---
title: "Dividends or buybacks: Lintner, Apple and Tesco compared"
description: "Firms adjust dividends only partly (Lintner); Apple spent $117bn on buybacks and $41.6bn on dividends. Three ways to pay out, and four questions for the board."
url: "https://staging.khoi.co.uk/en/blog/co-tuc-hay-mua-lai-co-phieu"
lang: en-GB
author: "Lê Quốc Khôi"
date: 2026-11-03
updated: 2026-11-03
topics: ["Corporate finance"]
tags: ["Dividend policy", "Share buyback", "Payout ratio", "Dividend signalling", "Capital allocation"]
---

# Dividends or buybacks: what Lintner, Apple and Tesco show

## Key takeaways

- Miller and Modigliani (1961): in an ideal market, payout policy does not change the share price once investment is given; a price can still move with a dividend because dividends carry information about profits (Miller and Modigliani, 1961).
- Lintner (1956) interviewed 28 American companies over 196 company-years: the first question was always whether to change the current rate, and firms closed only part of the gap to a target payout, most often around 50 per cent of earnings (Lintner, 1956).
- Between August 2012 and 26 March 2016 Apple spent $163.1bn on capital return: $117.0bn in buybacks, $41.6bn in dividends and $4.5bn in taxes on employee share awards (totals I added up from the company's table) (Apple Inc., 2016).
- Tesco's 2021 report shows a full-year dividend of 9.15p, an exception to its 50 per cent payout policy, plus a £5.0bn special dividend after selling its Asian business (Tesco PLC, 2021).

In March 2012 Apple announced plans to start paying a quarterly dividend and buying back shares, expecting to use about $45bn of domestic cash over the first three years for dividends, repurchases and settling employee share awards (Apple Inc., 2012). Tesco's 2021 report held the full-year dividend at 9.15p although its own policy gave 5.97p (Tesco PLC, 2021, pp. 20, 25). A CFO looking at this year's profit faces the same question: pay it out, keep it, or pay it out by repurchasing shares. Three public documents give three different answers.

## Miller and Modigliani say payout policy does not matter, so why does everyone discuss it?

Because their result holds in an assumed world, and that world shows what is left once the assumptions go. They set three conditions: perfect markets (no trader large enough to move prices, no transaction costs, no tax difference between dividends and capital gains), rational investors, and complete certainty about the future (Miller and Modigliani, 1961, pp. 411-412); later in the paper they relax certainty into an uncertain world, assuming symmetric market rationality (Miller and Modigliani, 1961, p. 428). A bigger dividend then forces the firm to issue shares to keep investing, the two effects cancel, and payout policy leaves the price unchanged (Miller and Modigliani, 1961, p. 413).

Two points in the same paper keep the topic alive. One is the "informational content" of dividends: where a firm has a long-established target payout, investors have reason to read a dividend change as news about profits, and the price moves, though in their view the price still reflects future earnings rather than the dividend itself (Miller and Modigliani, 1961, p. 430). The other is the clientele effect: each company attracts investors who like its payout ratio, and if the spread of payout ratios matches the spread of preferences, one clientele is as good as another for valuation (Miller and Modigliani, 1961, p. 431). See [dividend signalling](https://staging.khoi.co.uk/en/glossary#dividend-signalling) and the [Dividend policy theories](https://staging.khoi.co.uk/en/framework/dividend-policy-theories) framework.

## How do real companies set the dividend?

Lintner went and asked. He picked 28 companies out of more than 600 listed ones, interviewed two to five executives at each, and studied 196 company-years from 1947 to 1953 (Lintner, 1956, pp. 98-99). In no case did management ask "how much should we pay this year?" from scratch. They asked whether to change the current rate, and the thing being decided was the change from the existing rate (Lintner, 1956, p. 99). They used partial adjustment: move part of the way this year, and further next year if still warranted (Lintner, 1956, pp. 99-100).

Two-thirds of the companies had a fairly definite target payout, from 20 to 80 per cent of earnings, with 50 per cent the most common (Lintner, 1956, p. 102). Among the firms that adjusted gradually, the share of the gap closed each year ranged from one half down to one fifth or one sixth (Lintner, 1956, p. 103). He notes that the cost of capital generally did not enter into the dividend policies of the firms he studied, and found just a few cases of dividends raised ahead of a share issue to improve its terms (Lintner, 1956, p. 104). His model: the new dividend equals the old dividend plus a fraction of the gap between the target and the old level. It accounted for about 85 per cent of company-years, but the firms were chosen to span situations, so he says no statistical significance can be attached (Lintner, 1956, pp. 107-108).

A hypothetical example: every figure is mine, the formula is Lintner's. A packaging workshop pays VND 1.2bn in dividends, targets 40 per cent of profit, and closes one third of the gap each year.

| Year | Profit | Target dividend (40%) | Dividend paid | Actual payout ratio |
|:--|--:|--:|--:|--:|
| 1 | VND 4.0bn | 1.60bn | 1.33bn | 33% |
| 2 | VND 6.0bn | 2.40bn | 1.69bn | 28% |
| 3 | VND 3.0bn | 1.20bn | 1.53bn | 51% |

In year 3 profit halves and the dividend falls about 10 per cent. The owner need not reopen the dividend every year, but the payout ratio jumps to 51 per cent, above target. That is the price of stability. Lintner records that earnings fluctuate more than dividends (Lintner, 1956, pp. 103, 106) and that managements treated their liquidity position as far more flexible than dividend policy, a buffer between dividend needs and investment needs (Lintner, 1956, p. 105); linking the two into "the price they accepted" is my inference.

## Does Apple pay out through dividends or buybacks?

Mostly buybacks, and the two behave differently. The table comes from Apple's 10-Q for the quarter ended 26 March 2016; I added the two kinds of repurchase together and worked out the changes.

| Fiscal year | Dividends | Buybacks | Change in dividends | Change in buybacks |
|:--|--:|--:|--:|--:|
| 2013 | $10.6bn | $23.0bn | | |
| 2014 | $11.1bn | $45.0bn | +5% | +96% |
| 2015 | $11.6bn | $36.0bn | +4% | -20% |

Dividends follow a gentle slope; buybacks jump and fall back. In my reading that is what a [share buyback](https://staging.khoi.co.uk/en/glossary#share-buyback) is for: it creates no recurring commitment. In April 2016 Apple raised its dividend by 10 per cent, said it intended to raise it every year, and lifted the whole capital return programme from $200bn to $250bn (Apple Inc., 2016). From August 2012 to 26 March 2016 it had returned $163.1bn: $41.6bn in dividends, $117.0bn in buybacks, and the rest in taxes on employee share awards (Apple Inc., 2016). Buybacks were about 72 per cent of the total.

Two details belong with that. The stated main aim of the first repurchase programme in 2012 was to neutralise the dilution from employee share grants (Apple Inc., 2012), so part of the buying only stopped the share count rising. And the money did not come only from profit: the filing says cash would come from domestic cash, US operating cash flow and borrowings, with $71.3bn of notes and $8.0bn of commercial paper outstanding at quarter-end (Apple Inc., 2016). See the case file on [Apple's capital return](https://staging.khoi.co.uk/en/case/apple-capital-return).

## Tesco kept the dividend when its policy said lower

For the 2020/21 year Tesco proposed a full-year dividend of 9.15p, the same as the year before, and called it an exception to its policy of a 50 per cent payout, which would have given 5.97p (Tesco PLC, 2021, pp. 20, 25). From those two figures I infer an actual payout of roughly 77 per cent of the earnings used for the policy (my arithmetic, not a reported number). The reasons given: the importance the Board places on dividends, the resilience of the business in an exceptional year, and confidence in future cash flows (Tesco PLC, 2021, pp. 20, 25).

The report also shows the steadiness once broke. Tesco records a decision to reinvest heavily in its customer offer and "withdraw the dividend in 2015" to strengthen the UK business and the balance sheet, then reinstated it in 2017 (Tesco PLC, 2021, p. 85). After completing the sale of its Asian business in December 2020 it paid a special dividend of 50.93p per share on 26 February 2021 (about £5bn) and made a one-off £2.5bn contribution to the pension scheme (Tesco PLC, 2021, pp. 5, 20, 85). The company bought back no shares that year (Tesco PLC, 2021, p. 86). See the [Tesco case file](https://staging.khoi.co.uk/en/case/tesco-dividend-and-capital-allocation); the post on [Tesco leaving the US](https://staging.khoi.co.uk/en/blog/tesco-roi-my-khung-chien-luoc) covers a different decision by the same group.

## Four questions for the board

I assembled these four questions from the three sources above; no author proposes this table.

| Question | If yes | If no |
|:--|:--|:--|
| Is this profit recurring every year? | Steady dividend, set a target rule (Lintner style) | One-off payment (like Tesco's special dividend) or a buyback |
| Can you hold this level if profit halves? | Keep it; shareholders may read a cut as bad news | Set it lower, pay the excess through something flexible |
| Is anything more urgent than shareholders (pensions, debt, a project)? | Pay it first, like Tesco's £2.5bn into the pension scheme | Distribute the surplus |
| Who are the shareholders? | Listed company: signalling and clienteles both matter | Family company: you need a rule, not a signal |

The table fits poorly when the company needs capital for a large project. Lintner found such investment needs rarely changed the dividend pattern; projects beyond the available cash were re-examined to see whether outside capital was justified (Lintner, 1956, pp. 105-106). For an owner-manager with no share price to send a signal, what carries over is the rule: a target percentage, a speed of adjustment, and a limit for when profit collapses.

## Going deeper: what did I read and not read?

Miller and Modigliani (1961) appeared in the Journal of Business; I read an unofficial full-text scan, mainly pages 411 to 414, 428 and 430 to 431, and have not checked it against the publisher's version. Lintner (1956) appeared in the American Economic Review; I read up to page 109 in a public JSTOR copy and did not read the later tests against rival models. Neither copy discusses buybacks.

For Apple I read the 2012 press release in full and the dividend, repurchase, borrowing and liquidity sections of the 10-Q; the fiscal-year rows in my table are sums I made from the company's table. For Tesco I read the passages on dividends in the financial review and the remuneration report of the 2021 report, not the whole document. I have not read Bhattacharya, Miller and Rock, or Easterbrook, so the post does not use their signalling and agency theories. Comparing Apple with Tesco is an illustration only, since they differ in size, tax and shareholders; the tables are mine, not tests.

## Frequently asked questions

### Does paying a dividend make a company worth more?

In Miller and Modigliani's ideal economy (perfect markets, rational investors, certainty about the future to begin with), payout policy does not change the share price once the investment plan is given (Miller and Modigliani, 1961, p. 413). They add that a price can still move after a dividend change, because the dividend carries information about profits, not because of the payment itself (Miller and Modigliani, 1961, p. 430). Real markets have taxes and trading costs, so this is a starting point for questions, not a verdict for every firm.

### Why do companies hate cutting dividends?

In Lintner's interviews, many managements believed most shareholders prefer a reasonably stable rate and that the market rewards stability, so they avoided changes that might have to be reversed within a year or so (Lintner, 1956, p. 99). He also records a general reluctance to reduce an established rate (Lintner, 1956, p. 103). That is what managers at 28 American firms said in the 1950s, not a law for every market.

### Can buybacks replace dividends?

Apple's figures show two payments behaving differently: dividends rose steadily, by roughly 4 to 5 per cent a year, in fiscal 2014 and 2015, while buybacks swung between $23bn and $45bn (Apple Inc., 2016). That is my reading of one company, not general evidence. Apple itself said the first repurchase programme aimed to neutralise the dilution from employee share grants (Apple Inc., 2012).

### Should a company borrow to pay dividends or buy back shares?

In its 10-Q for the quarter ended 26 March 2016, Apple said the cash for dividends, repurchases and debt repayment would come from domestic cash, US operating cash flow and borrowings, while $208.9bn of cash and securities sat in foreign subsidiaries and was generally subject to US tax if brought home (Apple Inc., 2016). That is Apple's own tax position at that date. For a small business the question should start with whether recurring cash flow covers the payout.

## References

- Apple Inc. (2012) *Apple announces plans to initiate dividend and share repurchase program (19 March)*. Press release. Available at: https://www.apple.com/newsroom/2012/03/19Apple-Announces-Plans-to-Initiate-Dividend-and-Share-Repurchase-Program/.
- Apple Inc. (2016) *Form 10-Q for the quarterly period ended 26 March 2016*. SEC EDGAR (Form 10-Q). Available at: https://www.sec.gov/Archives/edgar/data/0000320193/000119312516559625/d165350d10q.htm.
- Lintner, J. (1956) ‘Distribution of incomes of corporations among dividends, retained earnings, and taxes’, *American Economic Review*, 46(2), pp. 97–113.
- Miller, M.H. and Modigliani, F. (1961) ‘Dividend policy, growth, and the valuation of shares’, *Journal of Business*, 34(4), pp. 411–433. https://doi.org/10.1086/294442
- Tesco PLC (2021) *Tesco Annual Report and Financial Statements 2021*. Annual report. Available at: https://www.tescoplc.com/media/757589/tesco_annual_report_2021.pdf.

---

Canonical HTML version: https://staging.khoi.co.uk/en/blog/co-tuc-hay-mua-lai-co-phieu