Dividend policy

In short

Dividend policy is the decision of how much of a company's earnings to pay out to shareholders and how much to retain for reinvestment, and in what form - cash, bonus shares or buyback. It matters because the payout is funded from the same earnings that finance growth, so every rupee paid out is a rupee not reinvested; the decision therefore interacts with financing and investment decisions. B.Com and M.Com syllabuses examine the three classic models (Walter, Gordon, Modigliani-Miller) with numerical problems, plus the legal framework under the Companies Act, 2013. It is standard ground in interviews for equity research, corporate finance, credit rating and FP&A roles at Indian firms, and in CA/CS/CFA papers.

Explained

The dividend decision and its vocabulary

• Dividend: a distribution of profit to shareholders, declared per share and paid in cash unless otherwise stated. In India dividends are declared as a rupee amount per share or as a percentage of face value, not of market price.
• Payout ratio = DPS/EPS (or total dividend/PAT). Retention ratio b = 1 − payout. The two are complements; the policy choice is essentially fixing one of them.
• Dividend yield = DPS/market price - a return measure for the investor, not a policy variable for the company.
• Interim dividend: declared by the Board during the financial year out of current-year profits. Final dividend: recommended by the Board and *declared by shareholders* at the AGM; shareholders may reduce but not increase it.
• Dividend per share (DPS) vs dividend rate: a 150% dividend on a ₹2 face-value share is ₹3 per share - a very common exam trap.
• Key dates: declaration date → record date (register of members closed; holders on this date are entitled) → ex-dividend date (first day the share trades without the dividend right; in India with T+1 settlement the ex-date is effectively the record date itself) → payment date. The share price is expected to drop by roughly the dividend on going ex-dividend.
• Types of dividend: cash dividend, stock dividend (bonus issue), special/one-off dividend, scrip dividend (promissory note), property dividend (rare in India), and liquidating dividend (return of capital, not profit).

Legal framework in India - Companies Act, 2013

• Section 123: dividend may be paid only out of (a) profits of the current year after depreciation, (b) accumulated past profits (free reserves), or (c) both; or out of money provided by the Central/State Government for a guaranteed dividend. Revaluation reserve and other unrealised or notional gains cannot be used - they are excluded from "free reserves".
• Depreciation as per Schedule II must be provided before declaring dividend; the Board may transfer any amount to reserves at its discretion (the old compulsory transfer slabs were dropped under the 2013 Act).
• Dividend out of reserves in a loss year is allowed subject to Rule 3 of the Companies (Declaration and Payment of Dividend) Rules, 2014: the rate cannot exceed the average rate of the last three years; withdrawal cannot exceed 10% of paid-up capital plus free reserves; the balance of reserves after withdrawal must not fall below 15% of paid-up capital; and losses of the current year must be set off first.
• Section 123(3): interim dividend may be declared out of surplus in P&L and current-year profits; if the company has incurred a loss in the current year up to the quarter preceding declaration, the interim rate cannot exceed the average rate of the last three years.
• Section 124: unpaid/unclaimed dividend must be transferred to a separate Unpaid Dividend Account within 7 days of the 30-day payment window; amounts unclaimed for 7 years go to the Investor Education and Protection Fund (IEPF), along with the shares.
• Section 127: failure to pay a declared dividend within 30 days attracts punishment for defaulting directors (imprisonment up to 2 years and fine of at least ₹1,000 per day of default), and the company must pay 18% p.a. simple interest; exemptions apply where non-payment is for reasons beyond the company's control or due to a legal dispute over entitlement.
• Dividend must be paid in cash (electronic mode permitted); a company cannot declare dividend if it has defaulted on redemption of deposits. Dividend-bearing shares that are partly paid-up rank only for the paid-up amount.
• Regulatory overlays: SEBI (LODR) Regulation 43A requires the top 1,000 listed companies by market capitalisation to formulate and disclose a Dividend Distribution Policy. Banks may declare dividend only if they meet RBI's eligibility criteria (capital adequacy, net NPA ceiling); insurers are governed by IRDAI norms. These thresholds are revised periodically - check the current circular.

Taxation of dividends in India

• Dividend Distribution Tax (DDT) was abolished by the Finance Act, 2020 (with effect from 1 April 2020). Dividends are now taxable in the hands of the shareholder at applicable slab/corporate rates, restoring the classical system.
• TDS under Section 194: 10% on dividend paid to a resident where the aggregate exceeds the prescribed threshold in a financial year (₹5,000 historically, raised to ₹10,000 by the Finance Act, 2025 - verify the rate and threshold for the relevant year). Non-residents are deducted under Section 195 subject to treaty rates.
• Section 80M: a domestic company receiving dividend from another domestic company may deduct it to the extent of dividend it distributes - removing cascading in holding-company chains.
• Buyback: was taxed on the company at ~20% plus surcharge under Section 115QA, with the receipt exempt for the shareholder. From 1 October 2024 this shifted - buyback proceeds are treated as deemed dividend taxable in the shareholder's hands, with the cost of acquisition treated as a capital loss.
• Consequence for policy: because dividends are now taxed at the investor's marginal rate while long-term capital gains are taxed lower, high-bracket Indian investors mildly prefer retention to payout - the buyback route lost its own tax edge on 1 October 2024. A live illustration of the tax-preference argument.

Theories: does dividend policy affect value?

(A) Relevance theories - dividend policy affects the share price

• Walter's model (1963). Assumes all-equity financing, no external funds, constant r and k, 100% retention or payout for ever, infinite life, constant EPS and DPS. The key driver is the relation between the firm's return on investment r and its cost of equity k.
  • *Growth firm* (r > k): value rises as payout falls → optimal payout = 0%.
  • *Normal firm* (r = k): payout is irrelevant → every payout gives the same price.
  • *Declining firm* (r < k): value rises as payout rises → optimal payout = 100%.
  • *Limitations:* no external financing is unrealistic; r and k do not stay constant (r falls as more projects are taken, k rises with risk); ignores business risk.
• Gordon's model (1962) / dividend capitalisation. The basic model reaches the same conclusions as Walter - price falls with payout when r > k, rises with payout when r < k, flat when r = k. Gordon's refinement adds the "bird-in-the-hand" argument: investors are risk-averse and discount distant, uncertain retained earnings more heavily than a certain dividend today, so k rises with the retention ratio and a higher payout can lower the required return and raise price. Assumes g = b×r and requires k > g.
• Signalling (information content) hypothesis. In an imperfect market, managers know more than investors; a dividend increase credibly signals confidence in sustainable earnings, a cut signals distress. This is why Indian boards are reluctant to cut dividends even in weak years.
• Clientele effect. Different investor groups prefer different payouts - retirees, trusts and insurance funds want steady cash; young high-tax investors prefer capital gains. A firm attracts a "clientele" matched to its policy, so sudden changes in policy upset the shareholder base.
• Agency cost argument. Paying out free cash flow reduces managers' ability to over-invest in empire-building projects and forces them back to the capital market for scrutiny.

(B) Irrelevance theory - Modigliani and Miller (1961)

• In a perfect market - no taxes, no transaction or flotation costs, perfect information, rational investors, a given investment policy - the value of the firm depends only on its earning power and investment policy, not on how earnings are split between dividends and retention.
• Arbitrage/ homemade dividend argument: if a firm pays dividends it must raise an equal amount of fresh capital to fund the same investment; the dilution to existing holders exactly offsets the cash received. An investor wanting cash can sell shares; one not wanting cash can reinvest the dividend. So the two are substitutes.
• MM formula: P₀ = (D₁ + P₁)/(1 + k). Combined with the external financing requirement, the number of new shares n = (I − (E − nD₁))/P₁, and total value nP₀ works out the same for any D₁.
• *Criticisms:* taxes differ between dividends and gains; flotation costs make external equity dearer than retained earnings; transaction costs make selling shares an imperfect substitute for a dividend; information asymmetry is real.

Payout policies used in practice

• Stable / constant rupee DPS: a fixed DPS maintained and raised only when sustainable - most common among mature Indian companies (ITC, Hindustan Unilever, Infosys-style). Investors read a cut as bad news.
• Constant payout ratio: a fixed percentage of each year's PAT; dividends fluctuate with earnings - rarely used alone because it transmits volatility.
• Stable dividend plus extra: a safe regular dividend plus a special dividend in good years - keeps the floor credible.
• Residual dividend policy: fund all positive-NPV projects from retained earnings at the target capital structure first; pay out only what is left. Theoretically sound, but produces an erratic dividend.
• Lintner's model (1956): managers partially adjust towards a target payout - Dₜ = Dₜ₋₁ + a(target payout × EPSₜ − Dₜ₋₁), where *a* is the speed-of-adjustment factor. This explains dividend "smoothing" observed in practice.
• Determinants of payout: level and stability of earnings; liquidity (dividends need cash, not just book profit); availability and cost of external finance; growth/investment opportunities; loan covenants and legal limits; shareholder expectations and tax position; inflation (replacement cost of assets); control considerations (dividend avoids dilution from new issues); industry norms.

Bonus issue, stock split and buyback

• Bonus issue (stock dividend): free shares from free reserves, securities premium or capital redemption reserve (Section 63). Face value unchanged, number of shares rises, reserves fall - no cash leaves the company and the shareholder's proportionate stake is unchanged. Theoretically value-neutral; valued as a signal and for improved liquidity. Cannot be issued in lieu of dividend, and not permitted if the company has defaulted on debt interest/principal or statutory dues.
• Stock split: face value is sub-divided (say ₹10 into ₹1); number of shares rises, reserves and capital unchanged. A bonus affects reserves; a split does not - a favourite one-mark distinction.
• Share buyback (Section 68): company repurchases its own shares. Limits - up to 10% of paid-up capital and free reserves by Board resolution, up to 25% by special resolution; post-buyback debt-equity must not exceed 2:1; a maximum of 25% of paid-up equity in a financial year; a 1-year cooling-off between buybacks; shares must be extinguished within 7 days. Routes under the SEBI buy-back regulations: tender offer (proportionate, with 15% reserved for small shareholders), book-building, and odd-lot - the open-market-through-stock-exchange route was phased out and closed from 1 April 2025.
• Why buyback instead of dividend: flexible (no signalling commitment to repeat), raises EPS and ROE by shrinking the base, and signals that the share is undervalued; promoters' stake rises if they do not tender.
• Right vs obligation: equity dividend is discretionary; preference dividend is a fixed rate and, if cumulative, arrears accumulate - but it is still payable only out of profits.

Key points

• Dividend decision - how to split earnings between payout and retention; the third of the three core financial decisions after investment and financing.
• Payout ratio - DPS/EPS; retention ratio b = 1 − payout; growth g = b × r.
• Dividend rate - always a percentage of face value, never of market price or book value.
• Walter's model - r > k → zero payout; r = k → indifferent; r < k → 100% payout.
• Gordon's model - same conclusion, with the "bird-in-the-hand" risk-aversion refinement; requires k > g.
• MM hypothesis - in a perfect market dividends are irrelevant; arbitrage and homemade dividends make payout and retention substitutes.
• Signalling - a dividend rise/cut conveys management's private view of sustainable earnings.
• Clientele effect - investor groups self-select into firms whose payout matches their cash and tax needs.
• Section 123 - dividend only out of current or past profits after depreciation; never out of capital or revaluation reserve.
• Sections 127 and 124 - declared dividend must be paid within 30 days; unpaid amounts move to the Unpaid Dividend Account and, after 7 years unclaimed, to the IEPF along with the shares.
• DDT abolished (FY 2020-21 onwards) - dividends are now taxed in the shareholder's hands, with TDS under Section 194.
• Stable DPS - the dominant practice; boards smooth dividends and resist cuts (Lintner's partial adjustment).
• Residual policy - pay only what remains after funding all positive-NPV projects; sound but volatile.
• Bonus issue - capitalises reserves, no cash outflow, proportionate holding unchanged, value-neutral in theory.
• Stock split - reduces face value; capital and reserves untouched (unlike a bonus).
• Buyback - Section 68 limits (10% Board / 25% special resolution, post-buyback D/E ≤ 2:1); taxed as deemed dividend for the shareholder from 1 October 2024.
• Cash matters - a profitable company with no liquidity cannot pay a dividend; a loss-making one may pay from reserves within Rule 3 limits.
• Ex-dividend drop - the share price is expected to fall by about the dividend amount on the ex-date.

Formulas and examples

Example 1 - Walter's model across three firms. EPS = ₹10, k = 12%, payouts of ₹0, ₹4 and ₹10 per share.

For r = 15% (growth firm), D = ₹4: P = [4 + (0.15/0.12)(10 − 4)]/0.12 = [4 + 1.25 × 6]/0.12 = 11.5/0.12 = ₹95.83

Conclusion: the growth firm maximises price at zero payout, the declining firm at 100% payout, and the normal firm is indifferent.

Example 2 - Gordon's model. EPS = ₹20, r = 18%, k = 14%, retention b = 60%.

• g = b × r = 0.60 × 0.18 = 10.8%
• D₁ = E(1 − b) = 20 × 0.40 = ₹8
• P₀ = 8/(0.14 − 0.108) = 8/0.032 = ₹250
• Now raise payout to 60% (b = 40%): g = 0.40 × 0.18 = 7.2%; D₁ = ₹12; P₀ = 12/(0.14 − 0.072) = 12/0.068 = ₹176.47.
• Price falls when payout rises because r > k - the Gordon result for a growth firm. (If b = 80%, g = 14.4% > k and the formula breaks down - a reminder that k > g is a binding condition.)

Example 3 - MM dividend irrelevance. A company has 1,00,000 shares at P₀ = ₹100, k = 10%, earnings for the year E = ₹10,00,000, investment required I = ₹20,00,000.

*Case A - no dividend (D₁ = 0):*

• P₁ = P₀(1 + k) − D₁ = 100 × 1.10 − 0 = ₹110
• Funds needed from outside = 20,00,000 − 10,00,000 = ₹10,00,000 → new shares = 10,00,000/110 = 9,090.91
• Value of old shareholders = 1,00,000 × 110 = ₹1,10,00,000

*Case B - dividend of ₹10 per share:*

• P₁ = 100 × 1.10 − 10 = ₹100
• Dividend paid = ₹10,00,000; outside funds needed = 20,00,000 − (10,00,000 − 10,00,000) = ₹20,00,000 → new shares = 20,00,000/100 = 20,000
• Old shareholders: dividend ₹10,00,000 + shares worth 1,00,000 × 100 = ₹1,00,00,000 → ₹1,10,00,000

Identical wealth in both cases - dividends are irrelevant under MM assumptions.

Example 4 - Legal check under Section 123 with Rule 3. A company with paid-up capital ₹50,00,000 and free reserves ₹40,00,000 makes a loss of ₹3,00,000 this year but wants to continue its dividend. Average rate of the last three years = 12%.

• Maximum rate allowed = 12% → ₹6,00,000 on ₹50,00,000 capital.
• Maximum withdrawal from reserves = 10% of (50 + 40) lakh = ₹9,00,000.
• Current-year loss of ₹3,00,000 must be set off first → cash available for dividend from the ₹9,00,000 drawn = ₹6,00,000. Within the rate cap, so a 12% dividend is permissible.
• Reserves after withdrawal = 40,00,000 − 9,00,000 = ₹31,00,000 = 62% of paid-up capital, comfortably above the 15% floor. Dividend of ₹6,00,000 (12%) is legal.

Example 5 - Buyback versus dividend on EPS. PAT = ₹60,00,000; shares = 10,00,000 (EPS = ₹6); market price ₹80; surplus cash ₹80,00,000.

• *Dividend route:* DPS = 80,00,000/10,00,000 = ₹8; shares unchanged, EPS stays ₹6; price falls to about ₹72 ex-dividend; shareholder taxed on ₹8 at slab rate.
• *Buyback route:* shares bought = 80,00,000/80 = 1,00,000 → remaining shares 9,00,000; EPS = 60,00,000/9,00,000 = ₹6.67 (+11.1%); at an unchanged P/E of 13.33 the price holds near ₹88.9 for those who stay.
• Caveat: Section 68 caps the buyback at 25% of paid-up equity in a year (10% here is fine), and since 1 October 2024 the proceeds are taxed as deemed dividend in the seller's hands - so the tax edge of a buyback has narrowed sharply.

Where it is used

• Equity research and buy-side analysis: forecasting DPS and payout for dividend discount models; screening "dividend aristocrat" stocks and PSU high-yielders (Coal India, ONGC, Power Grid, NTPC) where the Government of India's stake makes payout policy a fiscal matter.
• Corporate finance and treasury teams: board notes on declaring interim/final dividend, cash-flow testing, Section 123 and Rule 3 compliance, and drafting the LODR Regulation 43A Dividend Distribution Policy.
• Company secretarial and compliance roles: record date and book-closure intimation to the exchanges, the 30-day payment deadline, Unpaid Dividend Account and IEPF transfers, Form MGT-7/AOC-4 disclosures.
• Banking and insurance supervision: RBI's eligibility criteria (CRAR, net NPA, provisioning) decide whether a bank can pay a dividend at all; IRDAI applies solvency-based norms to insurers.
• Credit analysis and rating: a high payout drains the cash cushion, so rating agencies and lenders model payout as a leakage and often impose dividend-restriction covenants.
• Mutual funds and personal investing: IDCW (Income Distribution cum Capital Withdrawal) plans versus growth plans - the payout comes out of NAV, so it is not "extra" return; SEBI renamed "dividend plan" to IDCW in 2021 precisely to stop that misreading.

Interview questions

• Q: What is dividend policy and why is it a trade-off? A: It is the decision on how much earnings to distribute versus retain; because retained earnings are the cheapest source of internal finance, every rupee paid out must either be replaced by costlier external capital or forgo an investment.
• Q: What does Walter's model conclude for a growth firm? A: When r > k, every rupee retained earns more than shareholders' required return, so value is maximised at a zero payout ratio.
• Q: How does Gordon's model differ from Walter's if the conclusions are the same? A: Walter compares r with k mechanically; Gordon adds risk aversion - the "bird-in-the-hand" view that uncertain future dividends are discounted more heavily - and expresses growth as g = b × r.
• Q: Explain MM's dividend irrelevance in one line of logic. A: Paying a dividend forces the firm to raise an equal amount of new equity for the same investment, and the resulting dilution exactly offsets the cash received, leaving total shareholder wealth unchanged.
• Q: Which MM assumption breaks first in India? A: Taxes and flotation costs - dividends are taxable at the shareholder's slab rate after DDT's abolition, while issuing fresh equity carries real costs, so retention and payout are not perfect substitutes.
• Q: Why do companies rarely cut dividends? A: Because of signalling - a cut is read as a statement about sustainable earnings and usually triggers a price fall - which is why managers smooth dividends, as Lintner's model describes.
• Q: Difference between a bonus issue and a stock split? A: A bonus issue capitalises reserves and leaves face value unchanged; a stock split sub-divides face value and leaves capital and reserves untouched. Neither involves cash.
• Q: Can a company pay a dividend in a year it makes a loss? A: Yes, out of accumulated free reserves, subject to Rule 3 - the rate cannot exceed the three-year average, withdrawal is capped at 10% of paid-up capital plus free reserves, current-year losses must be set off, and remaining reserves must stay above 15% of paid-up capital.
• Q: Why might a board prefer a buyback to a special dividend? A: A buyback carries no commitment to repeat, lifts EPS and ROE by shrinking the share count, and signals undervaluation - though since 1 October 2024 the proceeds are taxed as deemed dividend, which has removed most of its tax advantage in India.
• Q: Company A is profitable but pays no dividend; is that bad? A: Not necessarily - if it has reinvestment opportunities earning above its cost of capital, retention creates more value than a payout; it becomes a concern when cash piles up without productive use.
• Q: What happens to the share price on the ex-dividend date? A: It typically falls by about the dividend per share, since a buyer from that date is no longer entitled to the dividend.
• Q: What is the clientele effect and what does it imply for changing policy? A: Investors self-select into firms whose payout suits their income and tax needs, so a sharp change in policy drives away the existing clientele and causes transitional price pressure even if the change is value-neutral.

Common mistakes

• Applying the dividend rate to market price or book value instead of face value - a 200% dividend on a ₹1 share is ₹2, not ₹2 per ₹100 of price.
• Using Gordon's formula when b × r ≥ k; the result goes negative or absurd, and the correct answer is to state that the model's k > g condition is violated.
• Forgetting that Walter's and Gordon's r and k must be in the same form (both decimals or both percentages) and that Walter uses E − D, not E, in the retained-earnings term.
• Treating a bonus issue or a stock split as a gain in wealth - the holding and total value are unchanged; only the number of shares and price per share change.
• Saying DDT is still payable by the company; it was abolished with effect from 1 April 2020 and dividends are now taxed in the shareholder's hands.
• Confusing declared with paid - the 30-day clock under Section 127 and the liability to directors starts from declaration, not from the board's recommendation.
• Assuming book profit means cash - a company can be profitable, legally eligible, and still unable to pay a dividend for want of liquidity.

Related topics

Cost of capital and capital structure · Sources of finance: equity, debt and hybrids · Time value of money · Capital budgeting · Mergers, acquisitions and valuation basics · Mutual funds and personal investing