Sunday, July 13, 2008

Paulson's rescue act & the need for better risk management regulations

The Treasury Secretary did what every one expected - step in and propose a unfettered vote of confidence on Fannie Mae and Freddie Mac. If the Fed could react with such promptness to save Bear Stearns (which i feel was right anyway), there's no way any one could expect the Fed/govt reaction to be muted for these behemoths which together buy/package almost 50% of the 12 trillion+ mortgage loans outstanding. Any delay would further depress already low markets and cause more trouble for the already-shattered housing market.

Having said that, I personally don't agree with the whole trend that such moves set - though given such a situation, there's pretty much nothing else that the Fed/govt could have done. The real trouble lies in the market mechanism which let things come to such a state. Of course, many questions remain on the role that risk management groups play in even mid and large tier investment banks - however in a quarterly-result driven market, little different could be expected in terms of market dynamics that force firms to sacrifice prudent risk managemenmt for profit-maxmimizing strategies backed by hollow VaR-backed 'risk management policies'. The real blame lies elsewhere though - as I probably have written multiple times over the past 6 months (!), one of the primary reasons has to be the lack of self regulatory (SRO-driven) or central regulatory supervision on credit rating agencies.

Moody's latest faux-pas related to rating errors on European constant proportion debt obligations is probably only the tip of the ice berg. In a scenario where rating agencies rate debt for the same investment banks which form their key client base for fee revenues, conflict of interest is the (only) name of the game. The other driving factor could lack of senior rating analysts to cover the vast volume of debt that gets issued in today's market. A small bunch of analysts using basic senior management interviews and review of the entity's internal VaR/other risk management limits cannot do justice given the ingenuity that exists in ths system. For example, as long as regulations allow hidden buckets like 'Level 3 assets' to park illiquid assets (no rule book can ANY WAY be fool proof), its very difficult to look at a balance sheet, conduct interviews, review adherence to firm-level exposure limits and decide on current solvency or credit worthiness. Forced adherence to stronger risk management policies leveraging tougher measures like Condition VaR or extreme scenario-based stress testing and modeling need to be made an industry norm - with penalties proposed for non-adhering members through higher capital requirements. The SEC should show some urgency to forge a strong SRO culture that would do multiple things:
a) define best practices w.r.t rating methodology for rating agencies, especially with respect to exotic and structured instruments,
b) code of busines guidelines for addressing the conflict of interest inherent in the model,
c) define guidelines address extreme/tail-end risks in internal risk managemnet frameworks
being some of the key.

We already have seen a good amount of laxness in interpreting and implementing capital allocation rules per Basel II rules in the US for example - unless this trend is reversed and a stronger SRO-driven risk management focus emerges, we would continue to see more systemic market crashes!