How to know if your fund manager is actually beating the market

"Beating the market" sounds like a yes-or-no question. It is — but only once you have fixed three things most people get wrong: which return you compare, which index you compare it against, and over what period. Get any of the three wrong and you will confidently conclude your manager is a genius or a dud, on evidence that proves neither.

This is a companion to our cornerstone on how investment return is actually calculated. The cornerstone establishes the two returns — the investment's and your own. This piece uses that distinction to answer the question underneath it: did the manager earn their fee?

The fair comparison, in one sentence

To judge a manager, compare the fund's time-weighted return, net of fees, against the total-return version of an appropriate benchmark, over the same long-enough window.[1] Every word in that sentence is load-bearing. Take them one at a time.

Time-weighted, not your personal return. This is the error that quietly ruins most amateur manager assessments. Your own money-weighted return has your contribution timing baked into it — and the manager does not control when you deposit. Judge them on your money-weighted figure and you reward or punish them for your decisions. If you added a large lump just before a rally, your personal return flatters a manager who may have lagged; buy in before a slump and you will damn one who did fine. Time-weighted return strips your timing out, which is exactly why the global reporting standards require it for manager performance.[1] Use it.

Net of fees. An index charges nothing. A fund charges. So the fund must clear the index after its own costs to have genuinely beaten it. A fund that matches the index gross has lost to it net — by the size of its fee.

Total return, not the price index. This one manufactures illusory outperformance and almost nobody notices. Most headline index levels you see quoted — the JSE All Share level, the S&P 500 number on the news — are price indices: they ignore dividends. Your fund, meanwhile, earns and reinvests dividends. Compare your dividend-earning fund to a dividend-ignoring index and you will credit the manager with the market's dividends. Always use the total-return version of the benchmark. (We cover the price-vs-total-return trap in full in a companion piece.)

Same window, long enough. Short horizons are noise; skill takes time to separate from luck. There is no single "correct" minimum window — be suspicious of anyone who asserts one — but a single good or bad year tells you almost nothing.

The errors hiding in "an appropriate benchmark"

"Appropriate" is doing a lot of work. The benchmark has to match the fund on three dimensions, and mismatches are where flattering comparisons come from:

• Style and asset class. A global-equity fund is not measured against the JSE All Share. A balanced fund needs a blended benchmark, not a pure-equity one — comparing a 60/40 fund to an all-equity index makes it look weak in bull markets and strong in crashes, both meaninglessly.

• Currency. Match the benchmark to your measurement currency. A global index quoted in rand mixes the investment's return with the rand's movement — that is a currency question, not a skill question, and it belongs to a separate analysis.

• Survivorship bias. Index and peer-group averages quietly drop funds that closed or merged, which inflates the survivors' apparent average. The bar you are clearing may be higher than it looks.

The test that answers "would I have done better in the index?"

There is a sharper, more personal version of the question, and it has a clean method. To ask "would I personally have done better just buying the index?", run your actual cash flows — same dates, same amounts — through the benchmark's total-return index, compute the benchmark's XIRR on those identical flows, and compare it to your own portfolio's XIRR.[2] Because both sides carry your timing, the timing cancels out, isolating the manager's selection and allocation. This is the personal analogue of Morningstar's dollar-weighted "Investor Return."[2] It answers a question the fund factsheet cannot: not "did the fund beat the index," but "did I, with my actual behaviour, beat the index."

A brief word on risk, because raw outperformance ignores it. A fund that beat the index but with far higher volatility may not have added value once you account for the risk taken. The first-pass adjustment is the Sharpe ratio (excess return per unit of volatility); the benchmark-relative version is the information ratio (active return over tracking error). You do not need to compute these to think clearly — just to remember that "higher return" and "better" are not the same sentence.

The South African benchmark that changed

If you benchmark SA equity, one detail matters and is easy to get wrong because it is recent. The FTSE/JSE Capped SWIX All Share (J433) — for years the default benchmark for SA general-equity funds — was terminated at the end of 2025, with the broader SWIX index family decommissioned from 2 January 2026 under the FTSE/JSE index harmonisation.[3] Its successor for capped, broad SA-equity benchmarking is the FTSE/JSE Capped All Share (J303). Any benchmarking you do dated on or after January 2026 should use J303 (or the FTSE/JSE All Share total-return, J203T gross of dividend withholding tax) — not the retired Capped SWIX. Always the total-return variant.

For global equity, the standard total-return references are the S&P 500 Total Return (developed-US), MSCI World (developed markets), or MSCI ACWI (developed plus emerging).[3] Again: the total-return version, in your measurement currency.

One pointer, not a full treatment: South African dividend withholding tax and capital gains tax on disposals change your after-tax return and influence whether you should use a gross or net-of-DWT index. That tax layer is real and we defer the detail to a dedicated piece — but it is worth knowing the gross figure is not the figure that reaches your pocket.

The point

Whether your manager beat the market is answerable — but only with the right number, the right index, and enough time. Use the fund's time-weighted return, net of fees, against the total-return version of a style-matched benchmark, over a window long enough to be more than luck. For the personal version of the question, run your own cash flows through the index and compare XIRR to XIRR. And if you benchmark SA equity, use the Capped All Share (J303) — the Capped SWIX it replaced stopped being calculated in January 2026. Most "my manager is beating the market" claims fall apart on one of these; the honest answer survives all of them.