Why Most Active Funds Lose — and It’s Arithmetic, Not Bad Luck

It feels almost self-evident that paying for expertise should pay off. An actively managed fund hires clever, well-resourced people whose entire job is to pick winners and dodge losers. Surely that beats a mindless index fund that just buys everything? For the large majority of funds, over the long run, the answer is no — and the reason isn’t that the managers are foolish. It’s arithmetic. Once you follow the logic, the result stops looking surprising and starts looking inevitable.

The argument that can’t be wrong

In 1991 the Nobel laureate William Sharpe laid it out in a short essay, The Arithmetic of Active Management. The argument needs no data at all — it’s true by definition:

1. Everyone, together, owns the whole market. Add up every investor’s holdings and you get the market itself. So the average return earned by all investors, before costs, is exactly the market’s return.
2. Passive investors get the market return, cheaply. An index fund simply holds the market in proportion, for a tiny fee.
3. Therefore active investors, as a group, must also earn the market return before costs. They own whatever the passive investors don’t — which is just the rest of the same market. The active crowd’s average, before costs, is the market return too.
4. But active management costs far more — higher fees, more trading, more tax. So after costs, the average actively managed dollar must earn less than the average passively managed dollar, by roughly the difference in their costs.

Notice what this argument does not claim. It doesn’t say no manager can beat the market — some always will. It says that in aggregate they can’t, and that the average investor in active funds is mathematically guaranteed to fall behind the average indexer by the cost gap. Skill can move winnings between active managers, but it cannot conjure extra return for the group as a whole.

The evidence agrees — and then some

If the theory is airtight, the data should confirm it, and it does. Long-running scorecards such as S&P’s SPIVA have repeatedly found that over fifteen-year horizons, the large majority of active funds — commonly around nine in ten in US large-cap categories — fail to beat their benchmark. Two effects make the real picture even worse than the raw averages suggest:

• Non-persistence. The funds that win in one period are mostly not the ones that win in the next. Past performance, as the disclaimer says, genuinely does not predict future results — so last year’s star is a poor guide to next decade’s.
• Survivorship bias. Funds that perform badly get quietly merged or closed, vanishing from the statistics. The surviving averages therefore flatter active management, because the failures have been swept off the table.

Together these mean that not only does the average active fund trail, but picking tomorrow’s winners from today’s leaderboard is close to a coin toss.

Where the money actually goes: fees

The cost gap at the heart of Sharpe’s argument isn’t abstract — it’s the fee on your statement, and it compounds with quiet ferocity. Suppose two investors each put $100,000 to work for 30 years, and suppose — generously — the active manager matches the market’s 7% before fees. One pays a near-zero index fee; the other pays 1% a year.

Fund	Net annual return	Value after 30 years
Low-cost index (~0.04%)	~6.96%	~$753,000
Active (1.00%)	~6.00%	~$574,000

Same gross return, one percentage point of fees — a gap of about $178,000, more than the original investment.

A “1%” fee didn’t cost 1%. It quietly removed a slice of every year’s compounding, and over a lifetime that slice grew into more than the stake you started with. You can run this on your own numbers with the investment fee impact calculator — it’s one of the most sobering calculations on the site. It’s also a close cousin of the volatility drag idea: a small annual leak, relentlessly compounded, dominates the long-run outcome.

When active still makes sense

None of this means active management is always a mistake. The arithmetic bites hardest in large, efficient, heavily-researched markets like US large-cap stocks. There are legitimate cases for active approaches: genuinely less efficient corners where information is scarce, specialised mandates, or a low-cost active fund with a real and durable edge. The bar is simply high — the fund must out-earn its index by more than its extra costs, reliably, after tax. Most don’t clear it, which is why “low-cost index by default” has become the sensible starting point rather than a slogan.

What to do with this

• Default to low cost. Between two funds tracking the same thing, the cheaper one wins by the fee difference — guaranteed. Treat the expense ratio as the first number you look at, not the last.
• Don’t chase last year’s winners. Strong recent performance is weak evidence of future performance, and the marketing leans on exactly the survivorship-flattered averages above.
• Judge on risk-adjusted return. Raw returns ignore how much volatility you stomached to get them; the Sharpe ratio scores return per unit of risk, a fairer comparison.
• Capture the market and rebalance. A broad, low-cost index, kept on target with periodic rebalancing, reliably delivers the market return minus very little — which, per the arithmetic, beats the average active investor.

The takeaway

The case for indexing isn’t pessimism about human talent; it’s arithmetic about costs. The active industry, in aggregate, earns the market return before fees and therefore trails it after — every year, by definition — and the evidence piles confirmation on top of the certainty. Some managers will always win, but you can’t reliably spot them in advance, and the fees you’d pay trying compound into a fortune you’d otherwise have kept. When the maths is this clear, the humble index fund stops looking lazy and starts looking like the informed choice.