Being open about supervisory aims January 4, 2013 at 7:20 am
Before Christmas, I had a fascinating exchange of comments with long-time and insightful reader Quality Mullet. It began with my observation
It seems to me that the great debate we are not having is what we want the system to look like. If supervisors said `here is what we want, broadly’, then everyone could critique rules on that basis. As it is, supervisors have (often disparate) ideas of what they are trying to achieve which are not publicly articulated. The rules are a result of compromises between supervisors and so are intended to push towards the desired objectives, but often they don’t because of unintended compromises, conflicts between objectives, or because they are either too strong or too weak. The whole paradigm of `do what we are thinking of based on these rules’ is flawed – here’s what we are thinking of, help us get there would be better, but it would require both honesty about objectives and the ability to articulate what they are.
Put tersely, I was suggesting that rather than write rules to achieve a desired end state, supervisors should first articulate what their desired end state is, and then write the rules. That way, we could all judge the rules by their efficiency at achieving the goal, rather than trying to figure out what the goal was (and not being sure we had got it right).
In a sense, then, I am suggesting that the whole dialogue is at the wrong level. We discuss rules, when these are mean simulacra of their desired effects on the financial system. We are talking about the wrong thing when we argue about, say, the new Basel ABS rules improve financial stability or not. We should first say, for instance, that we want banks to hold fewer ABS (with all that that implies). Once that is clear, we can debate how to do that effectively.
QM commented in reply
… what’s not much talked about is how the regulatory framework and system structure you want is actually a very very important social contract. In the UK it is presumed the social contract has to be tilted away from public support but the debate is lop-sided, ignoring what it means on a day to day basis to end users. Vickers to his credit set out some of the tradeoff of losing a contingent liability but gaining a costlier … service. No one has taken his analysis much further in the public policy / social contract space.
This for me is absolutely correct. There is a social contract, and regulatory goals articulate it. By saying, for instance, that we do not want capital markets trading to ever be able to take a retail bank down, we are expressing part of the contract under which society has decided to let banks operate. We should make this contract explicit in order first for it to be agreed and second to better judge what the rules should say.
This then leads to my last observation: because the contract is implicit, it is fuzzy. Supervisors have not been forced to articulate their vision of it, nor has that vision been critiqued. By keeping dialogue at the level of the rules rather than the aims, we allow an unhelpful ambiguity to persist. If we don’t know what the official sector’s aims are, how can we judge their success or failure?