Proposed Basel reforms do not reflect UK market, say lenders
Published: 14 March 2016 | Author: Bernard Clarke
Proposed changes by international banking regulators to the rules for assessing credit risk do not reflect the real underlying risk of those assets and would result in unduly harsh capital treatment of both prime residential and buy-to-let mortgages. Those are among the key conclusions of our response to the consultation document, Revisions to the standardised approach for credit risk, submitted last week.
Our submission argues that significant increases in risk weightings proposed by the Basel Committee on Banking Supervision (BCBS) for both prime residential and buy-to-let lending are not justified by historic losses. We therefore question the way in which risk weightings have been calibrated.
In current market conditions, mortgage funding is available and attractively priced, and UK consumers are enjoying some of the lowest rates ever. But capital requirements that are excessive relative to the risk of the underlying assets are likely to affect the cost and availability of mortgages.
In our response to the BCBS proposals, we argue that, if implemented in their current form, the changes would have “unintended and negative consequences” for prime residential and buy-to-let markets.
Our submission also says that the proposals fail to make sufficient allowance for the way in which mortgage regulation has already been reinforced in the UK. Now that we have stress testing of mortgage affordability by the Financial Conduct Authority, for example, we believe that the BCBS places too much emphasis on historic loss data in its proposals for calculating risk weightings.
Without taking into account the effects of reinforcing UK regulation, the changes proposed by the BCBS are being presented in a “regulatory vacuum,” our submission says.
Assessing the evidence
Our view that risk weightings being proposed by the BCBS for the standardised approach are too high is supported by evidence from lenders that have adopted the more sophisticated internal ratings-based approach to assessing risk.
Additionally, we are concerned about the committee’s proposals for the valuation of property. The BCBS is currently proposing that national supervisors should continue to be able to require lenders to adjust property valuations downwards. Once they have done so, values may subsequently be revised upwards again – but cannot be increased to a level exceeding the original valuation.
We think this approach is flawed. In effect, it means that property values can go down, but not up. The obvious problem is that the valuation affects the loan-to-value ratio of a mortgage. So, being able to update this to reflect true market value of a property allows a lender to operate on the basis of a clearer understanding of the potential risk associated with the loan, and the loss given default.
We understand that regulators concerned about mortgage market expansion based on rising property values might see benefits in being able to limit the potential for upward revaluation of property. But this, in our view, is outweighed by the problems caused by not making decisions based on the most accurate property valuation available. Failing to use the most reliable valuation may distort the amount of capital lenders are required to hold against a loan.
There are other ways in which a lack of flexibility in the rules prevents lenders from using property re-valuation to get a more up-to-date understanding of their real exposure to risk.
In practice, lenders are able to capture changes in the value of property at a number of different points in the lending cycle – not only when advancing the original loan, but also when offering a further advance or re-structuring debt. We believe that lenders should be able to use the updated market information captured at these points to revise the LTV ratio – and ensure the loan has a more accurate risk weighting.
Failing to resolve these valuation issues will have implications for borrowers. If, for example, lenders are not allowed to reflect the true value of the property, they might be prevented from agreeing to a further advance – even if it would be justified by an up-to-date property valuation. That would be detrimental both to the borrower and the firm.
Loan-to-value and capital
We are concerned that, even if lenders are allowed to update the valuation of the property and adjust the risk weighting accordingly, the proposed rules mean that even small changes in LTV could have disproportionate effects on requirements to hold capital. There are, in effect “cliff edges” in the proposed capital requirements – and these have the potential to distort the mortgage market.
The problem arises because the Basel committee is proposing a “slab” structure for risk weightings, rather than a marginal one. So, a loan at an LTV ratio of 81% would attract a risk weighting of 45% applying to the whole sum advanced, while a mortgage with an 79% LTV ratio would have a weighting of 35%.
We believe that lenders should be able to apply a blended risk weighting, reflecting real risk across their portfolio of loans.
In calculating the weighting that should be applied to an individual mortgage, we would like to see that loan broken down into a number of segments reflecting increasing risk. From this data, it would be possible to produce a composite risk weighting reflecting exposure to LTV as it increases. We believe that this would be a better measure of the overall risk associated with the loan.
Overall, we believe that more work is needed before the credit risk can be captured properly, and we would like to be able to contribute to the development of an appropriate methodology.
Introducing the changes
We are concerned about how the BCBS proposals would be implemented. Even if they are applied only to new loans, this process could distort the market – if the rules are flawed.
This effect could be magnified if the rules were applied to the whole of a lender’s back book. Even if there were no flaws in the rules, applying them to the whole back book would be difficult and expensive. This could be simplified by applying an historic LTV calculation based on the valuation when the loan was originated. But there is a risk that this would not produce a true reflection of current market value of the property, and therefore properly reflect credit risk.
The buy-to-let sector
There are fundamental differences between the prime residential and buy-to-let markets – and we believe that the proposed rules should reflect this.
Buy-to-let lending volumes are often more volatile than in the residential sector, but this does not necessarily imply greater risk. The buy-to-let market tends to respond more rapidly than the residential sector to rising interest rates, for example. But this may simply mean that buy-to-let is more efficient in adjusting to evolving market conditions and quicker to recover its equilibrium.
Lenders advance most buy-to-let loans on an interest-only basis. They also have different strategies for recovering arrears in the sector. In appropriate circumstances, for example, they are able to appoint a receiver of rent to ensure that payments made by tenants are used to pay off the mortgage.
Buy-to-let risks are therefore not the same as in the residential market. So, we continue to believe that data collected from lenders using the internal ratings-based methodology in the buy-to-let sector are likely to provide the best evidence of risk.
Using the right data
There is also a lack of robust data over the buy-to-let cycle, and this presents challenges in trying to establish appropriate risk weightings. Data on buy-to-let loans originated before the financial crisis is patchy. And this data reflects lending criteria that have changed significantly since the financial crisis – driven by a combination of factors, including regulatory intervention and a shortage of funding.
What is certainly clear, however, is that, for loans with an LTV ratio of between 60% and 80%, the risks associated with new buy-to-let mortgages are not two to three times greater than they were in the past. That is, however, what is implied by the BCBS proposal to raise the risk weighting for these loans from 35% to 90%.
As we have proposed for the residential sector, we would like to see individual buy-to-let loans broken down into segments, to get a more detailed picture of differences in risk as the LTV ratio rises. This would enable regulators to develop a more nuanced view of risk across the market, and apply appropriate risk weightings.
And as in the prime residential sector, we favour taking advantage of opportunities to re-value buy-to-let property so that weightings can be updated to reflect current risk.
Social housing – and self-build
The funding of social housing is important to UK lenders, and we welcome the proposal that this should be given the same risk weighting as prime residential lending. But great care is needed to ensure that the drafting of rules delivers this outcome.
The BCBS is, of course, seeking to draft global rules, and we are concerned that any poorly drafted or ambiguous rules could unintentionally impose higher risk weightings on the funding of social housing in the UK.
We have put forward proposals to try to avoid this. We have suggested that the rules for calculating risk weightings for prime residential mortgages should be extended to “loans to registered providers of social housing that are regulated nationally within an overall policy environment set by the government.” We are also seeking to ensure similar protection for the funding of social housing in devolved administrations within the UK.
The government in the UK is also keen to promote self-built property, which it sees as an important component of housing supply. We would therefore like to see the Basel proposals for prime residential properties extended to capture this.
We believe that the Basel committee’s proposals for reforming the standardised approach to assessing credit risk would have unintended and negative consequences for both prime residential and buy-to-let mortgage lending in the UK. There is a risk that the committee will be operating in a regulatory vacuum if it fails to take into account reforms that have already been introduced in the UK market.
Currently, we believe that the capital weightings being proposed are based on a flawed understanding of risk in the UK. If the committee introduces reforms that do not properly reflect UK market conditions, there is a risk of significant detriment for both consumers and firms.
Our submission suggests a number of reforms to the BCBS proposals, and we are keen to work with the Basel authorities to implement rule changes that reflect the real needs of consumers and firms in this country.