My opinion of crypto is unprintable, so I won't express it here. Paul Krugman has just given the bowdlerized version. But I'm not sure if the "derisking" logic makes sense. The logic of capital regulation is that only aggregate risk is important. If a bank wants to go to Vegas with a small stake for a positive NPV, prudential regulators should approve. How is crypto different? Balance sheet is balance sheet. Risk is riskiness multiplied by exposure. Eh?
The rebuttal would point to the criminality pervading crypto. Risk of being too closely associated with criminals is not the kind of risk Basel had in mind. The surrebuttal would point to private banking--an intensely scummy business if done profitably. (Visit Basel or Lugano for local color.). But supervisors still permit it.
What about the FDIC/OCC style microprudential supervision that would indeed object to a Vegas trip? They don’t want these banks to fail at all. The all-important A in CAMELS. Right?
A trip to Vegas is a less certain way to lose money over an economic cycle than CRE. Yet smallish banks are allowed to write a lot of CRE because it is the only way they can make money. The all-important "E" in CAMELS. (IMO, the all-important part of CAMELS is "M.")
Strong points. So your view is that there is no role for supervisor-imposed derisking? Or that if we go down this path we need to shut down CRE portfolios top to bottom?
I wouldn't mind de-risking for reputational risk if there were any structured way to evaluate it. "Your customers are scumbags" isn't very structured, and is too moralistic for my taste. And "drawing political heat"--although perhaps the basis for a rational structure--has a real heckler's veto problem, putting aside constitutional concerns. "Proximity to criminality" might be better. We already have this for politically exposed persons.
As far as microprudential derisking, I view CRE as one of the subsidies baked into our system of banking regulation--much like FDIC insurance for brokered deposits or credit card bank liabilities: both of which are nearly identical to commercial paper. I have no problem with controlling abuse of the subsidy through prudential supervision. It is inconsistent with the Basel capital regulation approach.
I guess one could argue that the big banks are smarter than the micro supervisors, but the micro supervisors are smarter than the little guys. Or that bank supervision is large: it contains multitudes.
My opinion of crypto is unprintable, so I won't express it here. Paul Krugman has just given the bowdlerized version. But I'm not sure if the "derisking" logic makes sense. The logic of capital regulation is that only aggregate risk is important. If a bank wants to go to Vegas with a small stake for a positive NPV, prudential regulators should approve. How is crypto different? Balance sheet is balance sheet. Risk is riskiness multiplied by exposure. Eh?
The rebuttal would point to the criminality pervading crypto. Risk of being too closely associated with criminals is not the kind of risk Basel had in mind. The surrebuttal would point to private banking--an intensely scummy business if done profitably. (Visit Basel or Lugano for local color.). But supervisors still permit it.
What about the FDIC/OCC style microprudential supervision that would indeed object to a Vegas trip? They don’t want these banks to fail at all. The all-important A in CAMELS. Right?
A trip to Vegas is a less certain way to lose money over an economic cycle than CRE. Yet smallish banks are allowed to write a lot of CRE because it is the only way they can make money. The all-important "E" in CAMELS. (IMO, the all-important part of CAMELS is "M.")
Strong points. So your view is that there is no role for supervisor-imposed derisking? Or that if we go down this path we need to shut down CRE portfolios top to bottom?
I wouldn't mind de-risking for reputational risk if there were any structured way to evaluate it. "Your customers are scumbags" isn't very structured, and is too moralistic for my taste. And "drawing political heat"--although perhaps the basis for a rational structure--has a real heckler's veto problem, putting aside constitutional concerns. "Proximity to criminality" might be better. We already have this for politically exposed persons.
As far as microprudential derisking, I view CRE as one of the subsidies baked into our system of banking regulation--much like FDIC insurance for brokered deposits or credit card bank liabilities: both of which are nearly identical to commercial paper. I have no problem with controlling abuse of the subsidy through prudential supervision. It is inconsistent with the Basel capital regulation approach.
I guess one could argue that the big banks are smarter than the micro supervisors, but the micro supervisors are smarter than the little guys. Or that bank supervision is large: it contains multitudes.