One of the recurrent themes in financial reform these days is that regulators need to get better about spotting systemic risk - or the risk that entire financial markets will suffer massive losses. One potential source of systemic risk is homogeneity in the investment and risk management strategies and portfolios of financial institutions. For example, if many different financial institutions buy the same kind of assets, you have herd behavior and a market boom. If, in reaction to a market decline, financial institutions begin selling the same assets to raise liquidity, the market goes into a tailspin.
How should regulation respond? I see three options. First, we can rely on traditional prudential regulations to regulate the types and concentrations of loans and investments that financial institutions make. There are a number of problems with this approach. Regulations can be gamed and can become quickly obsolete. This would require policymakers to continue to look for new forms of systemic risk and unhealthy homogeneity. And sometimes prudential regulations can also promote homogeneity, as banks would be restricted to similar kinds of investments.
The second option is to require greater disclosure of investment activity by financial institutions to regulators, who would then be responsible for spotting emerging systemic risks. This is the approach taken by the Committee to Establish a National Institute of Finance (http://www.cenif.org/). This approach would require regulators to devote massive resources (including computing resources) to map out emerging threats. One potential problem with this second approach is that the sheer scope of the task might overwhelm even well-financial regulators with access to high-quality information and powerful supercomputers.
This leads to the third approach, which I advocate, namely greater and more granular disclosure of financial institutions' investment and loan activity to the entire marketplace. Per my earlier post, I would like to apply concepts from the Open Source movement in software to financial regulations to allow the many minds of the marketplace to spot systemic risk. You could think about the difference between the second and third approach as a variation on Hayek's arguments for why market economies allocate resources in a superior fashion to planned economies.
The third approach would be a complement not a substitute for traditional bank regulation and inspection (in other words, backstopping the first and second approaches). Financial institutions might resist extremely granular and real-time disclosure of the investment opportunities because that would allow others in the marketplace to reverse engineer trading strategies and trade against the firms. So some mixture of a time lag, confidentially, and aggregation would be in order.
TrackBack URL for this entry:
Links to weblogs that reference Open Source and Spotting Systemic Risk: