Reading performance honestly

Everything in this chapter is about one question: when a fund reports a number, what is it a number about?

What the published figure already includes, and does not

SEBI is exact, and the sentence repays reading twice:

NAV declaration made by AMC/Mutual Fund on every business day is net of expenses, and consequently scheme performance disclosures made by Mutual Fund, which are based on NAV values of the scheme are also net of expenses but does not consider impact of load and taxes, if any.

So a published 14% is after the expense ratio. Good — you do not need to subtract it again, and a common error is doing exactly that.

It is before exit load and before tax. If you redeemed inside the load period, your return was lower. After tax, it was lower again. The fund's number is honest about the fund; it is not your return.

What you are entitled to see is set out too: half-yearly results with returns over six months, one, three and five years and since inception, and a dashboard on the fund's website with performance and key disclosures for every scheme it runs.

The start date is doing the work

A point-to-point return — the standard "3-year return" — is entirely determined by two dates. Move either and the number changes, sometimes enormously.

A three-year record beginning at a market low looks superb. The same fund measured from a year earlier, including the fall, looks ordinary. Nothing about the manager changed between those two sentences.

This is why the same funds appear in "best performing" lists after every rally and vanish after every correction, without anybody's skill changing.

The fix is rolling returns: instead of one three-year window, compute every three-year window that has occurred — starting each day, or each month — and look at the distribution. What you get is not a number but a shape: the best, the worst, the median, and how often the fund beat its benchmark.

The worst rolling three-year period is usually the single most useful figure about a fund, and it is the one no advertisement contains. It answers the question that actually decides your outcome: how bad has this been, for how long, for someone who bought at the wrong moment?

The benchmark has to be the right one, and the total return version

Two separate traps.

The right index. A fund's benchmark should be the honest comparison for its category. A mid cap fund measured against a large cap index will look brilliant when mid caps run — and that is a fact about the category, not the manager.

The total return version. A price index counts only price changes. A total return index adds the dividends the constituents paid. The fund received those dividends; they are in its NAV. Comparing a fund's return against a price index therefore credits the fund with dividends the benchmark was not given.

Over a decade with a 1.5% dividend yield, that gap is enormous — it is roughly the size of the entire fee of an active fund. Indian funds are now required to use total return indices for comparison, which closed a gap that had flattered the industry for years. If you are reading older material, or a third-party comparison, check which index it used.

Survivorship

The category average you see is the average of the funds that still exist.

Funds that did badly are merged into better ones or wound up, and their records leave the tables with them. What remains is a survivors' average, which is higher than what investors actually experienced.

This matters in one specific way: it makes "the average fund in this category returned X" a weaker claim than it looks, and it makes any comparison between a fund and its category average slightly generous to both.

Your return is not the fund's return

The last gap, and the largest for most people.

A fund's published return assumes one lump sum at the start, held throughout. You invested monthly, or added after a good year, or stopped during a bad one. Your money-weighted return can be far below the fund's time-weighted one, and the difference is entirely the timing of your own contributions.

This is the number you should actually track, and it is available: the consolidated account statement gives your cost and your value, and a XIRR over your own cash flows gives your own return. A fund's factsheet cannot tell you that, because it does not know when you invested.

The four checks

1. Rolling returns, not point to point — and look at the worst window.
2. Against the total return version of the right index, not a price index or a different category's.
3. Net of what you will actually pay — plan, exit load, tax.
4. Your own XIRR, over your own contributions.

The point

A published return is net of expenses and not of load or tax, and depends entirely on two dates somebody chose. Rolling returns against the total return version of the correct index remove most of the illusion, survivorship removes a little more, and your own XIRR is the only figure that is about you.