How share buybacks create value

A share buyback creates value for the owners who stay only when the company buys its stock for less than the stock is worth. The mechanics never change — fewer shares outstanding means each remaining share owns a larger claim on the same cash flow — but the mechanics alone are not value creation. PayPal Holdings, Inc. (NASDAQ: PYPL) spent $6.05B repurchasing about 86M shares in FY25, cutting its diluted share count toward 920M by Q1 2026, at a price near ten times earnings. Whether that built value or destroyed it turns on a single question: was ten times earnings below what the business is actually worth?

Why the price paid is the whole game

Most commentary on buybacks stops at "share count fell, so earnings per share rose," and treats that as automatically good. It isn't. A buyback is a use of owners' cash, and like any purchase it can be a bargain or a mistake depending on what you pay.

Buy below intrinsic value and the continuing owners capture the gap; buy above it and they hand that gap to the shareholders who sold. The per-share figure rises either way, which is exactly why per-share accretion is a poor test of whether a buyback was wise. The right test is the price paid against the value received — the same test you would apply to any investment the company makes.

How does a buyback actually move value?

A buyback moves value by changing how many owners share the same pool of cash, and by the price at which it shrinks that group. Take a company earning $100M of owner free cash flow, with 100M shares, trading at $10 — a market value of $1,000M, a 10% free-cash-flow yield, and $1.00 of owner free cash flow per share. Suppose the business is genuinely worth $15 a share. Management spends one year's cash, $100M, buying stock at $10 and retiring 10M shares.

Measure	Before	After $100M buyback
Shares outstanding	100M	90M
Owner FCF	$100M	$100M
Owner FCF per share	$1.00	$1.11
Value captured by stayers	—	$50M

Owner free cash flow per share rises about 11%, because the same $100M is now split across 90M shares rather than 100M. The value transfer is the second line: the company bought 10M shares for $5 less than they were worth, so continuing owners are richer by roughly $50M — the $5 discount on 10M retired shares. Nothing about the business changed; ownership simply concentrated at a favourable price.

Why does the price paid decide everything?

Reverse the price and the same buyback destroys value even as the per-share numbers still improve. If that business were worth only $8 a share and management bought at $10, continuing owners would lose about $2 on each of the 10M shares retired — $20M transferred to sellers — while free cash flow per share still ticked up to $1.11. That is the trap: the accretion is mechanical and always positive, so it can mask an overpayment.

The clean way to see a buyback is as the company reinvesting owners' cash at the free-cash-flow yield it pays. Buying stock on a 10% free-cash-flow yield is a 10% cash-on-cash return to the remaining owners — attractive if the business is stable and that beats the company's other options, poor if the yield is low because the price is high relative to worth. Low multiple and durable cash builds value; a high multiple, borrowed money, or a shrinking cash base puts it at risk.

What does that look like at PayPal?

PayPal's FY25 program is the large-scale version of the illustration: $6.05B of buybacks retired about 86M shares — close to 9% of the company — and pulled the diluted count down to 920M by Q1 2026, all at a price near ten times earnings. Paired with a newly initiated $0.14 quarterly dividend, PayPal returned very nearly its entire free-cash-flow base to owners in a single year. Whether retiring shares at roughly 10x earnings created value depends on whether that multiple sat below the company's worth, which is the argument I set out in the PayPal investment case rather than assert here.

What is not in doubt is the arithmetic that makes the case matter: at a low enough multiple, returning cash by buying shares turns flat business-level cash flow into rising cash flow per share, which is the mechanism behind a double-digit owner-FCF yield. The buyback is how the yield compounds.

The counter-argument

Buybacks earn their bad reputation honestly, because plenty of them destroy value. Managements buy their own stock most enthusiastically at the top, when cash is plentiful and prices are high, and cut buybacks in downturns when their stock is cheapest — the opposite of the discipline the maths rewards. Buybacks are also used cosmetically, to mop up the shares that stock-based compensation issues so that the net count barely moves while gross repurchases look impressive.

And a buyback funded with borrowed money raises the risk of the business to flatter per-share optics. Each of these is a real failure, and each is a failure of the same thing: paying the wrong price, or buying for the wrong reason. None of it argues against buybacks as such — it argues for judging them on price versus value and on the net change in share count, which is the discipline this whole framing rests on.

What would make this wrong

The PayPal version stops creating value if the company keeps buying above intrinsic value, or if the buyback is really just absorbing stock-based compensation so the net share count falls far less than the gross repurchases suggest. The figure to watch is the net diluted share count over time, not the dollars spent — $6.05B of buybacks that only holds the count flat is a very different outcome from one that cuts it 9%. The other thing to watch is the price paid against the value argued in the case: a buyback is only the good news the numbers imply for as long as the shares are actually below what the underlying owner free cash flow is worth.