Felix Salmon

Why there can’t be a cap on bank capital ratios

By Felix Salmon
November 5, 2009

I don’t understand much of the mathematics in Nassim Taleb’s new paper, co-written with Charles Tapiero. But still I fear that the paper’s main point is hidden in a blizzard of equations:



The point here is that the risk to the taxpayer associated with any given bank grows exponentially with that bank’s size. Ceteris paribus, a bank with $500 billion in assets is a lot riskier than a bank with $400 billion in assets, not only 25% riskier.

Treasury seems to consider too-big-to-fail to be a binary thing: either you are, or you’re not, and if you are, then you’ll have to get by with higher capital requirements. But if that kind of a scheme is implemented, then it automatically creates a strong incentive for any too-big-to-fail bank to grow, and grow fast. The bigger that a TBTF bank gets, the more moral hazard can be palmed off onto the taxpayer, while the bank’s own capital ratios don’t have to rise at all.

Treasury I think should consider a scheme where capital ratios rise steadily with the size, risk, and interconnectedness of financial institutions, rather than simply falling into one of two buckets. And there should be no cap on how high those capital ratios can get. If such a cap is put in place, then every big bank will simply be given an incentive to blow straight past it.

6 comments so far | RSS Comments RSS

you’re going to accept his conclusions without understanding his assumptions or logic just because he is taleb and can write fancy equations?

Posted by q | Report as abusive

I’ve suggested a progressive tax on bank size for funding the CFPA, but that’s to get Small Banks behind the plan. I’m still believe that TBTF will be worth whatever disincentives you try to put on these banks. I think that’s one of Mervyn King’s main points, and I agree:

“Capital requirements reduce, but not eliminate, the need for taxpayers to provide catastrophe insurance,” King said. He added it is “hard to see why” proposals such as those of former Federal Reserve Chairman Paul Volcker to separate proprietary trading from retail banking are “impractical.”

And, from another speech, here’s where much of the disagreement on Narrow/Limited/Utility Banking resides:

“The costs of this crisis are not to be measured simply in terms of its impact on public finances, the destruction of wealth, and the number of jobs lost. They are also to be seen in the lost trust in the financial sector among other parts of our economy. For a generation or more, businesses and families up and down the country were told, not least by the City, that the disciplines of the market economy were essential, even if painful in the short run, for greater prosperity in the longer term… We who work in the financial sector have much to do to regain the trust of those who work outside it… Change to the structure, regulation and indeed culture of our banking system is necessary. Blaming individuals is no substitute for acknowledging the failure of a system, of a certain type of banking. We have a real opportunity now to put that right, and regain the trust that has been lost.”

Leaving the impression that we didn’t have the stomach for tough reform, and so we did the best we could considering the fact that we’re leaving essentially the same system is in place, is toxic for the future if our govt’s finances go sideways.


the equations would be more coherant if you had included the part where he explained what W and Xi represent.


“Treasury I think should consider a scheme where capital ratios rise steadily with the size, risk, and interconnectedness of financial institutions, rather than simply falling into one of two buckets”

FWIW, this is exactly what the UK’s Financial Services Authority is proposing.

Posted by Ginger Yellow | Report as abusive

The paper was a very brief, rather humdrum application of a standard risk model. Hard to believe others have not worked through this. I’m not sure how much validity it really has.

Posted by Mike | Report as abusive

It seems that a binary-function threshold would be desired by the largest banks. Perhaps that is why it is being proposed.

My reasoning is as follows: there are roughly 19 banks that are too big to fail. (That is the number that underwent the “stress test.”) Those that are far above the threshold will be able to operate with higher leverage. This would give them a competitive advantage over the banks that are just over the threshold. As a result, the threshold would tend to act as a barrier to other banks that are trying to gain that competitive advantage. It would, in effect, be an anticompetitive practice, a practice that is not only allowed by regulators, but actually created by regulators. This strikes me as a particularly insidious form of regulatory capture.


Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/