The Paradox of Perfect Supervision

I do not think that is true, what I have called the paradox of perfect supervision. The opposite is more likely. The more supervision focuses on measuring and controlling risk, the more likely we get to a place we don't want to end up at.

Perfect supervision assumes that, with enough information and control, risk can be set at the optimal level. But when trying to control risk in complex systems, we might not eliminate it, but rather shift it or transform it — often in ways we cannot easily observe.

Attempts to contain risk too tightly can make the system more uniform, more correlated, and ultimately more fragile. It is in this sense that the pursuit of perfect supervision can become self-defeating.

There are three reasons.

The first is basic incentives. The more responsibility we give to supervisors, the more we end up outsourcing risk management and hence responsibility to them. When things go wrong, is it the fault of the bank or the supervisor? The private sector is very good at exploiting this for its own benefit.

The second problem is one of knowledge. The only risk we can measure with reasonable accuracy concerns day-to-day events — those that do not matter much in the grand scheme of things.

We cannot measure the risk of catastrophic outcomes such as the likelihood of banks failing or financial crises. By pretending we can, we create a culture of complacency by focussing attention on the least important eventualities, and in doing so, give plenty of space for the forces of instability to emerge elsewhere.

The third problem is one of system design. When the inevitable shock happens, a war or a virus perhaps, our prudent financial institutions have no choice but to get rid of risky assets. Because that's what the regulations tell them to do. In practice that means selling into a falling market, so the financial institutions act as crisis amplifiers.

At the root of this is that the culture of risk measurement and control is not up to the task at hand. A better way forward would be to borrow the basic principle of finance.
(These remarks were first given at our event Supervisors on Supervision — Culture in the Financial Sector on 22 September 2025, hosted by the Department of Finance in the Marshall Building, named after one of the most successful hedge fund managers in the United Kingdom.)

Successful investors diversify their portfolios. Finance 101.

Successful supervision should diversify the system they supervise. Diversified supervision allows institutions to respond differently to pressure, so the system as a whole can absorb shocks rather than amplify them.

I find it ironic that we do not apply the most successful principle of finance, diversification, to the supervision of finance.

The paradox of perfect supervision is that control seeks uniformity, while resilience depends on diversity.