Do higher minimum capital requirements for banks do any good? July 21, 2011 at 5:24 pm

Here’s the problem.

If we set a minimum capital requirement, it is just that, a minimum. Banks have to meet it. The capital that they use to meet that requirement is trapped: it can’t absorb losses, as if it does, the bank is below the minimum, and that is not allowed.

The market also plays a role here. If a certain level of capital is required, then the market will demand that banks have at least that much before they will allow them to roll their debt. So again this capital is not available to absorb losses.

There is an argument then that raising minimum capital levels does not improve the ability of banks to absorb losses at all. All it does is decrease the loss to the taxpayer after banks fail. (In this argument banks fail for liquidity reasons while still close to being well capitalized due to the market declined to roll their debt.)

Rather, what makes banks more likely to be able to continue as going concerns is a relatively low level of required minimum capital, with banks holding higher levels of actual capital above that. The excess over the minimum is actually available to absorb losses.

Food for thought, perhaps, as Basel III implementation begins in Europe.

6 Responses to “Do higher minimum capital requirements for banks do any good?”

  1. […] Do higher minimum capital standards do any […]

  2. One does need to set the intervention point above the deadweight loss that occurs in liquidation to protect the insurer. Also, for systemic firms, setting a higher default point provides options in terms of reducing the cost of the externalities of failure.

  3. Carter

    The way I see it, there are actually two points. The minimum capital should be at least the amount the market requires to continue to permit the institution to roll its debt. The bank (even in resolution) will always require that much capital. The intervention point is above that, and the amount of capital a bank has as a going concern is above that. The key observation is that capital can’t do two jobs – absorb losses AND retain confidence, so you need to separate out the capital for the former from that for the latter.


  4. But if capital is linked to Risk weighting that is calculated using different formulas, can increased minimum level of capital restore confidence?
    What about a bank that holds 100% of its portfolio in European sovereign bonds that are at risk of default (provided that those bonds are associated to 0% RW)?

  5. Prof – you ask ‘can [an] increased minimum level of capital restore confidence?’. The answer is that it shouldn’t, as that capital is not available to absorb losses. The answer to ‘can increased level of capital ABOVE THE MINIMUM restore confidence?’ is ‘yes’.

    The point about free sovereign bonds is a red herring: no one would argue that those exposures should have had zero RW, but given that they do, now is not the time to fix the problem. Rather now is when you want banks to have capital ABOVE THE MINIMUM (am I getting boring here?) to absorb whatever losses materialize on those bonds and indeed anything else that happens to go sour.

  6. […] is because failures are typically liquidity rather than solvency driven, and so (as we have argued before) required capital is not available to absorb losses, only capital above the minimum […]