Balancing the Banks: Global Lessons from the Financial Crisis Hardcover – May 9, 2010

The authors do an excellent job of presenting their case,which is that " Yes. VAR failed .However,a revised VAR could succeed ". Unfortunately,their argument ultimately collapses although they do not realize the contradictions between the different parts of their presentation in their book.Thus, "Take the example of the credit derivative market...But the uncertainty about the net amounts and their structure...suffices to scare the markets.As Cabolleso and Sinsek(2009)emphasize,the task of knowing not only whether one's counterparties are solvent but also whether one's counterparties counterparties are,their counterparties,and so on,becomes daunting in times of generalized stress".(p.35). This problem,of course,is an application of the beauty contest problem first discussed by J M Keynes in chapter 3 of the A Treatise on Probability(1921,TP) in his discussions of the measurement of probabilities and later in chapter 12 of the General Theory(GT,1936).Keynes later showed in the TP that ,unless the weight of the evidence,w, was equal to 1 and there were no nonlinear probability preferences,it is not possible to assign a point estimate and/or a unique probability distribution to represent the decision problem.Interestingly,Keynes is not mentioned in the book except in one arcane footnote .Neither is Mandelbrot ,Taleb, or Ellsberg.This is because all three of the authors are committed to relying on the basic Value at Risk (VAR)approach.Their hope is that VAR can be revised and improved.Thus, "There is another... reason why the focus on the probability of failure for individual banks may (?-author's query) have been inappropriate: the absence of systemic considerations.A 1 percent probability of failure means either that 1 percent of the banks fail every year or,alternatively,that the whole banking system fails every hundred years...Therefore ,it is crucial for regulators to find ways of discouraging herding behavior by banks...New indicators of risk must be designed based on correlation with aggregate activity rather than the absolute probability of failure."(p.116)The authors seem to have forgotten what they wrote on p.35.It is impossible to assign any kind of a point estimate ,using the standard normal or log normal probability distribution, to beauty contest problems.Keynes had already proven this over a hundred years ago.The view of the authors appears to be that the purpose of society and its economy is to serve the wishes of the bankers.Society has to come up with ways to regulate bankers effectively so that they can continue to play their derivatives game over time and extract profit without production.Warren Buffett had a better idea-ban financial derivatives permanently.Their purpose is purely speculative.They are used to bet on the probability of failure of different corporations and banks.No contribution is made to the production of goods and services.Hopefully,the authors will also come to realize that (a) they are dealing with the tail probabilities of Cauchy distribution and not normal or log normal distributions and (b) that the outcomes cost trillions in losses if you end up on the left tail .

This is a good book for professionals involved in reviewing banks. I found the chapter on lessons learned to be very useful. Concepts are well explained and easy to understand.