Bailout

Basel III: The incomplete capital buffer proposal

By Felix Salmon

The Basel Committee has started producing pretty detailed documents: today it released what it calls a “a fully fleshed out countercyclical capital buffer proposal“.

The idea of countercyclical capital buffers is a really good one. When credit is expanding faster than GDP, bank regulators slowly increase their capital requirements, signaling those requirements clearly one year in advance. The higher capital requirements serve three main purposes: they help to slow down credit bubbles, they make an economy’s banks stronger, and they offer a way out of the paradox of capital.

The paradox of capital is pretty simple: let’s say that a bank has a minimum capital requirement, and then suffers a series of write-downs. Because the write-downs come straight out of capital, the bank is left below the minimum. So it is forced to raise new capital right at the worst possible time to do so, or else fail. The minimum capital requirements, which were meant to make banks safer, end up making the entire system more precarious.

Enter countercyclical capital buffers. With them, banks increase their capital in good times, not bad. And then, in bad times, they disappear: regulators can (and indeed are encouraged to) abolish the buffers immediately, if there’s some kind of credit crisis. When write-downs eat into bank capital, they eat only into the buffer, which is no longer required, rather than the underlying minimum capital requirement.

The BIS proposals are well formed: banks are required to hold capital according to the jurisdictions in which their loans are made, for instance, rather than where their headquarters are. That makes perfect sense.

But for a document which purports to be “fully fleshed out”, the single most important thing is missing: any indication of how big these countercyclical capital buffers should be. Essentially, the BIS has released the uncontroversial bits of the proposal, and is kicking the can down the road when it comes to the biggest, toughest question.

Reading between the lines, the BIS seems to be thinking of buffers which max out at about 2% or 3% of assets. Is that big enough? Should it be bigger? Where’s the debate on the actual number, and is any of that debate happening in public? These are important questions, and it would be great to have a lot more transparency on how they’re being answered. It’s possible that litigating such things in public is not the best way of coming to a consensus — but at the same time, if they’re adopted in smoke-filled rooms full to bursting with bank lobbyists, they’re unlikely to have much credibility. Let’s hope the BIS keeps us all posted on how exactly these buffers are going to get set.

Full post as published by Macroprudential on July 16, 2010 (boomark / email).