I haven't done an analysis on this, but it feels like the financial sector has had more than its share of responsibility for scandal in the most recent economic "troubles." Which makes it even more tempting to regulate by compartmentalizing and dictating specific behaviors.
In the broadest terms, that would be a mistake.
Say you have a 17-year old son who wants to take the family car out at night. You're worried about alcohol abuse and aggressive driving-things occasionally associated with teenagers.
Do you say:
a. Absolutely not, and I don't care what Louie's parents are doing, you're not going out at night with the car until you're 18 / have your own car / etc. Period.
b. OK, but here are the ground rules, and if you violate them, here are the consequences; they will be severe and immediate, and I'll check randomly.
I think most of us would prefer b. If not, just add a year to the age. Eventually you're going to have to let the kid out at night.
More broadly, the question is: do you regulate by dictating behaviors, or combining audits with enforcement and sanctions?
You could answer this with ideology, or with a cost-benefit analysis. But there is one factor that I don't think gets mentioned enough. And that is trust.
Let's say an accounting firm is considered to have abused its relationship with its consulting branch. You could:
a. force accounting firms to sell their consulting businesses or build strong "Chinese walls (basically what Sarbanes Oxley did), or
b. increase the budget of the GAO, the Justice Departments' enforcement branches, and the sanctions for violation of rules.
Others know more than I about the costs of Sarbanes; let's just say it was high. But the real cost was that accountants won't get to rub noses with consultants. Firms won't have to develop their own policies. They have had to become compliance-driven, not outcome-driven.
The real trouble with structural regulation is that it removes the ability to evolve relationships, or trust, between business entities. Therefore it removes all possibilities for future economic improvement from trust.
Structural regulation is like putting up a concrete fence with your neighbor; it ain't coming down anytime soon. The opportunity cost is bigger than the out of pocket cost.
In a really excellent NYTimes piece-The End of the Financial World as We Know It by Michael Lewis and David Einhorn (I find them the best source these days for understanding what just went down)--they nonetheless plop for structural approaches.
They are probably right in part; revolving doors from government to industry lobbying, for example, is a pretty sure source of corruption. But structural solutions ought to come as last options, not first.
There are tons of laws governing the financial industry that simply get buried in process, language, bureaucracy, small print. They simply do not get enforced, and if enforced, they are toothless, and even then do not get publicized.
Before we build concrete walls, invest some money in creative auditing, enforcement, and sanctions that really bite. It at least leaves the door open for good players to do something really good with relationships. To grow up and drive right.
Link to original post