How Private-Fund Reporting Can Differ Across Managers

Reporting is a set of choices, so two reports of the same fund can both be honest and still hand a reader two different pictures. Which multiple is headlined and whether its denominator is named. Whether a rate of return is gross or net. How often an independent valuation is taken. Each choice changes what a reader can see, and knowing which choice was made matters more than preferring one of them.

Start somewhere small. The idea is not really about funds at all. A household sells the scooter it bought four years ago. Ask what it made and three honest answers are already waiting. The household got Rs 42,000 against the Rs 60,000 sticker price. The household got Rs 42,000 against the Rs 71,500 it actually paid once insurance, registration and four years of servicing went in. And it got Rs 42,000 against the Rs 80,000 it had set aside for the whole scooter idea in the first place. Nobody has lied anywhere in that paragraph. Three different denominators sat under one sale, and whichever one goes into the sentence is the one the listener will carry away.

A private fund report is that situation with a great many more zeroes and a great many more places to put them. Every figure in such a report is produced by a rule somebody chose, and a different rule produces a different figure from facts that have not moved by one rupee. That is not a defect to be policed away, and it is not something a reader repairs by preferring one style of presentation. Because a chosen rule sits under every figure, the first task with any report is to establish which rules produced the numbers in front of the reader, before deciding what those numbers say.

Why do two honest reports of the same fund look different?

Because a report is not a photograph. Somebody decided what goes inside the frame, and most of those decisions were written into a document long before the first quarter ever closed. Seven of those decisions come next. Every figure below belongs to one invented fund, Nilgiri Growth Partners Fund II, managed by Nilgiri Alternatives Advisors Private Limited, at the end of its Year 9 Quarter 2. That record date sits 8.50 years after the fund's final close.

One fund is not a sample. How often a choice gets made one way rather than the other is a question about hundreds of funds, and no single set of documents can answer it. What one set of documents can do is more useful anyway: name the choice at each point, state what this fund's own documents fixed, and describe the other form the same kind of term is capable of taking. Naming the choice makes the alternative visible without inventing anybody who took it.

Naming choices rather than ranking them changes what the word difference means. A reader who sets two reports side by side and finds them unalike is usually looking at two sets of choices rather than two levels of quality, and neither report will say which of the two is on the table. The distinction between two sets of choices and two levels of quality carries the whole argument, and the seven choices below are the places where the difference gets made.

Which multiple gets headlined, and does the report say which denominator it used?

Take the first choice on its own. The first choice costs a careless reader the most. At the end of its Year 9 Quarter 2, Nilgiri Growth Partners Fund II holds a total value of Rs 7,20,00,00,000. The total value is Rs 4,38,00,00,000 of cash already sent back to investors plus Rs 2,82,00,00,000 of value still sitting in five holdings that have never been sold to anybody. One number, and it is not in dispute.

Now divide it. The nine holdings cost Rs 4,00,00,00,000, so the fund is 1.80 times cost. The fee and the expenses were drawn from investors too, so investors actually paid in Rs 4,80,00,00,000 and the fund is 1.50 times capital paid in. Investors promised Rs 5,00,00,00,000 at final close, so the fund is 1.44 times commitment. Three multiples, one fund, one day, one unchanged Rs 7,20,00,00,000, and each of them is arithmetically correct. The choice a manager makes is which of the three becomes the headline multipleThe single multiple a report puts at the top, out of the several it could truthfully use., and whether the words after it say what it was divided by.

Watch how the arithmetic moves. As the denominator grows the multiple shrinks. The figure that flatters least is the one measured against the largest promise, and the figure that flatters most is the one measured against the smallest number in the set. A reader handed a bare 1.80 has been handed the top of that range with the scale removed. A reader handed a bare 1.44 has been handed the bottom of it, from the same fund, on the same afternoon.

Is the rate of return gross or net, and how would a reader tell?

The second choice is a single word, and its absence is invisible. A rate of return can be stated gross of feesBefore the management fee, the expenses and any carried interest have been taken out., meaning before anything the fund and the manager charged has come out, or it can be stated net to investorsAfter everything the fund and the manager charge has been taken out, which is what an investor actually keeps., meaning after all of it. Both are real numbers. Only one of them describes what an investor kept.

Nilgiri Growth Partners Fund I, an earlier invented fund from the same manager, is the only completed record here, and it carries both readings. Over its own ten years it reported a gross internal rate of return of 12.8 per cent and a net rate to investors of 11.1 per cent. The 1.7 percentage points between those two figures is not an error and not a rounding: it is the whole of what the fund and its manager were paid, expressed as a rate. The same fund shows the same story on its multiple, where 2.00 times gross became 1.80 times net, and that 0.20 of a turn on Rs 2,40,00,00,000 of capital paid in is exactly the Rs 48,00,00,000 of carried interest it eventually paid.

Nilgiri Growth Partners Fund II reports 8.3 per cent net at the end of its Year 9 Quarter 2, and that is the only rate on record for it. There is no gross figure to set beside it, so the same gap cannot be computed for this fund. The missing label is worth noticing on its own account. A report showing one rate and no label has not left a reader with a smaller number or a bigger one. Such a report has left them unable to say which kind of number it is.

How often is an independent valuation taken, and who marks in between?

Here is a small domestic version of the third choice. Two households each want to know what their flat is worth. One asks a registered valuer every year and, in between, writes down whatever the neighbours were quoting last month. The other asks the valuer every three years and writes nothing down in between. Neither household is being dishonest. But on a random Tuesday the two answers to the same question are a different age, and the age of a number is part of what the number means.

In Nilgiri Growth Partners Fund II the arrangement is fixed in the fund's own documents. Palani Valuation Advisors LLP, an invented limited liability partnership (LLP), acts as the independent valuation agentA party outside the manager, engaged to put a value on holdings that have no market price. and values every unsold holding once a year. The manager marks the same holdings each quarter in between. Kolar Fund Services Private Limited, also invented, is the administrator and strikes the net asset value. The valuation calendar decides how old the newest figure in the report is allowed to be, and it decides that before any figure exists.

Changing the interval changes nothing about the businesses and everything about the report. A term fixing a shorter interval puts an independent hand on more of the marks. A term fixing a longer one puts fewer. A term naming no independent party at all leaves every mark with the manager. Each of those is a form the same term can take. Which form is more common, and which is preferable, are questions a single fund's documents cannot answer. How a valuation is actually performed, and why a private mark moves later than a public price does, are both covered separately.

Is the capital drawn shown as one number or split into its parts?

The fourth choice looks like formatting and it is not. Nilgiri Growth Partners Fund II has drawn Rs 4,80,00,00,000 from its investors to the record date. The Rs 4,80,00,00,000 can appear in a report as one line, labelled capital drawn, and be entirely accurate. Or it can appear as three: Rs 4,00,00,00,000 invested in the nine holdings, Rs 70,20,00,000 of management fee and Rs 9,80,00,000 of fund expenses. Added together they come back to Rs 4,80,00,00,000 either way.

Now put the split beside the multiples from the first choice. The two meet here. The 1.80 times divides by the Rs 4,00,00,00,000 the holdings cost. The 1.50 times divides by the Rs 4,80,00,00,000 the investors actually paid. The difference between those two denominators is Rs 80,00,00,000, and that Rs 80,00,00,000 is the fee plus the expenses and nothing else. The split is the only form from which a reader can reach the gap between 1.80 times and 1.50 times without being told what it is. Left whole, the line remains correct and the gap becomes unreachable.

The fee figure does some work above, so here is where it comes from. During this fund's five-year investment period the management fee ran at Rs 9,80,00,000 a year, being 2.00 per cent of the Rs 4,90,00,00,000 of investor commitments, which is Rs 49,00,00,000 over the five years. From Year 6 the basis changes under the fund's own documents and the annual charge falls, adding Rs 21,20,00,000 to the record date, so the total drawn for the fee is Rs 70,20,00,000. How that basis is constructed is covered separately.

How far does the look-through go?

The fifth choice decides whether a reader sees a portfolio or a parcel. Nilgiri Growth Partners Fund II has Rs 2,82,00,00,000 of value still held at the record date, spread across five holdings that have not been sold to anybody. A report can carry that as one figure. A report can also carry it as five lines, each with a name, a cost and a carrying valueThe value a fund puts on a holding it has not sold, as at a stated date.. This fund's reporting takes the five-line form across all nine of the holdings it has ever had.

The five sitting inside that Rs 2,82,00,00,000 are holding 4, Bhavani Speciality Chemicals Private Limited at Rs 1,08,00,00,000; holding 6, Vaigai Edutech Private Limited at Rs 21,00,00,000; holding 7, Manjira Industrial Services Private Limited at Rs 39,00,00,000; holding 8, Kaveri Renewables Private Limited at Rs 81,00,00,000; and holding 9, Indravati Packaging Private Limited at Rs 33,00,00,000. 108 plus 21 plus 39 plus 81 plus 33 crore is Rs 2,82,00,00,000 exactly, so nothing has been added by naming them and nothing has been taken away by not naming them.

The deeper form gives a reader not a better total but a second question, and the shallow form gives a total nobody can interrogate. How far a look-throughShowing the underlying holdings beneath a total rather than the total on its own. can be pushed, and what it still hides once it has been pushed, is covered separately.

Is a public market comparison shown, and against what?

The sixth choice is the one most likely to be read as a fact about the world when it is really a fact about an arithmetic exercise. A public market equivalentOne fund's own cash flows carried forward at an index's return over the same dates, then compared with what the fund produced. takes every rupee a fund called and every rupee it paid back, carries each one forward at what an index did over exactly those dates, adds the value still held, and divides one total by the other. The method is the one set out by Kaplan and Schoar in the Journal of Finance in 2005, and naming the method is part of stating the figure honestly.

For Nilgiri Growth Partners Fund II over the 8.50 years to the end of its Year 9 Quarter 2, the contributions carried forward come to Rs 7,35,21,00,000 and the distributions plus the residual value come to Rs 7,71,39,00,000. Dividing the second by the first gives 1.0492. The report states that as 1.05. The comparison is against an invented reference broad equity index, which is not any real index. The index was set to 1,000.00 at this fund's final close and stood at 1,760.00 at the record date, which is 6.9 per cent a year over the 8.50 years.

Strip the index out of that paragraph and the 1.05 stops being a statement. A comparison is entirely defined by the thing it is against. Two arithmetic exercises run against two different indices produce two figures that cannot be set beside each other, and a report that shows a public market equivalent without naming what it used has printed a number in the shape of a conclusion. The 1.05 belongs to one fund over one period against one index, and it carries no claim at all about private and public markets in general, in either direction.

Is what went wrong named, or only carried inside a total?

The seventh choice moves no money whatsoever, and moving no money is precisely what makes it interesting. Nilgiri Growth Partners Fund II has had nine holdings and two of them are worth less than what they cost. Holding 5, Palar Foods Private Limited, cost Rs 35,00,00,000 and was written off in full at Year 6 Quarter 4, returning nothing. Holding 6, Vaigai Edutech Private Limited, cost Rs 30,00,00,000 and is carried at Rs 21,00,00,000, being 0.70 times its cost. The fund's own reporting names both.

Build the same schedule without either name. Holding 6's Rs 21,00,00,000 sits inside the remaining value and the four other unsold holdings carry Rs 2,61,00,00,000 between them, so the remaining value is still Rs 2,82,00,00,000, and 21 plus 261 crore is 282 crore. Holding 5 returned nothing, and nothing is what got added, so the realised proceeds are still Rs 4,38,00,00,000. Both schedules are arithmetically complete, both are honest, and not one rupee moves between them.

Naming the two holdings changes what the reader can do next. Given the names, a reader can ask what happened at holding 5 and what has been done about holding 6. Given the total alone, there is nothing in the report to ask a second question about, so no second question is available. The seventh choice is the shape of the whole argument in miniature: the choice did not alter a fact, it altered the surface area a reader has to work with.

Which other terms quietly decide what a report can show at all?

Seven choices carry most of the difference, but the same principle runs through terms that were settled long before anybody sat down to write a quarterly document. Each of three further terms fixes something a reader will later meet as a printed line, so each is worth naming.

The middle row is the reason the fee arithmetic here can be stated to the rupee. Because no side letter touches the fee, the annual charge in the investment period is a clean 2.00 per cent of Rs 4,90,00,00,000, being Rs 9,80,00,000, with no investor-by-investor adjustment sitting underneath it. Even the one side letter that relieves an investor of the fee has the manager bear that cost out of its own fee rather than the fund rebating it, so the fund's fee base stays whole. Change that single term and every fee figure printed anywhere in the report becomes a figure with a footnote.

What is not a choice at all?

Some differences between two reports have nothing to do with anybody's judgement. Separating those differences from the seven above stops a reader from criticising an arrangement for being the thing that it is. Nilgiri Absolute Return Fund, invented, is an open-ended vehicleA vehicle where money is subscribed and redeemed rather than committed up front and called over time.. Money goes in by subscription and comes out by redemption. There is no commitment anywhere in it and no capital call, so there is no capital paid in to serve as a denominator, so it cannot report distributions to paid in at all. The missing line is not somebody declining to show something. There is nothing there to show.

Nilgiri Direct Lending Fund I, invented, gives a second example of the same kind. The lending fund charges its management fee on capital actually drawn rather than on commitments, so its paid-in figure moves for a reason that has nothing to do with how the fee is presented. Both vehicles are covered separately. What matters at this point is only which bucket a difference belongs in.

The test is simple: ask whether the report could have been written the other way with the same facts underneath it. If it could, what is in view is a presentation choiceA decision about how to show a figure, which changes what a reader sees without changing any fact.. If it could not, what is in view is the structure of the vehicle, and no amount of wishing will make the missing line appear.

What does a reader do with seven choices and no ranking?

Seven questions, each yielding an answer rather than a preference. Seven answers are genuinely all. A ranking would need a sample, and one fund is not a sample. A statement about which choice is preferable would be a sentence with nothing underneath it. Such a sentence travels, and that makes it worse than silence.

The seven questions are the seven choices turned round. Which denominator is the multiple over. Is the rate gross or net. Who made the newest mark and how often. Is the capital drawn split. How far does the look-through go. Is a public comparison shown and against what. Is anything that went wrong named. Every one of those establishes which rule produced a figure, and not one of them asks whether the rule was the right rule.

A completed set of seven answers buys something specific. The seven answers establish what kind of number each figure is. A reader needs exactly that before reading any of them. Whether the fund is doing well or badly is a different question, and no set of answers to the seven settles it. A reader who can answer all seven for a report in front of them knows what they are looking at. A reader who cannot does not yet know, however confident the figures look.