The Lintner model (John Lintner, 1956) is a simple, influential framework for understanding how companies set and change cash dividends over time. Lintner showed—based on inductive study of 28 large, public manufacturing firms—that firms behave as if they have a long‑run target dividend (or target payout ratio) but adjust actual dividends only partially toward that target each period. The model explains the common empirical patterns of dividend stability and gradual change (so‑called “dividend smoothing”).
Key idea (plain language)
– Firms have a target dividend (usually derived from a target payout ratio applied to earnings).
– When earnings change, boards do not immediately move dividends fully to the new target; they close only a fraction of the gap each period.
– The fraction they close (the partial adjustment coefficient) governs the speed of change: the higher it is, the faster dividends move to the target.
Canonical formula
D_t = k + PAC × (TD_t − D_{t−1}) + e_t
where
– D_t = dividend in period t
– D_{t−1} = dividend in period t−1
– TD_t = target dividend in period t (commonly TD_t = θ × Earnings_t)
– PAC = partial adjustment coefficient (0 < PAC < 1)
– k = constant (intercept)
– e_t = error term
Equivalently:
D_t = (1 − PAC)·D_{t−1} + PAC·TD_t + k + e_t
Typical empirical interpretation (Lintner’s findings)
– Firms set a target payout ratio and prefer relatively stable dividends.
– When earnings change, boards adjust dividends gradually; Lintner’s estimates suggested PAC ≈ 0.3 (i.e., firms, on average, adjust about 30% of the gap per year), and target payout ratios were materially less than 100%.
– The model is descriptive (explains observed behavior) but can be used prescriptively.
Why the model matters
– Explains dividend smoothing and the persistence of dividends.
– Provides a simple forecasting tool for analysts and investors who want to estimate likely dividend changes.
– Gives corporate managers a quantitative way to phase in dividend changes without abrupt shocks.
How to apply the Lintner model — practical steps
1) Define the target dividend (TD_t)
– Most applications let TD_t = θ × Earnings_t, where θ is the target payout ratio.
– Decide whether to use net income, free cash flow, or another earnings measure based on your company and industry.
2) Collect data
– Time series of annual (or quarterly) dividends per share (D_t) and the earnings measure (E_t) for a reasonably long period (10+ years preferred).
– Note any one‑time adjustments, stock splits, share repurchases, or capital structure changes that could distort dividend history.
3) Choose a specification for estimation
Two common regression forms:
A. Delta form (closely follows Lintner’s statement)
ΔD_t = k + PAC × (TD_t − D_{t−1}) + e_t
If TD_t = θE_t, then ΔD_t = k + PAC·θ·E_t − PAC·D_{t−1} + e_t.
B. Levels form (convenient for ordinary least squares)
D_t = a + b·E_t + c·D_{t−1} + error_t
From the levels form, you can recover:
– PAC = −c (because the Δ form implies coefficient on D_{t−1} is −PAC),
– PAC·θ = b ⇒ θ = b / PAC (if PAC ≠ 0).
4) Estimate the parameters
– Run the regression (OLS) using historical data.
– From coefficients, compute PAC and θ (target payout). For example, if the estimated c = 0.7 then PAC = 1 − c = 0.3; if b = 0.15 and PAC = 0.3 then θ = 0.15 / 0.3 = 0.5 (target payout 50%).
– Check t‑statistics, R², and residuals for model fit and serial correlation. Consider using robust standard errors if heteroskedasticity is present.
5) Forecast dividends
– Given current earnings forecast E_t and last dividend D_{t−1}, compute TD_t = θ·E_t, then:
D_t = D_{t−1} + PAC·(TD_t − D_{t−1}) + k.
Illustrative numerical example
– Suppose PAC = 0.3, θ = 0.4 (target payout 40%), D_{t−1} = $30m, Earnings_t = $100m, and k=0.
– TD_t = 0.4 × $100m = $40m.
– Adjustment = 0.3 × (40 − 30) = $3m.
– New dividend D_t = 30 + 3 = $33m.
Practical advice for managers (how to use the model as policy)
– Set a clear target payout ratio (θ) reflecting long‑run cash‑flow expectations, investment needs, leverage policy, and shareholder preferences.
– Use a partial adjustment rule to phase in increases or decreases: choose a PAC that balances credibility (higher PAC → faster moves) with flexibility (lower PAC → more smoothing).
– Maintain a buffer of retained earnings or cash to avoid forced cuts; the Lintner model’s appeal is its explanation for why firms avoid frequent dividend cuts.
– Communicate policy: investors value clarity about targets and smoothing behavior.
– Consider repurchases alongside dividends: repurchases can be used to return excess cash flexibly while dividends maintain stable cash flows.
Practical advice for analysts and investors
– Estimate a firm’s PAC and θ using its history to forecast likely dividend changes after earnings surprises.
– Use the Lintner model to flag firms where dividend policy is inconsistent with earnings trends (e.g., a firm paying dividends well above its likely target may be at risk of cuts).
– Remember that non‑dividend payouts (buybacks) complicate interpretation; since buybacks have grown in importance, consider total payout when possible.
Limitations and caveats
– Sample bias: Lintner’s original sample was 28 large manufacturing firms; parameter estimates vary across industries, firm size, and eras.
– Model simplicity: it abstracts from taxes, agency problems, dividend signaling theory, liquidity constraints, leverage targets, and repurchases.
– Structural changes: firms may change target payout ratios over time; the model assumes a relatively stable long‑run target.
– Endogeneity and statistical issues: earnings are persistent and may be endogenous; naive OLS estimates can be biased. More sophisticated econometric approaches (instrumental variables, GMM) may be warranted for research quality work.
– In recent decades, share repurchases have become a major return channel, complicating pure dividend analysis.
Empirical extensions and evidence
– Lintner (1956) provided the original descriptive evidence of target payout and partial adjustment.
– Subsequent empirical work confirms dividend smoothing is widespread but shows wide cross‑sectional variation in PAC and target payout.
– Industry and country studies (for example, Raju & Rane 2018 for Indian metal companies) find the model often fits but requires adjustments for local corporate behavior, market regulations, and the role of buybacks.
References and further reading
– Lintner, John. “Distribution of Incomes of Corporations Among Dividends, Retained Earnings, and Taxes.” The American Economic Review, vol. 46, no. 2, 1956, pp. 97–113.
– Investopedia. “Lintner Model.” (source URL you provided).
– Raju, Guntur Anjana and Rane, Anjali. “Dividend Smoothing & Implications of Lintner’s Model: An Empirical Analysis of Indian Metal Section.” Inspira-Journal of Commerce, Economics & Computer Science, vol. 4, no. 1, 2018, pp. 41.
– Harvard Business School. “John V. Lintner Papers: Finding Aid.”
– New York Times. “John V. Lintner Jr., 67, Dies; Harvard Business Professor.”
Summary checklist (quick practical steps)
– Collect dividends and earnings series; clean data for structural breaks.
– Specify TD_t = θ × Earnings_t (or another target rule).
– Estimate regression in levels or delta form to obtain PAC and θ.
– Use PAC to forecast how quickly dividends will move to the target after earnings changes.
– Consider corporate context (cash, debt, buybacks, taxes) before applying the model mechanically.
– Communicate policy clearly if you are a manager; incorporate flexibility tools (repurchases) as needed.
– Walk through a worked regression example with a sample dataset (showing how to estimate PAC and θ step by step).
– Help you adapt the model to incorporate buybacks or free cash flow instead of earnings.
– Provide an Excel template that computes PAC, θ, and forecasts dividends from historical data.