Is it time to centralise capital calculation? February 1, 2013 at 3:45 pm

As many have noted (see here for Alphaville and here for Dealbreaker), the BIS study on market risk weights is out. To no one’s surprise, the results show that different banks calculate radically different capital requirements for the same portfolio. The report is full of embarrassing graphs like this. (It shows the variation of the three components of models-based trading book capital, VAR, stress VAR and IRC, for seven test portfolios (18-24).)

Hypothetical portfolio exercise

As Dealbreaker acidly puts it, banks rarely differ from each other by more than a factor of ten. It’s no wonder, then, that investors are losing trust in capital ratios. The answer is clear. Centralize and standardise capital calculation. Throw all those internal models away, and use one common, regulator developed approach. Now that a lot of progress has been made on trade reporting, the data infrastructure exists to do this — or least it wouldn’t be too hard to extend what does exist to do it. Developing the models would be a huge undertaking, but compared to having each large bank do it individually, a central infrastructure would be cheaper and more reliable, and anyway you could pick the best of individual banks’ methodologies. You could even spin out the var teams from four or five leading banks into the new central body — just don’t pick the bank whose IRC is less than 10% of the average answer…

8 Responses to “Is it time to centralise capital calculation?”

  1. LOL. So back to Basel 1!

    One could ELIMINATE regulatory capital ratios and make counter parties (and deposit insurers!) do their own solvency assessment. But of course I may be less willing to extend credit to some of these firms then, or charge them more! Once upon a time people believed in private market discipline.

  2. At the very least standardise VaR horizons and calibration timeframes. They’re just begging for regulatory arbitrage if they let banks cherrypick the timing.

  3. Regulatory arbitrage and single point of failure…

    Why not just simply open source the banks’ models? In your proposal we already have the input data, so if we force them to also disclose the code, then the market can analyse the models (which could lead to greater demand for your services).

    If you still need capital ratios, you could just do a weighted average of the models (as you’ve got the data and the code, you can run every bank’s model on every other bank’s data).

  4. Carter – I have respect for that view. You would, though, need to dramatically improve disclosure (and the penalties for false disclosure) for that old fashioned solution to work.

    GY – Yep, agreed.

    cig – Oh very nice, yes, there is no reason for the public model not to be open source. I imagine a utility much like Euroclear say, perhaps based in Basel, running a var model (and stressed var and irc) model for every bank that qualifies. Its methodology should indeed be completely transparent.

  5. […] Is it time to centralise capital calculation? It turns out that when you tell a bunch of banks to calculate their own risk-weighted assets, they come up with wildly different numbers for the same stuff. So, maybe that’s a problem. Here’s the thing though: you know why they come out with different answers? Well let’s refresh our memory on which this calculation involves: […]

  6. […] sardonic (and insightful) Rhymes With Cars and Girls and the less sardonic (but equally insightful) Deus ex Macchiato-discuss the fact that model-based risk-weighted capital calculations on the same assets differ […]

  7. […] leads in some cases to a wide distribution of results. This prompted Deus Ex Macchiato to comment ( that this indicates a dangerous or embarrassing divergence of outcomes, leading to lack of trust […]

  8. […] I want to distinguish two issues, as a prelude to another post on central capital calculations. […]