Replacing market failure with regulatory failure
This is the fifth in a series of posts about ideas for financial reform generated by the “Make Markets Be Markets” conference I attended yesterday in New York City on 3 Mar 2010. You can download all of the written presentations here.
Let’s talk about regulatory capture and financial reform for a few minutes. Arnold Kling has an interesting take on last week’s conference. He wasn’t there, but did see the videos online. In a recent post, he says:
What is ironic to me is that the Roosevelt Institute people can cite clear evidence of government failure while proposing more government. For example, Richard Cornell points out that bank soundness regulators have every incentive to be lax until it is too late, yet his proposed solution is to consolidate bank regulation. Cornell begins his talk by listing comic book characters who never seem to learn (Charlie Brown, Wylie Coyote). In my opinion, he could be talking about policy wonks who think that regulation can be reformed so that special interests can be curbed and regulators will act wisely.
In selective cases, and without consciously realizing it, these folks apply public choice theory. But they assume it away when making their proposals.
I think he makes a good point here. He’s saying in effect that many regulators are captured by the industries they are supposed to regulate. Everyone knows this – and many of the Roosevelt Institute presentations point this out. Yet, those very same presenters seem to think the solution is to just get different regulators who aren’t captured or change the structure of who does the regulating. Kling is suggesting that doesn’t work. If market failure is what caused the credit crisis to spiral out of control, you’re just replacing market failure with regulatory failure.
So, how do we deal with this problem?
Self-regulation. The ivory tower libertarian solution would be to de-regulate i.e. reduce regulatory oversight and permit self-regulation to rule the roost. I liken this to having citizen police officers (with guns) to police criminals. If an individual is caught stealing a car, the self-regulating community can apprehend the criminal and let the court system adjudicate his sentence. Quite frankly, I think this solution is nuts. This is the Libertarian paradise in Somalia – and it’s called anarchy. Moreover, it assumes the economy tends to equilibrium i.e. when a large bankruptcy disrupts the economy, eventually things will right themselves. This is a false neo-classical assumption.
Trust but verify. A better approach would be to operate a two-stage system that I will call “trust but verify.” Judge Stanley Sporkin was right when he voiced his frustration in Q&A at the end of the conference that you can’t have a good regulatory system without regulators who actually regulate. So, I agree with him that stage one is filling regulatory seats with real regulators, rather than anti-regulators like Alan Greenspan, who believed that markets could self-regulate fraud. But that still leaves you with a potential that those same regulators could be captured or that at a later date different regulators could be captured.
As a result, you need a second fallback mechanism. It’s called bankruptcy. The problem with having let Lehman Brothers fail was not that it failed, but that a mechanism was not in place to limit the damage of that failure. Of course, Lehman should have been bankrupted. That’s the price you pay in a capitalist society for failure – and it’s the reason we have bankruptcy laws. Now, for most firms, bankruptcy is not destabilizing to the entire national or global economy. However, large financial companies like Fannie Mae, Freddie Mac, Citigroup, AIG, and Bank of America are too big to fail (TBTF). We have to deal with this problem by breaking them up – or in the case of Fannie and Freddie, eliminating them. If you have another solution i.e. a failsafe TBTF resolution mechanism, please chime in.
So, I say install the right people as regulator and trust the regulator to her job. But verify that the job is being done by first allowing firms to fail and then replacing incompetent regulators.
In my view, any sort of “resolution authority” other than the current bankruptcy regime that we have now is just a euphemism for a bailout authority.
Many have suggested a model similar to the FDIC, but the purpose of the FDIC model is to provide a bailout to one class of creditor, bank depositors. What exactly could a FDIC like structure do to unwind big financial institutions without causing the same sort of harm caused by a traditional bankruptcy except by bailing our some or all of the creditors?
So I agree with you. Those TBTF must be broken up.
Greg,
I have floated an idea that some others have mentioned – a special bankruptcy court for large financial institutions. My thinking is that we could carve out specific capital parameters to avoid bankruptcy. The equity would take say all of the losses up to a point, after which point they take 75% of future losses and sub debt takes the rest by mandatory equity conversion, etc. The point being that you could prevent a bankruptcy by moving up the capital structure in a pre-defined way. I think I saw another proposal like this somewhere where the main question was how to prevent bankruptcy arbitrage. I’ll see if I can dig that up.
Barring something like this, you really are just going to have to break these banks up because resolution authority gives way to much power to the executive branch.
I like the contingent capital idea in theory, although I think there would be a number of practical difficulties over how you determine whether the conversion point has been reached and how you determine the conversion price. You could specify in advance the trigger point and a conversion formula, but someone like a bankruptcy judge or a regulator would have to have the authority to make the determination.
If you could get around the implementation issues, a contingent capital concept could help preserve the franchise or going-concern value of an insolvent institution which would lessen creditor losses. However, it still seems to me that there could still be big creditor losses even with contingent capital, or at a minimum great uncertainty about the extent of the creditor losses. In addition, I think with Lehman, a big part of the systemic issues was that once creditors were forced to accept that they could suffer losses, no one trusted anyone else’s balance sheet leading to a big run on the shadow banking system.
I still end up back at the same place that any sort of resolution authority wouldn’t solve the systemic issues, unless the resolution authority is going to also have bailout authority (which I suspect is the Obama administration’s real goal). So, I would still want to break up the TBTFs.
I think the systemic issues can also be lessened by strenghtening the balance sheets of financial institutions and by reducing the opacity of their balance sheets. So, more capital, more reserves, less leverage, no off-balance sheet stuff, and regulation of derivatives, repos and money market funds.