How Scheme Categories Work, and What They Constrain

A scheme category is a placement rather than a description. The Securities and Exchange Board of India (SEBI) defines the categories and the conditions attached to each; an asset manager chooses which one to launch a scheme in and then lives inside it. Every scheme sits in exactly one, the category limits what it may hold, and the name has to match. Each condition behind that is published by SEBI and moves over time.

A reader arriving here usually wants a list. Something like: here are the groupings, here is what each one means, here is the number attached to each one, now go and pick. Every one of those numbers is set by a regulator and every one of them has moved at least once, so a printed figure stops being merely out of date and becomes wrong. The shape of the framework does not move, and neither does the address where the values live. Both outlast every figure.

Start with the word itself. A scheme categoryThe group a mutual fund scheme is formally placed in, which carries conditions on what the scheme may hold and what it may call itself. is not an adjective somebody attached to a scheme. A category is a placementA formal assignment into a defined group, made under somebody else's rules, which carries obligations rather than opinions., made inside a categorisation frameworkThe published set of groups a scheme may be placed in, together with the conditions attached to each group. that SEBI wrote and that SEBI revises. A description can be argued with and a placement binds. The difference between the two is the whole of the matter.

Two invented schemes run through everything that follows. Girnar Asset Management Limited operates the Girnar Large Cap Equity Fund, an open ended equity scheme with net assets of Rs 4,200 crore and 120.00 crore units outstanding, and the Girnar Broad Market Index Fund, which tracks a broad index that is never named here. Kalyani Bhagat manages the equity scheme and Sohail Merchant heads operations.

Three things are settled elsewhere and none of them is rebuilt here. What a scheme is, who runs it and whose property its holdings are: settled where the structure of a scheme is taught. What a unit is and how value per unit is struck: settled where pricing is taught. What a purchase, a switch and an exit actually do to a holding: settled where transactions are taught. All three are taken as given, and one question remains: how a scheme got sorted onto a shelf in the first place, and what that sorting does to it afterwards.

What is a scheme category, exactly?

The shape of a school is identical and the stakes are lower, so think about a school for a moment. A school does not let each teacher invent a subject name and then examine on whatever they feel like. Somebody above the teacher writes down what the subject called physics has to cover, what a paper carrying that name may and may not ask, and what a school may call a course before it is allowed to put that name on a certificate. The teacher still has enormous freedom inside physics. The teacher may not call something physics when it is not.

A scheme category works the same way. The regulator writes down the groups, writes down what a scheme placed in each group has to look like, and writes down what such a scheme may call itself. An asset manager then launches a scheme into one of those groups. The manager writes the scheme's objective, its strategy and its marketing name, and the framework writes the boundary those three have to sit inside.

The word large cap in the name of the Girnar Large Cap Equity Fund is therefore a marketing name that has to match a placement, and never a definition. Girnar Asset Management does not decide what qualifies as large capitalisation. The framework settles that, and the classification list holdings are measured against is published by the Association of Mutual Funds in India (AMFI) at amfiindia.com. Both addresses appear in the references below.

Who decides which category a scheme goes into?

Both parties do two completely different jobs, and separating those jobs is the second thing that matters here. SEBI defines the groups and attaches conditions to each of them. Girnar Asset Management decides which of those groups to launch a scheme into, and from that moment lives inside the conditions attached to it. The manager's freedom is entirely in the choosing and never in the defining. A definition of a category belongs to SEBI and to nobody else.

The split is easier to feel in a housing society allotting shop space. The society decides that this row is for food, that row is for services, and writes down what counts as each. A shopkeeper picks a row and signs for it. The shopkeeper has every freedom about what to cook, how to price it and what to call the stall. The shopkeeper cannot take the food row and quietly run a tailoring business out of it, and cannot rewrite what the word food means in the society's own register.

The split has a consequence worth stating. When somebody states what a category means, the useful next question is not whether they sound confident. The useful next question is where they read it. A definition of this kind has exactly one home, and everything else is a copy of that home at some past moment, made by somebody with no obligation to update it.

Can one scheme sit in two categories at once?

No. A scheme is placed in exactly one, however mixed its holdings happen to look on any particular day. The single placement carries more consequence than readers expect.

Here is the everyday version. A restaurant on a street is registered under one licence even if the food it serves would fit under two. The licence it registered under, not the one an onlooker would have guessed from the menu, settles what it may do, what it must record and what it may advertise. Two restaurants with almost the same menu can be registered differently, and the difference is invisible from the pavement.

The consequence readers miss is that two schemes which look similar can sit in different categories and be constrained differently, so a reader comparing them on holdings alone is comparing across a boundary they cannot see. The holdings are the visible part. The boundary decides what the scheme is allowed to do next year, in a market nobody has met yet, when the manager would rather do something else. None of that can be read off a list of what the scheme happens to hold today.

The same point explains something that confuses people about the two Girnar schemes. Both are run by the same asset manager, checked by the same trustee company, held by the same custodian and recorded by the same registrar and transfer agent. None of that has any bearing on where either one is placed. Placement follows the scheme, not the manager.

What does a category actually constrain?

Four things, and it is worth naming them precisely because the reader who can name them stops looking to the category for anything else. A category constrains what the scheme may hold. A category constrains how much of that it must hold, expressed as a floor beneath which the scheme may not fall. A category constrains how narrow the permitted set may be, and how narrow that set is separates a broadly drawn category from a tightly drawn one. And it constrains what the scheme may call itself.

Every one of those four is a SEBI condition with a current value, every one of those values moves by circularA published instruction from a regulator that adds to or changes an existing rule, and applies from the date it says it does., and the current value of each of those conditions sits with SEBI. The reason deserves saying in plain words. A value printed here is correct until the day it changes and wrong every day after that, and nothing in the text would announce the switch. Absent is honest. Wrong is not, and wrong is what a stale figure becomes.

There is a second reason, quieter but sharper. A reader who is handed the value learns the value. A reader who is handed the name of the condition and the address learns the route, and the route keeps working after the value has moved. The first reader has to return and hope somebody updated the text. The second reader never has to return at all.

Why must a scheme's name match its category?

Because the name is the only part of a scheme that almost everybody reads, and the placement statement is a part that almost nobody reads. The asymmetry between the two is the entire justification for the rule, and once it is seen the rule stops feeling like paperwork.

Picture the packet of atta on a shop shelf. The ingredient panel on the back is complete, accurate and legally required, and perhaps one shopper in fifty turns the packet over. Every shopper reads the word on the front. So the front of a food packet is regulated far more tightly than the back. The front is where the belief is actually formed. A scheme's name is the front of the packet.

SEBI ties the name to the category for exactly that reason, so a reader who checks a scheme's name against the placement stated in its own document is using the naming ruleThe condition requiring a scheme's name to be consistent with the category it is placed in, so the name cannot suggest one thing while the placement says another. precisely as it was meant to be used. That check takes about twenty seconds, and no twenty seconds of reading in this whole sequence returns more. A name that suggests one thing while the scheme is placed somewhere else reaches the largest number of people with the least effort on their part. No other single disclosure failure does more damage.

What problem was the framework built to solve?

ComparabilityThe property that makes two things worth setting side by side, because they are measured on the same basis and constrained in the same way.. Not fairness, not safety, not performance. Comparability, a smaller and more useful thing than any of those.

Think about what the world looks like without it. Two schemes carry almost the same name. One holds one kind of thing, the other holds something quite different, and both names are defensible because nobody wrote down what either name has to mean. Now put their results side by side. The two things being compared were never the same kind of thing. The comparison produces a number, the number looks meaningful, and it is worth nothing at all. Worse, the only visible signal is the name, and the name is precisely the signal that has been rendered useless. The reader has no way to discover that.

A framework fixes that by making the name carry information again. Once two schemes with similar names are required to sit inside similar boundaries, setting one beside the other tells the reader something. But a category makes two schemes comparable, it does not make them alike, and those are two entirely different claims. Two schemes inside one boundary can be run in ways that have almost nothing in common, and that gap is the subject of everything below.

What does a category not tell a reader about a scheme?

Almost everything a reader actually wanted to know. A category does not state what the scheme costs or how large it is. A category says nothing about how the scheme is run inside its limits, and every decision a manager actually makes lives there. Nor does a category state what the scheme returned. And a category does not state whether a scheme fits any reader. No part of the framework was built to answer that.

Two schemes in one category can differ on every one of those, and a reader who treats a category as a verdict has borrowed authority the framework never claimed. The framework claimed a boundary. A boundary is a statement about the edges of what is permitted, and it is silent about everything inside those edges. The interesting differences live inside.

The two Girnar schemes make this concrete without needing to sit in the same category at all. The equity scheme charges 1.65 per cent of net assets a year and the index scheme charges 0.20 per cent. The equity scheme therefore charges 8.25 times what the index scheme charges. Over the stated year the equity scheme returned 13.4 per cent net returnA result computed from values that already carry the scheme's own charges, so nothing is deducted from it afterwards., and the index scheme returned 12.12 per cent net over the same year, a difference of 1.28 percentage points. Both figures are net, so they are on the same basis and can be set beside each other. Nothing in the categorisation framework produced either gap.

The arithmetic invites a conclusion it cannot support, so the limit deserves saying out loud. One year and two schemes settle nothing about any approach in general. A cost gap of that size tells nobody what the next year will produce.

What do the two schemes look like as placement cards?

Here is the worked instance, laid out as two cards in one table. Every arithmetic figure below is computed rather than quoted, and every rule-set field is left blank with the source written inside it. A blank field names which value a reader has to fetch and where from. A blank never goes stale the way a printed value does.

Work the two arithmetic lines yourself rather than taking them from the table. The habit is worth building. Rs 4,200 crore divided by 120.00 crore units is Rs 35.00 a unit exactly, and the division is what makes it a fact rather than a claim. Then 1.65 per cent of Rs 4,200 crore is Rs 69.30 crore for the year, and the base has to travel in the same sentence as the rate or the figure states nothing at all.

The record insists on one more line, and it is easy to skip. The equity scheme's 13.4 per cent is net of its own 1.65 per cent charge, while the stated benchmarkThe index a scheme states it measures itself against, which nobody can hold and which therefore carries no costs of its own. figure of 12.1 per cent carries no costs at all. An index is not investable, and nobody pays anything to hold one. So the headline gap of 1.3 percentage points is not like for like. Placed on one basis, the honest gap moves nearer three points. The additive route and the compounding route give different answers and neither is the truth, so no exact bridge is available. The finding survives both routes. A single figure does not.

What happens when holdings drift, or a definition moves?

Two different things go wrong here, they look identical from the outside, and neither one involves anybody doing anything wrong. Drift and redefinition turn a static picture of the framework into a moving one.

The first is driftA holding sliding outside the shape a rule requires because prices moved, rather than because anybody bought or sold anything.. A scheme's holdings are measured as proportions, and proportions move when prices move even if nobody at the scheme touches anything. One holding runs up, another falls back, and the shape of the whole portfolio quietly changes on a day when the manager placed no order at all. Kalyani Bhagat can sit still for a month and find the scheme has moved anyway. Holdings drift with prices rather than with decisions. A scheme can therefore sit outside its category's required shape without a single transaction having taken place.

The household version is exact. A household set out to keep its savings mostly in one place and a slice somewhere riskier. A year passes, the risky slice runs up hard, and the split is now nothing like the one the household chose. Nobody moved a rupee. The proportions moved by themselves, and any rule written in proportions is now being broken by a household that did exactly what it said it would do.

The second thing is stranger. A definition can be changed under a scheme that did not change at all. The regulator issues a circular, the shape a category requires is redrawn, and a scheme that fitted comfortably on Monday sits outside on Tuesday having done nothing. Both situations are handled by conditions SEBI sets. The window a scheme gets and the process it follows are values, and values move. A reader who meets either situation will find the current position at the source, which is the only place it is ever current.

Where does the current value of every one of these conditions live?

Two addresses, and the route from a named condition to its current value is two steps long. Step one is naming the condition. Step two is reading its current value at the source, on the day it is needed. The source is where the value is current, and nowhere else is.

The nine conditions are these. How many categories the framework defines and how they are grouped. The definition of each individual category. The least a scheme in any category must hold, and the permitted range around it. How narrow a permitted set may be. How many schemes one asset manager may run inside a single category, and the exceptions to that. The rule tying a scheme's name to its category. The requirement to name a benchmark and the conditions on which one may be named. The treatment of holdings that drift outside a category's shape, and how long a scheme has to correct it. And the process by which a category definition is changed at all.

All nine sit with SEBI at sebi.gov.in. The industry classification list that a scheme's holdings are measured against sits with AMFI at amfiindia.com, and the difference between the two is worth being precise about: SEBI makes rules and AMFI publishes an industry classification. A rule never comes from an industry body, and blurring the two teaches the reader to ask the wrong organisation. Nine conditions named, and the current value of every one of them sits with the rule maker.

Girnar Asset Management Limited, the Girnar Large Cap Equity Fund, the Girnar Broad Market Index Fund, Kalyani Bhagat and Sohail Merchant are invented.
Educational material. Not advice on any investment, tax, budget or market position.