Behaviour & decisions · South Africa
Why can your return differ from the fund’s published return?
Your return depends on how much money you had invested when gains and losses occurred. A fund's published return usually describes the investment's performance over a period, while your money-weighted return also reflects the size and timing of your contributions and withdrawals. A gap can reveal harmful return-chasing, but it can also arise from sensible saving or necessary spending. Calculate it before judging it.
The distinction matters when a fund advertises a positive return while your account shows a loss. Neither number is necessarily wrong. They may answer different questions. Before concluding that the manager, adviser or statement is at fault, reconcile the dates, cash flows, fees, currency and calculation method.
In this article
Two return measures answer different questions
Time-weighted return links the investment's returns across periods while removing the effect of external money coming in or going out. It helps assess the investment performance without rewarding or penalising the manager for a client's contribution timing.
Money-weighted return reflects the size and timing of the money actually invested. An internal rate of return, or IRR, is one way to calculate it. The rate reconciles the money paid in and received with the ending value over the relevant dates. CFA Institute's GIPS handbook explains these distinctions; using them here does not claim that a particular account or firm is GIPS-compliant. GIPS Standards Handbook for Firms, time-weighted returns and provision 2.A.29.
A simple rand profit is a third useful number: ending assets plus withdrawals, less contributions and starting assets, with the treatment of fees and income made explicit. It does not by itself tell you the annual return, because some money may have been invested much longer than the rest.
The fund gains 20%, but the investor loses R160,000
Consider an invented fund with a 50% return in year one and a 20% loss in year two. There are no separate fees, tax or distributions in this example; the returns are assumed to be the change in the investment value available to the investor.
A R100,000 investment held throughout becomes R150,000, then R120,000. The fund's two-year return is 20%: 1.50 × 0.80 − 1. Annualised, that is approximately 9.54%. Averaging +50% and −20% to get 15% a year would be incorrect.
Now give the investor a different cash-flow history. They invest R100,000 at the start and add R900,000 exactly one year later, after the first year's gain and before the second year's loss.
| Event | Calculation | Account value |
|---|---|---|
| Initial investment | R100,000 | R100,000 |
| End of year one, before contribution | R100,000 × 1.50 | R150,000 |
| Additional money at the one-year point | R150,000 + R900,000 | R1,050,000 |
| End of year two | R1,050,000 × 0.80 | R840,000 |
Total contributions are R1 million, but the account ends with R840,000. The investor has lost R160,000. Most of the money arrived too late to experience the first year's gain and was present for the second year's fall.
The annual money-weighted return is approximately −14.74%. At that constant annual rate, R100,000 invested for two years plus R900,000 invested for one year produces R840,000. The accompanying calculation solves that reconciliation directly.
The −16% obtained by dividing the R160,000 loss by total contributions is a cash-on-contributions comparison, not the annual IRR. Nor should the 20% cumulative fund return be compared directly with an annualised investor number. Put both rates on the same time basis before describing their difference.
Reverse the market sequence, keeping the cash dates unchanged
Now let the fund fall 20% in year one and gain 50% in year two. Its two-year return is still 20%, or 9.54% annualised. The investor again contributes R100,000 initially and R900,000 at the one-year point.
The first contribution falls to R80,000. Adding R900,000 creates R980,000 for year two, which grows to R1.47 million. The investor gains R470,000 on the same R1 million of contributions. The annual money-weighted return is approximately 41.19%.
| Same cash contributions; different return order | Fund annualised return | Investor annualised return | Ending account |
|---|---|---|---|
| +50%, then −20% | 9.54% | −14.74% | R840,000 |
| −20%, then +50% | 9.54% | 41.19% | R1,470,000 |
Investors do not always underperform their funds. Contributions can create a positive gap as well as a negative one. The arithmetic cannot tell us whether the investor displayed skill, had no choice about the date or was simply fortunate.
A measured gap is not automatically a mistake
Suppose the R900,000 was only available after an estate distribution or the maturity of another investment at the end of year one. The investor could not have placed that money in this fund at the start. Treating the first year's missed gain as an avoidable loss assumes access to capital that did not exist.
If, instead, the money had been available all along and the investor bought only after seeing a spectacular return, the same figures justify a different conversation. What prompted the decision? Was there an agreed allocation? Was a recent winner mistaken for a reliable forecast? The outcome alone does not establish the answer, but the transaction history identifies what to investigate.
Regular retirement contributions create another case. Later contributions naturally experience fewer months of returns than earlier ones. A retiree drawing money for living costs has a different cash-flow pattern again. Neither should stop a sensible contribution or an essential withdrawal merely to make an account's return resemble a fund factsheet.
The right counterfactual uses money that was actually available and respects the household's commitments. If R900,000 was held in another asset, a fair comparison includes that asset's return, risks and costs while it was held. Our fund-only illustration does not measure the investor's entire wealth or the outcome of an alternative cash strategy.
What does the research establish?
Morningstar's August 2026 Mind the Gap study estimated an 8.7% annual return for the average dollar invested in its US mutual-fund and ETF universe over the ten years ended 31 December 2025, against a 9.9% aggregate fund return. The approximately 1.2-percentage-point gap concerns that universe and period, not a South African household or a guaranteed benefit from advice.
The study pools fund assets and estimated monthly flows. It includes eligible funds present at the start, including later closures and mergers, but excludes funds launched after the starting date. Its average-dollar measure is not the return of the average individual investor. Morningstar also cautions that regular saving and rebalancing can create gaps without proving bad behaviour. Morningstar, Mind the Gap 2026, executive summary, overall table and methodology.
Research comparing active funds with an index answers another question. It should not be used as proof of the difference between an investor's experience and the very fund they own. Manager selection and investor cash-flow timing need separate evidence.
Reconcile a South African statement before changing the investment
For a concrete local example, Allan Gray's Balanced Fund page presents rand performance after fund fees, with distributions reinvested. Its disclosure says actual investor results can differ because of investment dates, reinvestment dates and dividend withholding tax. Those are defined measurement differences to investigate, rather than evidence by themselves of a poor decision. This example concerns the provider's reporting basis, not a recommendation of the fund. Allan Gray Balanced Fund: performance and investor disclosures.
Ask for the exact start and end dates, opening and closing values, every external contribution and withdrawal, and the return method used. Then check the fund class and currency. A dollar return for a foreign fund is not directly comparable with a rand account return; an offshore holding's rand result includes the currency movement unless that exposure is appropriately hedged.
Separate fund-level costs already reflected in the unit price from advice, platform or other account-level charges deducted elsewhere. Include distributions consistently: reinvested income is not the same as a new contribution from your salary, while income paid out needs to be recognised when reconciling what you received. An unexplained gap should be traced through these records before being labelled a timing effect.
For a household with several accounts, reconcile the consolidated result as well. A transfer out of one account may be a transfer into another, rather than money spent outside the portfolio. Counting it incorrectly can distort the household-level calculation. Keep genuine contributions, withdrawals, transfers and charges distinct.
The request can be specific: “Show the fund's comparable time-weighted return, my money-weighted return and the rand reconciliation, using the same dates and currency. Identify the part explained by external cash flows and the part explained by fees or other differences.” That is more useful than asking why your percentage is lower without first identifying which percentage it is.
ClearGauge's judgement: fix the process that caused the avoidable part
Where the evidence shows return-chasing, the remedy is usually a decision process that survives a tempting chart. Agree the role of each investment, how new money will be allocated and what information would justify a change. Record the reason for a material switch before seeing the subsequent return.
Where the gap comes from a necessary withdrawal or newly available capital, address the underlying plan rather than blame the transaction. Check the reserve and spending policy, or invest the new money according to the appropriate allocation. The lump-sum versus phasing question is relevant once capital is available; it cannot recover a return from before you had it.
An adviser should not claim the entire published research gap as a fee-justifying “value add”. Some flows are unavoidable, some gaps are positive and some are measurement differences. Ask what behaviour or cost the proposed service will address in your circumstances, how that work will be delivered and how its effect will be assessed without promising outperformance.
Common questions
Does a negative return mean the fund statement is wrong?
Not necessarily. Our first example produces a positive fund return and a negative investor return with correct arithmetic. Reconcile the cash flows and calculation bases before drawing a conclusion.
Can I improve my return by never withdrawing?
That is not a useful rule if the investment exists to fund spending. The relevant question is whether the withdrawal is planned, affordable and funded appropriately. Retaining unsuitable risk merely to avoid recording a sale can create a worse problem.
Should I compare my IRR with a different fund's headline return?
Only after aligning dates, cash flows, fees, currency and risk. A counterfactual investment comparison should apply your actual available cash flows to the alternative. A headline return over a different period cannot establish what you would have earned.
Sources and calculation basis
Sources were checked on 6 September 2026. The return examples use exact one-year intervals and an annual IRR; real dated transactions require an appropriate dated calculation. All return inputs are hypothetical, with no separate tax, fees or distributions. Morningstar's 2026 research is a US aggregate study, not local investor-performance evidence. The detailed cash flows remain available under technical data. This is general education and does not establish that a particular investor, adviser or manager made an error.
Download investor contributions and return order (Excel)
Inspect the fixed worked-example records in Excel, with readable headings and filters. The workbook preserves the article’s calculation data; it is not a personal planning calculator.