Friday, March 21, 2008


MBAs probably make good money for themselves. I don't think MBAs make money for the organization they are in. Any organization that has a large influx of MBA managers (replacing people actually competent and willing to do real work) begins to drown in procedural morass and a sea of paper shuffling. A fantastic example was Dell. It was completely taken over by the consultants from Bain, who ran the company into the ground. Google is the next example - it has been named as "most desired place to work" for MBAs, ahead of Mckinsey and Goldman Sachs. Luckily they had the common sense to insist that their CEO will be a person with scientific education, rather than merely bean counting, oops, MBA education.


The problem with a lot of MBAs - they are often competent enough to be useful members of society, but instead prefer to not do anything with their abilities. The confidence projected by a typical McKinsey consultant stems from their arrogance at being in McKinsey, not from their knowledge. One such person I met even went so far as to admit "I did a Phd in Philosophy and had no career option except to be a Management Consultant."


Jim Simons who is probably the most successful and richest hedge fund manager has started publishing papers in topology once again, at the age of 60. On his deathbed his main concern will not be that he did not do an MBA.

Wednesday, March 19, 2008

The casino on wall street

The whole thing came about because I was thinking how exactly are people making money. I don't mean salaries, I mean the firms paying the salaries and bonuses, where is all that coming from?

My model is to divide the world into two groups -

Group (1) includes banks and equity research groups who design the structured and complicated products, and then use that model to get some one to buy some thing and another person to whom they can sell something that is very similar, so they are then perfectly hedged and need accurate pricing, but they don't need price forecasting models because they don't care which way the market goes. Say you have an ideal IB, then this IB buys a bunch of mortgages from BofA, charges BofA a commission for that, then makes some complex model and sells packaged parts of it to, say pension funds, takes fees from all of those as well for providing them with investment vehicles, and all the while it is also enjoying the spread between how much it bought the mortgages from BofA for and how much it sold the components to the pension funds. So really what this IB is selling to its customers is the promise that "we are removing your risk" which is true - BofA got a lumpsum and removed the risk it had that mortgages would be worthless. Similarly the 3 funds that bought the 3 tranches got investments at their selected risk levels, so they are happy to pay fees also for "removing undesirable risk". So the IB only needs to make sure that what it takes on left side it can sell on right side for approx same amount. Today it might take a 500k mortgage and sell it in 3 tranches for 510k, tomorrow if that market crashes then it buys the 300k mortgage and sells it to someone at 305k. A glorified e-trade so to speak. IB doesn't care for the market direction just that there should be enough players in it, so the focus here is on designing more and more complex products to attract more and more players to try them out but not necessarily to accurately forecast anything because money making does not depend on forecast.

Group (2) are the poor fools who are actually trying to make real returns via trading/investing and have directional exposure. This includes all the 401k investors including you and me, as well as Warren Buffett as well a bunch of investment funds, mutual funds etc. These people might try to reduce their exposure to the direction of the market, but in the end they have to take a real position if they want to make real returns. Exposure to market risk is what creates returns in the long run. So here hedging is not enough, you need to model the whole universe and everything that goes on in it. The promise of the players here is "we will work hard for you to increase your money". The work hard part consists of econometric analysis, sifting through 10s of years of price data, volume and volatility data etc. to find "opportunities", as well as studying fundamentals to put into the models etc.

(1) is a glorified casino selling newer and fancier kinds of chips, (2) are the bettors. (1) will always make money because the rules are in its favor, (ie fees, spreads will always earn it money) whereas (2) will have a few people that are very good/lucky/both who make all the money and most will go bust.

All said and done, it is better to be a casino operator than a bettor; it may be boring but you make guaranteed return as long as you have visitors.