Sunday, June 22, 2008

Risk Management and Its Implications for Systemic Risk -- My Testimony to the Senate Banking Committee

On June 19 I testified before the Senate Banking Committee, at a hearing of the Subcommittee on Securities, Insurance and Investment. You can read my prepared testimony here.


  1. I think you are totally missing the boat here on all of your points in your testimony (but you are at least consistent with your book of which I have a similar view).
    1. VAR could easily be made to taking into account the increased correlation in extremes by mandating a more extreme correlation matrix (in the direction of positions on the books) when calculating tail probabilities, which would limit risk taking. This could easily be done even with historical data if risk managers included the Depression or Japan circa 1989 as an example. I was risk manager at a mortgage insurer and listened to the business's crazy justification for why house prices could not fall anymore if rates fell in 2005( which they did in the Depression and Japan). The argument was based on the hope or faith in the economic geniuses that run our economies. Well guess what rates in the US have fallen and house prices are dropping like rocks. (Similarily in the late 70's early 80's house prices rose as interest rates rose - again counter to the optimists selectively leaving out historical data).
    2. Government as liquidity provider of last resort, means you are hoping that someone (Government) without a profit motive (unlike Citadel) can take over a portfolio that has a high probably of rebounding due only to liquidity reasons. Without the profit motive it is doomed to be a repository of stuff that has a high probability of not recovering. Sounds like socialism for financial institutions to me.
    3. Mark-to-market should be mandated everywhere. The problem is that many entities (insurance companies for example) are NOT mark-to-market and so the price discovery process is delayed. Only triggers like rating downgrades (the recent MBIA Ambac downgrades wil force price discovery at last for these firms) result in instantaneous shocks rather than a gradual move. Of course if they were forced to Mark-to-Market they probably would not buy the stuff in the first place which would have helped avoid the liquidity bubble.

  2. Your testimony is a fascinating read.

  3. Another approach to mitigating the problem of spiraling declines is to prevent any one entity from owning so much of a market that if required to liquidate it would force the market significantly downward. With markets so global, it's difficult to imagine how such a limitation could be enforced, but perhaps US firms should be required to provide evidence that they are not in such a position in any market they enter.

  4. The current crisis is a spectacular example of a system accident where the regulatory system, (mark to market and inflexible capital regulations) acts as an accelerator of the downward spiral.

    You should write an article about it while the crisis is fresh.

  5. The video of the full hearing is available from this page:

    This should include questions and answers from all participants.

    As of today, there's no transcript.

    Charles P. Cohen

  6. Jean-Pierre BerlietJanuary 31, 2009 at 11:23 AM

    I share your views on the points that you made in your deposition, but feel that, collectively, we are too charitable to management of the firms that got themselves and us in the mess in which we are. My carreer in business strategy development, followed by significant work as a member of leading actuarial advisory firms has caused me to observe that;
    - much of what drives risk to solvency or to the value of a company is baked in the business strategies that firms pursue and their business models
    - naive reliance on "management by objectives" and "accountability enforcement" in performance management can actually increase risk by exacerbating conflicts of interests between managers and shareholders.( if the points are of interest, see my essay in the Risk Management monograph on Lessons from the Crisis, just published jointly on their werespective web-sites by the Society of Actuaries,the Casualty Actuarial Society and The Canadian Institute of Actuaries)
    Risk management frameworks that do not explicitly provide ways to address these challenges fall short of what is needed, even if they appear to "work"..
    In my judgement, the combination of naivete in the design of ERM (enterprise risk management) frameworks, inadequacies in the oversight of risk management, and the hubris with which senirr management and risk management professionals assume that they can actually identify, measure and manage risk made a significant contribution to creating the crisis. I wonder whether the Emperor might be half naked, or worse.