This may be joyriding in the obvious-mobile, but commissions and other fees/expenses are a big deal in trading. I pointed out a few months back that it could cost you about 1/6th of your profits in the speculative alternatives portfolio to use a broker with the wrong type of commissions structure. I want to explore that idea in a lot more depth now. It turns out that commissions put very fundamental boundaries on what sorts of trading you can make money at. They’re a fundamental part of the market’s structure in the same way that auction market dynamics and FIFO ordering are. Failing to understand this point cost me a lot of money along the way – certainly many thousands of dollars. Let me see if I can explain what I learned for that expensive tuition.
One of my favorite psychology ideas comes from the poker literature – specifically Mike Caro’s “threshold of misery”. Mike’s point, paraphrased, is that there’s a difference between how bad a situation is objectively and how bad people feel about their miserable circumstances. Sure, as a situation gets worse, we feel worse. But eventually we hit a point where they feel about as shitty as is possible. Analytically speaking the situation can get worse, but there’s only so bad you can feel about it. That boundary is the threshold of misery. Continue reading
In the wake of the economic scandals of 2008-2009 and more recent events like the JP Morgan’s “London Whale” losses the media has directed substantial scrutiny at trader compensation and the role it plays in trading losses. This discussion has focused on one point: traders who are paid primarily via bonuses are paid both for performance (which is arguably good) and for volatility (which can cause havoc).
Consider a trader managing a $100M portfolio who is paid $300K per year plus 15% of each year’s profits. Clearly this trader has what amounts to a cash settled call option on the portfolio. If the portfolio makes money, he gets 15%. If it loses money, he loses nothing. It’s easy to see how this could warp the trader’s behavior in the direction of taking excessive risk anytime the portfolio was close to break-even as the year end approached. Taking a 20% coin flip at the end of a break-even year would result in an expected profit of $1.5M for the trader (half the time he wins $3M, half the time he gets no bonus) and an expected loss of $1.5M for his employer/investors (who would win $17M after fees half the time, and lose $20M the other half of the time).
This phenomenon is undeniably real, and anyone designing the compensation packages for traders would be wise to take it into account. But I would argue the media discussion of the topic misses one central fact: traders don’t like to get fired. This rather common-sense aversion on their part goes a long ways towards explaining certain market events. More importantly it can directly put money in your pocket once you know what to look for. Continue reading