The Specialised Investment Fund: Where It Sits in India

The specialised investment fund is a vehicle category in the Indian market that sits between a broad pooled scheme and a single bespoke account. Who may put money in, what may be held inside it, how it must report and what it may charge are fixed by regulation rather than by whoever happens to run it, and every one of those conditions is published by the Securities and Exchange Board of India (SEBI) at sebi.gov.in.

Where those conditions are published matters as much as what they say. A vehicle categoryA named kind of arrangement through which money is managed, with its own set of rules attached to the name. Two arrangements in the same category answer to the same rulebook. is half shape and half rulebook. The shape does not move, so the shape can be taught. The rulebook moves, and an account that recites it from recollection offers something that looks exactly like a fact and may no longer be one.

Seven questions place any arrangement anybody describes, and they keep working after every rule change without a single threshold, minimum, charge limit or eligibility bar attached to them. Every one of those conditions is set in regulation and is read at the source.

The worked case throughout is the Anantara Multi-Asset Portfolio, an invented discretionary mandate of Rs 500 crore run by Faiz Ahmad Ansari for a charitable endowment whose investment committee is chaired by Rukmini Deshpande. Its stated shape is equity 60.0 per cent at Rs 300 crore, fixed income 30.0 per cent at Rs 150 crore and cash 10.0 per cent at Rs 50 crore, summing to Rs 500 crore exactly. Every figure that follows belongs to that mandate and to one stated twelve month period, and none of them describes the category itself.

Which statements about a vehicle stay true, and which must be looked up?

Two kinds of statement get made about any vehicle, and they behave completely differently. The first kind describes the delivery arrangementThe plumbing between a holder and their portfolio: whose name the securities sit in, whose money sits beside theirs, who takes decisions, what they can see and how they leave.. Who holds the securities. Whose money sits beside the holder's. Who takes the decisions. Which information the holder can see, and when. How the holder gets out. Each of those five is a structural fact about the plumbing, and each holds whether it is read today or in four years.

The second kind is a condition attached to the category by regulation. Who may put money in. Which assets may be held inside. Which charges may be applied. Which facts must be disclosed. Which registrations must be obtained. Every one of those is a rule, every rule can be amended, and none of them is discoverable by reasoning about how the plumbing works. An account that mixes the two teaches a recollection as though it were a mechanism, and a reader has no way to tell which half is in view.

The everyday version runs like this. Suppose a friend describes a chit arrangement running in her street. The shape of it can be learned in five minutes: who collects, who holds the pot, when each member draws, what happens if somebody stops paying. The shape of the arrangement is stable and can be reasoned about. Reasoning cannot work out whether the arrangement is registered, what the law currently permits it to collect, or what it must tell its members. The answers live in a rulebook, and the only honest thing anybody can say is where the rulebook is kept.

Where does the category sit between a wide scheme and a bespoke account?

Picture two ends of a range. At one end sits a broad pooledAn arrangement where many holders' money is combined and managed as one sum, so each holder has a claim on a share of the whole rather than on named securities. scheme, available to a very wide public and constrained tightly in what it may do, precisely because it is available so widely. At the other end sits a bespokeWritten for one holder. The limits come from that holder's own instructions rather than from a standard rulebook applying to everybody in the category. account written for one holder, where the limits are the ones that holder wrote. The Anantara mandate lives at that second end: its equity band of 50 to 70 per cent and its cap of 5 per cent of the portfolio on any single holding were set by Rukmini Deshpande's committee, not by a category rulebook.

Between those two ends is open space, and it is a two dimensional space rather than a line. How widely available an arrangement is, and how tightly what it may hold is constrained, are separate questions. A category like the specialised investment fund occupies a region in that space rather than a point, and its position is described entirely by which of the delivery questions it answers like each end. It can be placed without a single number.

Placing things this way works because the two dimensions explain each other. Wide availability is the reason a rulebook gets written: when an arrangement can be sold to anybody, the rules have to protect somebody who never negotiated them. When an arrangement is written for one institution with its own committee and its own advisers, that institution negotiates its own protection. A move along the availability axis predicts, without anything being looked up, that the constraint axis will move too.

What seven questions answer any vehicle, whatever it is called?

Seven questions do all the work, and not one of them needs a number. Who holds the securities. Whose money sits beside the holder's. Who decides what is held. Which information the holder can see, and how often. How the cost reaches the holder. How the holder gets out, and how quickly. Who writes the rules for the arrangement. A reader who can ask those seven of any vehicle does not need a condition memorised on their behalf.

The questions the seven do not ask matter as much. The seven never ask what the arrangement returned. Return is a different subject, and one stated year settles nothing. Nor do the seven ask what the arrangement costs as a rate, only how the cost reaches the holder. When two shapes are compared, the route matters more than the level. No delivery question can answer whether an arrangement is any good, so the seven do not ask. Seven structural questions, and every one of them can be answered by reading two documents and asking one supplementary.

Which of the seven do the parties settle, and which does regulation settle?

Six of the seven are settled between the parties in the terms of the arrangementThe agreement between a holder and whoever runs their money: what is permitted, what is reported, what is charged and how either side ends it.. Who holds the securities is a choice about custody. Whose money sits beside the holder's follows from whether the arrangement is pooled. Decision rights, visibility, how the cost reaches the holder and how the holder leaves are all negotiated, written down and signed. Six out of seven are settled that way, and the count explains why two arrangements in the same category can feel completely different to their holders.

Then the seventh does something the other six cannot do for themselves. In India the category's own rules set the outer bounds of what the parties are even permitted to agree about visibility, cost, exit and eligibilityWho is permitted to put money into an arrangement at all. Eligibility is a condition attached to the category rather than a term the parties negotiate.. The seventh question therefore partly answers the other six, and the negotiation happens inside a box somebody else drew. Choosing a vehicle category is therefore never purely a commercial decision taken between two willing parties: the category comes with a rulebook attached, and the rulebook was not at the table.

A household version makes the point stick. Two neighbours agree a rent, a notice period and who repairs the geyser. Six things they settle between themselves. Neither of them settles what the tenancy law of the state permits them to agree in the first place, and if the law says a notice period cannot go below a certain length, their handshake does not survive it. The agreement is real, and it lives inside a boundary drawn elsewhere.

What do the seven questions say about the Anantara mandate?

The seven can be run against a case already established. The Anantara Multi-Asset Portfolio is a bespoke, separately managed arrangement, so its seven answers are unusually clean, and watching clean answers is the fastest way to learn what a messier set would look like.

Holders skip question two, so it deserves a sentence of its own. Whether anybody else's money sits beside the holder's decides whether anybody else's behaviour can reach that holder. For this endowment the answer is nobody, and that is a structural fact rather than an assurance anyone gave.

Question four is where a bespoke arrangement differs most sharply from a pooled one, and the difference is not about generosity. The endowment has full look-throughBeing able to see the actual securities held rather than only a summary or a single combined value. A holder with look-through can recompute any weight. to the securities, which means it does not receive a concentration figure and trust it. The endowment computes its own. The largest holding of Rs 23 crore is 4.6 per cent of the Rs 500 crore portfolio and 7.7 per cent of the Rs 300 crore equity sleeve, and the holder can derive both numbers because it can see the holdings that produce them.

Question six looks trivial until two shapes are set side by side. Where the holder has a claim on a pool, getting out means disposing of that claim, and the speed is a term of that arrangement. Where the holder has the securities in its own name, as the endowment does, there is nothing to sell back to anybody: the mandate simply ends and the holdings stay where they are. The two exits differ in kind, and neither is faster in the abstract. The question therefore asks how the holder leaves rather than how fast alone.

Question five is the one holders underestimate, and the Anantara terms show why. The mandate's own commercial terms are a management fee of 1.25 per cent of assets, or Rs 6,25,00,000/- on Rs 500 crore, plus a performance fee of 15 per cent of the return above a 10 per cent hurdle. The stated year returned 14.2 per cent gross, so 4.2 points sat above the hurdle. On Rs 500 crore that is Rs 21,00,00,000/-, and 15 per cent of it is Rs 3,15,00,000/-. Together the cost of delivery for the stated year is Rs 9,40,00,000/-, or 1.88 per cent of assets. Each figure is this mandate's own term rather than a market rate, an industry level or anything a regulator sets.

Now watch what the route does. Because the cost is charged directly to the account, it comes out of the same gross return the holder is reading. Gross 14.2 per cent less 1.88 per cent leaves a net 12.32 per cent against a benchmark of 12.6 per cent, so a gross excess of plus 1.6 points becomes a net shortfall of minus 0.28 points. The portfolio beat its benchmark for the stated year and the holder did not.

The same result in rupees is harder to argue with, and holders find it easier to feel. The gross excess of 1.6 points on Rs 500 crore is Rs 8,00,00,000/-. The cost of delivery for the stated year is Rs 9,40,00,000/-. The difference is Rs 1,40,00,000/-, and Rs 1,40,00,000/- on Rs 500 crore is exactly the 0.28 points the percentage version showed. Two ways of writing one arithmetic, and neither is a verdict on the arrangement. The record contains no alternative to compare it with.

Both of those figures are correct and neither of them is the excess return on its own. Plus 1.6 points and minus 0.28 points describe one portfolio, one year and one benchmark, and they differ only in whether the cost of delivery has been taken out. The word gross or the word net therefore belongs in the same sentence every time.

What happens when the same seven are asked of this category?

Six of them get different answers, and those answers can be reasoned towards from the shape alone. The seventh stops the reasoning, and it should. For a specialised investment fund the rules are written by the regulator. Every condition hanging off the category is therefore a matter of published text rather than of negotiation, reasoning or recollection. Each of those conditions is left as a blank field in the form below.

The blank fields are not a draft left unfinished. A condition can be stated only as of a date, and the next amendment can move it, so a field filled in today asserts something nobody could know when it was typed. The destination is the same for all seven, which is itself worth noticing: when every blank on a form points at one office, the form shows where the subject actually lives.

What does a holder actually check, step by step?

Four steps, and the striking thing about them is that not one of them contains an answer. Read the category's current text at the regulator's own site. Read the specific arrangement's own constituent documentThe document that sets up a particular arrangement and states what it may do, what it will report and what it will charge. Two arrangements in one category can have very different ones.. Compare the two, because an arrangement is permitted to bind itself more tightly than its category requires and is never permitted to be looser. Then ask which of the seven questions this arrangement answers differently from whatever else is being considered.

A procedure is worth more than any condition stated in advance. A condition ages. A procedure does not. Step three is the one people skip. Reading two documents to find they agree feels like wasted effort, right up until the day they do not agree and the tighter one turns out to be the one that binds.

Why a stated condition is worth less than a destination

Say it plainly rather than apologetically. A minimum, an eligibility bar or a charge limit written from recollection does not merely become stale when it changes: it becomes wrong, and a reader who acts on a wrong condition is worse off than a reader who was told to go and look it up. Staleness is a small problem because it announces itself eventually. Wrongness announces nothing at all.

The uncomfortable part is this. A condition that has moved looks, in print, exactly like a condition that has not. Same typeface, same confidence, same authority. There is no visual difference between a sentence that was true when it was written and is still true, and a sentence that was true when it was written and stopped being true nine months ago. The reader has no signal to work with. The choice is not between a helpful figure and an unhelpful one but between a figure and a destination.

A wrong condition is the most damaging failure of the two, and routingNaming the question and the authority that publishes the answer, instead of writing the answer down. The destination stays correct after the answer changes. is the answer to it rather than an evasion. Routing costs the reader one visit to a website and buys them a fact that is true on the day they need it. Writing the condition down saves that visit and buys them a fact that was true on the day somebody typed it.

What does the vehicle category leave completely unchanged?

Almost everything that matters about the portfolio itself. A vehicle is a delivery arrangement and not a strategy. Move the Anantara mandate's Rs 500 crore from one category to another and the equity sleeve is still Rs 300 crore, the fixed income sleeve is still Rs 150 crore, the cash is still Rs 50 crore, the largest holding is still Rs 23 crore, and the top ten holdings are still Rs 155 crore, being 31.0 per cent of the portfolio and 51.7 per cent of the equity sleeve. Nothing in that list is settled by which category the money sits in.

The allocation still has to be decided, the constraints still have to be written, the concentration still has to be measured against a stated base, and the cost of delivery still has to be computed and disclosed. Every one of those is work somebody does, and none of it is done for them by the category. The vehicle decides how a portfolio reaches a holder and decides nothing whatsoever about whether the portfolio is any good.

How does a committee actually use this on a Tuesday?

Rukmini Deshpande's committee does not open a meeting by asking what category something falls under. The committee opens by asking the seven questions and writing the answers in a column, and that column is what gets compared. Then it asks step four: which of these seven does the arrangement in front of the committee answer differently from what the endowment already holds? Usually two or three of the seven differ and the rest are identical, and the discussion belongs entirely to those two or three.

Then the committee runs the arithmetic it can run, and stops where the record stops. For the stated year the alpha attributed to the mandate is 1.11 percentage points, or Rs 5,55,00,000/- on Rs 500 crore. The cost of delivery was Rs 9,40,00,000/-, so the fees exceeded the alpha by Rs 3,85,00,000/-, or 0.77 points. That is a real finding and the committee should say it out loud. The committee must not turn that finding into a verdict. Whether the arrangement was worth having depends on what the alternative would have returned and what it would have cost, and this record contains no alternative at all.

A household does the same work with a pen and no committee. Before agreeing to any arrangement for money, the household writes down who will hold what was bought, whose money it sits with, who decides, what will be shown and when, how the cost reaches them, and how the money comes back. Six lines. Then the seventh: who wrote the rules for this, and where is that written down? The seventh line is usually the one nobody in the room can answer, and finding it out is worth more than the first six put together.