Management Fee vs Carried Interest: Certain and Conditional

One is a charge for running the fund and arrives whether the fund makes money or not. The other is a share of profit and arrives only if profit arrives, after investors have had their capital back and a stated return on it. Nilgiri Growth Partners Fund II, invented, has been charged Rs 70,20,00,000 of the first to its record date and has paid nothing at all of the second.

The two payments separate much more easily before the rupees get large, so start with a street. A woman on a residential road runs a small tailoring business and has to be away for a year, so she asks a neighbour to run it while she is gone. She can have two completely different conversations with him about money, and confusing those two conversations is the mistake this whole subject turns on.

The first conversation is: what will it cost to keep the shutter up. Somebody has to open at nine, deal with the cloth supplier, pay the two tailors, chase the customer who has not collected her blouse, and sit there on the slow days when nobody walks in. Keeping the shutter up has a cost, and the cost is there in a month when the shop takes twelve thousand rupees and in a month when it takes nothing. The second conversation is completely different: if, at the end of the year, there is more money in the tin than she left behind, what share of that extra does the neighbour keep. The first payment is about the work of running the thing and the second is about how the thing turned out, and no amount of the first can ever turn into the second. The fee and the profit share are not two sizes of the same payment. Each has its own trigger.

A private fund pays its manager in exactly those two ways, for exactly those two reasons. The two are always described together because together they make up the manager's entire pay. Separately they behave nothing alike. One is certain and, on the fund this guide works on, shrinking. The other is uncertain and may never arrive at all. Most readers have met only the two headline percentages, and that is not meeting either payment properly.

What is each of the two payments actually for?

A comparison that begins before both sides are on the table teaches nothing. Take them one at a time and define each fully before setting them against each other.

The management feeA charge for running the fund, payable whatever happens. is a charge for running the fund. Not for winning, not for choosing well, not for selling at a good moment: for running it. Somebody has to be paid to look at three hundred businesses in order to buy nine of them, to send lawyers through the ones that get serious, to sit on the boards afterwards, to keep an administrator and an auditor and an independent valuer engaged, to answer twelve investors who each want something slightly different, and to keep doing all of that in a year when the portfolio produces nothing whatsoever. The charge falls due because a year has passed and the fund still exists. A year passing is the entire trigger.

Carried interestA share of profit, payable only if profit arrives. is a share of the gain. Carried interest is not payment for work, and treating it as deferred wages gets its behaviour wrong in every direction. The profit share is a claim on an amount that has to exist first. If the fund gives its investors back everything they put in and a stated return on top of that, and there is still money after all of that, the manager keeps a share of what is left. If there is nothing left, the manager keeps a share of nothing. A share of nothing is nothing. The work was done either way. The payment is not a reward for the work; it is a slice of a result.

Two consequences follow immediately and both are worth stating before any arithmetic. First, one of them is a known rupee amount and the other is an unknown that may be zero, so the two cannot be traded off against each other by anybody trying to judge whether a manager is expensive. Second, they answer to different questions. Of the fee: charged on what, and does that basis move. Of the profit share: after what, and has that condition been met. No other question about either payment gets a reader as far, and everything below answers these two properly on one invented set of numbers.

Before the numbers, one point of vocabulary, made once and then left alone. The words that appear everywhere in this subject, general partner, limited partner, limited partnership agreement, carried interest, come from a structure built in another country and then imported wholesale. The invented vehicles here are Indian and are settled as trusts under an indenture of trust. Nilgiri Trusteeship Services Private Limited, invented, is the trustee; Nilgiri Alternatives Advisors Private Limited, invented, is the investment manager and is the party paid both of the amounts worked here; Nilgiri Financial Holdings Private Limited, invented, is the sponsor. There is no partnership and no general partner as a matter of law here: the role a general partner would play is discharged by the manager and the trustee between them, and the contract is a trust deed and a contribution agreement rather than a partnership agreement. The economic vocabulary is still what the documents and the investors use, so it is the vocabulary used here too.

Which one arrives whatever happens, and how much of it has?

The management fee, and on the fund worked here it has arrived every single year since the fund began. Nilgiri Growth Partners Fund II, invented, is a closed-end growth and buyout fund managed by Nilgiri Alternatives Advisors Private Limited, invented. Twelve investors committed Rs 4,90,00,00,000 to it and the manager committed a further Rs 10,00,00,000 of its own, making Rs 5,00,00,00,000 in total. Its term is ten years from final close and its investment period is the first five of those. Dates run from that final close, so Year 9 Q2 means the second quarter of the ninth year of this fund's own life.

This fund's record date is the end of its Year 9 Q2, and between the start of Year 1 and that date Nilgiri Growth Partners Fund II has been charged Rs 70,20,00,000 of management fee. The Rs 70,20,00,000 is not a projection and not an average: it is the sum of nine years of charges, four of them at a rate applied to a basis that had already stepped down, and the last of them counted only half because only half of Year 9 had run when the record was struck. Charges after that date are not in this record.

Now hold that beside the second payment on the same fund, at the same date. Carried interest paid by Nilgiri Growth Partners Fund II to the end of its Year 9 Q2: nil. Not a small amount. Not an amount too early to see. Nothing at all. In the ninth year of a ten-year fund, on a portfolio that has already sold four positions and part of a fifth, the manager has received not one rupee of profit share, and exactly why becomes clear further on.

The contrast is the point, but a fund that has not finished can only ever show the first half of the story, and Fund II has not finished. Seeing the contrast properly needs a second fund. So the record offers a completed one. Nilgiri Growth Partners Fund I, invented, is a wound-up fund managed by the same manager. Fund I raised Rs 2,50,00,00,000, ran its full ten years, sold everything it held and closed. Across its whole life Nilgiri Growth Partners Fund I paid its manager Rs 48,00,00,000 of carried interest, and the whole of that amount fell in its last distribution.

Read those two funds side by side and the shape of the difference is unmistakable. Two traps sit right here, so read them carefully. The first is that Fund I and Fund II are two different funds. Fund I is wound up; Fund II is live and in its ninth year. Their money is separate, their investors are only partly the same, and their clocks are four years apart, so Fund I's Year 10 is not the same real moment as Fund II's Year 10. The second trap follows from the first: Fund II's Rs 70,20,00,000 of management fee and Fund I's Rs 48,00,00,000 of carried interest belong to two different pools of money raised from two different sets of investors at two different times, and must never be added, netted, or presented as one manager's total. Any sentence containing both figures has to name both funds.

The two funds together do legitimately show the arrival pattern. Fund II's fee arrived in Year 1, and in Year 2, and in every year since, including the four years after its investment period ended and including the half-year in which the record was struck. Fund I's profit share arrived once, at the very end, after nine years in which it arrived not at all. Fund I made four distributions across its life and the first three of them paid its manager nothing whatsoever of that amount; it all came in the fourth. The difference is between a payment triggered by time passing and a payment triggered by a condition being met.

What is the two per cent actually charged on?

Here is where most readers stop asking questions, and it is exactly one question too early. A rate on its own is half of a charge. The other half is the basis the rate is charged on, and the basis is where almost all of the interesting behaviour lives.

The amount a percentage is applied to is the fee basisThe amount the percentage is charged on, and it can change over the life.. Nilgiri Growth Partners Fund II contracts a management fee of 2.00 per cent a year. Two per cent of what? During the five-year investment period, of aggregate commitmentsThe total investors agreed to contribute. from investors, being Rs 4,90,00,00,000. Two per cent of Rs 4,90,00,00,000 is Rs 9,80,00,000, and that is what the fund was charged in each of Years 1, 2, 3, 4 and 5, giving Rs 49,00,00,000 across the investment period.

Notice the figure the basis is not. The basis is not Rs 5,00,00,00,000, the fund's total commitments. The manager committed Rs 10,00,00,000 of its own money to this fund alongside the investors, and it charges itself no fee on that. So the denominator under the 2.00 per cent is the investors' Rs 4,90,00,00,000 and not the whole Rs 5,00,00,00,000, and the difference between those two, though it is only 2.0 per cent of the total, is a real Rs 20,00,000 a year of fee that is never charged. There is a second such carve-out inside the investor side too: one of the twelve investors is a vehicle through which the manager's own senior staff invest, and it pays neither management fee nor carried interest. But the manager bears that cost out of its own fee rather than the fund rebating it. The fund's fee basis is therefore still the full Rs 4,90,00,00,000, and the arithmetic above is exact.

Now the discipline this guide keeps for the rest of its length. Nilgiri Growth Partners Fund II, invented, has been charged Rs 70,20,00,000 of fee to the end of its Year 9 Q2. As a share of what? Against total commitments of Rs 5,00,00,00,000 it is 14.04 per cent. Against the investor commitments of Rs 4,90,00,00,000, the basis the rate was actually written against, it is 14.3 per cent. Against the Rs 4,80,00,00,000 of capital this fund has actually drawn from its investors, it is 14.625 per cent. Three true statements about one amount, and the only thing that changed between them is the denominator. A share quoted without its denominator has said almost nothing.

The 14.625 per cent is printed to three decimals deliberately. Fund II's Rs 70,20,00,000 of fee over its Rs 4,80,00,00,000 of capital drawn, both to its record date, is 14.625 per cent exactly. The third decimal lands precisely on a half. Rounding it down gives 14.62 and rounding it up gives 14.63; describing it in one decimal, 14.6 and 14.7 are both defensible. Rather than make that choice silently and leave a reader unable to reproduce the figure, the convention used here is to print it unrounded and say why. Any share that lands exactly on a half gets the same treatment.

If the rate never moved, why did the charge fall by nearly two thirds?

Because the basis moved, and the basis was always going to move, on a day written into the contract before a single rupee was drawn.

Think about how a fee on commitments behaves once a fund stops buying things. In the first five years the manager is looking for companies, and it is looking on behalf of the whole Rs 4,90,00,00,000 that investors have promised, whether that money has been called yet or not. Charging on the promise makes a kind of sense while the promise is the thing being worked on. But at the end of Year 5 the investment period closes and the fund may no longer make new investments at all. From that point the manager is not searching; it is holding what it already has and trying to sell it. Charging on a promise nobody is going to draw against any more would keep the charge flat while the work behind it shrank.

So this fund's documents contain a step-downThe point at which the basis changes and the charge falls.. From Year 6, the 2.00 per cent is charged not on commitments but on the acquisition costWhat the fund paid for holdings it has not yet realised. of the holdings the fund has not yet realised, measured at the start of each year. The rate is untouched. The figure it multiplies is replaced.

Watch what that does. At the start of Year 6 all nine of this fund's holdings were still held and their cost was Rs 4,00,00,00,000, so the year's charge was Rs 8,00,00,000. By the start of Year 9 two positions had been sold, one had been written off, one had gone to an initial public offering and a further slice of another had been sold, and the acquisition cost still held had fallen to Rs 1,80,00,00,000. Two per cent of that is Rs 3,60,00,000. The rate in Year 9 is the same 2.00 per cent it was in Year 5, and the annual charge is Rs 3,60,00,000 against Rs 9,80,00,000, being 36.7 per cent of what it had been, a fall of Rs 6,20,00,000 a year, with not one term renegotiated to get there.

A charge that falls by nearly two thirds while the rate stands still is the single most useful fact about the certain payment, and it is invisible to anybody holding only the headline rate. The fee did not fall because somebody asked nicely, or because the fund underperformed, or because a regulator intervened. The fee fell because selling a holding removes that holding's cost from the basis, and the basis is what the percentage eats. A fee written this way shrinks as the portfolio empties, automatically, as a matter of arithmetic.

The total reconciles cleanly and it is worth checking rather than accepting. Five investment-period years at Rs 9,80,00,000 give Rs 49,00,00,000. The record date falls halfway through Year 9, so the four years after the investment period, on the stepped-down basis, give Rs 8,00,00,000, then Rs 6,40,00,000, then Rs 5,00,00,000, and then half of Rs 3,60,00,000, being Rs 1,80,00,000. The four amounts come to Rs 21,20,00,000. Rs 49,00,00,000 plus Rs 21,20,00,000 is Rs 70,20,00,000, the whole management fee charged by Nilgiri Growth Partners Fund II to the end of its Year 9 Q2 and the figure this walkthrough opened with. How that schedule is set out year by year, and what else moves with it, belongs to the treatment of how a private fund is structured and is covered separately.

One more thing about the certain payment before leaving it. The word certain earns its qualification here. Certain means the payment arrives whatever the portfolio does. Certain does not mean fixed. On this fund the annual charge has taken five different values across nine years without the rate ever changing, and a reader who assumed a flat 2.00 per cent of Rs 4,90,00,00,000 for the whole ten years would have described the later part of this fund's life wrongly by a factor of nearly three.

What has to happen before a single rupee of carried interest arrives?

Investors have to get their money back first, and then some. The condition is the whole of it, stated in one line. The rest of this section works out what the line means for the fund worked here.

Nilgiri Growth Partners Fund II has drawn Rs 4,80,00,00,000 from its investors. The Rs 4,80,00,00,000 is not only the money that went into companies, and it is worth pausing on. The drawn capital is Rs 4,00,00,00,000 of investments, plus Rs 70,20,00,000 of management fee, plus Rs 9,80,00,000 of fund expenses, and those three add to exactly Rs 4,80,00,00,000. The management fee is itself drawn capital, so investors have to be given the fee back too before any profit share can begin. The fee sitting inside drawn capital is one more reason the two payments cannot be traded off: the certain payment enlarges the amount that has to be returned before the contingent one can start.

Against that Rs 4,80,00,00,000, this fund has distributed Rs 4,38,00,00,000 in four cash payments. Every rupee of that has gone to giving investors back what they put in, and the fund is still Rs 42,00,00,000 short of even that. The fund has not begun to pay a return on top. So the manager's profit share is not small, or early, or partly accrued. The profit share is nil, and it will remain nil until a great deal more cash has actually reached investors.

How much more? The first rupee of profit share on this fund arrives at Rs 7,09,98,27,451 of cumulative distributions, or 1.48 times what investors paid in. The fund has distributed Rs 4,38,00,00,000. A further Rs 2,71,98,27,451 of cash has to reach investors before the manager receives anything at all. The cash sitting between those two figures, the order in which it is paid and how the share is finally struck are covered separately. The condition exists, it is measured in cash actually distributed, and this fund has not met it.

A genuine tension sits in this fund's record, and it is exactly what makes people mis-read a profit share. On one measure this fund is doing perfectly well: its total value is 1.50 times what was paid into it, counting the Rs 2,82,00,00,000 of holdings it has not sold at the value they are carried at. On the measure that actually governs the profit share, it has not returned the capital it called. Both sentences are true at the same time. A distribution waterfall pays on cash that has left the fund. A multiple counts an estimate of things nobody has bought yet. Rs 2,82,00,00,000 of this fund's stated value has never been sold to anybody, and a profit share cannot be paid out of an estimate.

What does the contingent payment look like when it does arrive?

Seeing the contingent payment arrive requires a fund that finished, and the record has exactly one. Nilgiri Growth Partners Fund I, invented, is wound up. Fund I raised Rs 2,50,00,00,000 across nine investors, ran a full ten years with no extension taken, sold all seven of the businesses it held, and closed. Its terms were the same shape as Fund II's: same fee rate, same profit share, same requirement that investors are made whole first.

Its investors paid in Rs 2,40,00,00,000 across its life. The fund distributed Rs 4,80,00,00,000 back to them and to its manager. The difference between those two, Rs 2,40,00,00,000, is the profit the fund made above the capital its investors had returned to them. Of that profit, the manager of Nilgiri Growth Partners Fund I kept Rs 48,00,00,000 across that fund's whole ten years.

Here a stated rate and a stated amount can be reconciled exactly, so do the check rather than take the percentage on trust. Fund I's Rs 48,00,00,000 multiplied by five is Rs 2,40,00,00,000. Not approximately: exactly, with no remainder. The manager's share was one fifth of the profit, exactly what 20.0 per cent means, and the four fifths left over, Rs 1,92,00,00,000, went to the investor class. An identity that closes to the rupee is worth more than a percentage somebody has quoted, and if it had not closed, one of the two figures would have been wrong.

Two things about that Rs 48,00,00,000 are worth more than the amount itself. The first is when it arrived. Fund I made four distributions across ten years. The first was in its Year 6, the second in its Year 8, the third in its Year 9 and the last at its Year 10 Q4, when it wound up. The manager received nothing of this payment from the first, nothing from the second and nothing from the third. The entire Rs 48,00,00,000 fell in the last distribution of a ten-year fund, in the tenth year, on the day the fund closed. Meanwhile the management fee on that same fund had been charged from its first year onwards, without interruption, the whole way through.

The second thing is that this outcome was not available in advance. On the day Fund II held its final close, Fund I was at the start of its fifth year and had made no distribution at all. Not a small one; none. Its first distribution came more than a year later. So at the moment Fund I's manager was raising its successor, the profit share that eventually came to Rs 48,00,00,000 was worth precisely nothing and might have stayed there. The wait is not an irregularity and not a warning about anybody. The wait is the ordinary consequence of a ten-year clock, and it is the plainest available demonstration that a contingent payment is contingent right up until it is not.

Is there a third flow, and where does it land?

There is, and it catches people out because it does not come from the fund at all. A manager can be paid by the companies the fund has a stake in, directly, for work done at those companies: a monitoring arrangement, or a fee for arranging a transaction. The money leaves a business the fund holds and arrives at the manager without passing through the fund's own accounts on the way.

Put it back on the street for a second. The neighbour running the tailoring shop also charges the cloth supplier a small commission for placing the order with him. The money never touches the shop's tin. But the shop still bought that cloth, so the shop has, in a roundabout way, paid part of the neighbour's commission. Whether that commission then comes off what the shopkeeper owes him at the end of the year is not a fact about commissions. The answer is a term the two of them either agreed or did not.

The mechanism that settles it is a fee offsetA reduction in the management fee by fees taken elsewhere.. On Nilgiri Growth Partners Fund II's terms, one hundred per cent of fees taken at portfolio companies is set against the management fee. To the record date, Rs 1,20,00,000 of such fees has been taken and the whole of it has been offset, being 1.7 per cent of the Rs 70,20,00,000 of management fee charged. So the manager received that money and the fund's own fee bill fell by the same amount. The offset does not change what the manager did at those companies; it changes who ultimately bears the cost of it.

An offset of less than one hundred per cent is also a term that exists. Where the offset is partial, the two amounts stop being equal and part of the fee taken at a company stays with the manager on top of the fund's fee. The terms set out here are Nilgiri Growth Partners Fund II's own. One invented fund's arrangement is not evidence about anybody else's, so what such terms usually are, in India or anywhere else, does not follow from them.

Why do the two headline percentages say the least?

Almost everybody who has heard of this subject at all has heard the phrase two and twenty. Two and twenty is short and memorable. Each half of it is a rate with its other half missing, so it is the least useful thing anybody could carry away.

What does the pair look like across a whole life rather than one year?

A year is the wrong unit for this comparison and it is the unit almost every discussion uses. Across a single year the fee looks like a fraction of a large number and the profit share looks like a possibility. Across a full ten-year life they behave like two different instruments altogether, and the analysis of that difference has a name: Metrick and Yasuda, in The Economics of Private Equity Funds, published in the Review of Financial Studies in 2010, set out how a manager's revenue divides between fee income and the share of profit over the life of a fund, and in their treatment the split itself is the object of study rather than the size of either part.

Described as objects, the two payments look nothing alike. The management fee on Nilgiri Growth Partners Fund II is a sequence of nine charges beginning in Year 1, none of which depends on any outcome, whose annual size fell from Rs 9,80,00,000 to Rs 3,60,00,000 as the portfolio emptied, and which reached Rs 70,20,00,000 in total by the record date. The fee has a known shape. The only real uncertainty in it is how fast holdings get sold, and that uncertainty only makes it smaller rather than larger, so most of the fee could have been drawn on the day the fund closed.

The profit share is not a sequence at all. On Fund II the profit share is one number, currently zero, and it will either stay zero or become something on a date nobody can name in advance. On Fund I, now finished, the profit share was also effectively one number, Rs 48,00,00,000, and it landed on a single day at the very end of a decade. Certain pay accumulates; contingent pay resolves. One is a running cost that the fund carries whatever happens, and the other is an event that either occurs once or does not occur.

Two consequences of that fall out immediately and they are the practical end of the comparison. The first is about time. A manager receives the certain payment throughout, so the certain payment is what actually keeps the lights on across the years when nothing is being sold. On this fund there were five consecutive years with no distribution at all. The second is about information. At any moment before a fund finishes, the certain payment can be stated to the rupee and the contingent one cannot be stated at all, and that asymmetry does not narrow gradually. On Fund I it stayed at nil through three of four distributions and then resolved in a single step.

None of that says either payment is high, low, fair, or well designed. The comparison says they are different objects with different behaviour, and that a description which merges them, in either direction, has described neither.

What would somebody reading a real set of fund terms actually do with this?

Fund terms are read most often by people who do not sign the cheque. Analysts at an institution that has been offered a place in a fund, junior members of an investment team asked to summarise terms, staff at a pension pool or an insurer or a bank treasury who have to explain to a committee what the manager will be paid, and students who will be one of those things within a couple of years all read the same document. Four questions, asked in the order below, get every one of them further than the two headline percentages do.

People reach for the word alignment when they discuss these two payments, and the moment they do, they have stopped describing and started judging. The mechanism can be described: one payment moves with the passage of time and the size of a stated basis, and the other moves only with cash returned to investors above what they put in. The two dependencies are different. Any conclusion drawn from having the two payments side by side is a matter for the reader.