The Building Societies Association of the United Kingdom (BSA) is pleased to respond to the Committee’s consultation on the design of a framework of capital floors based on standardised approaches. The BSA represents all 44 UK building societies, of which only three currently use an IRB approach. Building societies have total assets of over £330 billion and, together with their subsidiaries, hold residential mortgages of over £240 billion, 19% of the total outstanding in the UK. They hold over £240 billion of retail deposits, accounting for 19% of all such deposits in the UK. They employ approximately 39,000 full and part-time staff and operate through approximately 1,550 branches.
Building societies are domestic “non-Basel” mutuals, but are especially affected by any changes to the risk weighting of residential mortgages, whether as a result of capital floors, or revisions to the standardised approach which the vast majority of our members use.
The BSA supports in general terms the response from the European Association of Co-operative Banks of which the BSA is a member. We particularly welcome the EACB’s contribution given that the BSA itself, and its own membership, has limited experience of IRB. This response is therefore restricted to some high level challenge to the capital floors proposals: we do not attempt to address the technical questions in the CP.
Is a permanent aggregate capital floor really necessary?
The CP asserts (at paragraph 11) that capital floors are an integral part of the capital framework. But a permanent floor would in fact be an innovation, replacing the Basel II transitional floors which were originally time-limited, though subsequently extended. So the case for a permanent capital floor has to be established, not just assumed, especially when other measures – in particular, but not limited to, leverage ratios, address some of the same concerns that are produced as justification for the permanent capital floor.
The CP’s arguments at paragaphs 13 to15 on the complementarity of the leverage ratio and capital floors are unconvincing – especially the oversimplified table in paragraph 15. A more accurate summary would recognise that both leverage ratios and capital floors (claim to) address the same underlying concern over “model risk” resulting in insufficient capital.
This underlying issue manifests itself in various different ways, with the principal problems catalogued in paragraphs 13-14, though sometimes these problems are stated in a way that seems designed to bolster the thesis of complementarity. Not surprisingly, either remedy may be more or less effective in dealing with an individual problem, properly understood. For instance, the third bullet of paragraph 13 mentions horizontal inequity in risk weighted capital requirements. But the desired “more level playing field” is in overall capital requirements. Capital floors might deliver this by constraining very low risk weights produced by IRB models. A leverage ratio, or set of differentiated ratios, can deliver much the same outcome because they, and not the risk weighted requirements, will act as the binding capital constraint where IRB risk weights are too low. Moreover, not all the detailed problems described in these paragraphs are of equal importance.
The first bullet in paragraph 13 mentions RWA inconsistency, claiming that leverage ratios do not directly address the inconsistent assignment of risk weights. But if leverage ratios do in fact avert the risk that banks are under-capitalised, why does this (secondary) inconsistency matter ?
In conclusion, we think the analysis in paragraphs 13 to 15 understates the extent to which a leverage ratio regime already deals with the underlying problem of model risk, and overstates the extent to which a permanent capital floors might be necessary as a separate, additional measure.
Would other measures be more suitable ?
Awkwardly overlapping with the BCBS consultation on the capital floors idea is a separate discussion paper from the European Banking Authority on the future of the IRB approach, aimed at improving the operation of IRB models, particularly as to robustness and comparability. The EBA paper addresses some of the same perceived problems as the BCBS capital floors consultation, but canvasses a series of more limited measures that go more clearly with the grain of the IRB framework.
Nevertheless, the EBA work is expected to place a “substantial burden” on both banks and national supervisors in Europe. Would it not then be more sensible to aim to make similar improvements first, on a global level, and then consider if additional steps such as permanent capital floors are really needed ? Otherwise there will be a great waste of uncoordinated effort.
What are the likely unintended consequences ?
One of the principal – but, we assume, unintended – consequences of the regulatory rhetoric accompanying any move to permanent capital floors will be a general loss of faith in IRB models. What is all too readily overlooked in the regulators’ current moral panic over “model risk” is that IRB modelling, properly used, can represent better informed and therefore superior risk management. With hindsight, the regulators’ enthusiasm for IRB modelling through the introduction of Basel II, now seems excessive, even foolish.
But there is no need for the pendulum to swing back too far in the other direction. A pertinent question for the Committee to answer is – is IRB modelling still seen as a desirable development for individual banks (on grounds of better risk management) and if so, does that still deserve a modest capital incentive? Or are IRB models now a given feature of the environment ? Several of our larger members could well aspire to migrate to an IRB approach for credit risk. But most of the rhetoric from both national and international regulators over the last couple of years – whether the moral panic over models, or the prospective reduction or even elimination of any capital incentives - creates strong discouragement from that endeavour. Is that the intended outcome ?
In our comments on the Committee’s parallel consultation on the review of standardised approach to credit risk, we challenged that those proposals – contrary to the stated objectives – will in fact lead to a substantial across-the board increase in Pillar 1 capital requirements. The same concern clearly arises with regard to a permanent capital floor: on top of all the other capital measures in course of implementation, such a floor is likely to lead to major constraints on new lending. Is that, in part, the intended outcome, or just collateral damage ?
In this context, it is also pertinent to draw attention to a recent piece of work published by the PRA, as part of its Pillar 2 methodology for credit risk. Table A sets out a comprehensive set of benchmark risk weights for residential mortgage lending, designed to help assess whether the actual current Pillar 1 charge understates or overstates the true risk, and therefore whether adjustment through Pillar 2 is necessary.
The benchmarks are, for the most part, well below either the current or proposed standardised risk weights. While PRA is careful to state that these benchmark risk weights are not implied to be a better reflection of underlying risk than the (current) standardised approach, they certainly provide powerful challenge to any move to increase standardised risk weights and then use them to floor IRB risk weights. Res ipsa loquitur.
Our colleagues at the EACB have also drawn attention to a further danger. Capital floors applied to IRB users risk incentivising a shift from lower-risk to higher-risk assets and exposures, both within asset/exposure classes and between classes. A leverage ratio has the same tendency. But introducing both measures may make this problem especially acute.
There is also concern, again highlighted by the EACB, that introduction of capital floors could exacerbate competitive distortions, seriously damaging the mortgage finance market in Europe compared with (in particular) the US. Much of the risk of US home loans is in fact already borne by the US taxpayer through the implicit guarantees of the federal agencies Fannie Mae and Freddie Mac. In Europe, generally, no such major distortions operate. But introducing capital floors is likely to make the effect of the distortions worse.
Do the incremental benefits of a permanent capital floor outweigh the various downsides ?
We do not find that the CP has adequately made this case in principle, and – in relation to the downside of possible constraints on new lending – this cannot be assessed in the absence of calibration. In summary, we find that the CP has overstated the incremental benefits of introducing a permanent capital floor as an additional measure on top of one or more Basel III leverage ratios. Some of the residual problems – such as inconsistent assignment of RWAs – are anyway rather secondary once the main issue of undercapitalisation risk is deal with.
 The PRA’s methodologies for setting Pillar 2 capital, January 2015 ( Table A on page 10 ). http://www.bankofengland.co.uk/pra/Documents/publications/cp/2015/pillar2/cp115.pdf