently result in ‘double-sided erosion’ in the asset quality of standardized banks. That is, standardized banks would have incentives to target the riskiest securitization positions and not diversify their holdings by investing in higher quality exposures.

One of Fitch’s primary concerns about this potential double-sided erosion in asset quality is that standardized banks tend to be those institutions with more rudimentary risk management systems. Thus, standardized banks will have capital incentives to build portfolios of riskier positions without necessarily having the appropriate tools and modeling sophistication to understand and manage the risk of these portfolios. Clearly, it is undesirable from a systemic perspective if the less sophisticated standardized banks were to become warehouses for lower quality tranches as a result of Basel II.

More generally, it is uncertain how standardized banks will both identify and respond to these new regulatory capital incentives under Basel II. While banks under Basel II will inevitably make some decisions based largely on regulatory capital considerations, it is important for standardized banks, in particular, to continue developing their capacity for evaluating and managing the underlying economic risk they face.