Bank Risk Models

Q: Why bank risk models failed?
A: ....... Blame the models!?
....................................................
....................................................
.......Hell Yeah, Ya Right !!!

Why bank risk models failed by Avinash Persuad
This article calls for an ambitious departure from trends in modern financial regulations to correct the problem caused by subprime mortgage crisis in the United States. The main purpose is to deal with the failure of bank risk models implemented by financial institutions, banks and other financial intermediaries.

Lehman Brothers, with a long and famous legendary in contribution to U.S. economy for 150 years, filed bankruptcy in mid-September. The effects of falling down are so huge that it was well-known worldwide immediately, which economists, investors, academics called it as "Lehman Shock".

Greenspan (former Fed Chairman) and others raised question why risk models failed to avoid or mitigate the current financial meltdown.

Avanish Persaud of Intelligence Capital granted 2 answers, one technical and the other philosophical. He argued that "market-sensitive risk models" used by main players in the financial markets did work smoothly as it should be. The models assume that each user is the only person using them. Investors have the same data on the risk, returns and correlation of financial instruments and they use standard optimization models.

Profit-maximization theory discourages them to invest in un-favoured market. As a result, when risk models detect a rise in risking their portfolio (rise in volatility), they try to do the same thing at the same time with the same assets for the same purpose. Therefore, a vicious cycle ensues as a vertical fall in prices, prompting in more and more selling (sell it at a low price before the price get lower approach). Then excess supply leads to a further depreciation in prices.

Avinash has also raised one concern (to achieve effective risk models) about the paradox of the observation of areas of safety in risk models and the observation of risk. To keep it simple, paradoxically, the observation of areas of safety in risk models creates risks, and the observation of risk creates safety.

In the conclusion of the paper, Avinash Persaud granted some suggestions in terms of solution on market-sensitive risk models. He argued that, if people rely on market prices in risk models and in value accounting, they should do so on the understanding amid rowdy times central banks are to be buyers and sellers of the last resort of distressed assets to avoid systemic collapse. The asymmetry of being the only a buyer not a seller of last resort during during the unsustainable boom will only condemn them to cycles of instability.

Regulating ambition should be set now, while the fear of the current crisis is fresh and not when the crisis is over and the seat-belts are working again, Avinash recommended.


Blame the models by Jon Danielsson

• What is in a rating?

Dealing with the same question, Jon Danielsson suggested that understanding the paradox of "Quality of Ratings" helps understand both how the crisis happened and the frequently inappropriate response to it. He argued that the core of the crisis is the quality of ratings generated by sophisticated statistical models. It was the incorrect risk assessment provided by rating agencies, who underestimated the default correlation in mortgages (assumption of independent events of mortgage default).

Jon acknowledged that the rating agencies have an 80-year history of evaluating corporate obligations, which does provide a benchmark to assess the rating quality. Unfortunately, the rating quality of securities differs from those of other regular corporations, he confirmed.

• Foolish sophistication

Financial modelling changes the statistical laws governing the financial system. The reason is that market participants react to measurement and therefore change the underlying statistical processes. By the way, modellers are always playing catch-up with each others, which becomes pronounced when the system gets into a crisis. Jon Danielsson criticized that the endogenous risk (inside-model risk) of interaction between institutions in determining market outcomes works only when everything is under control/calm. In crisis it does not and that is when the models fail, he added.

• Demanding numbers

There are increasing demands from supervisors for exactly the calculation of such numbers as a response to the crisis, right now. Indeed, the underlying motivation is worthwhile trying to quantify financial stability and systemic risk. However, exploitation of bank s` internal models for this purpose is not an appropriate way to do.

• Conclusion
  

Of course, the current crisis took everybody by surprise in spite of all the sophisticated models are in place, all the stress and all the numbers. Jon Danielsson optimistically thinks that the primary lesson from the crisis is that the financial institutions that had a good handle on liquidity risk management came out best. Indeed the problem created by the conduits cannot be solved by models, but the problem could have been prevented by better management and especially by better regulations.

One of the most important lessons from the crisis has been the exposure of the unreliability of models and the importance of management. However to understand the products being traded in the markets and have an idea of the magnitude, risk, coupled with a willingness to act when necessary, supervisors and the central banks need even more sophisticated models with effective management and better regulations in place, Jon Danielsson elaborated.

In the subprime crisis, the key problem lies with the bank supervision and central banking, as well as with the banks themselves.

(Sources: Why bank risk models failed by Avinash Persaud, page 11~12 and Blame the models by Jon Danielsson, page 13~15 of Section 1: Why Did the Crisis Happened)