Equity Financing: What It Costs and What It Gives Away

Equity costs Sankalp Industrial Systems Limited, invented, 14.00 per cent a year, against 6.00 per cent after tax on its borrowing, and not one rupee of that cost appears anywhere in its accounts. An issue gives away something permanent: raising Rs 3,60,00,00,000 at Rs 90.00 a share creates 4,00,00,000 new shares and hands their holders a sixth of everything the business earns from then on.

One distinction carries the whole subject, and the accounts cannot show it. A lender is paid; a shareholder is not paid anything by the company merely for holding the share. Interest of Rs 48,00,00,000 leaves Sankalp every year, lands on a line, and is deducted before profit is struck. The shareholder gets no such transfer. The shareholder holds instead a claim on whatever is left, forever, and the cost of equity is the yearly return that claim must be able to throw off before anybody would have bought it in the first place. The cost of equity is not an entry, it is not invoiced, and no accountant will ever raise it. The missing entry is the entire reason equity feels free and is the most expensive money the company has.

What is a shareholder actually buying?

The shape is easier to feel outside finance. Two people help a shopkeeper open a tea stall. The first hands over Rs 50,000 and says: give me Rs 1,000 a month for five years and then my money back. The second hands over Rs 50,000 and says: I want half the stall.

The first person has bought a schedule. Every month the shopkeeper owes them a fixed amount whether the morning was busy or dead, and on the last day of the fifth year the arrangement is finished and they walk away. The first person's cost can be computed exactly before a single cup has been sold.

The second person has bought a fraction of the future, and there is no last day. They receive nothing on any fixed date. If the stall does badly they get nothing at all and cannot complain. Half of the thing is what they bought, so if in year nine the shopkeeper is running four stalls and a supply contract, half of that is theirs too. Nothing written down each month records what that half is costing. The cost is half.

A share in Sankalp Industrial Systems Limited is the second arrangement written at scale. The holder has no promise of any payment. The holder stands behind every supplier, every employee, the tax authority and every lender, and takes what remains after all of them have been served. Standing last in that queue is what makes the share a residual claim, and a residual claim is the only claim on the business with no ceiling and no floor.

So what does a holder require in exchange for standing last in that queue? A number that is nowhere written down: the annual return that makes taking the residual position worth taking rather than lending instead. For this company that number is 14.00 per cent, and it is an estimate rather than a reading.

So what does equity cost this company, and how would anybody know?

The figure for Sankalp is 14.00 per cent a year, and building it is a subject of its own, covered separately. The build takes the risk-free rateThe base yield a long-dated government security offers, which this worked example simply assumes and never presents as today's market level., an equity risk premiumThe extra a holder wants each year before taking on ownership rather than lending, counted in percentage points on top of the base yield. and this company's betaHow far a company's shares tend to swing when the whole market swings, expressed as a multiple of the market's own move., and combines the three. The three combined are the capital asset pricing modelA frame that turns a base yield, a market premium and one measure of sensitivity into a single required return figure., set out by Sharpe in Capital Asset Prices, Journal of Finance, 1964.

The three inputs are estimated where they belong, in the build of the cost of equity. Two of them are rates in per cent, and either would sit within a whisker of the earnings yield below without being related to it at all. The answer, 14.00 per cent, is what carries forward.

Most treatments skip the honest part. The 14.00 per cent is quoted to two decimal places, and the precision belongs to the arithmetic rather than to the estimate. Compare it with the cost of the borrowing. Sankalp borrows at a blended 8.00 per cent before tax, and anybody holding the three loan documents can check that figure in a minute.

Weight each rate by its balance and the blend comes to exactly 8.00 per cent. Every rate in that table is this invented company's own contracted rate and none of them is a statement about what borrowing costs in India. The assumed effective tax rate of 25.0 per cent is also this company's own assumption and not any statutory figure. Apply it, and the after-tax cost of borrowing is exactly 6.00 per cent.

Nothing equivalent exists on the equity side. No shareholder has signed anything. There is no balance to weight and no rate to read off a document. The 14.00 per cent is assembled from an assumed base rate, an assumed premium and an estimated sensitivity, and a reader who wants to argue with it has to argue with three judgements rather than check three contracts. Both numbers are printed to two decimals in this subject and only one of them earns the decimals.

How much more expensive is equity, once it is put in rupees?

Set the two side by side. Borrowing costs Sankalp 6.00 per cent after tax. Equity costs it 14.00 per cent. Equity is more than twice as expensive as debt, and of the two, the expensive one is the one with no line in the accounts.

Rates are easy to nod at and hard to feel, so put the equity figure in rupees. Sankalp's market equity is Rs 18,00,00,00,000, being 20,00,00,000 shares at Rs 90.00. A 14.00 per cent required return on that is Rs 2,52,00,00,000 a year. Rs 2,52,00,00,000 is the amount the business has to be capable of producing, year after year, for the people holding the residual claim to have been right to hold it. Against that sits an interest bill of Rs 48,00,00,000 which the company writes down, pays out and deducts before profit.

The larger of the two numbers is the invisible one. The required return is more than five times the interest bill. And because it appears nowhere, a board can raise money by issuing shares, look at a profit and loss account that is entirely unchanged apart from a slightly better looking profit line, and genuinely believe it has found cheap funding.

Which figure do people reach for instead, and why is it wrong?

Faced with an unobservable 14.00 per cent, an analyst under time pressure looks for something with the same shape that can be read off a screen. There is one, and it is the wrong one.

Sankalp earns Rs 6.90 a share and trades at Rs 90.00. Put the first over the second and 7.67 per cent falls out. The earnings yield is observable, it is expressed in per cent a year, it moves when the share price moves, and it looks exactly like a cost of capital. Read the other way up it is the price to earnings ratio of 13.04 times, the same fact stood on its head.

The earnings yield is not the cost of equity, and it is not any kind of cost. An earnings yield is one year of accounting profit measured against today's price. Nothing in its construction asks a holder anything, so it says nothing whatever about what a holder requires. The cost of equity is what a holder needs the claim to deliver over a life that has no end at all. One is a backward-looking ratio of two published numbers; the other is a forward-looking requirement. The two differ here by 6.33 percentage points, and the observable one is roughly half the answer.

Why does that matter beyond tidiness? Because a hurdle rate is the thing a company measures a project against. Set the hurdle at 7.67 per cent and almost every proposal on the table clears it. Set it at 14.00 per cent and a great many of them do not. Swapping the earnings yield in for the cost of equity is not a small error carrying a small consequence. The swap builds a company that waves through very nearly everything put in front of it.

What does an issue actually give away?

Cost is only half the subject. The other half is what changes hands, and it is easier to get wrong because it does not look like a payment at all.

Suppose Sankalp needed Rs 3,60,00,00,000 and chose to raise it by selling shares at Rs 90.00. The issue creates 4,00,00,000 new shares. The count rises from 20,00,00,000 to 24,00,00,000.

The last two rows carry the lesson. Read them together. Profit did not move, and could not: the money is sitting still by assumption. The number of claims dividing that profit is what moved. Earnings per share lands at Rs 5.75 where it stood at Rs 6.90, a fall of Rs 1.15 on every share. The fall is 16.67 per cent, precisely the slice of the company the new holders now hold.

The two figures agree by construction rather than by luck, and holding on to why is the cleanest way to hold on to what dilution is. With profit standing still, the fraction of the company handed over and the fraction by which earnings per share drops are one division written out twice.

The qualifier about putting the proceeds to work belongs in the same breath as the arithmetic. Real proceeds are raised for a reason and usually earn something. The left-hand side survives whatever the money earns: the sixth of the company handed over does not come back if the project goes well. Dilution of earnings can be undone by earning more. Dilution of ownership cannot be undone at all.

When does the giving away stop?

Here is where a rate comparison stops being useful, and it is the part most treatments leave out.

The Rs 48,00,00,000 of interest is not forever. Each borrowing has an end written into it. Tranche 1, the secured term loan of Rs 3,00,00,00,000, has exactly one repayment day written into it: the last day of Year 5, when the whole balance comes due together. Tranche 2, the debentures of Rs 2,00,00,00,000, runs on to the last day of Year 7, a full year beyond where the explicit forecast stops. Tranche 3, the working capital facility, is different again: it is renewed each year rather than running to a maturity of its own, and it is the tranche the Rs 25,00,00,000 of new borrowing is drawn on each year, so it stands at Rs 2,25,00,00,000 by the end of Year 5.

A share has no maturity date, so the sixth of the company handed to new holders takes a sixth of Year 6, a sixth of Year 20 and a sixth of whatever the business turns into. The comparison between 14.00 per cent and 6.00 per cent quietly assumes both claims last equally long, and they do not. One of them runs for as long as the company does.

Notice also that the two arrangements fail in opposite directions. A borrowing that ends is a borrowing somebody has to be repaid, on a named day, whether or not the money is there. A claim that never ends is never a repayment problem. The permanence is the cost and it is also the comfort. Neither source is simply better than the other.

Is there already an example of this in the company's own accounts?

Sankalp is already living with equity it gave away, so the issue does not have to be imagined to see what given-away equity does.

Sankalp holds 75.0 per cent of Sankalp Coatings Private Limited, invented. Because it controls that subsidiary, the group statements pull in the whole of the subsidiary's revenue, costs and cash flow, not three quarters of them. The profit that results, though, is not all of it the group's. The remaining quarter, carried in the statements as a minority interestOn a consolidated statement, the slice of a subsidiary's profit and net assets that never belonged to the parent's shareholders in the first place., belongs to somebody else, and it has to come back out before the group can say what its own holders earned.

So the group's profit for the year is Rs 1,44,00,00,000, and Rs 6,00,00,000 of that is stripped out and handed to the outside holder of the subsidiary. The remainder, Rs 1,38,00,00,000, is the profit attributable to owners, and it is the figure the earnings per share of Rs 6.90 is built on.

Every property the Rs 6,00,00,000 line has is a property of equity given away. Look hard at what the line actually is.

Four answers, four properties, and each is a property of a share rather than of a loan. The minority's slice is a permanent deduction running quietly inside a company that has never sold a share to anybody, and it is the clearest demonstration available of what an issue does.

There is a second route to the same Rs 1,38,00,00,000 that is worth running as a check. Begin at net operating profit after tax (NOPAT) of Rs 1,80,00,00,000, remove after-tax interest of Rs 36,00,00,000, then remove the minority's Rs 6,00,00,000, and Rs 1,38,00,00,000 is what remains, exactly, with no residue at all. Two claims came out on the way: one belonging to lenders, with a rate and an end, and one belonging to outside shareholders, with neither.

Why does equity come last, and can it be seen here?

If equity is the most expensive source and the one that gives most away, companies would be expected to reach for it last. Companies do reach for it last, and the pattern has a name and a mechanism behind it.

Look at how Sankalp funds its growth. Every forecast year absorbs Rs 1,00,00,00,000 of net new invested capitalMoney put into the business over and above what depreciation already replaces, measured after the year's movement in working capital., and the record says where each rupee of it comes from.

The borrowing row takes gross debt from Rs 6,00,00,00,000 up to Rs 7,25,00,00,000 by the close of the forecast. The third row is empty, and it stays empty right across the five years.

Nobody announced a policy. The forecast simply shows internal cash first, borrowing second, and equity nowhere. Myers and Majluf described that same order in Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have, Journal of Financial Economics, 1984.

The mechanism underneath matters as much as the order, and it is not about cost at all. Managers know more about their own business than outside investors do. If they believe the shares are worth more than the market is paying, selling shares means selling something cheap, and they would rather borrow. If they believe the shares are dear, selling them is attractive. Investors work that out too. So an issue arrives carrying information about what the people inside believe, entirely separately from whatever the company says it will do with the money.

What does an issue signal, whatever the company says?

Follow that mechanism one step further and something uncomfortable falls out. If managers issue shares more readily when they think the shares are dear, then the act of issuing is itself evidence about what they think.

The everyday version: a cousin who has run a catering business for fifteen years offers to sell a quarter of it to a relative who knows nothing about the order book. The cousin knows everything. The cousin's willingness to sell at that price is information, and it would be foolish to ignore, however warmly the offer is made and however good the stated reason.

An announced issue therefore carries information about what the people deciding believe, whatever the stated use of the proceeds. That is why an issue is read the way it is, and it is the mechanism sitting underneath the funding order rather than a separate idea bolted on beside it. The claim is a narrow one. Managers who issue need not be dishonest at all, and what a market does in response cannot be read off the decision either. The argument is only that the decision to issue is made by people with better information, and that the people subscribing know it.

What does equity buy that borrowing cannot?

Everything so far has been the cost side, and stopping there would leave half the subject untold. Equity is expensive and permanent, and there are four things it does that no lender will do.

Equity has no maturity date, so there is no instalment anybody has to meet and no repayment to arrange. Equity carries no fixed payment, so a bad year costs the shareholder a return rather than costing the company a default. Equity takes no security, so no assets are pledged against it and no charge is registered anywhere. And equity does not reprice against the company: the sixth handed over is a sixth whatever happens to the business next year. The rate a lender charges is a rate a lender can revisit.

Put the two lists next to each other and the useful thing becomes visible. Borrowing and equity are not two prices for the same object. The two are different objects, and choosing between them is choosing which set of consequences to carry.

One last note on how dilution behaves. Dilution is a straight line: double the issue and the new shares double, exactly, every time. Nothing in it curves, and nothing in it repays being dragged back and forth. The one relationship in this subject that genuinely bends is worked out separately.