Moral Hazard for Investment Bank Traders in the Era of Credit Expansion

In days of yore we had investment banks. These were banks that were not only engaged in bringing new securities to market, the traditional realm of investment banks, but were also buying and selling securities for the banks own account: a practice called proprietary trading.

The trader is allotted a certain percentage of total firm capital to trade. He draws a monthly salary and a bonus as a percentage of any trading profits. If he loses money he is fired.

So for the individual trader the following moral difficulties exist. On the one hand, if he makes a good trade he will pull down a large bonus for the year. This bonus will occur at the end of the year and will be based on any "booked" profit that the trader generated for the firm. But this can be calculated from positions that are still open, and thus still exposed to the risk inherent in any financial decision.

If he doesn't make any profit for the firm he will be paid his monthly salary until he is fired. While a firing has significant opportunity costs associated with it, we're assuming that there is little to no actual monetary loss to the trader associated with his firing (especially in relation to the total amount of proprietary funds the trader is allocated).

So how does the individual trader maximize the expected value of his job? To answer this we will assume that the trader has no actually edge on the financial market he trades: the probability of success (profit) in a year is 50/50.

In this case the trader has the incentive to take as large and as volatile a trade he can. In markets, assuming more risk is usually compensated by having a larger range of upsides and downsides. Normally this increased risk is avoided as the downside outcome is just as scary as the upside is tempting. But we must remember that the worst thing that can happen to a trader at a bank is get fired, which costs him relatively little.

In a financial transaction for an individual investor, he has the chances of both making and losing money. Thus if his chances for success are also 50% and his possible upside profit is A and his downside B, then he has an "expected value" of 0.5*A - 0.5*B. If A = B then the expected value is zero. This is the case usually confronting your average investor, but not so for proprietary traders at an old investment banks. In their case, B = 0 (or close to it), thus making their expected value always positive.

So to maximize their jobs (in terms of expected value) the trader must make A as large as possible, and the only way to do that is to take as large and as risky a bet as the bank will allow him. While this is rational for an individual trader, for the bank this is a recipe for disaster. No trader has any incentive to make sustainable long term bets for the good of the firm. Everyone is making short term decisions that maximize the year end bonus they will receive.

It is important to point out that I am in no way saying that these traders are evil. Any self-interested person would make similar decisions if presented with the same tradeoffs. Why wouldn't you? Nobody is going to get a prize for self-sacrifice at a bank.

I myself am a proprietary trader, although of a much different sort. My contract states that I can be sued by the broker/dealer (my boss) for any losses I may incur while using firm capital, which makes me much more judicious in choosing what strategies I employ! I believe that this model will take hold elsewhere on the street, from prop trading desks to hedge funds. Making managers and traders accountable for the downside of their actions is imperative if we are to fix the current problems we face today. Restoring confidence in the street starts with reforming the internal incentive infrastructures for managers and employees.

Oil Making a Bottom

I am in no way long. I'm not going long. This is in no way an endorsement to buy anything in this market. But I am going to call this a bottom in oil and most other commodities for at least 6 months. If i'm right: expect a huge rip off the bottom that is in no way sustainable.

How the Stock Market Works 2: Some Simple Statistical Arbitrage

So I'm gonna attempt to explain some statistical arbitrage (or stat arb, in the truncated parlance of the street). Depending on your experience level, your reaction to this post should range from "duh" all the way through "well uhh" straight through to the "what the fu..." Its OK, you're gonna make it, I promise. Just stick with it and you will come out a smarter person on the other side.

In finance, there are many assets which seem to be driven by the same process. The fate of a collection of oil companies will rise and fall with the price of oil. A parent company may make record profits only if its partially owned subsidiary company gets a new contract. Or it could be something as simple as company A owns a 20% stake in company B. For whatever reason, there are plenty of examples of how the price of two seemingly different assets are linked together. When an educated investor sees two stocks moving together, he should know that there is an opportunity for profit.

Asset prices that move in concert are known as "cointegrated." Although this term has a technical definition, the intuition is straightforward. If the price of an asset rises, we would expect the price of its cointegrated partners to be rising as well. If prices are not moving together, it is because of some temporary anomaly. Thus an investor can profit by buying one cointegrated asset and selling another.

In particular, suppose we have two related assets- A and B- and we observe A rising and B falling. The arbitrageur can then sell A and buy B. Now he has a "market-neutral" position. Theoretically he does not care what direction the market as a whole takes because he is both long and short.

The investor above is betting only that the spread between the two cointegrated assets will narrow (long and short at the same time is known as a spread) . This will occur if A decreases more than B decreases or if A increases less than B increases.

For example, suppose we're long 100 shares of B and short 100 shares of A. If the market increases, its likely that our two stocks will also increase. We will make profit if the money we make in one side of the spread is more than we're losing on the other. If A rises 2 points and B rises 2.5 points, we will have lost 200 on the short position and made 250 on the long position, netting out to a profit of 50 dollars. Else, if A rose 3 points, we would lose 50 dollars.

It is important to note that a trader can make a spread out of any two assets. You could go long gold and short gasoline for example. But it is the idea of cointegration that makes the above spread so powerful. If two assets are truly related, then the spread will eventually narrow, netting the stat arb a profit.

The risks are twofold. The first is that the spread will increase so much in the short term as to bankrupt the arbitrageur before the relationship returns to normal. This is a very real risk, especially if leverage is involved. The second risk is that the relationship will cease to exist. Perhaps company A dumped its holdings of company B. Maybe company A had a corrupt CEO who embezzled billions and takes the equity to zero overnight. Either way, the fortunes of the two companies could diverge quite strikingly.

On the whole, the cointegration trading described above is useful only for calm and stable markets. But risk profiles can be adjusted for any market. In a benign market, this strategy would seek small but frequent intraday returns. The stat arb would thus make alot of trades to capitalize off of low volatility in the size of the spread itself. In a crazy market like we have now, the number of trades would be cut down dramatically. Spreads have gotten very volatile, and opposing small moves could send one to the poorhouse. Waiting for spreads to widen significantly is the only way to implement this strategy in a high volatility marketplace.

What the Data Can't Tell You: Plucking the string of the market

Image via Wikipedia
So much time and effort has been placed in the pursuit of analyzing past stock market data sets to mine for statistically significant anomalies. This is a great process for determining what would have made you money in the past, but is lousy for predicting how the strategy will perform when large amounts of money are employed to capture the discovered anomaly. Oftentimes, especially in statistical arbitrage, the presence of one or two more large players can tip a once profitable strategy into a money pit.

The key to arbitrage is being the first or second to find it... and staying small. Scaling up a strategy is hard, but making small amounts of sure money is easy for those patient enough. Perfecting a stable of small but consistently profitable arbitrage strategies is the key to long term income.

A successful arbitrageur should be able to be both statistician and wildcatter. We must pluck the string of whatever market instrument we have assembled and listen to what sound results. We must model the interaction of our strategies with the market as judiciously as we derive the theoretical arbitrage itself. Otherwise, taking a strategy from theory to practice is impractical and useless.

This financial crisis has taught that even the most statistically inclined of us must have a sense of just going with the flow. Make a decision. See what happens. Go from there. Use the tools that have worked before but with new assumptions.