Integrated Dividend Discount Valuation Workshop

From a Dividend Forecast to a Stock Value

A dividend discount valuation turns a forecast of shareholder cash flows into an estimate of value today. In a multistage case, the forecast may have several years of changing growth before dividends settle into a stable long-run pattern. The calculation is manageable when each dividend is assigned to the right year, the terminal value is placed at the end of the explicit forecast, and every cash flow is discounted to the same valuation date.

The central modeling judgment is not just arithmetic: a stable terminal growth rate must be plausible for a continuing business, and it must be below the required return under the usual Gordon model assumptions. Small changes in that long-run growth assumption can have a large effect on estimated value, so the final calculation should be tested rather than treated as a precise prediction.

Definition: Let $D_t$ be the dividend paid at the end of year $t$, and let $k_e$ be the annual required return on equity. A multistage dividend discount model estimates the stock value $P_0$ by discounting forecast dividends and the value of dividends beyond the explicit forecast period. If the explicit forecast ends at year $T$, and stable growth begins after year $T$, the terminal value $P_T$ is the value at time $T$ of dividends from year $T+1$ onward.

The Gordon Terminal Value and Its Conditions

If dividends grow at a constant rate $g$ indefinitely after year $T$, then $D_{T+j}=D_{T+1}(1+g)^{j-1}$ for $j\geq 1$. Under the usual dividend model assumptions, the terminal value is the Gordon growth value:

P_T=\frac{D_{T+1}}{k_e-g}.

The formula requires a continuing stream of dividends, a constant growth rate after year $T$, and a required return that makes the discounted dividend series converge. For the geometric series representation, the exact convergence condition is $\left|(1+g)/(1+k_e)\right|<1$, with $k_e\ne -1$. Under the usual assumptions $g>-1$ and $k_e>g$, the ratio is between zero and one, so the series converges. In ordinary applications, $k_e>g$ is therefore essential; a nonzero denominator by itself is not enough to guarantee convergence.

The model’s timing follows directly from the cash flows: $P_T$ includes dividends beginning with $D_{T+1}$, not the dividend $D_T$ already paid at time $T$. To obtain today’s value, add the explicit dividends and discount the terminal value from time $T$:

P_0=\sum_{t=1}^{T}\frac{D_t}{(1+k_e)^t}+\frac{P_T}{(1+k_e)^T}.

This formula assumes annual dividends paid at year-end and a constant annual required return for the forecast. If payment dates or required returns differ, the cash flows must instead be discounted according to their actual timing and applicable rates.

A Reliable Valuation Sequence

A complete calculation is easier to audit when forecasting, terminal valuation, and discounting are kept separate. The terminal dividend is especially important: first forecast the dividend immediately after the explicit period, then use it in the Gordon formula.

Set the timeline.
Identify the valuation date, the last explicit forecast year $T$, and the year in which stable growth starts.

Forecast each dividend.
Apply each year’s stated growth rate to the dividend from the preceding year. Do not apply a growth rate to the wrong year’s dividend.

Calculate the terminal value.
Forecast $D_{T+1}$, check the stable-growth assumptions and convergence condition, and calculate $P_T$.

Discount to time zero.
Discount every explicit dividend by its own number of years and discount $P_T$ by $T$ years.

Test the assumptions.
Recalculate value under reasonable alternative stable growth rates or required returns, keeping the dividend timing consistent.

Worked Example: Valuing a Stock Through a Growth Transition

A hypothetical stock has just paid $D_0=\$2.00$. Forecast dividend growth is 12% in year 1, 12% in year 2, and 8% in year 3. Beginning after year 3, dividends are expected to grow at a stable 4% annually. The annual required return is $k_e=10\%$. Calculate $P_0$, showing the forecast, terminal value, and discounting.

Method and timing. Forecast the three explicit dividends in sequence. Stable growth begins after $D_3$, so the terminal value is at time 3 and uses $D_4$. The common mistake to avoid is using $D_3$ in the Gordon formula or discounting $P_3$ for four years.

Forecast the explicit dividends.

\begin{aligned} D_1&=2.00(1.12)=2.24,\\ D_2&=2.24(1.12)=2.5088,\\ D_3&=2.5088(1.08)=2.709504. \end{aligned}

The first stable-growth dividend is $D_4$, so

$$ D_4=2.709504(1.04)=2.81788416. $$

Find the terminal value at time 3. Since $k_e-g=0.10-0.04=0.06>0$, and the usual assumptions $g>-1$ and $k_e>g$ hold, the Gordon series converges.

P_3=\frac{2.81788416}{0.10-0.04} =\frac{2.81788416}{0.06} =46.964736.

Discount all values to time zero. The discount factors are $1.10$, $1.10^2=1.21$, and $1.10^3=1.331$:

\begin{aligned} P_0 &=\frac{2.24}{1.10} +\frac{2.5088}{1.21} +\frac{2.709504}{1.331} +\frac{46.964736}{1.331}\\ &=2.0363636+2.0733884+2.0356905+35.2853013\\ &=41.4307438. \end{aligned}

Rounded to the nearest cent, $P_0=\$41.43$. As a check, the terminal value is the value at time 3, so it is discounted by $1.10^3$, just like a year-3 cash flow. The separate year-3 dividend is included once as $D_3$, while $P_3$ represents dividends from year 4 onward.

Why the Gordon Formula Holds

The terminal formula is the sum of the discounted dividends after year $T$. Dividing each dividend by its discount factor gives a geometric series whose first term is $D_{T+1}/(1+k_e)$. Each subsequent term is the preceding term multiplied by $(1+g)/(1+k_e)$. When the absolute value of that ratio is less than one, the infinite geometric series has sum equal to its first term divided by one minus its ratio:

\begin{aligned} P_T &=\frac{D_{T+1}}{1+k_e} \left[ 1+\frac{1+g}{1+k_e} +\left(\frac{1+g}{1+k_e}\right)^2+\cdots \right]\\ &=\frac{D_{T+1}}{1+k_e} \cdot \frac{1}{1-\frac{1+g}{1+k_e}}\\ &=\frac{D_{T+1}}{k_e-g}. \end{aligned}

The final equality follows because $1-(1+g)/(1+k_e)=(k_e-g)/(1+k_e)$. The derivation shows why the convergence condition matters: if the series ratio has absolute value at least one, the infinite sum does not converge to the stated value, even if the denominator $k_e-g$ is nonzero.

Testing the Stable Growth Assumption

The stable growth rate is an assumption about the period beyond the explicit forecast, not a rate that should be selected merely to produce a desired price. Because $g$ appears in the denominator $k_e-g$, a higher stable growth rate generally increases the terminal value, and the impact becomes especially pronounced as $g$ approaches $k_e$ from below. Scenario analysis makes this sensitivity visible.

Worked Example: Comparing Stable Growth Scenarios

A hypothetical company has forecast dividends $D_1=\$1.80$, $D_2=\$2.00$, and $D_3=\$2.10$. After year 3, dividends grow forever at a stable rate. The annual required return is 8%. Calculate the value under stable growth assumptions of 2%, 3%, and 4%, and compare the results.

Method and common mistake avoided. The explicit dividends are the same under all three scenarios. For each scenario, calculate $D_4=D_3(1+g)$, calculate $P_3=D_4/(0.08-g)$, and discount $P_3$ by $1.08^3=1.259712$. Do not use $D_3$ as the terminal numerator or discount $P_3$ by four years.

First, discount the explicit dividends.

\begin{aligned} \frac{D_1}{1.08}&=\frac{1.80}{1.08}=1.6666667,\\ \frac{D_2}{1.08^2}&=\frac{2.00}{1.1664}=1.7146776,\\ \frac{D_3}{1.08^3}&=\frac{2.10}{1.259712}=1.6670477,\\ \text{Explicit-dividend present value}&=1.6666667+1.7146776+1.6670477\\ &=5.0483920. \end{aligned}

Now calculate each terminal value and its present value. In each case $g<k_e=8\%$, and $g>-1$, so the standard stable-growth convergence condition holds.

\begin{aligned} g=2\%:\quad D_4&=2.10(1.02)=2.142,\\ P_3&=\frac{2.142}{0.08-0.02}=35.70,\\ \frac{P_3}{1.259712}&=\frac{35.70}{1.259712}=28.3398110,\\ P_0&=5.0483920+28.3398110=33.3882030;\\[4pt] g=3\%:\quad D_4&=2.10(1.03)=2.163,\\ P_3&=\frac{2.163}{0.08-0.03}=43.26,\\ \frac{P_3}{1.259712}&=\frac{43.26}{1.259712}=34.3411827,\\ P_0&=5.0483920+34.3411827=39.3895747;\\[4pt] g=4\%:\quad D_4&=2.10(1.04)=2.184,\\ P_3&=\frac{2.184}{0.08-0.04}=54.60,\\ \frac{P_3}{1.259712}&=\frac{54.60}{1.259712}=43.3432404,\\ P_0&=5.0483920+43.3432404=48.3916324. \end{aligned}

Rounded to the nearest cent, the three values are $\$33.39$, $\$39.39$, and $\$48.39$, respectively. The explicit-dividend present value stays fixed; the increase comes from the terminal value. This comparison does not establish which growth assumption is correct. It shows how much the estimate depends on the assumption and why a defensible stable rate matters.

Check the Terminal Date Before Discounting

A terminal value is located at a date; it is not itself a dividend. If explicit forecasts end with $D_T$, stable growth starts with $D_{T+1}$, and $P_T$ is discounted by $T$ years. This timing convention prevents both an omitted explicit dividend and a duplicated terminal dividend.

Worked Example: A Timing Check With a Simple Stable Period

A hypothetical stock has just paid $D_0=\$1.00$. Its year-1 dividend grows by 20%, its year-2 dividend has zero growth, and dividends grow at 0% forever after year 2. The annual required return is 10%. Find $P_0$, and explain why discounting the terminal value by three years is incorrect.

Method and common mistake avoided. Treat year 2 as the last explicit year. Stable growth begins with $D_3$, so calculate $P_2$ from $D_3$ and discount that value by two years. The error to avoid is treating a time-2 value as though it were located at time 3.

The explicit dividends are

D_1=1.00(1.20)=1.20,\qquad D_2=1.20(1.00)=1.20.

The first dividend in the stable period is $D_3=1.20(1.00)=1.20$. The stable-growth assumptions hold: $g=0\%$, $k_e=10\%$, and $k_e>g$.

P_2=\frac{D_3}{k_e-g}=\frac{1.20}{0.10-0}=12.00.

Discount the year-1 dividend by one year, and discount both the year-2 dividend and terminal value by two years:

\begin{aligned} P_0 &=\frac{1.20}{1.10}+\frac{1.20}{1.10^2}+\frac{12.00}{1.10^2}\\ &=1.0909091+0.9917355+9.9173554\\ &=12.0000000. \end{aligned}

Thus $P_0=\$12.00$. The terminal value is at time 2 because it values dividends from year 3 onward. Discounting it by three years would move it one period too far into the past and understate the value.

Common Errors and Practical Interpretation

A multistage value depends on both the forecast and the valuation conventions. The following checks catch errors that can materially change a result:

Using the wrong terminal dividend. If stable growth starts after year $T$, the Gordon numerator is $D_{T+1}$, not $D_T$.
Misplacing the terminal value. A terminal value calculated at time $T$ is discounted by $T$ periods, not $T+1$.
Double counting. Include each explicit dividend once; the terminal value includes only dividends after the explicit forecast ends.
Ignoring convergence. Check $\left|(1+g)/(1+k_e)\right|<1$. Under the usual assumptions, verify $g>-1$ and $k_e>g$, rather than checking only that the denominator is not zero.
Reporting false precision. Forecasts of long-run growth and required return are assumptions. A value with many decimals is not evidence that those assumptions are known precisely.

A useful final check is to reconcile the result as the sum of two components: the present value of the explicit dividends and the present value of the terminal value. Then perturb the stable growth rate or required return while holding the remaining inputs fixed. If the estimate changes sharply, that sensitivity is part of the valuation conclusion, not a reason to conceal the alternative results. These calculations value the equity cash flows represented by dividends; they should not be confused with measures based on operating cash flows or total firm value.

Key takeaway: Forecast the dividend immediately after the explicit period, value that stable-growth stream at the end of the explicit period, and discount every component from its own date. Confirm convergence and test alternative assumptions before interpreting the result.

Check Your Understanding

Work each problem from the stated timing and assumptions. Amounts are hypothetical.

A stock has just paid $D_0=\$1.60$. Dividend growth is 10% in year 1 and 5% in year 2. If $k_e=9\%$, find $D_1$, $D_2$, and the present value of those two dividends. (Answer: $D_1=\$1.76$, $D_2=\$1.848$, present value $\approx\$3.17$.)
A stock has just paid $D_0=\$2.00$, and its year-1 dividend grows by 10%. From year 2 onward, dividends grow at 2% forever. If $k_e=8\%$, calculate the terminal value at time 1 and $P_0$. (Answer: $P_1=\$37.40$, $P_0=\$36.67$.)
A stock has just paid $D_0=\$1.50$. Growth is 12% in year 1, 8% in year 2, and 4% thereafter. If $k_e=9\%$, calculate the value at time 2 and then $P_0$. (Answer: $D_1=\$1.68$, $D_2=\$1.8144$, $D_3=\$1.886976$, $P_2=\$37.73952$, $P_0\approx\$34.83$.)
A stock has just paid $D_0=\$2.00$. Its next dividend is $\$2.20$, and from the following year onward dividends grow at a stable rate. Calculate $P_0$ at $k_e=8\%$ for stable growth of 2%, then for 3%. (Answer: at 2%, $D_2=\$2.244$, $P_1=\$37.40$, and $P_0=\$36.67$; at 3%, $D_2=\$2.266$, $P_1=\$45.32$, and $P_0=\$44.00$.)
A model forecasts dividends through year 3, then assumes stable growth of 8% forever with $k_e=7\%$. Is the Gordon terminal value valid under the standard dividend-model assumptions? State the convergence check. (Answer: No. The ratio is $(1.08)/(1.07)>1$, so its absolute value is not less than 1; the infinite series does not converge.)

What You'll Learn