Hi all,
This is my first post in Teluglobe. Though it might sound cliched, I feel privileged to be able to post in Teluglobe. Teluglobe is a great platform not only for all Gulutes, but also for people in general, to know about so many things from financial markets to history of Telugu culture and language to education in US and in general. I’m very thankful to Mohan garu for letting me a part of this effort.
I saw this talk in the Research channel this weekend and remembered Mohan gari comments about quants and their role in the current financial crisis. The speaker is Andrew Lo, a professor at the MIT Sloan School of Business. Hope it will not put off the regular readers but I personally feel all of us have to know what the problem was. The talk might sound slightly too academic to some of us but if you listen with some patience you can easily understand the issues at hand. The thing I liked about it was it’s simple way of explaining the problem using basic math/probability. The speaker makes some very interesting points.
Are Mathematical Models the Cause for Financial Crisis in the Global Economy? from jojimbo on Vimeo.
The following are some of my thoughts (I borrow some of these from the e-mail discussion I had with Mohan garu) :
Prof. Lo’s allusion to Charles Purrow’s Theory of “Normal” Accidents in industrial processes and how one can prevent them can be similar to my comment here. Human behavior or systems involving humans cannot “completely” be captured in math models. Obviously not all the role-players or their roles can be included in their complete complexity in this dynamical system. So certain “simplifying” assumptions are made. But it is important to check whether these assumptions are robust by means some kind of a “sensitivity analysis”. More so, if these models have far-reaching effect on the lives of so many people. In this case, they assumed that the values of the AAA bond and BA bond are independent and also don’t depend on whether the housing values are growing or not. As Mohan garu most of the time points out the economists consider, “Real world is a special case for the economists and the quants”. Technically, in this case statistically, this might be true in that the outcome of a day’s financial market is only one realization of a vastly varying market behavior – one realization of a statistical distribution to be more accurate. But when these instruments or models are being used at the expense of life savings of so many people, the economists/quants or whoever have to behave with so much more responsibility. Imagine the NASA sending a space shuttle without a thorough testing the thrust generated by it’s rocket booster. And this is only one space shuttle, what about so many people’s lives ?
The speaker tries to put it very softly (in a karra viragakunda pamu chavakunda fashion) when he says that humans are not omniscient and the CEOs sometimes may only be human in neglecting risk and focus only on profitability. In that case, I’m not very sure why the CEO must receive such “super human” bonuses ? And as the speaker points out, how would a cut-throat capitalist react when a rare conscientious quant points out that there is a problem with the CDO market model ? So although Prof. Lo makes a suggestion that more smart people should be involved in the process of developing and applying these models in the market, I feel that social awareness/consciousness are more important. And if I doubt such people would end up at Wall Street.
Hence, I think, regulation is the only way forward. And like Prof. Lo points out there must be a financial safety authority in the mold of National Transportation Safety Board. This authority will serve as an expert system and try to explain and analyze in detail why did each economic downturn happen and what could have been done to prevent it. It is interesting though that Charles Purrow was a sociologist who was analyzing the accidents caused in industrial processes. How many of the quant or CEOs are willing to work with sociologists ? Probably they should be involved with the government in framing the regulatory process.
Sandeep Garu,
Good video. Looking forward to see more of your posts.
Sandeep,
Welcome to TG and thanks for posting. I have something to say about Andrew Lo and the views he expressed here. It is a bit crazy at work for now, when I find some time, I will sneak in here and complete the thoughts.