The biggest problem for regulators is “capture”—a situation that happens when their activities are influenced excessively by those they are supposed to be regulating. In the case of banks, you can imagine how this happens.¹ A bright, young regulator receives a call from a vice chairman of the bank he would quite like to work for later. The vice chairman explains that there are some complex issues at stake and to help clarify them he has asked an eminent economist to write a report explaining why a regulatory proposal is economically inefficient. The vice chairman adds that if the offending regulation is enacted, the bank will have to move its profitable operations to the British Virgin Islands, with the loss of hundreds of local jobs.

This is a crude caricature. The capture is always gentlemanly, sophisticated, and subtle. One way to prevent it would be to make it less appealing for bankers to employ regulators by imposing a long period of “quarantine” before a former regulator can join a bank. But there is still the issue of pay. The pay differential between the regulator and the regulated is one of the causes of capture.

I propose an instant and cost-free solution. Before a bank supervisor considers any regulation, he should ask what it would look like if he were captured, against the public good, by the largest banks. He should then circulate the “captured version” to as many people as possible. Knowing your adversary is one thing, but knowing his agenda is better still.

What would Basle II look like if the committee had been captured by the large banks? First, the regulation would be very complex. Complexity is the avenue of capture. The simpler things are, the more you can see what is going on. A five hundred-page rule book, for example, is an intimidating barrier to entry for nonbankers. Complexity also makes regulation less easy to enforce.

Second, we would expect the regulation to focus more on internal processes and less on outcomes. Big banks can more easily carry the cost of sophisticated internal risk assessment than banks that are small, new, or operating in developing countries.

This is exactly what Basle II looks like. There is complexity where there should be simplicity. There is also a lower capital requirement (a subsidy) for those with sophisticated internal risk assessments, without much attention to whether these assessments work. Internal processes on their own do not keep banks away from bad lending. Indeed, if financial institutions use similar internal processes, it can lead to financial instability.

Today’s sophisticated internal risk assessments work via daily, price-sensitive risk limits. They require a bank to reduce exposure to risk when the probability of losses increases as a result of falling or more volatile asset prices. When a handful of banks use these systems, everyone may be better off. But if every bank uses them, and they have similar positions, when falling prices cause one bank to hit its risk limit, other banks hit theirs, too. As many banks try to sell the same asset at the same time, prices plummet and volatility increases, ­causing more banks to hit their risk limits.

As long as market participants herd, which they have been doing for as long as markets have existed, the spread of sophisticated risk systems based on the daily evolution of market prices will spread instability, not quell it.

Price-sensitive risk limits also add to herding. Financial crashes occur shortly after the point of greatest optimism, when prices are high and risks appear low. This is one of the reasons why the large banks, with their sophisticated internal risk systems, have been caught up in every market cycle. They lost considerable amounts during the dot-com bubble and on companies with crooked accounting. They may be about to do so again on their syndication of collaterised debt obligations—the next bubble to burst.

As their price-sensitive risk limits have forced them to offload these bad investments at the same time, their actions have triggered cascading declines in equity prices, undermining their own positions and those of pension funds and insurance companies.

None of the big banks has failed—yet. But we do not know what would have happened if the implicit government guarantee had not been there. Even with that guarantee, the markets are uneasy about banks. Their bonds have been downgraded and their bondholders have suffered capital losses. It is odd that the proposed response is to reduce the regulatory capital required of these systemically important banks.

Basle II will amplify the credit cycle and add to instability—a particular hardship for small companies and developing economies. Financial crashes occur because of a collective abandonment of common sense by the market. The history of finance, from the tulip mania of 1637 to the dot-com bubble, is full of such lapses. Only a captured regulator could conclude that an industry of such systemic importance, so prone to mutual self-delusion, is ready for more self-regulation.

___________________

¹This article was first published in the Financial Times on October 17th, 2002. Reproduced with permission.