Facts about financial regulation
By pretty popular acclaim, the two banking systems and their regulators which have come through the crisis with their reputation closest to intact are those of Canada and Australia. With Sweden perhaps close behind. All three of these regulators share a specific policy approach. Can you guess what it is?
They more or less explicitly endorse a cartel of a small number of major banking groups in the domestic market. This ensures that a) the banks are consistently strongly capitalised, because they are profitable, b) they have a reduced incentive to expand lending too rapidly, because margins are high enough to allow them to earn a mid-teens return on their enlarged capital base, c) they have a reduced incentive to take risks in non-core business areas, because doing so would put at risk the "franchise value" of their core domestic lending business and d) they tend to have a much more respectful and obedient relationship with their domestic regulators because they know that the regulators are on the same side. France used to have a version of this more or less formalised (they called it the "ni-ni" regime - no charging for bank accounts, no payment of interest on current accounts) until they caught a bit of neoliberalism.
A profitable banking system is a safe banking system, it's known throughout the world regulatory community. Now, how are we going to fit this insight into our new post-crisis regulatory framework?
(POSSIBLE QUICK-WAY SOLUTION: We can regulate bonuses! That will make the banks more profitable! WHY IT DOESN'T WORK: Sadly, any cost improvements of that sort get competed away quite soon in a banking system that looks like the UK or USA rather than Canada or Australia).
(I actually think Australia is riding for a fall, for largely unrelated reasons. Australia, Canada and Sweden all have incipient real estate bubbles, which have been retarded but not eliminated by tough and prescriptive regulation. It will be interesting to see what happens if and when they bust, as this will destruction-test, one way or another, the idea that there was anything at all that could have been done at a regulatory level to prevent or mitigate our own crisis).
How does this fit with Spain? Where the much-promised banking catastrophe continues to not happen. I should probably bait Ed "a bear - like Russia" Hugh about that.
ReplyDeleteAs you know, my own theory about this is that a lot of the crap in all the other European banks is actually crap that the Spaniards managed to sell off before the crash.
Hmm, two Canadian banks in the top 10 for re-hypothecation.
ReplyDeleteSpain on the cusp of the model and has done a lot better than it deserved to given the craziness of the real estate bubble. Spanish banking is obviously in a hell of a state and is the largest source of damage to the sovereign credit, but it soldiers on
ReplyDelete"A profitable banking system is a safe banking system"
ReplyDeleteThis seems not so much an insight as a tautology; in my limited memory a banking system is only deemed unsafe when we wake up one morning and find its projected collective balance sheet has gone crimson.
Perhaps if there were a definition of "safe" for a banking system, other than "wot we puts money in so it won't vanish," we'd be pondering some other constraint that reality places on regulatory thinking.
Then again, the habit of regarding banks as great big marble safes is one that probably shouldn't be entirely abandoned, either.
There's a really interesting whole debate on the subject of return on equity in the banking industry; Andy Haldane thinks that most of the bad behaviour that took place was motivated by unrealistic aspirations of making a mid-teens return. Haldane's solution is (as far as I understand it) to make RoE targets illegal, which I don't think will work at all - my alternative is to make it so that the target RoEs aren't unrealistic.
ReplyDeleteI do think this debate is important though, because it is (or ought to be) pretty uncontroversial that a big part of the problem in the financial sector was to do with a heightened level of competition for market share - in as much as there's anything to the Claudio Borio "excess elasticity" view, this is the problem at the root of it.
Actually it's just struck me that this is a less trivial point than I thought. If (as a lot of people, not necessarily including me) believe, we want to have a banking sector that doesn't expand credit limitlessly in response to demand, then we are going to have to find some way of making sure that we don't get a load of new entrants to the market to take up the slack. If we're doing that, then we're running a cartel of banks. So if we're thinking about a "new financial infrastructure", then people do need to start thinking about the fact that this will be a cartel and what do we want to do about that.
Late to this, some slightly random thoughts:
ReplyDeleteIsn't this quite close to John Kay's "banks should be more like utilities" proposal?
There seems to be an obvious problem when considering the regulation of quasi-utility banks as compared to actual electricity & gas utilities. What actual utilities do is quite simple, so if you see one that's earning high RoE, it's fairly easy to find out why. I'm not sure the same is true with banks. Is the bank earning high RoE because it's taken on more risk, or because it's just better than its peers at managing either opex, or risk, or both?
Is this an argument for really fiercely constrictive banking regulation? I think it might be. I know that precautionary principle arguments should be deployed with caution, but since the regulator can never really be sure quite what the banks on its patch are up to, much better to be safe than sorry.
But I know next to nothing about banking regulation, and this is just from-first-principle doodling.
This comment has been removed by the author.
ReplyDeleteNowadays maximum company of govt and private commercial run their professional business using digital ML system. Digital banking system is one of the main example for this machine learning business management system.
ReplyDelete