Saturday, 13 April 2013

What Financial Regulators Can Learn from Network Theory

When regulators seek to identify systemically important financial institutions (SIFIs), they tend to focus on an institution’s size and connectedness. But this approach mises an important dimension of systemic risk, according to Imre Kondor, Stefano Battiston, Giorgio Fagiolo, and Alan Kirman.
This team of economists and physicists emphasizes that systemic risk is a complex and collective problem, not something that can be read off the balance sheet of individual units. They take into account all kinds of indirect influences – managers in business schools and quants at trading desks comprise a social network whose members consume more of less the same information – to investigate how strong correlations can arise between seemingly unrelated institutions at dispersed areas of a network. Strongly correlated clusters of institutions, they say, are what regulators need to watch out for.


This is a very interesting line of thoughts…my problem though is that there are three of them, because that makes them, albeit small, into a de-facto net-work, and so, de facto, it could also turn them into a very small mutual admiration club, something which effectively puts a lid on what they can produce.
For instance, it is amazing how we correctly want to assure the existence of independent central bankers, but then do not worry at all about how independent our central bankers really are.

RE: Strongly correlated clusters of institutions, they say, are what regulators need to watch out for.
Yes; a couple of years ago I proposed a countervailing solution called a virtual interactive think tank for macro prudential regulation, a kind of Internet supported "college of wise and connected" that would have the comprehensiveness and depth of understanding needed to understand and counter those risks.
Val Samonis

Our financial system is designed not to fall prey to financial contagion (a form of network problem).
The financial system is based on the FDR Framework which combines the philosophy of disclosure with the principle of caveat emptor (buyer beware).
Under buyer beware, investors know they are responsible for all losses on their investments. As a result, they have an incentive to both use the information disclosed and, more importantly, to limit the size of their investments to what they can afford to lose given the risk of the investment.
When buyers limit their risk to what they can afford to lose, it eliminates financial contagion.
Our ongoing financial crisis has revealed that one of the toxic byproducts of opacity is financial contagion.
In this case, a lack of transparency combined with cheerleaders (think regulators for banks and rating firms for structured finance securities) led market participants to believe that the risk of these investments was less than it truly was. This led to investors over-investing.
One example of an investor that over-invested was our financial institutions. They took on more exposure to each other than they could afford to lose and maintain a positive book capital level.

Yes, Network Theory, and similiar notions concerned with complexity are fine as far as they go.However, what if it were actually possible to continually monitor the economy in real-time? Could this not give a far more accurate understanding as to the actual mechanics of economics itself? This is explained in my evolving project of Transfinancial Economics found at the p2p foundation. Moreover, a degree of commercial confidentiality could also be respected.

No comments:

Post a Comment