The Investment Policy Statement and What It Governs

An investment policy statement is the written document that fixes what a portfolio is for and what it may not do, before any holding is chosen. The document states the return objective, the risk objective and the constraints, and every later decision is taken inside it. A portfolio without one has decisions but no standard against which any of them can be judged.

A piece of ordering hides inside that definition, and it decides everything else. The document comes first. The committee writes it while the portfolio is still empty, and no other moment finds every person in the room without a position to defend. A committee asked to agree a limit on a holding it already carries is no longer discussing policy; it is discussing that holding. An investment policy statementThe written document a holder of money agrees before any of it is invested, setting out what the portfolio is for and what may not be done with it. is written before any holding is chosen precisely because its job is to constrain choices nobody has made yet.

The running example is the Anantara Multi-Asset Portfolio, an invented discretionary mandateThe authority a holder of money gives to somebody else to run it, together with the limits on that authority. The word covers both the permission and its edges. of Rs 500 crore run by Faiz Ahmad Ansari for an invented charitable endowment, whose investment committee is chaired by Rukmini Deshpande.

What is an investment policy statement, and who writes it?

The holder of the money writes it, and it binds the person running the money. The direction matters, and it is the first thing readers get backwards. The manager may draft the words, may argue for a wider limit, may explain why a particular line is expensive to hold to. The manager does not agree it. In the Anantara arrangement the endowment agrees the document through an investment committeeThe group inside a holder that takes the decisions the holder cannot take one by one, meets on a schedule, and records what it decided. chaired by Rukmini Deshpande, and Faiz Ahmad Ansari then runs the money inside it and reports back against the same lines.

Here is the everyday version, and it is closer than it looks. A household hands the monthly grocery money to whoever does the shopping. Everybody in that household understands that the shopper decides which vegetables to buy and does not decide whether to spend the rent on vegetables. Nobody writes that down. The amount is small and the memory is short. Raise the amount to a wedding budget spread over eight months and the same household starts writing things down: this much for the caterer, nothing from the fixed deposit, no advance to anybody without asking first. The written note is a policy statement in every respect except the vocabulary.

One more thing about authorship settles a question readers raise early. The Anantara example uses an institutional holder, but the document is not an institutional instrument. The same five slots are written for a private holder, in the same order, with different contents in each slot. Nothing in the structure of a policy statement changes with the size or the type of the holder; only the numbers written into the slots change.

What does it mean to say a document governs a portfolio?

How an Investment Policy Statement Governs a Portfolio

Governing is not a metaphor here and it is not an annual event. Governing means that every proposed action is tested against the written lines before it is taken, and that an action failing any one test is not taken however attractive it looks on its own merits. The test is arithmetic, it takes a minute, and it happens on ordinary days when nothing is wrong. The document is consulted before the trade, not produced afterwards to explain it.

Watch one proposed action go through the Anantara lines. The largest holding currently stands at Rs 23 crore, and Faiz Ahmad Ansari wants to sell Rs 3 crore of one equity name and put the proceeds into it. The money stays inside the equity sleeve, so the portfolio total does not move: Rs 500 crore before, Rs 500 crore after. The equity weight does not move either, so the range line is untouched. But the largest holding becomes Rs 26 crore. Rs 26 crore on a Rs 500 crore portfolio is 5.2 per cent, and the written line is 5 per cent. The action is not taken. Nobody has to argue about whether the name is a good one. The test does not ask that question.

Notice what governing does to the argument in the room. Without the document, refusing that action means telling a colleague that the name is not worth Rs 26 crore. A refusal of that kind is a judgement, and therefore an argument. With the document, the refusal is a subtraction anybody at the table can do. BreachA written line that has been crossed. The word describes the state of the portfolio against the document, and says nothing at all about whether anybody did something wrong. is a word about arithmetic, not about character, and that is exactly what makes it usable between people who have to keep working together afterwards.

Why is a return objective on its own only half an objective?

Because a return figure standing alone can always be reached by taking more exposure, and a risk figure standing alone can always be met by holding cash and doing nothing. Each one, by itself, has an answer so easy that it destroys the question. Put side by side, the easy answers both disappear at once: exposure cannot rise without moving the risk line, and a retreat to cash abandons the return line. The pair is the objective, and either half written on its own is a wish rather than an instruction.

The household version of this is uncomfortably familiar. Someone says they want their savings to grow faster. The wish is a return objective with nothing beside it, and it has an immediate answer: put all of it into the most volatile thing available. Nobody means that, and that is the point. Their real meaning only becomes sayable when the second half arrives, usually in the form of a sentence about what must still be there in March when the school fees fall due.

Where do the objectives and the constraint lines come from?

How to Set Portfolio Objectives and Constraints

The objectives come from the purpose of the money and the date it is needed. The purpose and the date are genuinely the whole of it, and they are why the first conversation in a drafting room is never about markets. An endowment that funds a fixed programme every year and a household saving for a purchase in four years are answering the same two questions with different facts. Out of those facts fall a return figure and a risk figure, together.

The constraints come from somewhere else entirely, and from four separate places rather than one. The law's requirements on this holder, the cash needed and the date it is needed, the effect of the holder's tax position on a realised gain, and whatever is true of this holder and of nobody else: these are four sources, not a checklist to be worked through and ticked. Each of the four is covered separately. Each behaves differently once it is written down. One thing matters at this stage: the four are separate origins, so a drafter who asks one question four times has produced one constraint written four ways.

The third row deserves a moment. The Anantara record simply does not contain a tax position, so the honest entry is that nothing was supplied rather than a plausible sentence written to fill the row. A blank can be noticed and a guess cannot, so a policy statement with a blank in it is more governable than one with a guess in it.

What are the three kinds of constraint, and how does each one bind?

Sort them by how they bind and the whole sequence becomes easier. A range constraintA written line that binds on a total rather than on any one item, saying that some aggregate share must stay between two stated edges. binds on an aggregate: equity between 50 and 70 per cent of the Anantara Multi-Asset Portfolio says nothing about any individual name. A size constraintA written line that binds on each item separately, capping what any single holding may be, and saying nothing about what the items have in common. binds on each item taken separately: no single holding above 5 per cent says nothing about the total. An eligibility constraintA written line that binds at admission, deciding whether a thing may be held at all rather than how much of it may be held. binds on admission: no unlisted holdings decides whether a thing may be present, not how much of it may be.

Now the consequence most readers miss, and it is the reason the sorting is by binding rather than by subject. Prices move the numerator and the denominator together, so a range constraint and a size constraint can both be crossed while nobody in the building does anything at all. An eligibility constraint can only be crossed by an admission, or by something already held losing the quality that admitted it. One kind of line can go wrong overnight. Another kind cannot go wrong without a decision.

Put a number on the overnight case. The overnight case is far less exotic than it sounds. Take the Anantara Multi-Asset Portfolio at its policy weights: equity Rs 300 crore, everything else Rs 200 crore, total Rs 500 crore. Nobody trades for a month. The equity sleeve rises 16.0 per cent, a rise of Rs 48 crore, and the other sleeves do not move. Equity is now Rs 348 crore and the total is Rs 548 crore, so the equity weight is 348 divided by 548, or 63.5 per cent. The mandate never authorised a move from 60.0 to 63.5, and nobody made one.

What does a written constraint still permit?

A limit is a ceiling, and a reader almost always reads it as a plan. The five Anantara lines are short. Four of them reduce, with no further definition needed, to a rupee figure a manager can check today. Setting them out first and then computing what each one leaves open is the whole exercise.

Start with the policy weightThe share of a portfolio the holder decided each asset class should carry. It is a decision taken in advance, not an observation, and the actual weight drifts away from it as prices move. and the range around it. Equity is written at 60.0 per cent, or Rs 300 crore on a Rs 500 crore portfolio, and the range runs from 50 to 70 per cent, or Rs 250 crore to Rs 350 crore. The written figures are the policy weights, and the actual weights drift away from them between one rebalancing and the next. The corridor is twenty percentage points wide, or Rs 100 crore of equity exposure the manager may move without a single approval and without crossing anything, and the 60.0 per cent policy weight sits exactly ten points from each edge, so the discretion is symmetric in points.

The last line on the drawing is worth stopping for. The line is the base trap in miniature. Rs 250 crore to Rs 350 crore is the corridor measured at a constant Rs 500 crore total. If instead the equity sleeve alone moves and the other Rs 200 crore stays put, the arithmetic is different: equity reaches the 70 per cent edge only after a rise of about 55.6 per cent, at Rs 466.67 crore against a new total of Rs 666.67 crore, and it reaches the 50 per cent edge after a fall of about 33.3 per cent, at Rs 200 crore against a total of Rs 400 crore. Both descriptions of the same line are correct. The two descriptions answer different questions, and sliding between them without saying which base is in use would mislead a reader about how much room the manager really has.

Now the size cap, the line most likely to be misread as a promise about spread. Five per cent of Rs 500 crore is Rs 25 crore. Twenty holdings at Rs 25 crore each is Rs 500 crore exactly, so the cap permits the entire portfolio to sit in twenty names and forbids nothing about any of them once they are inside. Move to the equity sleeve and the answer changes. The cap is written against the portfolio, and the sleeve is smaller than the portfolio. Rs 300 crore divided by Rs 25 crore is twelve, so the cap fills the equity sleeve at its policy weight with twelve holdings. At the top of the equity range the sleeve is Rs 350 crore, and Rs 350 crore divided by Rs 25 crore is fourteen.

There is a further piece of headroomThe distance between where a portfolio actually sits and the written line it is measured against, expressed as a figure rather than as a feeling. arithmetic here that catches careful readers out. The largest Anantara holding is Rs 23 crore against a Rs 25 crore cap, so the obvious headroom is Rs 2 crore. Rs 2 crore divided by Rs 23 crore is 8.70 per cent, and that answer is wrong. The holding is part of the total it is measured against. If the holding rises on its own while everything else stays still, then at a rise of about 9.15 per cent the holding stands at roughly Rs 25.11 crore, the portfolio total stands at roughly Rs 502.11 crore, and 25.11 divided by 502.11 is 5.00 per cent exactly. The line is reached at 9.15 per cent, not 8.70.

The second route is the one nobody watches. Hold the largest holding perfectly still at Rs 23 crore and let the other Rs 477 crore fall by about 8.39 per cent, to Rs 437 crore. The total is now Rs 460 crore, and Rs 23 crore on Rs 460 crore is exactly 5.00 per cent. The cap has been reached by a holding that did nothing, in a portfolio where the manager did nothing. Nothing demonstrates more plainly that a size line binds on a ratio rather than on a decision.

What do the five lines actually leave open?

Set the permitted against the actual and the picture changes. Under the Anantara cap, the ten largest holdings could together be Rs 250 crore, or 50.0 per cent of the Rs 500 crore portfolio and 83.3 per cent of the Rs 300 crore equity sleeve. The ten largest actually hold Rs 155 crore, or 31.0 per cent of the portfolio and 51.7 per cent of the sleeve, at an average of Rs 15.50 crore each, or 3.10 per cent of the portfolio. The remaining eighteen names in a sleeve of twenty eight hold Rs 145 crore between them, averaging Rs 8.06 crore. Rs 250 crore divided by Rs 155 crore is 1.61, so the document permits a top ten 1.61 times the size of the one the portfolio has.

The finding is worth stating in plain words. Over the stated twelve month period the Anantara cap was never anywhere near binding. The largest holding sat at 4.6 per cent against a 5 per cent line and the top ten sat at 51.7 per cent of the sleeve against a permitted 83.3 per cent. A reader who assumed the cap was what kept this portfolio spread out would have credited a written line with a result that Faiz Ahmad Ansari produced by choosing positions, and the document supplied none of it.

What does the document deliberately not contain?

Three things, and the omissions are as deliberate as the inclusions. There is no list of holdings. There is no view on any market. There is no forecast. A policy statement that carries a list of names has confused the constraint with the position, and the confusion is expensive rather than merely untidy: the positions change constantly and the document is written not to. Put a holding list inside a policy statement and every ordinary trade becomes an amendment to the governing document. A document amended by every trade governs nothing.

Who may change it, and by what route?

The holder changes it, through the same body that wrote it, in writing, with the change dated and the reason recorded. The procedure is short, and it exists to protect one distinction. A limit rewritten because the portfolio had already crossed it is not a change of policy at all; it is a record of a breach, and unless the date and the reason are written down at the time the two become indistinguishable within a year.

The everyday shape of this is a household that decides in January to keep three months of expenses in a savings account, spends down to two months by August, and then agrees in September that two months was always the sensible figure. Nothing dishonest has happened. Everybody in the conversation believes the new figure. But the household has lost the ability to say whether it kept its rule or moved it, and that is the ability the whole document existed to create.

What does a portfolio without a written policy lose?

Not discipline in the abstract, and not seriousness. Something much more specific: the ability to tell a decision that was inside the agreed limits and turned out badly from a decision that was never authorised at all. The two decisions are completely different while they are being taken and completely identical once the year has gone badly. Both are now just a loss with a person attached to them.

Every holder eventually has this argument, and it is worth naming before it happens. Without the document, an authorised decision and an unauthorised one look exactly alike afterwards, and afterwards is precisely when the distinction matters most. A manager who worked inside every written line and had a poor year can say so and be checked. A manager with no written lines can only say that it seemed reasonable at the time. The claim is true and unfalsifiable, and therefore worth nothing to either side.

How does anybody use this on an ordinary Tuesday?

Four kinds of reader use the same document for four different purposes, and none of them is reading it for pleasure. Rukmini Deshpande, chairing the committee, reads it to know which items on the agenda are decisions and which are merely reports: anything the manager did inside the lines is a report, and anything requiring a line to move is a decision. The sorting is done before the meeting starts, and it is the reason a two hour agenda becomes a forty minute one.

An analyst looking at a discretionary mandateAn arrangement in which the manager takes each decision and acts on it, rather than recommending it and waiting for the holder to agree. from outside reads it as the boundary on what the record can possibly mean. A portfolio that may hold between 50 and 70 per cent equity has up to twenty points of exposure difference available to it in any year, so a comparison of two such portfolios is partly a comparison of where each one sat inside its own corridor. A lender assessing the same endowment reads the liquidity line and nothing else. Rs 50 crore of cash at the policy weight is the part that answers the lender's question.

And a household does the same work with a single sheet of paper. Write down the purpose of the money, the date it is needed, and the two or three things that must not be done with it. The writing takes twenty minutes and feels unnecessary. The twenty minutes are not wasted: the value arrives eighteen months later, on the day somebody asks whether a decision was outside the plan or merely unlucky, and there is a dated sheet of paper in the drawer that answers.