Thank you for inviting me to speak at Global ABS 2015.
For the past couple of years the Bank of England has been working with the ECB to consider what can be done to foster the development of robust and well-functioning securitisation markets in the EU.
Last year we published a joint discussion paper1
and received helpful feedback from a wide range of commentators.
One theme was support for defining the characteristics of simple, transparent and comparable securitisation – more on that in a moment.
More recently, we published a joint response2
in support of the European Commission’s consultation on securitisation and look forward to the Commission’s forthcoming legislative proposals.
A key objective of the Capital Markets Union in Europe is a financial system in which stable market-based funding complements bank lending as a source of finance for the economy. Securitisation can be an important part of that stable, market-based funding, alongside other markets such as equities, bonds and private placements. Banks can use securitisation to diversify their funding and transfer risk on underlying loans. Importantly, non-banks can also finance lending through securitisation.
Banks and non-banks can use securitisation to help match the maturity of their liabilities to their assets.
This is particularly important in the context of housing finance.
Residential mortgage-backed securitisation (RMBS) was the largest part of the EU market before the financial crisis and the subsequent decline in the overall market is primarily explained by the fall in RMBS issuance. A large part of that decline may be explained by fundamentals – banks have cheaper ways to finance themselves at present.
But, in the long run, I think we will need robust RMBS markets so that at least part of the risk on long-term housing lending is financed from long-term savings rather than short-term bank deposits.
I prefer to focus on the long-term goal of building sustainable securitisation markets rather than reviving securitisation in the short run. Comparing current issuance with pre-crisis levels misses the point that the pre-crisis market was fragile, based on investment by leveraged funds and bank treasuries. Building a stable market will require a broader, real money investor base.
The market also needs to be based on genuine risk transfer not regulatory arbitrage, such as synthetic sales of thin mezzanine tranches intended to maximise the reduction in regulatory capital at minimum cost.
I hope the future will be transactions in which banks look to sell down the capital structure, subject to meeting risk retention requirements. Transferring the entire risk on a portfolio will be the only way to achieve a reduction in the exposure measure used for the leverage ratio. And the new Basel risk-weighted capital requirements announced in December also move in the direction of incentivising transfer of senior as well as mezzanine tranches of portfolios.
I said I would come back to Simple, Transparent and Comparable (STC) securitisation.
Unfortunately, we have done our best to confuse people by using a bewildering range of acronyms and terms for this project – STC, SST, STS, qualifying, high quality etc.
Today I am using the STC term developed by the taskforce of global banking and securities regulators that I co-chair.
But I can reassure you that these will be brought together into a single set of criteria in European legislation.
I believe the intention is that they will then be applied consistently across different sectors – for example, banks, insurers and funds – and different regulations – for example capital and liquidity.
I see the aims of the STC criteria as threefold. First, to help investors. An STC securitisation should meet the ‘What You See Is What You Get’ principle. The structure should not have hidden traps or complications, allowing investors to focus on analysing the credit quality of the underlying assets.
But, importantly, the STC criteria are not a credit opinion on those assets and STC will certainly not mean ‘risk free’. Second, to help issuers.
As well as potentially helping to broaden the range of investors, the STC criteria should make risk transfer more robust. And third, to help regulators. Setting risk sensitive capital requirements for securitisation tranches is challenging. One solution is to use credit ratings but the shortcomings of that approach were exposed during the crisis. Another is to use a regulatory formula capturing dimensions of risk such as the credit quality of the underlying pool, tranche seniority and maturity. In our view, including a differentiation based on STC in that capital calculation helps to capture other important dimensions of risk related to structure, transparency and governance.
Because one of our objectives is to broaden the investor base, it is important that the STC designation makes things simpler for investors, particularly non-banks. For example, they should be able to place greater reliance on it when conducting their own due diligence on matters covered by the STC criteria, such as risk retention.
Investors would thus be in a position to concentrate with more confidence on other, existing due diligence requirements, such as in relation to the creditworthiness of the underlying assets and cash flows characteristics.
The implementation process for the STC designation is therefore important.
In our view, the primary obligation should be on the originator to attest compliance with the criteria.
But this needs to be in the context of a regulatory regime with effective supervisory oversight and sanctions.
We must also be careful not to label all non-STC transactions as ‘bad’. There should be a continuing place in the market for many transactions that do not qualify for the STC designation, including synthetic transactions, managed portfolios and pools that do not meet the granularity requirements.
Turning to capital requirements, looking at historical losses on securitisation tranches, we think the calibration of bank capital requirements that the Basel Committee published in December is broadly right.
But there is a case for some lowering of capital requirements for STC transactions on the grounds of lower structure risk.
A stronger argument can be made that Solvency 2 standardised capital requirements for EU insurers are still too high, especially at longer maturities - although it should be noted that these will not apply to insurers using internal models.
For banks and insurers, part of the issue with securitisation capital requirements is the comparison with covered bonds, which tend to be treated favourably in EU regulation. Issuers looking to raise secured funding may therefore see covered bonds as a more cost effective alternative to securitisation.
Covered bonds have a legitimate role in the market as a source of stable long-term funding. But securitisation has the advantages of risk transfer and lower encumbrance of underlying assets. We would support moves to put securitisation and covered bonds on a more level playing field.
To conclude, we support the European Commission’s current work on an EU framework for securitisation as part of the longer-term objective of growth in stable, market-based financing markets alongside bank lending.
A uniform set of criteria for STC securitisation can play an essential role in de-stigmatising European securitisation, helping the market to develop on a sustainable track and attracting a broader investor base.