Venture Capital: Funding Risk That Banks Will Not Take

Venture capital buys minority equity in unlisted companies that have no security to pledge and no contracted cash to lend against, so a bank cannot fund them. The money arrives in rounds, each one setting a price per share and issuing new shares. A venture fund's own investors commit capital first and pay it only when the fund calls it.

A lender is repaid out of contracted cash and is secured on something it can take if the repayment does not come. A company with neither of those can offer a funder only a share of whatever it becomes. So the money is priced as equity instead of as debt, staged across separate occasions so that each tranche is priced against what the last one bought, and turned back into cash only when somebody else buys the shares. Every number in this subject is an arithmetic consequence of that one substitution. Take the substitution and the arithmetic follows without effort; miss it and nothing else in this subject will sit straight.

What does a bank actually lend against, and why does an early company have neither of those things?

Here is the everyday version, at a scale that can be felt. A jeweller walks into a branch and asks for money against stock sitting in a safe. There is a thing in a room, it can be counted, and if the money is not repaid the bank can take it and sell it. Now a schoolteacher walks into the same branch with a plan to build a study app. There is a laptop, a rented desk and an idea. Nothing on that list can be taken and sold to cover a shortfall, and nothing on it has contracted to pay anybody anything on any date.

A lender needs two things at once, and an early company usually has neither: something to take if the money is not repaid, and contracted cash to be repaid out of. Notice that neither of those is a judgement about quality. The branch is not saying the study app is a poor idea. The branch is saying that a loan is a specific instrument with a specific shape, and that shape needs a security and a repayment source. A lender that writes the loan anyway is not being brave. The lender is writing an instrument whose downside it cannot describe. Prudential rules exist for that reason, and credit assessment is a subject in its own right.

The difference runs wider than security alone. Read the four rows below together rather than one at a time. A loan is secured, repaid out of contracted cash, due on a written date, and silent about who runs the business. Venture capitalMoney paid into an unlisted company for newly issued shares, with no repayment date. reverses every one of those four at the same time, and no single row is the whole of the difference.

What replaces the security when there is nothing at all to pledge?

If a funder cannot be secured and cannot be repaid on a date, the only thing left to buy is a share of the company itself. A share of the company is what venture capital buys: newly issued equity in an unlisted companyA company whose shares are not quoted on any exchange, so there is no daily price for them., usually a minority stakeA holding too small to decide anything on its own., with no repayment date and no coupon. The funder stops being a creditor with a claim and becomes a holder with a percentage. Nothing is due. There is no schedule to breach and nothing to demand. The money has gone into the company and what came back was a line on a share register.

The swap from creditor to holder changes what every question after it is about. A lender asks whether it will be repaid, and can answer from the contract. Nobody promised a holder of shares repayment, so a holder cannot ask that question at all and asks a different one instead: what will this holding be worth to somebody else, and when might somebody else want it. Buying control of a company rather than a minority of it is a different transaction with different levers, and it is covered separately. A minority holder has a say and cannot decide.

Here is the test that does the work, and it is worth running on any business described in these terms. The two questions in the drawing below are asked in that order. Both questions are about the shape of the business, not about how promising it is. A company can be excellent and fail both, and a dull company with a warehouse and a signed supply contract passes both without effort.

Why does the money arrive in rounds instead of all at once?

Consider lending a nephew money to open a tea stall. Nobody hands over the whole of what the stall might eventually need on the first morning. The cart is paid for, a season is watched, and only then does the second location come up. Not out of stinginess, but because after that season something is known that was not known before, and whatever goes in next can be priced against what was learned.

Staging the money is a pricing device before it is a discipline device. Each roundOne occasion on which a company issues new shares at an agreed price. is priced against what the previous round bought, so the funder is never asked to price ten years of an unlisted company at one sitting. The funder is asked to price the next eighteen months, then look again. The company gets the same benefit in reverse: it sells a slice for what it is agreed to be worth now, rather than a much bigger slice for what it was agreed to be worth when it was three people and a laptop.

The illustration below is one invented company, Aravalli Learning Systems Private Limited, across three of its own years. Read only two things off it: how much money came in at each round, and what the post-money figure was afterwards. Who held what after each round is set out separately. Notice the shape of the money coming in. The money is not one payment but three, spaced across Aravalli's own clock, each one larger than the last, and each one arriving after the last had been spent and something had been found out.

How Venture Capital Rounds Work: where do the price and the new shares come from?

A round has exactly four moving parts and only one of them is negotiated. The pre-moneyThe value agreed for the company before the new money arrives, in rupees. is agreed between the company and the buyers, usually in a term sheet before any register moves. Feld and Mendelson, in Venture Deals, 2011, treat a term sheet as a document to be read structurally rather than argued line by line, and at that level a term sheet answers three questions: what is being priced, who is buying, and what changes hands. Everything after the pre-money is arithmetic that anybody with the share register can reproduce.

The price per share is the pre-money divided by the shares already in issue, and the new shares are the raise divided by that price. Two divisions and nothing more. The trap sits in the denominator of the first division, and it is worth slowing down for: the divisor is the shares that exist before the round, not the shares that will exist after it. Using the later number gives a lower price. The lower price then fails to reproduce the post-money figure everybody has agreed to, and an afternoon goes into looking for the mistake somewhere else.

What did one round actually do to one company's share register?

Mechanism means little until it runs on a company. Watch it run on one invented company. Nilgiri Venture Fund I is an invented venture capital fund with Rs 1,50,00,00,000 of commitments and eighteen investments, managed by Nilgiri Alternatives Advisors Private Limited, also invented. One of those eighteen is Aravalli Learning Systems Private Limited, whose seed round in its own Year 1 is worked below to the share.

Before the round, Aravalli had 80,00,000 shares in issue: 70,00,000 held between its two founders and 10,00,000 sitting in an employee option poolShares set aside for employees who may not have been hired yet.. The pre-money agreed was Rs 24,00,00,000. So the price per shareThe pre-money divided by the number of shares already in issue. is Rs 24,00,00,000 divided by 80,00,000, being Rs 30.00. The raise of Rs 6,00,00,000 at Rs 30.00 buys 20,00,000 new shares. Nilgiri Venture Fund I took 15,00,000 of them for Rs 4,50,00,000 and an angel syndicate of six individuals took the other 5,00,000 for Rs 1,50,00,000.

The round sold 20,00,000 of the 1,00,00,000 shares that existed afterwards, being 20.0 per cent, for 20.0 per cent of the post-money. The post-money is worth checking two ways, because a figure that reconciles in both directions can be defended. Route one: the pre-money of Rs 24,00,00,000 plus the raise of Rs 6,00,00,000 is Rs 30,00,00,000. Route two: 1,00,00,000 shares multiplied by Rs 30.00 is Rs 30,00,00,000. Same number. The agreement between the two routes is not luck. The two divisions were built to produce it, and two routes that disagree mean the wrong denominator has been used somewhere.

One more figure, and this is the one most often mis-stated. The Rs 4,50,00,000 that Nilgiri Venture Fund I put into Aravalli is 3.0 per cent of that fund's Rs 1,50,00,00,000 of commitments. Commitments are the denominator, and the denominator has to be named every single time the figure is quoted: 3.0 per cent of Rs 1,50,00,00,000 of commitments is not 3.0 per cent of deployed capital. Commitments and deployed capital are different denominators that produce different percentages from the same rupee amount, and a percentage travelling without its denominator is how an account ends up contradicting itself two paragraphs later.

Why is a post-money figure an arithmetic result rather than an appraisal of the company?

A post-money figure looks like a valuation, gets reported like a valuation, and is not one. Nobody appraised Aravalli to produce Rs 30,00,00,000. Two numbers were agreed, a pre-money and a raise, and the third fell out of adding them. The post-moneyThe pre-money plus the money raised, which is also the share count afterwards times the price. is the pre-money with the raise added. No judgement of any kind sits between the two ends. The same figure arrives by multiplying the share count afterwards by the price one buyer paid, the same journey by the other road. The price was itself a division of the agreed pre-money, so the figure is still not an appraisal.

The distinction matters because of what people then do with the figure. A post-money figure gets treated as a market price for the whole company, and it is nothing of the sort: it is the price at which a slice was sold on one day to a small number of buyers who negotiated it, applied arithmetically to every other share on the register including shares nobody offered to buy. The pool shares in Aravalli are counted at Rs 30.00 apiece in that Rs 30,00,00,000, and nobody offered a rupee for them. A share count multiplied by a round price has to be named for exactly what it is, a share count multiplied by a round price.

Commitment vs Capital Call: what is the difference, and why does it matter?

Here is the everyday version, and it is exactly the same shape. A shopkeeper signs a lease and promises the fitter Rs 3,00,000 for the shelving when the lease is signed. The promise is real, it is written, and the fitter can rely on it. But the Rs 3,00,000 is still in the shopkeeper's account, not his, and it moves on the day he sends the bill. Anybody looking at that bank statement in the morning would see the money and would be wrong to conclude that nothing is spoken for. Anybody looking at the promise would see an obligation and would be wrong to conclude the fitter has been paid.

A commitmentA signed promise to pay money to a fund when it asks for it. is a signed promise to pay; a capital callThe notice a fund issues that turns part of a commitment into cash. is the demand that turns part of that promise into cash. Nilgiri Venture Fund I has Rs 1,50,00,00,000 of commitments. The Rs 1,50,00,00,000 is a stack of signed contribution agreements. A stack of promises is not a balance, and no commitment figure says how much cash the fund holds on any given morning. When the Aravalli round was closing, the fund issued a call, the cash arrived from its own investors, and the fund paid Rs 4,50,00,000 to Aravalli for 15,00,000 shares. Before that call went out, nothing had moved and nothing needed to.

The call notice itself is a short and unglamorous document, and looking at one is the fastest cure for the confusion. The notice has one job: to name an amount, a purpose and a date. The schedule of calls across a whole fund's life, and the shape that schedule traces, are covered separately. The distinction that matters is between the promise and the cash, worked on the one call that funded the Rs 4,50,00,000 subscription.

Where does that same promise-then-call relationship run twice over?

Here is the part that catches careful readers, and it catches them because they have understood half of it properly. In this transaction the commitment and the call do not happen once. Both happen twice, at two different levels, on the same day, and the fund sits on a different side of the arrangement each time. At level one the fund's own investors have promised money to the fund and pay when the fund calls; at level two the fund has promised money to a company at a round and pays when the round closes.

Follow the cash for Aravalli's seed round in its Year 1 and the two levels come apart cleanly. The fund cannot pay Aravalli out of nothing, so it issues a call to its own investors for the amount it needs. The called Rs 4,50,00,000 arrives. The fund then pays Aravalli Rs 4,50,00,000 at the closing, and 15,00,000 shares go onto Aravalli's register in the fund's name. Two promises, two moments of payment, one afternoon. A reader who sees only the second level cannot say where the cash came from; a reader who sees only the first cannot say what it bought.

What does the fund hold after the round, and when does it see any cash?

After the closing, Nilgiri Venture Fund I holds 15,00,000 of the 1,00,00,000 Aravalli Learning Systems Private Limited shares then in issue, or 15.0 per cent. The entire position is a share count and a percentage. There is no coupon, no dividend contracted, no repayment date and no cash of any kind until somebody buys those shares from the fund. The holding can be written up or written down in the fund's own reporting, and none of that writing produces a rupee. Cash arrives on one occasion only, a sale, and what a sale actually produces is covered separately.

An unlisted holding behaves differently from a listed one for exactly that reason, and the point is worth stating plainly rather than treating as an inconvenience. There is no daily price because there is no daily buyer. There is no partial exit because a share register is not an order book: a quarter of a private position cannot be sold on a Tuesday because there is nobody standing there to buy a quarter of it. Ang, in Asset Management, 2014, argues that illiquidity is a characteristic to be reckoned with in its own right rather than a nuisance attached to an otherwise ordinary holding, and that framing is his. The practical consequence is narrow and concrete: the only decision the fund fully controls is the one it makes on the way in.

Why does a venture portfolio carry so many positions rather than a few large ones?

Think of ten shops in one shopping mall. Each is a separate business with a separate owner. All ten share one driver, so if the mall loses its anchor tenant every one of them has the same bad year. A person who runs ten shops in ten different towns has a different problem and a different protection. The count of positions is one thing; what they have in common is another; and a portfolio built as many small holdings is answering the first question, not the second.

The reason a venture fund holds many positions is not diversification in the ordinary sense but the fact that nothing can be trimmed afterwards. Nilgiri Venture Fund I holds eighteen investments against Rs 1,50,00,00,000 of commitments. A private register has no buyer standing by for a third of anything, so the fund cannot sell a third of Aravalli next quarter. So the weight of each holding is decided once, at entry, and lived with until a whole position changes hands. Deciding a weight once is a very different discipline from an approach where a position can be reduced on any working day, and it is the reason the entry decision carries so much of the weight. How much any one position matters as the count of positions rises is a measurement question and is covered separately.

How does somebody actually use any of this on a working day?

Three people meet this material for real, and they use it differently. The analyst on a private capital desk is given a set of round documents and a share register and asked one question: does the arithmetic tie. The analyst rebuilds the price from the pre-money and the shares already in issue, rebuilds the new shares from the raise and that price, then adds up the register and checks it comes to the total the documents claim. When it does not tie, the fault is almost always a denominator, and the second most common fault is a share count taken from the wrong date.

The person doing diligence on an unlisted company asks a different question: what will the register look like after the next round, rather than after this one. Diligence reads the same four moving parts forward. And the reporting and operations side has the least glamorous job and the one that goes wrong most visibly: it has to hold the register afterwards, reconcile a call notice against what actually arrived in the bank, and answer, in a quarter's time, exactly what was called, what it bought and what is still promised. All three of those jobs are the same two divisions and one distinction, read from different sides of the same afternoon.