Checking Dani Rodrik’s blog I’d noticed a new post that’s really pretty sharp. First, Dani had noticed an interesting post at Vox’s Global Crisis Debate by Katherina Pistor. I liked Katherina Pister’s post at Vox too. In fact, I’d say that Dani actually underestimates the potential of the thinking that underlies that post. The key excerpt:
The major argument against standardization as the cure all for financial crisis, however, is not that the wrong model was chosen. Nor is it the most common critique of legal standardization, namely that one model does not fit all. Instead, the idea that effective market regulation can be achieved by standardizing rules and regulations on the most successful model at the time is deeply flawed for the following reasons. First, it treats legal institutions as endowments and ignores the need for maintenance and adaptation not only to local conditions, but also to future change. Second, it creates the illusion that a given market is institutionally sound and thereby disguises problems that may trigger future crises. Third, the selection of ‘best practice’ models tends to reward regulatory regimes based on simple quantitative outcome variables, such as market size, even when market size may be the product of a bubble, while ignoring volatility and other risk factors.
So here’s my response to Dani (cross-posted over at his excellent blog also):
And here is my response:
I would tend to think that Catherine Pister’s comments have several other positives.
First, having different regulatory schemas in different countries, as opposed to a regulatory “monoculture” sounds like an excellent way to achieve a more compartmentalized economic structure that is inherently safer. Think of the analogies with natural systems, and how disease can and do run rampant in a monoculture.
Second, it is entirely possible that there exists regulatory schemes that work well only because of local conditions and would be very bad policies for another location. In that case, a consistent regulatory structure might only be able to be consistently BAD.
Third, regarding your concern of someone being able to somehow arbitrage the difference and thereby take advantage of this scheme of “a thousands flowers” of regulation, I suspect that you are right that some form of capital controls are needed but it is entirely possible that the necessary controls are inherent in the market structures that would arise should Catherine’s comments be acted upon. The ability to arbitrage depends on some kind of external reference structure and supra-national and very liquid markets they have to be. Compartmentalize the topology, and arbitraging isn’t so easy. In theory, one could have arbitraged the different regulatory environments in the 13th century for example, and made handsome profits selling nails in Greenland and Walrus Ivory in England but no one did, did they?
Another note: When Katherina says “The scale of the global crisis and its fallout for people around the globe calls for a new regulatory model. Rather than treating laws and regulations as fixed endowments that incorporate lessons from the past, a process-oriented strategy is needed that emphasizes continuous monitoring and adaptation of regulatory responses to changes in the market place.” her reasoning gets just a little slippery–the responsiveness criteria is at odds with the diversification criteria, and present regulations do incorporate important lessons that shouldn’t be forgotten (remember Glass-Steagall?? I lot of people wish they did!) and therefore I’d say tread lightly with that responsiveness criteria….
Oh, and one other note: here’s a post over at Global Guerillas about exactly that kind of decentralized (and therefore massively parallel) process, very similar to what Pister’s post at VOX is getting at:
Open source insurgencies (both global guerrillas bent on disorder and resilient communities focused on stability), the general description for movements where many small groups with different motivations cooperate to move the war forward, have a big advantage over traditional hierarchies (both conventional militaries and guerilla movements formed in the 20th Century). They adapt well to highly ambiguous, complex, and rapidly changing environments.
Open source insurgencies solve these problems by (I’ll draw a revised OODA process for this when I find the time):
- Decision making processes run in parallel. The larger the number of loops, the faster it goes.
- Multiple responses by a diverse group of participants (anybody can participate) generates a wide variety of hypothesis and decisions. Many of which, work.
- Sharing and rampant copying of the processes (i.e. the hypothesis and orientations) that work yields strong horizontal propagation. It scales, potentially even to a global level.