Watering down Basel III's not a sop to the banks

But it is still a hallmark of some worryingly misguided thinking.

A man walks down the banks of the Rhine in Basel, Switzerland. Photograph: Getty Images

The changes to the Basel III international banking regulations have been widely reported as a sop to financiers. But what actually happened?

The Basel regulations are about the stability of the banking system. When the third Basel accord comes into effect this year, it will introduce strict new requirements how leveraged-up banks can be, as well as mandating that they hold enough liquid assets to cover all of their cash outflows for a month. The idea is that by requiring these safety nets, the amount of revenue banks can make is curtailed, but so too is the risk that they will go belly-up in the event of another crisis.

The problem with Basel III is that reducing the amount of leverage a bank is allowed to use is the same as reducing the number of loans it is allowed to make, assuming its available capital stays the same. Reducing the number of loans is sort of what we don't want to happen, what with much of the developed world still being deep in depression and businesses clinging to survival by the skin of their teeth.

In fact, as the NYT's Andrew Ross Sorkin writes, the chances of a leverage induced crisis are quite low.

The change in Basel has been painted, by none-other than Mervyn King, as a trade-off. We thought that the big risk would be another bust; but now we know the big risk is a dead recovery. So lets water down the regulations. King said:

Since we attach great importance to try to make sure that banks can indeed finance a recovery, it does not make sense to impose a requirement on banks that might damage the recovery.

But the problem is, it's not Basel's leverage requirements that have changed. It's the liquidity ones. And they are a lot more important to implement sooner rather than later.

Leverage requirements are important in case we find ourselves in a situation like 2008, where the value of the assets banks are holding drops precipitously. Banks suddenly find themselves much poorer than they thought they were, and a wave of failures sweeps through the system. But we are a long way from the sort of bubble which is required for leverage requirements to be needed. First we need a recovery.

Liquidity requirements, on the other hand, guard against bank runs. And bank runs are a symptom of lack of faith in the system – something which remains very real today. The dilution of Basel now delays the implementation of those requirements, meaning that the risk of bank runs won't be actively fought until 2019; and it also weakens the very requirements themselves, allowing banks to claim a far larger pool of assets as "liquid capital".

Felix Salmon points out that what's really happening is that Basel III has become the latest in unconventional central bank actions:

The committee has clearly determined that if you’ve run out of ammunition in terms of interest rates and quantitative easing, then when you’re searching around for some other monetary-easing tool, regulations are a reasonable place to look. And I really don’t like that precedent. Monetary policy should be entirely separate from bank regulation, even if central banks should properly perform both roles. With the ink barely dry on the Basel III agreement, now is no time to start diluting it for the sake of some hypothetical temporary future marginal boost to growth.

It's important to point out that the actual changes may not be that bad. Alphaville's Lisa Pollack argues that there's a fair amount of whinging which ignores that the weakened regulations are still perfectly perfectly capable of fighting a liquidity crisis. But the principle of the change is still concerning. Regulators decided what would be the best and safest way of running banks, and then changed their mind based, not on new evidence that they could achieve the same safety with less stringent regulations, but on completely different criteria. That bears the hallmarks of the thinking which got us into this problem in the firs place.