Trading Costs: How Commissions and Overtrading Eat Returns
Every trade has two prices: the one on the ticket and the one in the fees. The first is enormous and visible; the second is small and relentless — and the research finding is that the small one decides more portfolios than the big one.
The study with the unforgettable title
Barber and Odean (2000) — Trading Is Hazardous to Your Wealth — analyzed 66,465 households at a large discount broker from 1991 to 1996. Households on average turned over 75% of their portfolio annually, and the aggressive tail much more. Sorted into quintiles by turnover, the most active fifth earned an annual 11.4% net return while the market returned 17.9% — a gap of six and a half percentage points a year, produced not by picking terrible stocks (gross returns were roughly ordinary) but by the tolls paid crossing the road so often: commissions and bid-ask spreads.
Two honest caveats. The data predates zero-commission brokerages, so the explicit-fee part of the toll is smaller today — though spreads, payment-for-order-flow economics, and taxes remain, and the study’s deeper result was about turnover, not the fee schedule. And it describes averages across a large population; it cannot say what any particular trader’s costs did.
The arithmetic, which hasn’t changed
What makes costs vicious is that they are certain and compounding while gains are neither. Three mechanically true statements:
- Costs raise your break-even before anything happens. Buy 100 shares at $50 with $5 commissions each way and your break-even is $50.10 — the stock must rise 0.2% for you to have accomplished nothing. Each round trip resets that hurdle. Fifty round trips a year at 0.2% is a ~10% annual headwind, which is roughly the gap Barber and Odean measured.
- Costs subtract from the base that compounds. A 1% annual cost drag doesn’t cost 1% of your final wealth — over 30 years at 7% growth it costs about a quarter of it, because every dollar taken early forfeits its own future compounding. The CAGR calculator makes this visceral: compare 7% and 6% over 30 years and look at the end values.
- In aggregate, active trading is zero-sum before costs and negative-sum after. Sharpe (1991) proved this in one page of arithmetic: the average actively-managed dollar must earn the market return minus its costs, because all the dollars together are the market. Individual outcomes vary; the average cannot.
Watching your own toll meter
The position calculator treats fees as first-class inputs — flat or percentage, each side separately — precisely so the toll is visible rather than folded into vibes. Enter a realistic sequence of trades with your actual commission structure and compare total invested against the sum of your lots: the difference is what the road charged. Then set the fees to zero and watch the break-even and ROI move; the gap between the two runs is your personal Barber-Odean number, measured on your own behavior.
The evidence doesn’t say “never trade” — it can’t, and neither can we. It says the meter runs whether or not the trade works out, and that the most reliable improvement most active traders can make costs nothing: fewer, more deliberate round trips.