Shareholder returns explained — why a buyback only counts once shares are cancelled
Shareholder returns are what a company hands back to its owners through dividends or share repurchases, and the two work differently: a dividend puts cash in the account and is taxed on receipt, while cancelling repurchased shares reduces the share count so each remaining share owns more
The three lines
- Total shareholder return ratio = (dividends + buybacks) ÷ net income
- A repurchase alone does not reduce shares outstanding — cancellation does
- Samsung Electronics' 2025 total return ratio was 43.6%, or 19.3 trillion won
Key questions
- What do shareholder returns mean?
- Everything a company gives back to shareholders out of what it earned. There are two main routes. A dividend distributes cash directly. A buyback has the company purchase its own shares in the market, and cancelling those shares removes them permanently. Dividends plus buybacks, divided by net income, gives the total shareholder return ratio.
- Which is better, dividends or share cancellation?
- It depends on the holder. A dividend delivers cash but is taxed on receipt, and in Korea it counts toward the financial income threshold that triggers comprehensive taxation. Cancellation delivers no cash but raises per-share value, and the benefit is realised as a capital gain when the shares are sold. Investors who need regular income favour dividends; long-term holders who reinvest generally favour cancellation.
- Why didn't the share price move after the buyback announcement?
- Because a repurchase and a cancellation are different events. Repurchased shares sit on the balance sheet as treasury stock and can be sold back into the market, used for employee compensation, or issued as merger consideration. Markets price that possibility, so a repurchase alone is not read as a permanent reduction in share count. Only cancellation reduces shares outstanding, lifting earnings per share and book value per share.
- Is a higher shareholder return ratio always better?
- No. What matters is the source of the funds. Returns paid out of free cash flow are sustainable. Returns funded by borrowing shift interest costs onto next year's earnings, and returns funded by cutting capital expenditure surface as lost competitiveness several years later. In capital-intensive industries like semiconductors, a sharply rising return ratio can signal that management has run out of things worth investing in.
When the KOSPI rose 3.68% to close at 6,579.04 on August 12, 2026, one of the reasons analysts gave was "shareholder return expectations." There was no earnings release and no filing, and Samsung Electronics rose 6.68% on the expectation alone.
So what is the expectation about? Shareholder returns reach a share price by more than one route, and the routes differ in timing, tax and signal.
1. The two channels
Shareholder returns cover everything a company hands back to its owners out of what it earned. There are two main channels.
| Method | What the holder receives | Shares outstanding | When taxed |
|---|---|---|---|
| Dividend | Cash | Unchanged | On payment, as dividend income |
| Buyback | Nothing directly (better supply) | Unchanged | — |
| Cancellation | Nothing directly (higher ownership) | Reduced | On sale, as capital gain |
The important line in that table is the one splitting buyback from cancellation. Coverage often merges them, but their effect on share count is opposite.
The headline ratio combines the first two:
Total shareholder return ratio = (dividends + repurchases) ÷ net income
2. Why repurchase and cancellation are not the same
When a company buys its own shares, those shares become treasury stock. They have not disappeared. They have been put in a drawer.
Shares in that drawer go one of three ways.
- Cancellation — destroyed. Shares outstanding fall
- Resale — sold back into the market. Share count reverts
- Use — employee compensation, or consideration in an acquisition
That is why markets do not react to a repurchase the way they react to a cancellation. As long as routes two and three remain open, a buyback has not reduced the share count — it has parked it.
Cancellation changes the arithmetic.
| Item | Before | After 10% cancellation |
|---|---|---|
| Net income | 100bn won | 100bn won |
| Shares outstanding | 10m | 9m |
| Earnings per share | 10,000 won | about 11,111 won |
Earnings did not rise by a single won, and per-share figures improved by 11%. Cancellation is the only mechanism by which a shareholder's claim grows without the shareholder receiving anything.
3. Where the two diverge: tax and cash flow
| Item | Dividend | Cancellation |
|---|---|---|
| Cash received | Yes | No |
| Taxed when | On payment | On sale of shares |
| Tax type | Dividend income tax | Capital gains (where applicable) |
| Counts toward comprehensive financial income | Yes | No |
| Suits | Regular income needs | Long holding and reinvestment |
In Korea, dividends are withheld at source and, above an annual financial income threshold, roll into comprehensive taxation. Cancellation triggers nothing at the time; whether tax arises is determined when the shares are eventually sold.
Which means the retiree living on dividend income and the investor reinvesting on a twenty-year horizon are served by different mechanisms. There is no general answer about which is superior.
4. The numbers, using Samsung Electronics
| Item | Value |
|---|---|
| 2024–2026 policy | Return 50% of free cash flow |
| 2024–2025 cash dividends | 20.9tn won (regular 19.6tn + special 1.3tn) |
| 2024–2025 repurchase and cancellation | 8.4tn won |
| 2025 total dividends | 11.108tn won |
| 2025 repurchases | 8.189tn won |
| 2025 total shareholder return | 19.297tn won |
| 2025 total return ratio | 43.6% |
| 2025 dividend payout ratio | 25.1% |
| 2026 H1 cancellation approved | 16tn won |
Two readings matter here.
The gap between payout ratio (25.1%) and total return ratio (43.6%). Those 18.5 points are what repurchases contributed. Looking only at dividend payout understates the total return by more than half.
The policy references free cash flow, not net income. Free cash flow is operating cash after capital expenditure. In semiconductors, where capex is enormous, net income can be strong while free cash flow is thin — which means a heavy investment year mechanically shrinks the pool available for returns. The policy is honest about that trade-off rather than hiding it.
5. So what was being priced on August 12?
Not anything in the table above. The market was positioning for an additional programme larger than what has been announced, and that positioning met heavy foreign buying.
No such announcement has been made. What is filed runs through the 16 trillion won cancellation approved for the first half of 2026. Everything beyond that is brokerage forecast and market inference.
Which suggests a reading hierarchy for shareholder-return news generally.
| Tier | Reliability | Example |
|---|---|---|
| Regulatory filing | Settled | "16tn won cancellation approved" |
| Stated company policy | High | "50% of free cash flow to returns" |
| Brokerage forecast | Reference | "Special dividend of 30tn won possible" |
| Market expectation | No basis | "A record programme is coming" |
The August 12 move was driven by the bottom two rows. Expectations from the bottom rows revert if the top rows never confirm them. The session is covered in "KOSPI closes at 6,579.04."
6. A high ratio is not automatically good
One last correction to a common assumption. The total return ratio is not a metric where higher is better. The source of the money decides the meaning.
- Paid out of free cash flow — sustainable
- Funded by borrowing — interest costs reduce next year's earnings
- Funded by cutting capex — surfaces as lost competitiveness in a few years
The third is the one that matters most here. In semiconductors, batteries and AI infrastructure, where capital expenditure determines market share, a return ratio that jumps suddenly can mean the company has run out of places worth investing. In a mature industry a high ratio is a virtue; in a growth industry the same number can be a warning.
The same 43.6% means different things depending on where in the investment cycle it was produced.
7. What is unresolved
No additional 2026 return has been filed. The special-dividend and large-buyback scenarios circulating in the market are brokerage forecasts, and this article deliberately leaves those figures out of the body.
The 43.6% depends on computation method. When repurchase and cancellation fall in different years, the year in which each is recognised changes the ratio.
The tax discussion is structural only. Actual burden depends on total financial income, account type, and current tax law. Individual figures need to be checked against a holder's own circumstances.
Sources
- Samsung Electronics IR — shareholder return policy
- ZDNet Korea — Samsung AGM approves 16 trillion won share cancellation
- Financial News — Korean memory makers set to expand shareholder returns
- Investing.com — Samsung announces 110tn won investment plan, maintains return policy
- Financial News — Return expectations drive heavy foreign buying (August 12)