What is a mutual fund?

When your money reaches a scheme it is merged into the pool and you are allotted units at that day's NAV: ₹50,000 into a scheme priced at ₹20 buys 2,500 units (illustrative), recorded against your folio by the registrar. A unit is not a share. It carries no boardroom vote and no claim on any single security; it is a proportionate beneficial interest in the whole portfolio. If the scheme holds sixty shares and some government paper, every unit owns the same microscopic slice of all of it, which is how a few hundred rupees ends up diversified across positions no retail order book would ever sell in such small pieces. Whether you enter in one lump sum or spread the same money across many dates through a systematic investment plan, or SIP, you are buying the identical units; only your average entry price differs.

NAV is not a quote discovered by trading; it is computed, once per business day, from the closing books. Take everything the scheme owns marked to that day's closing prices, add income accrued, subtract liabilities and the expenses accrued for the day, and divide by the units outstanding. Every input is disclosed or derivable, which makes NAV one of the few prices in finance an investor can audit from public documents. And the phrase that governs the rest of this guide sits quietly inside that formula: subtract the expenses accrued for the day. The fee is not a bill that lands in your inbox. It is a deduction that happens inside the price, every single day, before you ever see the number.

Work the example once by hand. On Tuesday the scheme has assets of ₹804 crore, liabilities of ₹4 crore, net assets ₹800 crore, and 40 crore units, so the NAV is ₹20.00. On Wednesday the holdings gain one percent, lifting assets to ₹812.04 crore; the day's expense accrual, at a 1.8 percent annual ratio, is ₹800 crore times 0.018 divided by 365, about ₹0.04 crore, or roughly ₹4 lakh. Net assets become ₹808 crore and the NAV is ₹20.20. The market gave one percent; you received one percent minus one day of cost. That subtraction runs silently on every business day the scheme exists, which is precisely why the size of the annual ratio matters far more than its small appearance suggests.

One consequence kills a persistent myth. A scheme with an NAV of ₹20 is not "cheaper" than one at ₹500. The NAV level only records how long the scheme has existed and how its past went; ₹10,000 buys 500 units of the first and 20 units of the second, and a one percent portfolio move changes both investments by exactly ₹100. New fund offers marketed at "₹10 a unit" trade on precisely this confusion. What differs between two schemes is never the smallness of the unit price. It is what they hold and what they charge, and only the second of those is genuinely in your control.

A mutual fund in India is not a company you buy into. It is a trust, and the regulations deliberately force its functions apart so that no single entity ever both manages your money and holds it. The sponsor promotes the fund and seeds the asset management company (AMC), the business that actually runs the portfolios and earns the fee. The trustee holds the assets in trust for unitholders and polices the AMC; the regulations require a strong majority of trustees to be independent of the sponsor. A SEBI registered custodian, which cannot be controlled by the sponsor, keeps the securities. The registrar and transfer agent (RTA) maintains the ledger of who owns which units. Five roles, one pool, and a wall between the hand that decides and the hand that holds.

The practical consequence is the one investors most often miss: the scheme's assets belong to the trust, not to the AMC, and they sit outside the fund house's own balance sheet. If the AMC's business fails, is sold, or loses its licence, the portfolio does not become an asset in anyone's insolvency; the schemes are transferred to another AMC or wound up under trustee and SEBI supervision, with the proceeds belonging to unitholders throughout. This structure cannot stop your NAV falling with the market. It exists to stop a different failure entirely: the person managing your money making off with it, or their creditors reaching it.

Around that trust sits the rulebook. Under the SEBI (Mutual Funds) Regulations, the regulator registers the trust, trustees, AMC and custodian, caps what schemes may charge, fixes the category definitions, and mandates the disclosures that let you audit a fund at all: the daily NAV, the monthly portfolio, a monthly risk review, and the cost lines. AMFI, the industry body, administers distributor registration and publishes the aggregate data quoted at the top of this guide. Regulation constrains conduct and disclosure; it is deliberately silent on outcomes, because no rule can promise a return the market has not produced.

Until late 2017, Indian fund names were largely marketing. A single fund house could run half a dozen near identical equity schemes under different evocative names, and "balanced" or "multi cap" meant whatever the offer document quietly said. SEBI ended that with its categorisation circular of 6 October 2017: every open ended scheme must now fit exactly one of 36 categories arranged under five groups, and an AMC may run only one scheme per category, with narrow exceptions for index funds, fund of funds, and sectoral or thematic funds. The label on a fund became a contract about what it must hold, which is the thing that makes any comparison between two schemes meaningful at all.

The circular's quiet masterstroke was fixing the vocabulary. Large cap now legally means the 1st to 100th company by full market value, mid cap the 101st to 250th, and small cap the 251st onward, with AMFI republishing the ranking every six months, and each category carries binding portfolio minimums, so a large cap fund must keep at least 80 percent in large caps and a small cap fund at least 65 percent in small caps. Two "corporate bond funds" from different houses must now play by the same definition. A second computed label, the riskometer, is re evaluated every month from the actual portfolio and mapped to one of six bands, so a debt scheme that quietly buys riskier paper sees its dial climb whether or not the marketing changes.

All of that describes what a fund holds. None of it describes what a fund costs, and none of it decides what you keep. Two large cap funds can sit in the identical category, holding almost the identical companies, and hand two investors materially different corpuses over a working life, for one reason that has nothing to do with the label on the box. That reason is the subject of the rest of this guide.

The annual cost of running a scheme is its total expense ratio (TER): the management fee, the registrar, the custodian, the auditors, and in a regular plan an ongoing distribution commission, all expressed as a percentage of assets and, as the NAV ledger showed, deducted daily inside the price. Two features make it the most important number on the page. It is certain, charged every year whether the fund beats the market, matches it, or trails it, and it compounds, because every rupee the fee removes is also a rupee that never earns a return again. Alpha, the manager's outperformance, is uncertain and may never arrive. The fee arrives without fail.

Percentages this small look harmless, which is exactly why they deserve arithmetic. Take, purely to isolate the cost effect, a portfolio that compounds at 12 percent a year before costs; the number is an assumption for the sums, not a forecast and not a promise. Put ₹10 lakh into that one portfolio through two wrappers, a low cost fund charging 0.5 percent a year and a higher cost one charging 2.0 percent, roughly the real world spread between a plain index fund and a typical actively managed regular plan, and let both run. The cheap wrapper compounds at 11.5 percent net, the dear one at 10.0 percent. At ten years they are ₹3.76 lakh apart. At thirty years, on the same ₹10 lakh in the same portfolio, they are ₹87 lakh apart.

Read the year thirty bars slowly, because they carry the whole thesis. The ₹87 lakh gap is not a market outcome that a clever fund might reverse; it is baked in the moment the fee is set, and it grows with certainty every year the money stays invested. Even a difference of a single percentage point, the sort investors wave away as a rounding error, does most of this damage on a long enough horizon. This is why the fee you pay is the return you have already given away, and why it deserves more scrutiny than any past performance table, which describes a decade you cannot buy.

The same arithmetic decides the single most consequential tick box on any mutual fund form, and it is one the investor controls entirely, instantly and for free. Since 1 January 2013, under a SEBI circular of 13 September 2012, every scheme must offer a direct plan: the identical portfolio, run by the identical manager, minus the distribution commission that a regular plan pays to the intermediary who sold it. The two plans publish separate NAVs, and the direct NAV compounds faster by roughly the commission, commonly somewhere between 0.3 and 1.25 percentage points a year on equity schemes. It is the expense ratio story again, in its purest form: one fund, one manager, one portfolio, two prices.

A regular plan is not a swindle. The commission pays for distribution and, sometimes, for advice a household genuinely uses and values, and for an investor who would otherwise not invest at all, that service can be worth its price. But the arithmetic should be seen before it is paid, because no other single choice an investor controls entirely and for free moves a twenty five year outcome by this much. Choosing the plan type is a one time tick that then repeats its effect, silently, on the whole balance every year for the life of the holding.

Set against those certain costs is the thing the industry actually markets: alpha, the promise that a skilled manager will beat the market by enough to more than pay for the fee. Sometimes one does. The uncomfortable part is what has to be true of the group. In 1991 the economist William Sharpe set it out as plain arithmetic. Before costs, the holdings of all active investors added together are simply the market, so their average return before costs must equal the index. It cannot be otherwise, because they are the index. After their higher costs, the average actively managed rupee must therefore trail the index by roughly its extra cost. This is an accounting identity, not an opinion about anyone's skill.

The evidence lines up behind the arithmetic. Independent scorecards that track active funds against their benchmarks over multi year periods have repeatedly found that a majority underperform, and the share that lags tends to grow the longer the horizon; the precise figures move each year, so treat them as a direction to verify at the source rather than a fixed number. None of this says a good active manager is impossible. It says two harder things: you cannot identify next decade's winner reliably in advance, and the fee that funds the search is charged in full whether the search succeeds or not.

You cannot buy next decade's winner. You can read this decade's cost. One is a guess; the other is printed on the form.

That asymmetry, an uncertain benefit paid for with a certain cost, is the entire reason low cost index funds exist, and it is why the sober version of fund selection spends its energy on the controllable lever. It is not that active management is worthless. It is that its average product, after fees, is the market minus the fees, and the fees are the one part you can know today with total confidence.

Cost has one more component that never appears in the expense ratio, and it too is a certain drag on what you keep: tax. How a mutual fund is taxed depends on what it mostly holds, and the line that matters is whether it is equity oriented, meaning at least 65 percent in Indian equity, or not. An equity oriented fund is taxed like a listed share; a debt oriented fund is not. The table sets out the framework as of 17 July 2026, following the changes in the July 2024 budget and the April 2023 rule for debt funds.

Read across the rows and the pattern is the tax code's, not the fund's. For an equity oriented scheme, units held twelve months or less produce short term gains taxed at 20 percent, and units held longer produce long term gains taxed at 12.5 percent, with total long term gains up to ₹1.25 lakh a year exempt. The same rule covers a tax saving ELSS fund, which is equity oriented and carries a three year lock in. For a debt oriented scheme, for units bought on or after 1 April 2023, there is no special long term rate at all: the gain is added to your income and taxed at your slab, whatever the holding period. A hybrid fund follows whichever side its equity share puts it on. Surcharge and cess can apply on top, and the treatment depends on your circumstances.

The tax code, in other words, quietly rewards the investor's horizon over the trader's: a longer hold in an equity oriented fund reaches the concessional long term rate, while frequent switching realises short term gains at the higher rate and resets the compounding each time. That is one more reason the difference between trading and investing is partly a tax difference, and our guide to trading and investing taxation works the classification through in depth. For the purpose of this guide, the point is narrow and of a piece with everything above: tax, like the expense ratio and the plan commission, is a certain subtraction from what the market gives you, and the parts of it you influence, chiefly how long you stay, are behaviour, not stock picking.

Step back and a mutual fund resolves into what it really is: the delegation of execution, not of judgement. The trust holds the assets, the AMC runs the book, the RTA keeps the score, and the regulator watches all three. What remains yours is everything that actually decides the outcome, and it is a short list. You choose the category, which is to say the kind of risk you are willing to own. You choose the cost, the expense ratio and the direct plan over the regular one, which is the certain drag you can make small. And you choose your staying power, whether you remain invested through the falls that every equity NAV delivers, or sell into them and buy back after the recovery. Notice what is not on the list: the return. That belongs to the market, and no fee, fund or forecast can hand it to you in advance.

In a fund, cost is the return you keep and patience is the return you earn. Neither is the fund manager's to give you.

So the honest way to read a mutual fund is almost the opposite of how it is sold. The brochure leads with past performance and a promise of skill; the arithmetic says performance is the part you cannot control and cost is the part you can. Read the primary documents before committing capital, let the evidence set the position, and treat the low, certain cost as the closest thing to a free lunch the market offers. That habit, evidence first and exposure second, is exactly what the method we teach is built around, in funds no less than in trading.

• SEBI (Mutual Funds) Regulations, Regulation 52. Caps the total expense ratio a scheme may charge, on a slab that steps down as assets grow: for open ended equity schemes about 2.25 percent on the first ₹500 crore of daily net assets, down to about 1.05 percent beyond ₹50,000 crore, with debt schemes lower. Stated as of 17 July 2026; confirm the current caps at the source. sebi.gov.in
• SEBI circular CIR/IMD/DF/21/2012, 13 September 2012. Mandated a direct plan in every scheme from 1 January 2013, with no distribution commission charged to it: the source of the direct versus regular cost gap worked in this guide. sebi.gov.in
• SEBI circular SEBI/HO/IMD/DF3/CIR/P/2017/114, 6 October 2017. Categorization and Rationalization of Mutual Fund Schemes: the five groups, thirty six categories, one scheme per category rule, and the market capitalisation definitions used here. sebi.gov.in
• AMFI monthly industry data. Industry assets under management of about ₹81.58 lakh crore as of 31 May 2026, and monthly systematic investment plan contributions above ₹30,000 crore. amfiindia.com
• Union Budget July 2024 and the Finance Act 2023. The listed equity, and hence equity oriented fund, long term rate of 12.5 percent over a ₹1.25 lakh exemption and short term rate of 20 percent, and the taxation of debt oriented funds at the investor's slab rate for units bought on or after 1 April 2023. Stated as of 17 July 2026; verify current rates. indiabudget.gov.in
• William F. Sharpe, The Arithmetic of Active Management (1991). Establishes that before costs the average actively managed rupee earns the market return, so after costs it must trail the index by its extra cost: the identity behind the active versus passive figure. Independent benchmark scorecards corroborate it empirically; confirm the latest figures at the source. stanford.edu