Buybacks, and why per-share figures rise on their own

Cash out, shares gone

A buyback is a company spending its own cash to buy its own shares and cancel them. Nothing is created. Cash leaves permanently, the share count falls, and every share still outstanding becomes a larger slice of a business that now holds less.

That is the whole mechanism, and it separates a buyback from the two corporate actions it gets filed alongside. A split and a bonus divide the same company into more pieces and move no money at all — the four events and the test that sorts them are set out separately. A buyback moves real cash across the company's boundary, in the same direction as a dividend and by a different route.

The shares do not go into a drawer. Indian company law requires them to be extinguished, and that requirement — not the purchase itself — is what makes the reduction in the share count permanent rather than reversible. Some markets allow a company to hold repurchased shares as treasury stock and put them back into circulation later; where that is permitted, the share count can be un-reduced at management's convenience, which makes the reduction a loan, not a gift.

So two things happened at once, and almost all the confusion in this subject comes from reading only one of them. The company got smaller by exactly the cash it spent. Your fraction of it got bigger. Neither of those is a gain on its own, and whether the pair of them adds up to one depends entirely on the third section of this article.

The denominator moved. The profit did not

Start with the figure a buyback is usually credited with improving, because the improvement is arithmetic rather than commercial.

A company earns ₹80 crore a year and has 10 crore shares outstanding. Profit divided by share count — earnings per share — is ₹8. It now buys back 1 crore shares and cancels them. The same ₹80 crore is divided among 9 crore shares, which is ₹8.89. Earnings per share rose 11% and not one additional rupee was earned. The numerator sat still and the denominator got smaller.

The same applies to every per-share measure at once: book value per share, cash flow per share, dividend per share for the same total payout. A reader looking at a five-year per-share series and seeing a steady climb is looking at two effects added together, and the series alone cannot tell them apart.

There is a cost sitting just outside that calculation. The cash spent was doing something — earning interest at the very least — and it stops. If the buyback was funded by borrowing, the profit line takes an interest charge that was not there before. Either way, the ₹80 crore is not quite ₹80 crore next year, so a buyback improves earnings per share by slightly less than the share count suggests, and occasionally not at all.

Now the failure mode, which is common enough to deserve a name and is invisible if you read announcements instead of share counts. Call it the treadmill buyback: a company buys back shares in the market while simultaneously issuing new ones to employees under stock option plans, and the outstanding count at the end of the year is flat or higher than at the start. The cash left the company, the leavers were paid, and the remaining holders got no larger slice at all. The buyback funded the dilution instead of reversing it.

The check takes a minute and needs no analysis. Pull the share count from the last five annual reports and read it as a series. If cash was spent on buybacks and the count has not fallen, the money bought something — but it did not buy you a bigger fraction of the company.

The price paid decides who gained

Here is the part the coverage skips, and it is the reason a buyback is neither good news nor bad news until you know one number.

Work it on a model small enough to hold in your head. A company's business is worth ₹9,000 and it is sitting on ₹1,000 of cash it has no use for, so the whole thing is worth ₹10,000. There are 100 shares, which puts ₹100 behind each of them. It spends the ₹1,000 buying back 10 shares at ₹100 and cancels them. What is left is the ₹9,000 business across 90 shares, every holder still has ₹100 a share, and nothing moved between them.

Change one number. The company pays ₹125 a share instead, spending ₹1,250 to buy the same 10 shares. What remains is ₹8,750 across 90 shares, which is ₹97.2 a share. The holders who tendered received ₹125 for something worth ₹100. The holders who stayed funded the difference out of their own per-share value, and their fraction of the company rose while the value behind it fell.

Run it the other way and the sign flips. If the company buys at ₹80, it spends ₹800 for 10 shares, leaving ₹9,200 across 90 shares, or ₹102.2 each. The holders who left took ₹80 for ₹100 of business, and the ones who stayed kept the difference. A buyback is a transfer between the two groups, and the price sets its direction.

Which reframes the announcement completely. A premium over the market price is not a reward for participating and not evidence that the shares are cheap — it is the rate at which value moves from the holders who stay to the holders who leave. The premium is reported because it is easy to compute. The comparison that decides the outcome is between the price paid and what the business is worth per share, which nobody can compute exactly and everybody has to estimate.

That estimate is a judgement, not a fact, and the honest position is that two careful people will disagree about it. What is not a judgement is the structure: a company buying its own shares is making the same decision an outside buyer makes, with the difference that it is spending money that already belonged to the sellers.

Two routes, two different mechanisms

A company can get to the same end state — cash out, shares extinguished — by two quite different roads, and the difference matters more to a shareholder than the headline size does. Which routes are open at any time is fixed by SEBI's buyback regulations rather than by the company, and that framework has been revised, so treat the availability of a route as something to check rather than assume.

In a tender offer the company makes a formal offer to every eligible holder: a fixed number of shares, at a fixed price, over a stated window. Entitlement is settled on a record date — the day the register of holders is read to see who qualifies — exactly as it is for a bonus or a dividend. You receive a letter of offer, and taking part is a positive act, submitted through your broker before the window closes. Offer more shares than the company set out to buy and each holder's tender is accepted in proportion, so the rest come back to you.

How that submission reaches you is your broker's choice and no part of the offer — a form inside the app for one, an emailed link for another, a phone call for a third. A cut-off a broker sets for its own processing is a broker's deadline, not the offer's, and the two need not fall on the same day.

In an open-market buyback the company simply becomes a buyer on the exchange. It places orders through a broker over a period, at whatever the market is doing, up to a maximum price and a maximum total amount announced in advance. There is no record date, no letter, no form and no entitlement. Anyone selling during the window might be selling to the company or to another investor, and because the order book does not show who is on the other side, the seller cannot tell which.

Read the last row twice, because it is the difference that survives everything else. A tender offer prices the transfer described in the previous section openly: one price, one date, and any holder can compare it with their own estimate of value and act. An open-market buyback spreads the same transfer across weeks of ordinary trading, at prices nobody chose in advance, and no holder is ever asked.

One more asymmetry follows from the ceiling. A tender offer names a number of shares and a price and holds them open, so a shortfall means holders did not tender rather than that the company thought better of it; an open-market authorisation is permission to spend up to an amount, and a company that announces one and then buys very little has moved its share price on an intention. Which of the two happened is checkable after the fact against what the company reported to the exchanges, and it is worth checking, because the announcement is what most people read and the completion is what mattered.

What each route means if you do nothing

The two routes put a shareholder who takes no action in genuinely different positions, and the difference is not obvious from either announcement.

• Tender offer, declined — you have made a choice, whether or not it felt like one. You have chosen a larger percentage of a company with less cash in it over a known amount of money at a known price. That is a defensible position and often the right one, but it was a decision, and the letter sitting unopened made it.
• Tender offer, ignored entirely — the same outcome by a different route, with one thing lost. Because acceptance is proportionate, tendering is rarely a decision about the whole holding; a holder who tenders and is scaled back keeps whatever was not accepted. Ignoring the offer forgoes an option that was partial, not total.
• Open-market buyback — there is nothing to decline. Every holder who does not happen to sell during the window is a non-participant by default, and the company acts on their behalf at prices they never saw. The only way to participate is to sell in the market, which you could do on any other day too.

Which is why the small-shareholder reservation only exists on the tender side, and why it changes the arithmetic there. SEBI's regulations reserve part of a tender offer for holders whose holding sits below a stated value on the record date. Because that pool is set aside, the two categories are subscribed and scaled separately, and the acceptance ratio in one carries no information about the other — which is what makes the identical offer a materially different proposition for a small holder and a large one. Which category applies is fixed by the holding on the record date and by nothing done afterwards.

There is no equivalent on the open-market side, because there is no category and no date. That is the practical summary of the two routes: one asks you a question you can price, and the other does not ask.

A capital-allocation decision, not a verdict

Strip away the mechanics and a buyback is one answer to a question every profitable company faces each year: what to do with cash the business does not need.

The realistic options are few. Reinvest it in the business. Pay down debt. Buy something else. Pay it out as a dividend. Or buy back shares. The first three keep the money inside the company in some form; the last two send it to shareholders, and the distinction between those two is largely one of who receives it and when. A dividend goes to every holder in proportion, automatically. A buyback goes only to the holders who sell, and leaves the others with a larger claim instead of cash.

So a buyback is neither a signal of confidence nor an admission of stagnation, though it is routinely reported as one or the other. It is a choice, and it can be a good one or a bad one on exactly two counts: whether the money genuinely had no better use inside the business, and whether the price paid was below what the business was worth. Get both right and the remaining holders are better off. Get either wrong and the cash bought nothing.

The trade-off is permanent and rarely stated. Cash returned does not come back. A company that buys back shares in a good year and then needs capital in a bad one raises it by issuing shares — often at a lower price than it paid to retire them, because bad years and low prices arrive together. The two transactions reverse each other in share count and do not reverse each other in money.

Which is the useful question to carry away, and it is a question about a company's history rather than about any single announcement: over ten years, has the share count fallen, and did the cash that reduced it come from operations or from borrowing? The answer sits in the cash flow statement next to the share count, and a company that has retired shares out of cash it actually generated has done something different from one that has done it out of debt.

Reading a buyback after it has happened

Almost everything above is checkable after the fact, from figures the company already publishes, and the check is more informative than the announcement was.

Four things, in the order they answer the argument of this article. The share count across five or ten years, to see whether the reduction survived the option plans. The cash balance and the debt before and after, to see where the money came from. The price paid, which is a single figure for a tender offer and an average for an open-market programme. And the share price range over the buying window, against which that average tells you whether the company bought into weakness or into strength.

Two of those four also change how a company's market capitalisation should be read across the event. Price multiplied by share count falls when shares are extinguished, which is correct — the company really is smaller — but a per-share series over the same period will be rising for the same reason. Reading the per-share series against the share count is what separates the two effects.

FNOTrader's Market Pulse scanner charts NSE and BSE price history alongside breadth and relative-strength measures across the listed universe, which covers the price side of that check; the share count and the cash flow come from the company's own filings. Nothing in this article, and nothing in that scanner, says whether a buyback makes a company worth owning. What the mechanics establish is narrower and worth holding onto: cash left, shares were cancelled, per-share figures rose without the business changing, and the price paid decided which group of shareholders paid for which.

How the money received in a buyback is taxed in the hands of a shareholder is set by the Income-tax Act in force for that year, and that rule has been rewritten more than once. What you know about the tax on an ordinary sale should not be carried across and assumed to apply here. Establish which rule governs the year in question before treating any offer price as a net figure.

Common questions

Continue reading

More in Investing Basics · App: Market Pulse Stocks Scanner · Definitions: glossary · Free tools: calculators · All: every article