How to trace a Capital Call and Distribution

A capital call runs from the manager's notice, through each party's pro rata share, into the fund's bank account, and out to whoever is being paid. A distribution runs the same road backwards. Both are worked on one unchanging share of one base, and both finish on the same document. The last stop of each trace is the investor's own capital account statement.

Most people meet a private fund's arithmetic in the wrong order. Somebody is handed a notice, or a quarterly statement, or a spreadsheet with thirteen rows in it, and starts at the top and reads downwards until something stops making sense. Reading a notice from the top is not tracing it. Tracing means picking one amount, following it across every record that touches it, and being able to say at each stop what should reconcile there and against what. This walkthrough sets out the order in which to do that, in both directions, on one invented fund and one invented investor.

The distinction matters more than it sounds. A commitment, the running account a fund keeps for each investor, the way a distribution gets split into its contracted tiers and the reason any of that machinery exists are all separate subjects, taught separately, and this walkthrough uses every one of them without opening it up. The route supplied instead has ten stops. Each stop names the document or record that carries the money at that point, and each stop names the check a reader should be able to run before moving on. A trace with no checks in it is a story. A trace with checks in it is evidence.

Why trace one rupee rather than read the whole notice?

Because a whole notice hides its own errors and a single rupee cannot. Consider a residents' association in a building of thirteen flats, collecting for a lift repair. The treasurer writes to everybody: the repair costs a certain amount, and each flat pays its share according to the size of the flat. Nobody in the building reads that letter and checks it. Each resident looks at the line with their own flat number on it, sees a figure that seems about right, and pays it.

Now imagine one resident who does something different. She takes her own figure, divides it by her own flat's share of the building, and gets a percentage. Then she takes the total repair cost, works out what that same percentage of it would be, and compares. If those two agree, the treasurer took the share on the whole building. If they do not, something is wrong, and the most likely something is that the treasurer left a flat out of the total. Perhaps the association's own caretaker flat, a flat nobody thinks of as a resident. One resident, one figure, two minutes, and the error is either found or ruled out.

The resident's two-minute check is the whole method, and the private fund version is identical in shape and larger in consequence. The worked case throughout is Nilgiri Growth Partners Fund II, an invented fund, and one of its investors, a domestic life insurance company referred to here as investor 1.

What are the two traces, and why are they the same arithmetic?

There are exactly two directions money moves between a private fund and the people who put money into it, and each of them has five stops. Money goes out from an investor when a capital callA notice requiring investors to send committed money. is issued, and money comes back to that investor when the fund has received cash it can pay away. Nothing else happens. There is no third direction and no fourth.

One number makes both traces workable, and both traces run on it. Every party in this invented fund is drawn and paid pro rataIn proportion to each party's share of the total., meaning in proportion to its share of the total that everybody promised. A party's share of the total does not move over the life of the fund. The share is fixed the day the fund closes, and it is the same number on the first notice and on the last distribution. Once one party's share of the total everybody promised is in hand, both traces become the same multiplication, run once in each direction.

The map below comes before anything else. The left column is the money going out and reads downwards. The right column is the money coming back and reads upwards. Reading upwards is the only way to see that it is genuinely the return journey rather than a second outbound one. Each stop names the record that carries the money there, and the check to run before moving to the next stop.

Where does a capital call start, and what does the notice actually say?

A capital call starts with a document, and the document is short. Nilgiri Alternatives Advisors Private Limited, invented, is the manager of this fund, and it is the manager that issues the notice. The notice is not a request and it is not an invoice. The notice is the exercise of a right the fund already has under a contract every investor signed before the fund closed, and it converts a promise that has been sitting on paper into money that has to move.

The fund's own records call the whole episode a drawdownThe event of calling and receiving that money., meaning the notice, the thirteen payments it produced and the thing those payments bought, taken together as one numbered event rather than as a letter alone. The trace here uses the second drawdown of Nilgiri Growth Partners Fund II. Its notice was issued in the fund's Year 1 Q3 and called Rs 55,00,00,000. A date here is always written the way this invented fund's own documents write it, counted from its final close, so Year 1 Q3 means the third quarter of the first year of the fund's life and nothing else. The fund has seventeen such calls across its whole life, and drawdown 2 is the second of the seventeen.

Four things sit on the face of a notice like this one, and the reading order matters because each one is the input to the next. First, the total being called, here Rs 55,00,00,000. Second, the purpose, here the fund's entry into Sahyadri Diagnostics Private Limited, an invented company and this fund's first holding. Third, the schedule of amounts, one line for each party, holding a given investor's own figure. Fourth, where and by when to send it. The period is this invented fund's own contracted term and not a rule anybody else is bound by, so it is described and never quantified.

The single most useful habit at this stop is to read the total and the purpose together before looking at any individual line at all. A notice that calls Rs 55,00,00,000 and says it is for an investment of Rs 55,00,00,000 reconciles. A notice that calls Rs 45,60,00,000 and says it is for one investment of Rs 35,00,00,000 has something else in it, and on this fund that something is the year's management fee and operating expenses riding alongside the investment in the same call. Drawdown 7 of this fund is exactly that, and reading its purpose line first is the difference between understanding the notice and being surprised by it.

So drawdown 7 of this fund is one letter carrying two purposes, and reading its purpose line before its total is what stops a reader treating Rs 45,60,00,000 as the price of a company. Nine of this fund's seventeen calls include fee and expenses at all, and three of those nine carry it alongside something the fund was buying. Those three are where the reading goes wrong. The trace behaves identically on every one of them: the amount is still called on one rate, still lands in one account, and still splits between what it bought and what it paid for.

How is one investor's share of a call worked out?

By one division and one multiplication, and the whole difficulty is in the division. Divide the amount being called by the baseThe total the share is taken on, which here is total commitments., meaning the total everybody promised, and the result is a rate. The rate multiplied by any one party's own promise gives that party's line on the notice.

The base of this invented fund is total commitments of Rs 5,00,00,00,000. The base is made of two parts and both belong in it. Twelve investors promised Rs 4,90,00,00,000 between them, and the manager promised Rs 10,00,00,000 of its own money, being 2.0 per cent of the total. The manager is the thirteenth party. Its money is funded in cash exactly as everybody else's is, and it is called on exactly the same rule. The manager's money is not a token, not a fee arrangement and not exempt.

So the rate on drawdown 2 is Rs 55,00,00,000 divided by Rs 5,00,00,00,000, giving 0.11. Not 0.1099 and not 0.1101: fifty five over five hundred is eleven over a hundred exactly, so every one of the thirteen parties is called 11.0 per cent of its own commitment and not a single line on the notice needs rounding. That is unusual and it is why this particular call is the one to learn on. Investor 1 promised Rs 1,00,00,00,000, so investor 1 is called Rs 11,00,00,000. The manager promised Rs 10,00,00,000, so the manager is called Rs 1,10,00,000. Same rule, same rate, different promises.

The thirteen lines of drawdown 2, set out so the sum can be run

Here is the whole schedule of that one notice, worked to the rupee. The schedule is set out in full for one reason only. A trace that cannot be checked is not a trace. One line read first, then the column added, shows whether the total lands where it should. The interactive control that splits a notice across these same parties is covered separately; this is the pencil-and-paper version, and pencil and paper is what a reader usually has to hand.

Every figure in that right hand column is one multiplication, and every one of them lands on a whole number of rupees. The twelve investors come to Rs 53,90,00,000, being 11.0 per cent of their own Rs 4,90,00,00,000, and the manager's Rs 1,10,00,000 makes up the difference to Rs 55,00,00,000 exactly. Nothing is left over and nothing has to be absorbed by anybody.

What happens to the money once it arrives?

The money stops being thirteen payments and becomes one balance, and then it leaves again. The fund's bank account is the only stop where the money is briefly in one place with one number on it, and it is the shortest part of the trace in time and the part where the most useful joins live.

Investor 1 pays Rs 11,00,00,000 out of its own bank account. Twelve other payments arrive over the same window, from the twelve other parties, and every one of them is the same rate applied to a different promise. The record at this stop is the fund's bank statement, sitting alongside the cash book kept by Kolar Fund Services Private Limited, an invented administrator. The check here is the plainest of the ten: the thirteen credits on the bank statement should equal the thirteen lines on the notice, name by name and rupee by rupee. If a line is short, somebody has not paid. If the total is right but a line is wrong, two parties have been swapped. If the total is more than the notice, something arrived that was not called for.

Then it leaves. Drawdown 2 was called to buy Sahyadri Diagnostics Private Limited, this fund's holding 1, and the entry cost Rs 55,00,00,000. The record here is the fund's own payment instruction and the transfer documents for the shares. The check is that Rs 55,00,00,000 went out against Rs 55,00,00,000 that came in, with nothing sitting in the fund's account afterwards that this notice paid for.

The join most readers walk straight past

Money leaving a fund to buy a company can land in two completely different places, and which one it landed in is a fact about the transaction rather than about the fund. The money can go to the people who already held the shares, in which case they are selling and the company itself receives nothing at all. Or it can go to the company, issuing new shares in exchange, in which case no existing shareholder receives anything.

Nilgiri Growth Partners Fund II did both, and comparing the two calls is the cleanest way to see why the question matters. Holding 1's Rs 55,00,00,000 bought shares the founding shareholders already held, so not one rupee of investor 1's Rs 11,00,00,000 reached Sahyadri Diagnostics. The money reached the founders. Holding 4's Rs 50,00,00,000, called on drawdown 6 at the fund's Year 2 Q4, was subscribed for new shares in Bhavani Speciality Chemicals Private Limited, invented, so all of it reached the company and not one rupee reached a shareholder. Investor 1's line on that second notice was Rs 10,00,00,000, being the same 20.0 per cent of the amount called.

A reader who does not have this distinction cannot read either transaction correctly, and the record that settles it is not the capital call notice at all. The record that settles it is the transfer document against the record of new shares issued, and a company's own filings on such things sit with the Ministry of Corporate Affairs at mca.gov.in.

What changes on the investor's own record once the call is met?

Two lines move and one does not, and getting that straight is the last stop of the outbound trace. The record is the capital accountThe record of what one investor has put in and taken out. statement that investor 1 receives each quarter, one of six things this invented fund's documents contract to send its investors.

Before drawdown 2, investor 1 had already met drawdown 1, the Rs 13,10,00,000 called at the fund's Year 1 Q1 for the fund's organisational expenses and its first year of fee and expenses. Investor 1's share of total commitments is 20.0 per cent, and every call is drawn on that same share, so its line on that first notice was Rs 2,62,00,000. So investor 1 arrives at drawdown 2 with Rs 2,62,00,000 paid in and Rs 97,38,00,000 of its promise still to be called.

After the call is met, its paid in figure is Rs 2,62,00,000 plus Rs 11,00,00,000, being Rs 13,62,00,000. Its unfunded commitmentThe part of a commitment not yet called., meaning the part of its promise nobody has asked for yet, falls from Rs 97,38,00,000 to Rs 86,38,00,000. Both lines moved by exactly Rs 11,00,00,000 and they moved in opposite directions. The commitment itself stayed at Rs 1,00,00,00,000 and did not move at all. That is the signature of a capital call on a statement and it is the thing to look for: a promise does not shrink when it is called, it simply has less of itself left uncalled.

There is a trap sitting right next to this figure and it has caught readers repeatedly, so it is worth stating flatly. At the record date this fund has an unfunded commitment of Rs 20,00,00,000 at the fund level, being Rs 5,00,00,00,000 promised less Rs 4,80,00,00,000 drawn. Investor 1's own unfunded commitment at the same date is Rs 4,00,00,000, being its Rs 1,00,00,00,000 promised less the Rs 96,00,00,000 it has paid in. The two figures are measured at two different levels, so they are five times apart and both are correct. Name the level in the same sentence as the number, every single time.

What is the check that catches it?

Add the investor lines and compare them with the notice. On a correctly struck schedule they must come to less than the notice total, and the difference must be exactly the manager's own share. On drawdown 2 the twelve come to Rs 53,90,00,000 and the notice is Rs 55,00,00,000, so the difference is Rs 1,10,00,000. The difference is precisely 11.0 per cent of the manager's Rs 10,00,00,000 commitment. Twelve investor calls that sum to the whole notice are the error announcing itself out loud, and it takes about fifteen seconds to run.

The two failures look different, so run the check in the other direction as a second test. Take any one party's line, divide it by that party's own commitment, and see whether the answer is the same for two different parties. Investor 1 at Rs 11,00,00,000 on Rs 1,00,00,00,000 is 11.0 per cent. Investor 12 at Rs 55,00,000 on Rs 5,00,00,000 is 11.0 per cent. The manager at Rs 1,10,00,000 on Rs 10,00,00,000 is 11.0 per cent. If those three disagree, the schedule is not pro rata at all, and something other than a base error is going on.

Which way does the same trace run for a distribution?

Backwards along the same road, with one extra stop in the middle that has no equivalent on the way out. Money coming back starts at a buyer rather than at a manager's decision, and it has to satisfy a condition before any of it moves. Money going out satisfies no such condition.

The trace uses distribution 2 of this same invented fund. In the fund's Year 7 Q2, holding 1 was sold in full to a buyer and Rs 2,03,00,00,000 of cash reached the fund. The record at that first stop is the completion statement for the sale sitting against the fund's own bank statement, and the check is that the cash received equals what the sale documents say was payable. The check is call stop 3 with the direction reversed. A receipt feels self evidently fine in a way a payment does not, so this is the check people skip.

Note the gap between the sale and the payment out. The cash arrived in Year 7 Q2 and the distribution was made in Year 7 Q3, the quarter after. The one-quarter gap is this invented fund's own practice on every one of its four distributions, and it is not a rule, a requirement or a market convention. The gap is simply what this fund did. A reader looking only at the paperwork that came with the money would put the sale in the wrong quarter, so a trace must record the gap.

What has to be true before a distribution can be made at all?

The contracted order has to be applied to the money first. Every rupee arriving at a private fund is subject to an order of payment written into the fund's own documents, and that order decides who gets what before anybody gets anything. This is an Indian vehicle settled as a trust rather than a partnership, so on this fund the record is the trust deed and the contribution agreement. The general partner's role that the global vocabulary of this subject refers to is discharged here 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 trace stops at that order and hands the money to it. Working the order out is no part of the trace. What the order contains, how many steps it has, what each step is called, what each pays and in what sequence are a separate subject, taught in full separately. A procedure establishes where to stop and what to hand on, and this is the place.

The trace needs one fact from that order, and the fact changes the words on the statement rather than the arithmetic. At this fund's record date, every rupee of the Rs 4,38,00,00,000 that has ever gone back to its investors has been return of capital, and the fund is still Rs 42,00,00,000 short of returning everything it has called. So the character of the money on investor 1's statement is capital coming back, not profit. Whether that character matters for anything beyond the statement is a question for the tax authorities. Where the conduct of this kind of vehicle is set at all, it is set by the Securities and Exchange Board of India at sebi.gov.in.

The last three stops, and what the notice looks like coming back

Once the order has been applied, a distribution noticeThe document telling an investor money is coming back. goes out, and it is the mirror image of the capital call notice in shape. The distribution notice carries a total, a source, a schedule of amounts with one line per party, and a date. Investor 1's line on the notice for distribution 2 is Rs 40,60,00,000, being 20.0 per cent of Rs 2,03,00,00,000. The 20.0 per cent is investor 1's share of the same total commitments of Rs 5,00,00,00,000 that produced its 11.0 per cent call earlier, and the two percentages are not in conflict: 11.0 per cent was a rate on a notice, and 20.0 per cent is a share of a fund.

Then the money arrives, and the record moves to investor 1's own bank statement and the remittance advice travelling with the payment. The check is the same one as call stop 3 in reverse: the cash in against the line addressed to it. And then the statement is written up. Money coming back does not restore anybody's right to call it again, so distributions received rises by Rs 40,60,00,000 and unfunded commitment does not move at all. The two traces stop being symmetrical at that line, and the difference is worth holding: a call moves two lines, a distribution moves one.

Every line in that right hand column is the same multiplication as every line in the capital call schedule, run on the same share. There was no distribution at all in Years 1 to 5 of this fund and none in Year 9 to the record date, a fact about this one invented fund's own record and nothing more.

Where does each trace end, and what is left behind?

Both end on one sheet of one document, and that is the single most practical thing in this walkthrough. Investor 1 receives a capital account statement each quarter, and every line on it was written by one of the two traces or by neither. Reading that statement as the output of two traces turns it from a list of figures into something that can be interrogated.

Six lines matter. The commitment of Rs 1,00,00,00,000 was fixed the day the fund closed and neither trace touches it. Capital contributed of Rs 96,00,00,000 is the outbound trace, run seventeen times, once for each of this fund's drawdowns. Unfunded commitment of Rs 4,00,00,000 is the same trace seen from the other side, and it is the mirror of the line above it rather than an independent fact. Distributions received of Rs 87,60,00,000 is the inbound trace, run four times.

The fifth line is the one no trace reaches, and it is the line a reader should slow down at: investor 1's share of the fund's residual value, Rs 56,40,00,000, being 20.0 per cent of the Rs 2,82,00,00,000 the fund still holds. No money has moved to produce that figure. Nobody has paid it. The figure is what five companies are carried at, and none of them has been sold to anybody. The trace method has nothing to say about it, and saying so is the honest answer rather than a gap in the method. The sixth line, total value of Rs 1,44,00,00,000, is simply the fourth and the fifth added together, and it puts investor 1 of Nilgiri Growth Partners Fund II, invented, at 1.50 times its Rs 96,00,00,000 paid in as at the end of that fund's Year 9 Q2.

How does somebody actually use this in a working week?

The people who run this trace for a living almost never run all ten stops. Something did not tie, so they run two or three stops on one figure, and knowing the whole route is what lets them jump straight to the stop where the answer will be. Here is what that looks like for four different readers, and none of them is doing anything the ten stops above do not already cover.

Somebody in an investing institution's own team receives a capital account statement each quarter and has to sign off that it is right before it goes into the institution's own books. Such a reviewer does not recompute the fund. The reviewer takes the paid in figure, subtracts last quarter's paid in figure, and checks that the difference equals the notices paid in between. Then they take the distributions figure and do the same against the payments received. Both traces terminate on the lines being compared. Call stop 5 and distribution stop 5, run together as a two line check, therefore catch almost everything. When it does not tie, the route tells them where to look: a difference that is exactly one party's share points at a base error, and a difference that matches a quarter's fee points at a call whose purpose line they read too quickly.

Somebody in a fund administrator's team, at a firm doing the work Kolar Fund Services Private Limited, invented, does for this fund, runs the trace in the other direction and much more often. An administrator produces the schedule rather than checking it, so the working check is the one at call stop 2: strike the rate, apply it to every commitment, add the lines, and confirm the total equals the notice with nothing left over and nothing absorbed. On a call like drawdown 2 that closes to the rupee. On a call whose ratio is not exact, it will not, and then the fund's own documents decide who carries the odd rupees.

Somebody sitting on an investment committee, reading a fund report once a quarter with forty minutes to spare, uses the trace differently again. A committee member is not reconciling anything. The question being asked is which lines on the report a trace could reach and which it could not. Four of investor 1's six lines can be walked back to a bank statement. No money has moved to create the residual value line, so that line cannot be walked back at any price, by anybody. A reader who can sort a statement into those two piles in thirty seconds is asking better questions for the remaining thirty nine minutes.

And the fourth reader is the one who will never see a private fund at all. The residents' association and the lift make the same point. The method transfers exactly: take one figure, find the rate it implies, name the base that rate was taken on, and check that everybody's figures add back to the total with the right amount left over for whoever is not on the list of residents. The method works on a shared repair bill, on a wedding cost split between two households, on a small business where three partners put in different amounts, and on a Rs 5,00,00,00,000 fund. The scale changes and the arithmetic does not. A rate whose denominator cannot be named is a rate that has not been checked.

What does this trace not establish?

Four things, and each of them is somebody else's subject rather than a gap. A trace follows money and never judges it, so the method does not establish whether the call was a good idea. The contracted order and its steps are taught separately, and the trace hands the money to that order and stops, so the trace does not establish how a distribution gets split. A trace reaches only amounts somebody actually paid, so it does not establish what any holding is worth. And a tax position is a question for the tax authorities rather than for a trace.

Private funds differ from one another on every period and practice a trace passes through. The quarters, the seventeen calls and the one-quarter gap between a sale and a payment are Nilgiri Growth Partners Fund II's own contracted terms, and another fund's documents will say something else on each of them. Where the registration, reporting and conduct of this kind of Indian vehicle are set at all, they are set by the Securities and Exchange Board of India at sebi.gov.in, and that text changes.