Forecasting and terminal value: valuing the years you cannot forecast

A business is meant to last, and a valuation has to put a price on all of its future cash. But no one can forecast a company's sales, margins and investment for fifty years, and a forecast that tries gets less trustworthy with every year it adds. The usual answer is to split the future in two. The first few years, commonly five to ten, are forecast one at a time. Everything after the last forecast year is valued in a single step, called the terminal value, using a simple rule for how the cash flow behaves from then on.

The split solves one problem and creates another. Because the terminal value stands for all the years beyond the forecast, it is usually the largest piece of the total, and the assumptions behind it, two numbers, carry most of the answer. This lesson shows how it is calculated, why it is so large, and how to treat it with proper suspicion. It uses a round figure of $100 billion of free cash flow to the firm today so the effects are easy to see, and returns to Apple at the end.

In the first part, free cash flow to the firm in each year is forecast, then discounted to today at the cost of capital, as in the lesson on present value. Here, cash flow grows at a steady rate for five years. Each year's cash is divided by (1 + the rate) raised to the number of years.

The first line is the formula for a growing perpetuity, from the lesson on annuities and perpetuities, applied to the cash flows from year N + 1 onward. The second line brings that lump sum back from year N to today. The result is added to the present value of the explicit forecast, and the total is the enterprise value of the business.

A worked case. With 6% growth for five years, a year 5 cash flow is $100 billion × 1.06⁵ = $133.8 billion. At a discount rate of 9% and long-run growth of 3%, the terminal value at year 5 is $133.8 × 1.03 ÷ 0.06 = $2,297.3 billion. Brought back five years at 9% it is worth $1,493.1 billion today.

The discount rate must be higher than the growth rate. If growth were equal to or above the rate, the formula would divide by zero or by a negative number: cash flows would grow as fast as they are discounted, and their sum would never settle. In practice the gap decides the answer. At 9% and 3% the terminal value is 17.2 times the next year's cash flow; at 9% and 4% it is 20.8. The calculator keeps the rate at 6% or more and growth at 4% or less so the gap never closes.

Growth after the forecast is also limited by the economy. A business cannot grow faster than the economy it lives in for ever, because it would eventually be larger than the economy. Sources that discuss the method commonly put the long-run rate at 2% to 4%, and in the range of what the economy might grow in nominal terms. A figure above that is a claim that the company will take an ever-larger share of everything, and needs a reason.

Set the growth in the five forecast years, the long-run growth after year 5, and the discount rate. The left page shows the explicit forecast; the right page shows the terminal value and how much of the total it makes up. The business starts with $100 billion of free cash flow to the firm. Cash arrives at the end of each year.

The table repeats the starting case, 6% growth for five years from $100 billion, with different discount rates and long-run growth. Each cell shows the enterprise value in billions of dollars and, in brackets, the share that comes from the terminal value.

The other common way to find a terminal value is to assume the business could be sold at the end of the forecast for a multiple of its cash flow or profit. The terminal value is then the multiple times the year N figure. The two methods describe the same thing from different sides. At 9% and 3%, the perpetuity formula already implies a multiple of 17.2 times the next year's cash flow, which is 17.2 times the year 5 cash flow. Going the other way, a multiple of 20 times cash flow at a 9% discount rate implies long-run growth of 3.8% a year for ever.

The check is useful in both directions. A perpetuity assumption that implies an exit multiple far above anything the company or its industry has traded at is suspect. A multiple taken from today's market for a business growing fast, applied to a business that by year 5 should be growing slowly, imports growth that will not be there.

Apple's free cash flow to the firm in fiscal 2025 was $86,264 million, from the previous lesson. Its enterprise value, at the market value of its shares held outside the company on 28 March 2025 plus debt less cash and marketable securities at fiscal year end, was $3,219,668 million. That is 37.3 times its free cash flow to the firm. In the worked case, the terminal value is a multiple of 17.2 times the year 5 cash flow.

Since the market already values Apple at 37.3 times this year's cash flow, a terminal multiple of 17.2 in five years can only support today's value if cash flow has grown a great deal by then, or if the terminal assumptions are more generous than 9% and 3%. The next two lessons use that gap. One builds a full valuation of Apple and sees what it comes to; the other turns the market value around to ask what growth it assumes. Neither says Apple is cheap or dear. They show what each opinion about it requires, which is a better start than the opinion.

Most of the value in a discounted cash flow model comes from the terminal value, the value of everything after the explicit forecast. The table shows how much, with an illustrative set of assumptions applied to a business with Microsoft's latest after-tax operating profit of $105,869 million: five years of growth at the rate in the first column, then 3% a year for ever, a cost of capital of 9%, and the share of profit reinvested in the first five years set at Microsoft's actual 33.8%. The terminal value assumes a return on new investment of 20%. The table is an illustration of the method and not a valuation of Microsoft.

The terminal growth rate is the rate at which the cash flow grows for ever, and it cannot exceed the growth of the economy in which the business operates. A forecast in a currency with higher inflation has a higher ceiling in nominal terms, and one in a low-inflation currency a lower ceiling. The example is hypothetical. If the long-run nominal growth in dollars is 3.0% with inflation of 2.5%, then in a currency where inflation is 0.5%, the same real growth is a nominal 0.99%. A yen forecast with a terminal growth rate of 3% would be assuming real growth of about 2.5% a year for ever, which is a demanding assumption.

A common rule of thumb has a practical form: set the terminal growth in the same currency as the forecast, at or below the risk-free rate in that currency, because that rate reflects expected inflation and growth. A multinational that earns in many currencies should keep each currency's cash flows separate, or convert everything into one currency using forecast exchange rates that respect these inflation differences.