Share Dilution: How Ownership Gets Diluted and How It Is Measured

Dilution is the fall in each existing share's slice when more shares exist. Anjani Stationers, an invented stationer, has options over 25,000 shares against 4,00,000 in issue. Options over 25,000 shares are 6.25 per cent of the count, yet if exercised they would cut earnings per share by only 5.88 per cent, from Rs 7.50 to Rs 7.06. The two percentages differ because their denominators differ, and that gap is the single most misread thing about dilution.

Underneath that sits a division. Earnings per share puts one year's profit on top and a number of shares underneath. Dilution never touches the top. Dilution works entirely on the bottom, and it works there because somebody holds a right to receive shares that do not yet exist. So every question about dilution is a question about a denominator, and denominators behave in a way that catches people out.

Two published numbers carry this. Anjani Stationers Private Limited reported profit after tax of Rs 30,00,000 in year two on 4,00,000 ordinary shares of Rs 10 each. The published share capital is Rs 40,00,000 and the published basic earnings per share is Rs 7.50. Neither figure moves. The one new thing is a grant of options over 25,000 shares to senior staff. None has been exercised, so the share capital line and the count are exactly where the balance sheet left them.

What does dilution actually mean?

DilutionA fall in the proportion that each existing share represents, caused by the number of shares going up rather than by anything happening to the business itself. is a fall in proportion, and it is worth starting somewhere small enough to hold in mind. Four cousins run a printing job together and split whatever it earns four ways. A fifth cousin joins, and from that day the split is five ways. Each of the original four falls from a quarter to a fifth. The extra cousin is the whole mechanism. None of that says any of the four is worse off. Whether they are depends entirely on what the fifth cousin brought with him. Most explanations of dilution quietly skip that separate question.

Dilution is a fall in proportion, and a fall in proportion is not by itself a fall in worth. Almost every misreading on this subject comes from letting those two clauses collapse into one. Hold them apart.

Now put the case entity's numbers on it. Imagine a holder who has 1,00,000 of Anjani Stationers' 4,00,000 shares, 25.00 per cent of the business. The share optionsA right granted to somebody, usually a member of staff, to buy a fixed number of new shares at a fixed price at some point in the future. The right can be left unused. over 25,000 shares are held by senior staff and have not been exercised, so today that holder still has 25.00 per cent. If every option were taken up tomorrow, the count would become 4,25,000 and the same 1,00,000 shares would be 23.53 per cent. The holder did not sell anything. The shares in the demat account did not change. The denominator underneath them did.

The two ways of stating that fall are both correct and are different numbers. In percentage points, 25.00 becomes 23.53, a fall of 1.47 points. As a proportion of what the holder had, 1.47 over 25.00 is a fall of 5.88 per cent. The same 5.88 per cent recurs in the earnings per share arithmetic, where it is not expected.

What actually causes a share count to rise?

Four things, and it is worth naming all four because they feel very different from the inside even though the arithmetic treats them identically. The first is share-based payment, Anjani Stationers' own case: options granted to staff who may one day take them up. The second is the conversion of a convertible instrument, where a lender's claim turns into shares. The third is an issue of shares for cash, whether to a small number of investors or to the public. The fourth is an issue of shares to buy something, where the seller of a business is paid in shares rather than in money.

In all four the count rises, and what makes them different is not the arithmetic but what came into the business in exchange. Cash from a share issue lands on the balance sheet the day it arrives and can be pointed at. A business bought with shares lands the same way. The staff gave service already performed for their options, plus the exercise price if and when they take the options up. Service is real but far harder to point at on a single line. A conversion brings in the disappearance of a liability. The disappearance is also real and also easy to miss. It shows up as something leaving rather than as something arriving.

Anjani Stationers has only the first of the four. There are no convertible instruments on this balance sheet, no preference shares, no debentures and no shares bought back. The other three are worked below as labelled hypotheticals, and none of them changes the published equity of Rs 1,42,00,000 or the published count of 4,00,000.

How is diluted earnings per share computed?

Slowly. The two divisions look so alike that people stop reading after the first one, and that is where they go wrong. Start with what is published. Basic earnings per shareProfit attributable to ordinary holders divided by the number of ordinary shares actually in issue over the period, counting nothing that might come into existence later. takes profit after tax of Rs 30,00,000 and divides it by the 4,00,000 shares actually in issue, giving Rs 7.50. Rs 7.50 is the figure on the income statement, and it is a fact about the year that finished.

Now the second division. Diluted earnings per shareThe same profit divided by the share count that would exist if every instrument capable of producing new shares, and which would reduce the figure, produced them. asks a different question: what would this same profit look like spread over the count that would exist if the 25,000 option shares came into existence? Nothing about the year's trading changes if a member of staff exercises an option, so the profit does not move. Only the denominator moves, from 4,00,000 to 4,25,000. Rs 30,00,000 over 4,25,000 is Rs 7.0588, reported as Rs 7.06.

Two things in that table are worth stopping on. The numerator is identical in both columns. Dilution is not a profit event. And nothing has been exercised, so the share capital line on the balance sheet is still Rs 40,00,000 on 4,00,000 shares. The diluted figure is not a record of anything that happened; it is the same completed year divided by a count that does not yet exist. That is why it sits beside the basic figure rather than replacing it.

One more mechanic belongs here even though it does not bite on this case. Shares that existed for two months of a year cannot carry twelve months of profit. So where a business issues shares part way through a year, the count used in the division is a weighted averageA share count adjusted for how much of the period each share existed, so shares issued in the last month of a year add far less to the denominator than shares in issue all year. rather than the closing number. Anjani Stationers issued no shares during year two, so its weighted average and its closing count are the same 4,00,000 and the point never surfaces in the arithmetic above. Meet a business whose count moved mid-year and it surfaces immediately.

Why does earnings per share fall by less than the count rises?

Because the two percentages are not measuring the same thing against the same base, and a single line settles it. The rise in the count is the new shares divided by the count at the start. The fall in earnings per share is the new shares divided by the count at the end. Same numerator of 25,000 shares, two different denominators, and the second denominator is always the larger one, so the second percentage is always the smaller one.

Check it on the case. The rise is 25,000 over 4,00,000, or 6.25 per cent. The fall is 25,000 over 4,25,000, or 5.88 per cent. And the long way round gives the same answer: Rs 7.50 less Rs 7.0588 is Rs 0.4412, and Rs 0.4412 over Rs 7.50 is 5.88 per cent. Two routes give one number. They are the same relationship written twice.

The same relationship gives a formula simple enough to run mentally. If the count rises by a proportion, the fall in earnings per share is that proportion divided by one plus itself. A rise of 6.25 per cent divided by 1.0625 gives 5.88 per cent. A rise of 25 per cent divided by 1.25 gives 20 per cent. A rise of 100 per cent divided by 2 gives 50 per cent. The relationship is not linear, it is reciprocal, and the gap between the two percentages widens as the issue gets bigger.

Read the last two columns down the table rather than across it. The gap between the two percentages is 0.37 points on the real case, 5 points at a quarter, and a full 50 points when the count doubles. Treat the two as interchangeable and the error stays invisible while the numbers are small and becomes enormous exactly when the stakes rise. That is the worst possible property for an error to have.

Which instruments count in the diluted figure, and which are left out?

Only the ones that would make the figure worse. The whole test is that blunt, and the polite form hides it: an instrument is counted only if counting it reduces earnings per share. If counting it would raise earnings per share, it is anti-dilutiveDescribes an instrument whose inclusion would raise earnings per share rather than lower it. Such an instrument is left out of the diluted figure entirely. and it is left out.

For an option the test comes down to the exercise priceThe fixed price at which the holder of an option may buy the new share. The holder chooses whether to pay it, and will not pay more than the share is worth to them.. Anjani Stationers granted its options at a price below what a share is currently worth to the people holding them. So the grant is treated as dilutive and the 25,000 shares go into the count. Now take the mirror image as a clearly labelled hypothetical. Suppose the exercise price sat well above what a share is worth. Nobody sensible pays Rs 120 for something they could buy for Rs 100, so nobody would exercise, no new shares would appear, and adding them to the denominator would describe an event that will not happen. Options priced that high are left out, the denominator stays at 4,00,000, and the diluted figure equals the basic figure at Rs 7.50.

The diluted figure is therefore deliberately a worst case among realistic outcomes rather than an average of them. That is the design, and it explains something readers often find odd. A business can grant a great many options and still publish a diluted figure identical to its basic one. On the day of measurement none of the options would be exercised. The test also means the diluted figure moves as circumstances move, without any grant or cancellation taking place. Nothing about the instrument changed; only whether exercising it makes sense did.

Does dilution always leave an existing holder worse off?

No. The honest answer runs in both directions, and that is why this part of the subject is so often taught badly. Go back to the four causes. In every one of them the holder's proportion falls, and the arithmetic of that fall is identical. Only what the business received differs. So the question splits cleanly into two, and only the first of them is what dilution measures.

The first question is about proportion, and dilution answers it completely. A holder with 1,00,000 of Anjani Stationers' shares moves from 25.00 per cent to 23.53 per cent if the options are exercised, and that is a fall of 5.88 per cent in their slice whatever anyone thinks of the business. The second question is about worth, and dilution says nothing about it at all. Dilution measures proportion and is silent on worth. Answering the worth question means valuing the business, a separate exercise.

Look at what that separation permits. Take a clearly labelled hypothetical in which the same 25,000 shares are issued to buy something that earns, and profit rises from Rs 30,00,000 to Rs 33,00,000. The count is 4,25,000 either way, and the holder's slice is 23.53 per cent either way, down by exactly the same 5.88 per cent. But earnings per share is now Rs 33,00,000 over 4,25,000, or Rs 7.76, above the Rs 7.50 the holder started with. The proportion fell and the earnings sitting behind each share rose. Both statements are true at once, and neither is an opinion.

Now the other direction. The fairness runs both ways. If shares come into existence and nothing arrives that earns, the same profit is genuinely split more ways and the holder's position is worse in the plainest sense. An issue that brings in nothing is the case worth watching for, and it is why the note describing what was received matters more than the count itself. A household analogy holds it: two brothers splitting a shop's takings become three, and whether the third brother's arrival was good for the first two depends entirely on whether he brought a second shop with him or only an appetite.

How does somebody reading a set of accounts actually use this?

Differently depending on what they are trying to decide, and it is worth being concrete about three of them. The basic count describes a business that may not exist by the time a forecast lands. So an analyst building a forecast of per-share figures works from the diluted count, not the basic one. The working method is to rebuild the division rather than to discount the answer.

A lender looks at all of this and mostly walks past it. Dilution moves no cash, changes no covenant that is written on debt to earnings before interest, tax, depreciation and amortisation, and does not touch Anjani Stationers' borrowings, its interest cover of 11.9 times or the guarantee it has given for Chitra Binding Works. A facility written per share is unusual, and only there does dilution matter to a lender. Vaidehi Rao, as finance controller, would still be asked about it in a lending conversation, and the honest answer is that the option grant changes the ownership picture and leaves the debt picture untouched.

Somebody deciding whether to put money into a private business reads it hardest of all. The option pool is the difference between the slice they are being offered and the slice they will eventually have. The number worth extracting is never the two earnings per share lines alone but the note behind them. The note says how many instruments are outstanding and what would bring them into existence. Two businesses with identical basic and diluted figures can carry completely different pools if most of one pool is currently priced too high to be counted.

Anjani Stationers Private Limited, Chitra Binding Works Private Limited and Vaidehi Rao are invented.
Educational material. Not advice on any investment, tax, budget or market position.