Dividend theories and policy
Is dividend policy relevant to value? Walter and Gordon say yes; Modigliani-Miller say no in perfect markets. Lintner shows firms smooth dividends towards a target payout.
Theories at a glance
| Theory | View | Core idea |
|---|---|---|
| Walter | Relevant | Payout should depend on the firm's return on investment (r) versus the cost of equity (k) |
| Gordon | Relevant | Investors discount distant, uncertain gains more: a bird in the hand is worth two in the bush |
| Modigliani-Miller | Irrelevant | In perfect markets a rupee of dividend cuts the share price by a rupee; investors can make 'home-made' dividends |
| Residual theory | Passive | Pay out only what is left after funding all positive-NPV projects |
| Signalling (information content) | Relevant in practice | Managers know more; a dividend rise signals confidence and a cut signals trouble |
| Tax preference | Favours low payout | If capital gains are taxed more lightly or later, investors prefer retention |
| Clientele effect | Neutral overall | Investors choose firms whose payout suits their tax and income needs; firms should keep policy stable |
| Lintner | Behavioural | Firms set a long-run target payout and move towards it slowly; they avoid cuts |
Assumptions NET asks about
| Model | Key assumptions |
|---|---|
| Walter | All investment financed from retained earnings; r and k constant; all earnings either paid out or retained; infinite life |
| Gordon | All-equity firm; retained earnings the only finance; r, k and retention ratio constant; k > br (= g); no taxes; infinite life |
| Modigliani-Miller | Perfect capital markets; no taxes or flotation costs; investment policy fixed; perfect certainty (later relaxed) |
Worked example 1: Walter's model
E = ₹10 a share, D = ₹4 and k = 10%. Find the price by Walter's model if r = 15%. Then find the price at 0% and 100% payout.
- P = [D + (r/k)(E − D)] ÷ k = [4 + (0.15 ÷ 0.10)(10 − 4)] ÷ 0.10.
- = [4 + 1.5 × 6] ÷ 0.10 = 13 ÷ 0.10 = ₹130.
- 0% payout (D = 0): [0 + 1.5 × 10] ÷ 0.10 = ₹150. 100% payout (D = 10): 10 ÷ 0.10 = ₹100.
- Here r > k (a growth firm), so the lower the payout, the higher the price.
₹130 at 40% payout; the optimum is zero payout (₹150). If r < k the optimum flips to 100% payout; if r = k, payout does not matter.
Worked example 2: Gordon's model with retention
E1 = ₹10, k = 12% and r = 15%. Find the price by Gordon's model if the retention ratio b is 60%, and then if it is 40%.
- b = 0.6: g = br = 0.6 × 0.15 = 9%; D1 = E1(1 − b) = 10 × 0.4 = ₹4.
- P = 4 ÷ (0.12 − 0.09) = 4 ÷ 0.03 = ₹133.33.
- b = 0.4: g = 0.4 × 0.15 = 6%; D1 = 10 × 0.6 = ₹6; P = 6 ÷ (0.12 − 0.06) = ₹100.
₹133.33 at 60% retention vs ₹100 at 40%. With r > k, more retention raises value, the same direction as Walter.
Worked example 3: MM irrelevance
A firm has 1,00,000 shares at ₹100 each and ke = 10%. It expects net income of ₹10,00,000 and plans to invest ₹20,00,000. Show that paying a dividend of ₹5 a share does not change the firm's value.
- MM: P0 = (D1 + P1) ÷ (1 + ke). With D1 = 5: P1 = 100 × 1.10 − 5 = ₹105. With no dividend: P1 = ₹110.
- New money needed = I − (E − nD1). With the dividend: 20,00,000 − (10,00,000 − 5,00,000) = 15,00,000, so new shares = 15,00,000 ÷ 105 = 14,286 (rounded).
- Without the dividend: 20,00,000 − 10,00,000 = 10,00,000, so new shares = 10,00,000 ÷ 110 = 9,091 (rounded).
- (n + Δn)P1 = old shares × P1 + new money: 1,05,00,000 + 15,00,000 = 1,20,00,000 with the dividend; 1,10,00,000 + 10,00,000 = 1,20,00,000 without.
- Value = [(n + Δn)P1 − I + E] ÷ (1 + ke) = (1,20,00,000 − 20,00,000 + 10,00,000) ÷ 1.10 = ₹1,00,00,000 in both cases.
Firm value stays ₹1 crore. Each shareholder has ₹110 of wealth either way: ₹5 dividend + ₹105 price, or a ₹110 price.
Worked example 4: Lintner's model
Last year's DPS was ₹4. This year EPS is ₹12, the target payout is 50% and the speed of adjustment is 0.4. Find this year's DPS.
- Target DPS = 50% × 12 = ₹6.
- D1 = D0 + c × (target DPS − D0) = 4 + 0.4 × (6 − 4) = 4 + 0.8 = ₹4.80.
₹4.80: the dividend moves only part of the way to the target, which is why dividends look smooth.
Determinants and forms of dividend policy
- Legal rules: under the Companies Act 2013 (s.123), dividends are paid out of profits; loan covenants may also restrict them.
- Liquidity: profit is not cash, so a profitable firm may still be unable to pay.
- Stability of earnings and growth opportunities: growth firms retain more.
- Cost of and access to outside finance; flotation costs favour retention.
- Shareholders' preferences, taxes and control (new issues dilute control).
- Inflation and the need to replace assets.
- Stability patterns: constant dividend per share, constant payout ratio, or a small regular dividend plus extras.
- Forms: cash dividend, bonus shares (stock dividend), stock split and share buy-back. Bonus shares and splits do not change total wealth.
NET asks Walter and Gordon price numericals, the r-versus-k optimum payout, MM's assumptions and arbitrage logic, and theory-author matching (bird-in-hand, signalling, Lintner's partial adjustment).
Typical MCQ: “Walter's model: r = 8% and k = 10%. Which payout maximises the share price?” 100%: when r < k, shareholders can earn more elsewhere, so the firm should pay out everything. Trap: zero payout applies only when r > k.
Typical MCQ: “Who called the bird-in-hand argument a fallacy?” Modigliani and Miller: in their view a firm's risk comes from its investments, not from how it splits earnings.
Walter: r vs k decides. Growth firm (r > k) retains, declining firm (r < k) pays all, normal firm (r = k) does not care. MM: dividend + new price = old wealth.