Whoops! Ring-fencing retail banks could backfire | FT Alphaville

Whoops! Ring-fencing retail banks could backfire

Regulatory snafu anyone?

The UK’s Independent Banking Commission (IBC) recommended in April that banks start ‘ring-fencing’ their retail operations so that large banks are able to fail without endangering depositors. That is, so-called ‘universal banks’ that provide both investment banking and retail operations will have to have retail subsidiaries, with separate and sufficient ‘ring-fenced’ capital to cover their own liabilities.

Small problem. Analysts say there might not be enough retail deposits to actually cover UK retail loans. In fact, they say, retail ring-fencing could end up harming Britain’s depositors.

First, though, some background. A centrepiece of current financial reform is the idea of burdensharing or bail-ins — that is, investors in failed bank debt will have to swallow some losses on their investment. That idea has helped encourage banks to issue more secured funding, such as covered bonds — or debt with a direct claim over a set of specific assets, typically high-quality mortgages or public sector loans.

Increased covered bond issuance (and also repos) is a very useful form of post-financial crisis funding, but it’s also having an interesting effect on investors in other types of bank debt, including depositors which typically rank on the same level as senior unsecured in the event of a failed bank. Every covered bond issued by a bank effectively subordinates the claims of senior and subordinate unsecured debt holders, leaving fewer assets available to settle their claims in the event of a bank failure.

This asset ‘encumbrance’ issue has been attracting the interest of some regulators and rating agencies recently, with the IBC also mentioning it in its interim report. South Africa’s banking authority, for instance, said earlier this month that it wouldn’t let banks start issuing covered bonds because they would subordinate the interests of bank depositors. Can you see where we’re going with this?

Over to Deutsche Bank’s Bernd Volk, plus banking analysts, in a piece of recent research:

There are not enough UK retail deposits to cover UK retail loans

A common misperception in media coverage on UK banking is that retail deposits are used to fund ‘casino’ investment banking, leaving depositors vulnerable. However, deposits alone are not enough to fund UK retail banking itself. The pertinent figures are shown in Figure 25: UK Retail represents 17% of group RWAs, 36% of group loans, and 40% of group deposits at the three listed domestic banks, Barclays, LBG and RBS. The banks listed below have a £150bn loan/deposit gap. The only UK retail division fully funded by deposits is HSBC.

Deutsche estimates that a £150bn funding gap is way too big to be filled by depositors alone — and on the liability side, the government is trying to get banks to lend more, not less. So, they figure, the funding gap for ring-fenced retail units will have to be met by funding through things like covered bonds:

Covered bonds lower depositor protection

CBs and securitisations are a material source of flow funding for UK institutions with the volume of UK covered bonds amounting to c.£119bn at Feb 2011 backed by £192bn of mortgages … £113bn mortgages back RMBS at Barclays, LBG and RBS.

Given that these funding instruments are tied to mortgages, we envisage that any retail ringfence would mean that covered bonds and securitisations would be part of the retail entity, and used to fill much of the £150bn funding gap.

So far, so good. But the pools of assets which back RMBS and covered bonds – and which are therefore not available in the first instance to meet depositor claims – would be a significant portion of the new retail balance sheet, which could mean that, instead of increasing depositor protection, a retail ring-fence could lower depositor protection compared with before.

We illustrate this using Barclays’ balance sheet below … Under the current ‘universal bank’ model, loans encumbered for covered bonds and RMBS form just 2% of assets, whilst deposits are 24% of liabilities. Depositors have recourse to all the unencumbered assets on the balance sheet (98%) in the event of failure and insurance for deposits up to £85,000 (FSCS levy for the industry for this cover is £217m for 2011).

If we carve out UK retail banking operations from the group (UK retail total assets + the UK assets from Barclaycard), the balance sheet looks significantly different (Figure 31): deposits are now 81% of liabilities whilst the total ‘unencumbered’ assets falls to 80% of the balance sheet. If we ignore the benefits of overcollateralization in the CB pool, in the event of failure there are insufficient unencumbered assets to pay the deposit base. Perhaps the simplest way to think about this is to say that the stated loan deposit ratio of the UK retail operation at Barclays is 123%, but if we exclude the residential mortgages funded by covered bonds and RMBS, this falls to 98%.

The IBC has already said it doesn’t think covered bonds should become bail-inable. So the problem remains; more (safer) secured funding means fewer assets left over for depositors and taxpayers.

Deutsche Bank figures the ICB may end up solving this predicament by placing a cap on the amount of covered bonds banks are allowed to issue — say, a percentage of total deposits or assets. Either way, the Deutsche note is a nice example of the complexity of current financial reform.

You cure one problem (‘Too Big To Fail’ banks) and another one crops up (encumbrance).

Related links:
The covered bond craze – IndexUniverse
Towards the limits of covered bond bank funding – FT Alphaville
Heavy (covered bond) encumbrance – FT Alphaville
Covering up the capital structure – Deus Ex Macchiato