Reinvest, dividends, or buybacks

A company has had a good year. After paying every wage, every supplier, every tax bill, there is £100,000,000 of profit left over — pure surplus, sitting in the bank, belonging to nobody in particular yet. The chief executive walks into the boardroom. The question on the table is short and it is the most consequential one any successful company ever asks: what do we do with the money?

There are exactly three answers. Not five, not a clever fourth. Three. And the headlines that report this decision — “company returns billions to shareholders,” “firm announces record buyback” — almost always blur which one is happening and what it really does to the people who own the shares.

The three doors

The first door is reinvest. Plough the £100m back into the business — new factories, new products, more salespeople, research. The bet is that this spending grows next year’s earnings. You spend the pie to bake a bigger pie.

The second door is dividends. Hand the £100m straight out to the shareholders as cash. A cheque per share. The company keeps nothing; the owners get money they can spend or put elsewhere. You cut the pie and serve it.

The third door is buybacks. The company goes into the market and buys its own shares, then cancels them. The earnings don’t change, but they’re now split across fewer shares. You don’t grow the pie or serve it — you shrink the number of slices.

Same £100m, three completely different effects. Hold this scenario fixed so the numbers stay honest: the company earns £200,000,000 a year, has 100,000,000 shares, and trades at £20 a share. So each share earns £200m ÷ 100m = £2.00. That £2.00 figure — earnings per share, or EPS — is the number to watch.

Dividends: cash now

The simplest door. Pay the whole £100m out as a dividend. That’s £100m ÷ 100m shares = £1.00 per share, landing in shareholders’ accounts as cash.

What happens to EPS? Nothing. The company still earns £200m, still has 100m shares, so EPS stays £2.00. The owners are richer by £1 a share in cash, but the per-share earnings of the business are untouched. The money simply left the company and went to its owners. Clean, visible, no trickery.

Buybacks: the one that confuses everyone

Now the same £100m, but spent buying shares. At £20 a share, £100m buys 5,000,000 shares. The company cancels them. Shares outstanding fall from 100m to 95,000,000.

The earnings haven’t moved — still £200m. But now they’re divided across fewer shares:

EPS = £200,000,000 ÷ 95,000,000 = £2.105.

EPS rose from £2.00 to about £2.105 — up roughly 5.3% — and the company earned not a penny more. This is where people go wrong. A rising EPS looks like the business got better. It didn’t. The pie is exactly the same size; it’s just cut into fewer slices, so each remaining slice is bigger. Every shareholder who held on now owns a slightly larger fraction of the same company. That is the entire mechanism: a buyback concentrates the existing earnings into fewer hands, it does not create new earnings.

So is it good? Only if the shares were not overpriced when bought. If the company paid £20 for shares genuinely worth £20, the continuing holders are fairly served. If it paid £20 for shares worth £12 — overpaying with the owners’ own money — it has quietly transferred value to the shareholders who sold, and the ones who stayed are worse off. A buyback is value-neutral at best and value-destroying when the price is wrong. It is never free money.

Reinvest: growing the pie

The third door is the only one that can make the company actually bigger. Spend the £100m on projects. Assume good projects return about £0.10 of new yearly earnings for every £1 spent — so £100m buys roughly £10,000,000 of extra future earnings.

Future earnings then become £200m + £10m = £210m, across the unchanged 100m shares:

Future EPS = £210,000,000 ÷ 100,000,000 = £2.10.

A similar lift to the buyback’s £2.105 — but reached a different way. The buyback shrank the slice count; reinvestment grew the whole pie. And there’s a catch from earlier in this course: reinvesting only beats paying out if the project clears the hurdle rate, the cost of capital. Spend £100m on projects returning less than that hurdle and you’ve destroyed value — the owners would have been better off with the cash to invest themselves. So the honest order is: reinvest if and only if you have projects that beat the hurdle; if you don’t, hand the money back, by dividend or buyback.

On the whole

“Apple returns $90 billion to shareholders.” “Shell announces a buyback.” These headlines describe a company at the boardroom moment — surplus cash, three doors. The phrase “returns to shareholders” usually means dividends or buybacks: the company is admitting it has no project worth more than the owners’ own next-best use of the money. That is not failure; mature companies run out of pies worth baking, and handing the cash back is the responsible thing.

But notice what a buyback quietly does to the EPS line that the same headline celebrates. The number goes up while the business stands still. If you hold an index fund, you own slices of thousands of these companies, and the slices being concentrated are partly yours. The board is deciding, on your behalf and without asking, whether to grow your pie, serve it, or simply cut it into fewer pieces — and whether the price it pays to do the cutting is one a careful owner would have paid.