05 February 2011

Running a Bank with Almost No Capital

In his speech to The Swedish Society of Financial Analysts on 1 February 2011, Stefan Ingves, the governor of Sveriges Riksbank, well summarised the basic problem with the current regulations in terms of allowing the banks to hold far too little risk-bearing capital. Indeed, it is possible to run a bank with less than 1% capital of total assets!

Consider a simple calculation. If a bank applies the most advanced Basel II approaches when calculating its risk weighted assets (RWA) for credit risk, for market risk and for operational risk, its risk weighted assets can form less than 15% of its total assets. This is the case e.g. with such big players like Deutsche Bank and UBS (data as of 30 September 2010). The capital requirement for Tier 1 Capital in Basel II is 4% of risk weighted assets and Core Tier 1 Capital (share capital and retained earnings) can be even lower, let’s say 85% of risk weighted assets. This effectively means that Basel II allows a bank to be run with capital as low as 15%*4%*0.85%=0.51% (!) of total assets. Add some Pillar 2 stress testing buffer, and the result can still be less than 1%. Of course, for doing that the bank has to declare a low risk profile; however, less than 1% seems far too little for common sense no matter the arguable risk profile.

Let’s now take a real example: UBS. We use the data from Q3 2010 report (the latest data available when the analysis was performed). UBS’s total assets as of 30 September 2010 were CHF 1 460 509m. Its total RWA at the same time was CHF 208 289m. Thus, risk weighted assets formed ca. 14.3% of total assets. So regulatory minimum capital requirement for UBS in terms of Tier 1 Capital effectively was 0.57% (14.3%*4%) of total assets, and Core Tier 1 Capital requirement 0.48% of total assets (0.57%*85% when assuming that Core Tier 1 Capital forms 85% of Tier 1 Capital).

In reality, on 30 September 2010 UBS had Tier 1 Capital in the amount of CHF 34 817m and Core Tier 1 Capital in the amount of CHF 29 579m. Thus, Basel II Tier 1 Capital ratio was 16.7% and Core Tier 1 Capital ratio 14.2%. Looks strong, doesn’t it? In fact, Core Tier 1 Capital formed only 2.03% of UBS’s total assets, which makes me question if the bank is really sufficiently capitalised. Furthermore, meeting Basel III non-risk-based Tier 1 leverage ratio of 3% of total exposures seems more than questionable without additional capital or accounting tricks (as you can see, based on Q3 2010 data, the Tier 1 Capital divided by total assets was only 2.38%). No wonder that in its feedback to BIS regarding Basel III in 16 April 2010, UBS has argued against including cash and cash-like instruments into the calculation of leverage ratio, against taking off-balance sheet items into account with 100% credit conversion factor, against non recognition of accounting netting etc.

The fact that the calculation of risk weighted assets under the current regulatory framework is a bit strange, has also been noted e.g. by Standard & Poor’s and by Sveriges Riksbank, arguably the world’s oldest central bank. The former has introduced its own methodology for calculating the Risk Adjusted Capital of financial institutions. The later has analysed the capitalisation of Swedish banks in international comparison just based on Standard & Poor’s framework and not based on Basel II capital ratios (see p. 56 of the Riksbank’s Financial Stability Report 2010:2). Notably, UBS doesn’t look good in this comparison: its Risk Adjusted Capital Ratio is just two comma three or something percentages.

To summarise: it’s almost unbelievable, but Basel II has allowed running a large international bank with almost no capital. Some major too-big-to-fail or even too-big-to-save European banks have used this opportunity, as a result of which are now undercapitalised. Nothing will happen until banks make profits, but a double dip scenario or new economic downturn would be a real trouble.


  1. Edward C D Ingram11 June 2011 at 19:50

    In my coming Book, I will argue that for the UK and the USA and similar economies aiming to have 1% inflation, and 3%p.a. economic growth the nominal interest rate for Housing Finance will rise above 7% before inflation is contained at that level. The formula worked in correctly predicting the rate increase put in place by the Fed pre-crisis.

    The argument is based on the historic relationship between interest rates and nominal GDP growth rates, or Average Earnings Growth.

    With some lenders still offering loans at well below that 7% figure, how do they think that borrowers will cope with the resulting 20% - 50% rise in their level of payments? And what about house values? They are still inflated.

    My book sets out a solution if anyone wants to listen.

  2. Really , surprised to know this. Is it possible? Great. Thanks for such informative post.

  3. It’s really unbelievable .Your post is really informing about the miracle but you have proved it reallity by showing examples in my eyes..

  4. Thank you for your comments. Honestly, I'm surprised that people are so surprised about this disclosure. On the other hand, it's understandable because everyone cannot and does not need to be an expert in these issues. What worries me, is that many of the banks' CEOs do not seem to get it either (yes, it's about details, but the devil lies in details). Ok, maybe they are getting it, but they do not let you know.