26 May 2013

Basel III: To Be Suspended and Potentially Dismissed

The so-called Brown-Vitter Bill, the draft legislation which will be known as "Terminating Bailouts for Taxpayer Fairness Act of 2013" or the "TBTF Act” should U.S. congress enact it (and there is a good chance that it will), effectively prohibits implementation of Basel III capital standards for the U.S. banks. In fact, Basel II/III should be dismissed not only in the U.S. but it ought to be dismissed all together for the reasons of being overly complex and inconsistent with the real world. The new international standard needs to be much shorter, much simpler and to the point.

Andrew G Haldane, Executive Director for Financial Stability at Bank of England, has made a strong case about the excessive complexity of Basel III capital calculations in his speech at the Federal Reserve Bank of Kansas City’s 36th economic policy symposium (Jackson Hole, Wyoming, 31 August 2012). The accompanying research paper entitled “The dog and the Frisbee” provides a considerable analytical support to his views. Short summary of the key findings is as follows.

For the starters:
  • Length of the Basel III document is 616 pages; regulations adopting Basel III in the U.S. and in Europe are topping 1,000 pages each.*
  • For calculating regulatory capital ratios, the parameter space of a large bank’s banking and trading books could easily run to several millions.
  • By 2011, regulatory reporting had reached a magnitude of 2,000+ columns in an Excel spreadsheet. (Authors of the paper are noting that fortunately, the column capacity of Excel had expanded sufficiently to capture the increase.)
  • Number of financial supervisors is increasing in line with the complexity of regulations.
  • Complying with Basel III requirements translates into over 70,000 full-time jobs in European banks, and to tens of thousands full-time positions in the U.S. banks.
When you’d imagine all the work that these numbers assume, you’d probably ask: “Is the complexity of financial regulations really worth it?”

The irony is that complex model-based risk-weightings (the primary source of complexity in the Basel framework), supposedly improving risk-sensitivity of the regulatory framework, appear not to add a bit to the detection of bank weaknesses. Other way round: based on the analysis of real banking data, the authors of the paper find that risks of complex banks are better captured by the simple leverage ratios with assets equally-weighted – a 1/N rule – than by granular, risk-sensitive capital measures.

They estimate that for “true” models to start outperforming much simpler ones, longer observation periods would be needed – much longer. One of their analyzes suggests an observation period of 400-1,000 years to justify Basel II/III type of capital calculations. Yet the exact length of the observation period is even not that important at this point; what matters is that the data required to motivate the use of complex models is well beyond everyone’s reach. (In my opinion, no amount of data would be enough given the very nature of social systems.)

This conclusion is supported by the complexity theory and a number of references to the relevant literature.

In summary, Haldane draws a parallel between the Tower of Basel and its near-namesake Tower of Babel, both creating ever-increasing confusion. He argues for “less is more” principle and I agree.

Given the above, it looks odd indeed that implementation of Basel III is generally being promoted as an important part of the solution to the current problems in banking sector by the IMF, by the central bankers, by the financial supervisors, by the rating agencies...

It looks odd even if it can easily be explained by the motivations of big banks (running a bank with almost no capital while being able to claim solid capital ratios), rating agencies (use of their ratings in the regulatory capital calculations), and staff employed for implementing and supervising complex regulatory frameworks (they have their rather well-paid jobs). After all, admitting being wrong is not that easy for anyone who has been involved long enough.

Key message: Calculation of regulatory capital ratios based on Basel II / III frameworks has to be suspended and potentially dismissed as detrimental. At least until there is no compelling proof that complex regulatory capital framework actually is superior to a simple leverage ratio (possibly including some sort of very robust risk weighting), that simple leverage ratio should serve as a basis for capital requirement. In that sense Brown-Vitter Bill in the U.S., although far from being a solution to all banking and regulatory issues, is a welcomed step.

Last but not least: moving to a much simpler regulatory framework should not be viewed as moving backwards; in fact, something simple and to the point is always or almost always more difficult to create than its complex (and less useful if not useless or even harmful) alternative. Hopefully knowing this provides at least some encouragement to the financial regulators also in Europe to re-consider financial regulation on a very fundamental level.

Bankers and bank employees, especially in risk control and compliance functions, have to draw their own conclusions respectively.

* We should note that in addition to capital, Basel III also addresses liquidity; however, capital-related issues are still more than a good half of it.


  1. From my experience, people who like Basel III, usually are either

    1) people who wrote it;
    2) people who never read it.

    Having read Basel III requirements, I quite agree that dismissing them altogether would make the world a better place.

  2. I further found a good piece by the FDIC Vice Chairman Thomas M. Hoenig: "Basel III Capital: A Well-Intended Illusion": http://www.fdic.gov/news/news/speeches/spapr0913.html