What a Fund Manager Actually Decides, and What They Do Not

A fund manager is on the payroll of an asset management company and exercises whatever discretion the scheme's written documents leave open, and nothing beyond it. Inside those limits the manager selects what the scheme holds. Outside them, other parties hold the securities, keep the unit register and compute the value of a unit, and the charge is a term of the document.

Start with the word most readers skip. A fund managerA salaried worker whose job is choosing what a pooled scheme holds, inside limits drawn up by other people. holds delegated discretion. Somebody else handed it over, in writing, and somebody else can take it back. The word delegated changes how everything below reads. A reader who pictures a person choosing freely with a pool of money will get the next five questions wrong in the same way. The position is not ownership of a pool, it is permission to decide certain things about a pool that belongs to other people.

So the useful question is not what a manager is good at. The useful question is what the permission covers. A mandateAll the limits binding a manager at one moment, arriving in writing from several places at once. is the whole of that permission, built up here step by step: where its limits come from, who is above the manager in the line of accountability, what parts of a scheme were never in the manager's hands at all, and what stays put on the day the manager walks out.

One invented case carries every figure in this guide. Girnar Asset Management Limited, an asset management companyThe employer of a scheme's staff, paid a fee out of the scheme in return for running it., runs the Girnar Large Cap Equity Fund, an open ended equity scheme with net assets of Rs 4,200 crore and 120.00 crore units in issue. The same company also runs the Girnar Broad Market Index Fund, a scheme written to follow a broad index. The index scheme appears below only because its mandate leaves a narrower slice of discretion open than the equity scheme's does. Kalyani Bhagat runs the equity scheme as its fund manager, and operations across the company are headed by Sohail Merchant.

Three things are taken as already settled and are not rebuilt here. The first is a company, and what it means for one to be supervised. The second is an expense, and how it sits against a set of assets. The third is a portfolio, understood as a collection of holdings run against a stated objective, with any return measured over a period and on a stated basis. Everything that follows sits on top of those three, and what it adds is structural rather than technical: who is allowed to decide what.

What is a fund manager doing on an ordinary Tuesday?

A fund manager turns up to a salaried job. The employer is the asset management company, and the manager is one of its staff in exactly the way a branch head is staff at a bank. Across a working day the manager reads, meets people, keeps track of what the scheme already holds, and decides whether to buy something, sell something or leave the position alone. The activity itself is ordinary office work. The unusual part is that the money being decided about was never the manager's and never the employer's either.

Here is the everyday version, and it is worth holding on to because it survives every complication later on. Picture one member of a household sent out with three things settled in advance: a list drawn up by somebody else, a spending cap they may not go past, and the only two shops where that spending may happen. The shopper now makes real decisions: which brand, what size, what to skip when something is out of stock. The decisions matter and they are genuinely the shopper's. But nobody in the household would say that person has become the person whose money it is. The freedom is real and it is bounded, and both halves of that sentence have to be held at once.

Translate it straight across. Girnar Asset Management Limited employs Kalyani Bhagat to run the Girnar Large Cap Equity Fund. She decides what the scheme buys and sells. The written objective was fixed before she arrived, and it can be changed only by the process the scheme's own documents lay down. She does not decide the purpose of the scheme. She cannot widen her own limits. And the discretion she holds was delegated to the position rather than granted to the person. Delegation to a position rather than a person is why the position can be refilled without anything else moving.

Where does a mandate come from, and how many sources feed it?

A mandate is not one document and not one rule. A mandate is three separate sets of limits arriving from three separate places and applying at the same moment. The first comes from the regulator, the Securities and Exchange Board of India (SEBI). SEBI attaches conditions to what a scheme of a given kind may do at all. The second comes from the scheme's own document, a promise made to the people who put money in. The third comes from the employer, who adds internal rules of its own for reasons that have nothing to do with either of the first two.

The next point decides how the whole thing reads. Three limits applying at once do not average out and do not take turns: the narrowest of them binds, and the other two become irrelevant for as long as it does. If the regulator permits something, the scheme document does not, and the employer has no view, the answer is no. If the employer's internal rule stops at a point well inside both of the others, the answer is that internal point. A reader who assumes the binding limit is always the regulator's will keep opening the wrong document, and will keep being surprised by what a scheme did not do.

SEBI's layer sits above the other two. Conditions attach to who may manage a scheme, to what a scheme document must contain, and to what a scheme of a stated kind may hold. Every one of those exists, and conditions of that sort get revised, so the current position is the text published at sebi.gov.in on the day the answer matters. The rules an employer adds internally are the employer's own business and are not published in the same way at all.

Who does a fund manager answer to, and who can remove them?

The shape of the line is the answer, so trace it rather than assert it. Kalyani Bhagat answers to Girnar Asset Management Limited. The company employs her and can end that employment. Girnar Asset Management in turn answers to a trustee companyA separate company whose duty is to the people holding units, and which supervises the asset manager on their behalf.. The trustee company is a different company with a different set of people, and its duty does not run to the asset manager at all. Its duty runs to the people holding units. The trustee company appears by role only, unnamed, as do the auditor, the distributor, the custodianThe party that holds a scheme's securities and settles what it trades, kept separate from the manager on purpose. and the registrar and transfer agentThe party that maintains the register of unit holdings and processes purchases and redemptions..

Read in one direction, the line looks like a reporting chart. Read in the other, its purpose becomes visible. The purpose of the line is simple: whoever chooses what a scheme holds must never also be the last word on whether those choices stayed inside the limits. Somebody above the manager can ask that question, and somebody above that person has a duty to the holders that makes asking it their job rather than a favour. Take one link out and the structure stops working. The dependence of each link on the next is what makes this a structure and not an organisation chart.

The line does not say how often the questions get asked, what evidence gets looked at, or what happens when an answer is unsatisfactory. The machinery of supervision is real and is covered separately. Only the direction of the arrows matters at this point: outward and upward from the manager, ending at the people whose money it is.

Which parts of a scheme are not the manager's to decide?

The count is the lesson, so count them. There are four, and each one sits with a party that is not the manager and not the manager's employer acting through the manager.

First, where the securities are held. The custodian holds them, and the separation is deliberate: the person deciding what to buy is not the person holding what was bought. Second, the ledger showing which holder has which quantity of units. The ledger sits with the registrar and transfer agent. The register runs across every folioA single account on the register, carrying whatever quantity of units one holder has. in the scheme. Third, what one unit is worth. Nobody decides that at all; it is computed, by valuing what the scheme holds and dividing by the units in issue. Fourth, what the scheme charges. The charge is a term written into the scheme's own document, and any limit on it is a SEBI matter rather than a monthly choice somebody makes.

A reader who arrived thinking the manager runs the whole scheme has just lost four of its working parts, and none of the four is small. Between them they cover custody of everything the scheme holds, the entire ownership record, the price at which every purchase and every exit happens, and the largest recurring charge on the pool. The remainder in the manager's hands is genuinely important and genuinely narrow: choosing, inside the mandate, what the pool holds.

What do the divisions look like on one invented scheme?

A division shown is a division a reader can check, and a division quoted is one they have to trust. So put the case on the table and do the arithmetic in the open. The Girnar Large Cap Equity Fund has net assets of Rs 4,200 crore and 120.00 crore units in issue. Divide the first by the second. Rs 4,200 crore over 120.00 crore units is Rs 35.00 for one unit, and that one lands exactly, with nothing left over in either direction.

The second division is untidier, and the untidiness is worth showing rather than hiding. The scheme is held across 3,80,000 folios. Rs 4,200 crore spread over 3,80,000 folios is about Rs 1,10,526/- for an average folio, and the word about is doing real work: the exact quotient runs on past the rupee. Multiply the rounded figure back out and 3,80,000 folios at Rs 1,10,526/- comes to Rs 41,99,98,80,000/-. The multiplied back total falls Rs 1,20,000/- short of the Rs 4,200 crore the scheme actually holds. The rounding went down, and the residue is where it went.

Nobody decided either of those two divisions. Both belong in an account of what a manager decides for exactly that reason. The value of a unit falls out of the valued holdings and the unit count. The average folio falls out of the pool and the number of accounts on the register. A manager who wanted either figure to be different could not make it so by wanting it, and the average folio is not even a figure the scheme uses for anything. The average folio is here because it shows the shape of the holder base: about a lakh of rupees each, on average. A lakh each is a household sum rather than an institutional one.

What does one unit carry in charges across a year?

The Girnar Large Cap Equity Fund carries an expense ratioThe running charge taken out of a scheme's own assets across a year, quoted as a percentage of those assets. of 1.65 per cent, measured against net assets. A ratio without its base states nothing, so state the base out loud, always. Take 1.65 per cent of Rs 4,200 crore and the year's charge is Rs 69.30 crore, or Rs 69,30,00,000/- written out in full rupees.

Brought down to one unit, the figure becomes something a reader can feel. Rs 69.30 crore spread across 120.00 crore units is Rs 0.5775 a unit for the year. The other route checks it, and the habit is what is worth taking away rather than the number: Rs 0.5775 measured against a unit value of Rs 35.00 is 1.65 per cent, back to the rate the working started from. Two routes, one answer, no rounding anywhere in the chain. When those two disagree, something in the arithmetic has slipped and the check has caught it.

Kalyani Bhagat did not set that 1.65 per cent, cannot alter it by deciding to, and does not receive it. The charge is a term of the scheme's own document, it pays for a set of working roles of which managing the holdings is only one, and whether there is any ceiling on it is a SEBI matter, published at sebi.gov.in. The figure does show scale: fifty seven and three quarter paise a unit a year, against a unit worth Rs 35.00. A charge that size is small enough to be invisible on a statement and large enough to be Rs 69.30 crore once.

What did the scheme return, and on which basis?

Across the one stated year, with the value of a unit at the start set against the value of a unit at the close, the Girnar Large Cap Equity Fund returned 13.4 per cent. The 13.4 is a net returnA return measured after a scheme's costs have already been taken out of its assets, rather than before.: the value per unit at both ends of the year was already struck after the 1.65 per cent had come out, so nothing further is deducted from 13.4. Over those same twelve months the benchmark it states came in at 12.1 per cent. Both figures cover the same single year, and both belong to the same single scheme.

Set the two side by side and the gap looks like 1.3 points. Now say the awkward thing out loud. An index is not a thing anybody can hold, and nobody is charged a paisa for following one. A scheme's figure carries its costs and an index figure carries none, so subtracting the second from the first is not a like-for-like reading at all. Put both on one basis and the picture moves. Add the 1.65 per cent straight back and the scheme's costless equivalent is about 15.05 per cent. The gap against 12.1 is then about 2.95 points. Because a running charge accrues rather than arriving once, backing it out by compounding instead of adding gives about 15.30 per cent, and a gap of about 3.20 points.

Neither route deserves an exact bridge. The finding survives both routes, and the finding is the useful thing: the honest gap is near three points rather than 1.3, more than twice the headline. One more check goes unresolved and is worth naming rather than assuming. A scheme's return includes income the scheme received, and an index may be computed on prices alone or with income put back in. The record here does not say which of the two the stated benchmark uses. The question stays open rather than being quietly settled.

How much of that one year result did the manager decide?

Every reader arrives with this question, and it deserves an honest answer rather than a flattering one. Two things produced the observed figure and they are not the same thing. The mandate decided the field: the whole set of outcomes that were available to the scheme at all, fixed in writing before the year started and fixed by people other than the manager. The manager decided inside that field: which of the permitted things happened, and when.

One year on one scheme cannot separate those two contributions, and no amount of staring at 13.4 per cent will make it do so. The figure is a single observation of the two working together. Splitting it would need a record this one does not have: a run of periods rather than one, more than a single scheme, and something that holds the field steady while the choices vary. Separating them properly is a measurement question of its own and is covered separately.

Nothing in the record says the manager was skilled, and nothing in it says the manager was not. The invented record simply does not carry the evidence for either sentence. The record does support the structural point, and the structural point is the one worth leaving with: the field was drawn first, by a document, and that document is the part a reader can actually go and read.

If the manager walks out tomorrow, what actually moves?

Readers arrive with this question most often and almost never see it answered structurally, so here is the structural answer. Suppose Kalyani Bhagat resigns from Girnar Asset Management Limited tomorrow. Walk the vehicle part by part and check what moved. The scheme still exists. The trustee company is still there with the same duty to the same people. The registrar and transfer agent still keeps the register across the same 3,80,000 folios. The custodian still has the securities in its hands. The written objective still says what it said. The written policy still says what it said. And the pool of assets still belongs to the people holding units. The pool always did, and a resignation is not a transfer.

One thing changed: the person exercising the discretion, and that is the entire list. Somebody else, named by the employer, now works inside the same written limits that the previous holder of the position worked inside. Both of them found those limits already drawn. Neither of them could widen them.

The reason a mandate is written down at all, rather than left as an understanding between capable people, is precisely so that the vehicle does not depend on any one person staying. A structure that collapsed when one employee left would not be a structure, it would be an arrangement. Writing the limits down is what converts the second into the first, and it is also what makes the departure a personnel event rather than an event about anybody's money.

What can a written mandate never do?

A careful reader is most likely to overreach at this last step. A mandate fences in what is able to happen at all. The mandate draws edges around what the scheme may hold and what it may do, and those edges are real and enforceable. The mandate does not decide which of the permitted things actually happens, and it makes no statement at all about how any of them turns out.

A reader who takes a tight mandate as a floor under the outcome has read a boundary as a promise, and a boundary has never been one. A fenced field works the same way. The fence marks where the animals cannot go. The fence says nothing about where inside the field the animals will stand, and it certainly does not say the grass is good. Narrowing the fence narrows the possibilities; it does not fix the result.

So the honest use of a mandate is the modest one, and it is still the most useful thing a reader can do. Reading it shows what the set of possible outcomes even contains. No account of what happened last year can show that much. Last year happened under whatever limits applied then. The document is the part that decides what is available. The document is also, conveniently, the part that is written down and can be read before anything is decided.

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.