Continuation Vehicle vs Traditional Exit

Both routes end with holding 4 leaving Nilgiri Growth Partners Fund II, invented, and cash reaching anyone who wants it. Hold the holding and the price still, and the arithmetic is the same figure twice. The two routes differ on who set that price, and on the decision the investor is asked to make. One route this fund has taken four times; the other it has never taken.

The strangest thing about this comparison comes first. Miss it, and everything that follows turns into a contest. The two arrangements set out here are not two options a reader is being invited to weigh. One of them is a record of things that actually happened inside one invented fund: four separate transactions, four separate quarters, four separate buyers. The other has not happened. The second arrangement could be built on a holding this fund still has, and nothing in this record says it was ever proposed, priced or considered. A comparison between four things that happened and one thing that did not is not a scoreboard, and treating it as one is the single most common way this subject goes wrong.

So the comparison here is narrower and more useful. Take one holding, fix one price, and run both arrangements at that price. Because the number was held still throughout, everything that then differs between them is a real difference in the arrangement rather than a difference in the number. Six differences fall out. Five of them are structural. One of them, the one everybody expects to be the whole story, turns out to be no difference at all.

What is being compared here, and what is being held constant?

A comparison that starts contrasting before it has finished defining is just an argument with tidy columns. Take the two sides one at a time.

Side one is a traditional exitSelling a holding outright to somebody outside the fund.. The fund sells a holding outright to somebody who is not the fund and not the manager. The outside buyer looks at the business, decides what it is worth to them, and pays. Nilgiri Growth Partners Fund II, invented, has done this four times to the end of its Year 9 Quarter 2, and the four transactions produced Rs 4,38,00,00,000 of cash between them. The four transactions are set out below. The count matters later.

In every one of those four, the asset left the fund and somebody who was not the manager decided what it was worth. There is a fifth event in this fund's record that people sometimes count with them, and it is worth separating now: holding 5, Palar Foods Private Limited, invented, was written off in full in Year 6 Q4 and produced nothing. The write-off is an exit in the sense that the position left the portfolio. Nobody paid, so nobody set a price on it. So: five exit events across five holdings, four of them with a third-party price attached.

Side two is a continuation vehicleA new fund set up to buy a holding from an existing fund the same manager runs.. A new fund is raised, that new fund buys the holding from the old one, and the same manager runs both. Existing investors are offered a choice: take the cash the sale produces, or move their exposure across into the new vehicle. How a continuation vehicle is built, who stands where, the conflict inside it and the process built to answer that conflict are all covered separately. One sentence matters for this comparison, and it is the honest one: the manager is on both sides, setting the price it sells at and the price it buys at.

And now the sentence that governs everything below. Nilgiri Growth Partners Fund II has never done this. Not on holding 4, not on any holding, not once. Every continuation vehicle described here is a counterfactualA worked description of something that did not happen, run to see what it would have done.: an arrangement worked out on paper against a holding the fund still has. Nilgiri Growth Partners Fund II never proposed one and never priced one.

Two things are then held still so the comparison can say anything at all. The first is the holding: holding 4 of Nilgiri Growth Partners Fund II, invented, which is Bhavani Speciality Chemicals Private Limited, invented, entered in Year 2 Q4 at Rs 50,00,00,000 with a follow-on of Rs 10,00,00,000 in Year 5 Q2, so Rs 60,00,00,000 of cost, carried at Rs 1,08,00,00,000 at the record date, being 1.80 times its cost. The second is the price: Rs 1,08,00,00,000 under both. Note carefully that fixing the price is an assumption made for teaching, not a finding. Fixing the price is the device that isolates everything else.

Who sets the price in each of the two?

Picture a household selling a flat it has lived in for eleven years. The household has a number in its head, and the number is generous. It always is. Then a buyer walks in who has to find the money from a bank, has looked at four other flats that week, and would rather pay less. The two argue. Whatever they eventually shake hands on is a price, and it is a price precisely because somebody on the other side was trying to pull it down and failed to pull it all the way. Now imagine the same household selling the flat to itself, moving it from one name in the household to another name in the household, and writing a number on the paper. The written number is also a number, and it is not the same kind of object.

The difference between the two numbers has a name. A price reached by two parties with opposing interests is called an arm's lengthA price reached by two parties with opposing interests in it. price, and what makes it one is not the size of the number but the presence of somebody who wanted it to be smaller. How a number came to be the number it is, rather than the size of the number itself, is price formationHow a price came to be the number it is, as distinct from what the number is., and price formation is the axis the whole comparison turns on.

In each of the four transactions in the record above, price formation ran through somebody outside. A buyer already operating in the same industry decided what holding 1 of Nilgiri Growth Partners Fund II, invented, was worth to it and paid Rs 2,03,00,00,000 in Year 7 Q2. Another fund decided what holding 2 was worth and paid Rs 63,00,00,000 in Year 6 Q3. In none of those four did the manager get to write the number down on its own.

The record contains one small, uncomfortable illustration of what that outside pressure does, and it is worth stating carefully because it cuts in an unhelpful direction for anybody hoping for a winner to be picked. At the end of Fund II's Year 7, holding 3 was carried at Rs 1,20,00,00,000, being 2.00 times its Rs 60,00,00,000 of cost. In Year 8 Q1 it left the fund for Rs 1,50,00,00,000, being 2.50 times. The fund's own reported value for that holding moved by half a turn on the day somebody outside actually paid, and nothing about the business changed on that day. What that shows is not that carrying values are wrong. The move shows that a carrying value and a transacted price are two different kinds of number, made from different information on different timetables, and that the second one is made in a room with somebody arguing in it.

Under a continuation vehicle, there is no such room. The manager decides what the selling fund receives and the manager decides what the buying vehicle pays, and those are the same decision written twice. The conflict is structural rather than a matter of anybody's intentions. The arrangement is built with an answer to it, and that answer is covered separately. The plain shape, and the point at which the comparison stops, is this: a third party paying its own money has tested the price, and a manager selling to itself has not, whatever process surrounds it. That is a statement about who was in the room, not a statement about which number would be higher.

What does the investor actually get to decide?

The criteria are numbered as they are in the grid above. Criterion four is taken last on purpose. Criterion four is the one that comes out the same under both, and it lands better once the differences are in place.

Here is the everyday version of criterion two. A tenant rents a flat and the owner sells the building. In one version the sale simply happens, the deposit comes back, and nobody asks the tenant's opinion about anything. In the other version a letter arrives saying the building is moving to a new entity, that the tenant may take the deposit and go or sign a fresh agreement to stay on, and that an answer is needed within three weeks. The second version has given the tenant something. The letter was also not asked for, it costs an evening of reading, and if the tenant does nothing at all something still happens.

Under a traditional exit, an investor of Nilgiri Growth Partners Fund II, invented, decides nothing. The manager sells, the cash reaches the fund, and the fund distributes it in the following quarter. The fund did exactly that four times: Rs 63,00,00,000 went out in Year 6 Q4, Rs 2,03,00,00,000 in Year 7 Q3, Rs 1,50,00,00,000 in Year 8 Q2 and Rs 22,00,00,000 in Year 8 Q4, each one the quarter after the money arrived, each one in cash, and each one announced by a notice signed by Farida Contractor for the manager. Not one of those notices asked an investor a question. A distribution notice is an announcement, not a form.

Under a continuation vehicle the investor is handed an electionThe choice an investor is formally asked to make, by a stated date.: take the cash the sale produces, or rollMoving an investor's exposure into the new vehicle instead of taking cash. the exposure into the new vehicle instead. Both answers are available and both are real. The election is the difference on criterion two, and it is worth reading slowly. A choice sounds unambiguously like a good thing until what it actually is becomes clear. The election is work that lands on an investor who was not expecting it, with a deadline attached, on documents describing a vehicle that was not there last quarter, and with terms that are new rather than the ones already agreed. For the twelve investors of this fund, that work would land on all twelve, from the domestic life insurance company holding 20.0 per cent of the fund down to the charitable trust holding Rs 10,00,00,000. The reading is the same for both of them.

And there is a quieter point underneath. Investors 7 to 12 of this fund hold Rs 85,00,00,000 between them, being 17.0 per cent of the Rs 5,00,00,00,000 of total commitments and 17.3 per cent of the Rs 4,90,00,00,000 committed by investors. Six separate organisations, none of them large in this fund, each of which would have to run the same review on the same documents by the same date as the largest. The cost of an election does not scale down with the size of the holding.

Where are the manager and the business once each one has closed?

Criterion three is the one that is easiest to picture and hardest to argue with. After a sale to a third party the relationship is over. Holding 1 of Nilgiri Growth Partners Fund II, invented, was sold in Year 7 Q2, and from that quarter the fund had no shares in Sahyadri Diagnostics Private Limited, invented, no seat at its board, no list of matters it could block, and no monthly pack landing in an inbox. The business went on existing and the fund stopped being part of it. An asset leaving means exactly that: the manager's work on that company is finished and cannot be revisited.

After a continuation vehicle, nothing about that is true. The business carries on and the manager carries on with it, in the same rooms, with the same people, doing the same work. The paperwork around the work changes: a new vehicle with its own life, its own fee arrangement and its own carried interest arrangement. None of those three is fixed anywhere in this record. A sale ends a relationship and a continuation vehicle renews one, and a renewal on new terms is a thing an investor has to read before it can know what it has agreed to.

One thing is genuinely at stake here, and it is worth naming without leaning on it. The same manager that decided the holding was worth Rs 1,08,00,00,000 would go on managing it, and would be paid for doing so under an arrangement negotiated at the same moment as the price. The price decision and the fee arrangement sitting in one transaction are the whole reason the arrangement carries a process around it. That process, its independent valuation, its opinion on fairness, its committee consent and its cash option, is covered separately.

What has to happen before either one can complete?

Criterion five is about work, and it is where the asymmetry becomes almost physical. In a sale to a third party, the work of testing the price is done by somebody who is paying for the privilege. A buyer sends its own people, hires its own advisers, spends its own weeks, and does all of that before it has any transaction at all, knowing it may walk away with nothing to show for the spending. Nobody does that for fun. The seriousness of the investigation is bought and paid for by the fact that the investigator is about to hand over money.

Under a continuation vehicle the party on the buying side is the party on the selling side, and so nobody is in that position. So the arrangement has to construct, at its own cost, something that stands where that investigation would have stood. The substitute's contents, who signs it and what it is worth are covered separately. One route gets its price tested as a by-product of somebody else's self-interest, and the other has to build a substitute on purpose. Whether the substitute does the same job is exactly the question that separate treatment handles.

And what is the buyer actually buying?

Criterion six sounds like a trick question and is not. In both cases the buyer is buying the same company: Bhavani Speciality Chemicals Private Limited, invented, entered by Nilgiri Growth Partners Fund II, invented, in Year 2 Q4 at Rs 50,00,00,000, topped up with Rs 10,00,00,000 in Year 5 Q2, and carried at Rs 1,08,00,00,000 at the record date. Same revenue, same customers, same people, same building. Nothing about the business is different across the two versions.

Who wanted the company differs. In the four transactions this fund actually completed, somebody outside looked at the business and decided they would rather have it than their money. The outside buyer's preference is information. The preference is not a valuation and not a verdict on quality, but it is a fact about the world that is present in one case and absent in the other. Under a continuation vehicle, the buyer wants the business because the buyer is the party that has been running it. The reason is perfectly coherent and a completely different kind of fact.

Set the two transactions out as documents and the point becomes almost boring. Boring is the best thing that can happen to a point like this.

Which figure is identical under both, and why does that matter?

Criterion four, taken last. Criterion four is where the description stops and the calculation starts. The result is the least dramatic thing here, and that is the point.

The scene, in whole rupees. At the end of Fund II Year 9 Quarter 2, the record date for everything here, Nilgiri Growth Partners Fund II, invented, has drawn Rs 4,80,00,00,000 from its investors, has distributed Rs 4,38,00,00,000 back to them, and still holds five positions carried at Rs 2,82,00,00,000 between them. Total value is therefore Rs 7,20,00,00,000, being the Rs 4,38,00,00,000 already received plus the Rs 2,82,00,00,000 estimated. Against the Rs 4,80,00,00,000 of capital paid in, distributions stand at 0.9125 and residual value at 0.5875, and those two add to exactly 1.5000. The denominator in all three of those is the Rs 4,80,00,00,000 actually paid in, and it is worth saying every single time. The same fund on the same day also reads 1.80 times against the Rs 4,00,00,00,000 the holdings cost and 1.44 times against the Rs 5,00,00,00,000 committed. Three true numbers, three denominators.

Now let holding 4 leave at Rs 1,08,00,00,000, and assume for the moment that every investor takes cash. Distributions become Rs 4,38,00,00,000 plus Rs 1,08,00,00,000, being Rs 5,46,00,00,000. Residual value becomes Rs 2,82,00,00,000 less Rs 1,08,00,00,000, being Rs 1,74,00,00,000, made up of holdings 6, 7, 8 and the remaining 60 per cent of holding 9 at Rs 21,00,00,000 plus Rs 39,00,00,000 plus Rs 81,00,00,000 plus Rs 33,00,00,000. Against the same Rs 4,80,00,00,000 of capital paid in, that is 1.1375 and 0.3625, and 1.1375 plus 0.3625 is 1.5000 exactly. Total value has not moved at all. Nothing was created or destroyed. An estimate turned into cash.

The four decimal places matter. Written to two places the convention used across this record gives 0.91, 1.14, 0.59 and 0.36, and the addition check stops working on the printed figures even though it still works in the arithmetic. Carried unrounded, 1.1375 plus 0.3625 lands on 1.5000 without argument.

Every line in that table is the same under a sale to a third party and under a continuation vehicle, to the paisa, and it is the same because the price was fixed at Rs 1,08,00,00,000 on both sides before the arithmetic started. One further consequence follows: at Rs 5,46,00,00,000 of distributions against Rs 4,80,00,00,000 of capital called, the fund would have returned everything its investors paid in, with Rs 66,00,00,000 over. Money beyond that point is governed by the order of payment written into the fund's own documents, and the order of payment is covered separately.

A reader thinking about an institution rather than a fund wants to see the money land somewhere. Take it down to one investor. Investor 1 of this fund is a domestic life insurance company holding 20.0 per cent of the Rs 5,00,00,00,000 of total commitments. The insurance company has paid in Rs 96,00,00,000 and has received Rs 87,60,00,000 back, and its share of the Rs 2,82,00,00,000 still held is Rs 56,40,00,000. Under either route it receives 20.0 per cent of Rs 1,08,00,00,000, being Rs 21,60,00,000. Its distributions become Rs 1,09,20,00,000 and its share of what remains falls to Rs 34,80,00,000, and Rs 1,09,20,00,000 plus Rs 34,80,00,000 is Rs 1,44,00,00,000, and its total value was already there: 1.50 times its Rs 96,00,00,000 paid in. Same rupee, same day, either route.

So where is the difference, if the arithmetic is the same?

In three places, and none of them is a rupee figure.

The first is the election. Under a continuation vehicle, investor 1 may decline the Rs 21,60,00,000 and roll instead, in which case it receives nothing now and holds an interest in a new vehicle. A sale to a third party offers no such option, and the cash simply comes. A route that can produce two different outcomes for two investors who started identical is a different kind of object from a route that can only produce one.

The second is where the manager is afterwards, already settled under criterion three. The third is the one the arithmetic is structurally incapable of showing, and it is the reason the table above is honest rather than reassuring: under a sale, the Rs 1,08,00,00,000 would have been argued about by somebody whose own money was leaving. Under a continuation vehicle it would not have been. Two numbers can be identical and still not be the same kind of number, and the arithmetic cannot show which kind is in front of the reader.

What if the fund simply extends its term instead?

There is a third thing a fund can do when it runs out of time with an asset still on the books, and it belongs here because readers reach for it the moment the first two are laid out. Nilgiri Growth Partners Fund II, invented, has a term of ten years from its final close, ending at the end of its Year 10. At the record date, the end of Year 9 Quarter 2, six quarters of that remain and five holdings are unsold. Its own documents allow two term extensionAdding a year to the fund's life under the terms its own documents already set. steps of one year each: the first at the manager's election with the prior written consent of the investor advisory committee, the second requiring the consent of investors holding more than half of commitments by value. Neither has been taken.

Notice what an extension does and does not do. An extension does not move the asset. An extension does not produce a price, from anybody, of any kind. Cumulative distributions stay at Rs 4,38,00,00,000, distributions against the Rs 4,80,00,00,000 of capital paid in stay at 0.9125, residual value stays at 0.5875, and the total stays at 1.5000. An extension moves the deadline rather than the holding, and an investor receives exactly nothing from one. That is not a criticism of it. Time is a real thing to buy, and the fund's Rs 20,00,00,000 of unfunded commitment exists precisely to pay the fee and expenses of the quarters that remain. An extension is simply a different kind of answer to the same problem, and that is why it sits beside the comparison rather than inside it.

Put the three side by side on the only two measures that separate them and the extension stops looking like a middle option and starts looking like what it is: a different axis altogether.

Which one number is missing from this comparison?

How does somebody on the receiving end of the letter actually read this?

Picture an analyst at investor 5 of Nilgiri Growth Partners Fund II, invented, a fund of funds that committed Rs 50,00,00,000, being 10.0 per cent of the Rs 5,00,00,00,000 of total commitments. A notice arrives describing a proposed continuation vehicle for holding 4 and asking for an answer by a date. The analyst has perhaps two weeks and a queue of other work. The work consists of four separations, and they are worth setting out because they are the same four every time this shape of letter lands.

The first separation is the one this whole comparison exists for. The arithmetic question and the price question are different questions, and the arithmetic question takes about four minutes. Distributions of Rs 4,38,00,00,000 plus Rs 1,08,00,00,000 over Rs 4,80,00,00,000 of capital paid in gives 1.1375; residual value of Rs 2,82,00,00,000 less Rs 1,08,00,00,000 over the same denominator gives 0.3625; they add to 1.5000, and the fund already stood there. Done. Every minute after those four belongs to the price question, and a reader who spends the fortnight re-checking the arithmetic has spent it on the part that was never in doubt.

The second separation is between a valuation and a price. The carrying value of Rs 1,08,00,00,000 was struck by machinery inside the arrangement: an independent valuation agent signing annually, the manager marking between those, and an administrator striking the net asset value. The machinery is real work by people with professional duties, and it is still not the same event as somebody handing over money. The fund's own record shows the two are different objects. Holding 3 was carried at 2.00 times its cost at the Year 7 year end and left at 2.50 times in Year 8 Q1.

The third separation is between the old arrangement and the new one. A roll is not a continuation of anything the investor already agreed to. A roll is a new vehicle with its own length, its own fee arrangement and its own carried interest arrangement, and the fact that the underlying business is familiar is precisely what makes it easy to skim. The fourth is the deadline and the default: what happens to an investor who does nothing at all is written somewhere in the notice, and it is not always the answer the reader would have picked. Alongside all of that, the fund's own governance has a committee of investor representatives, chaired for investor 1 by Meera Sathe, whose consent is required on conflicts of this kind, and what that committee does and cannot do is set out separately.