The map · Part One · The method · Chapter 7 of 23

The reverse DCF

Do not ask what a business is worth. Ask what the price is assuming, and whether you believe it. Same arithmetic, run backwards, and a different kind of question comes out. About 10 minutes · on the 10-minute route.

Here is the usual way to value a business. You guess how fast its cash flow will grow. You add up the cash you expect it to throw off for the next ten years, discounted because money later is worth less than money now, and add something for the years after that. Out comes a number. You compare the number to the price. That procedure has a name, discounted cash flow, usually shortened to DCF.

It has a well-known problem. The number that comes out depends almost entirely on the growth you guessed, and the growth you guessed depends, if you are honest, on what you already wanted the answer to be. "What is it worth?" is a question that invites you to deceive yourself, because any answer can be reached by choosing the guess.

The inversion

Now run the same arithmetic backwards. Start from the price the market is quoting today. Ask: what growth rate would the business need, for the next ten years, for that price to be exactly fair? The arithmetic is identical; you are solving for a different unknown. Out comes not a value but an assumption: the growth the market is currently betting on.

That number can be judged. "Is this business going to grow its cash at 14 percent a year for a decade?" is a question a person who has read the reports can have an opinion about. It has a shape. You can look at what the company has done for the last ten years, what its market is doing, what it would have to sell to get there, and say: I do not believe that. Or: I believe more than that. "What is it worth?" is unanswerable. "Do I believe that?" is a judgement, and judgements are what people are for.

An analogy

A used car is advertised at a price. You could try to work out what the car is "worth", and you would find that the answer depends on how many years of trouble-free driving you assume. Or you could ask the other question: at this price, how many years of trouble-free driving is the seller asking me to believe in? If the price only makes sense if the car runs another twelve years without a repair, and it has 200,000 kilometres on it, you have your answer, and you got it without ever needing to know what the car is "worth".

Try it

NAME A produces 10 per share in spare cash a year. Fix the arithmetic the way this project fixes it: ten years of growth, then a terminal growth of 2.5 percent a year, all discounted at 9.5 percent. The only free number is the growth rate. Here is what various prices are assuming.

Read the table from the right. At 100, the price assumes the cash barely grows. At 250, it assumes growth that very few businesses sustain for a decade. The method does not tell you which is true. It tells you what you would have to believe, and hands the believing back to you.

How this project fixes the arithmetic, and what it admits

The inputs are the twelve-month spare cash from the filings, the company's net debt, and a share count the company has stated. The horizon is ten years, the growth after that 2.5 percent. The discount rate is 9.5 percent, and the method is unusually candid about what that number is: not an estimate of what the market demands, but the owner's own hurdle, set at what an index fund is expected to return over the long run plus two to three points for carrying the risk of one company. It is the same for every company, which knowingly demands a little more of a Swedish business than an American one, and the report prints the bias rather than hiding it. In the ruling's own words, the rate "will never be verified against anything. It is a preference and says so."11Ruling E29. The engine: vss/valuation.py. Every buy price is printed with its value at a rate half a point lower and half a point higher, so the reader sees how much the number leans on the preference.

Where the idea comes from

Reading the growth embedded in a price, rather than forecasting growth to find a price, was worked out and popularised by Alfred Rappaport and Michael Mauboussin in Expectations Investing, first published in 2001. Their summary of the method is two steps: read the price, then anticipate the revision.22Rappaport and Mauboussin, Expectations Investing (Harvard Business School Press 2001; revised edition Columbia 2021). Their site states the method as 'Read the price. Then anticipate the revision.' Aswath Damodaran teaches the same inversion under the name implied growth. This project borrows the inversion whole. Where it departs is in what happens next: Rappaport and Mauboussin compare the market's assumption with the analyst's; this project requires the analyst's view to be on record before the market's number is even calculated. That is the next chapter.

The exact rule

Framework §5 as rebuilt by rulings E28 and E29: the reverse DCF is the sole engine; two older methods, comparing to past multiples and to peers, are shown beside it as context and never averaged in. Inputs: filed twelve-month free cash flow, net debt, a stated share count. Horizon ten years, terminal growth 2.5 percent, hurdle 9.5 percent, solved by bisection for the growth rate that makes the equity value per share equal the price. Code: vss/valuation.py, function implied_growth.

Sources for this chapter

1. Rappaport and Mauboussin, Expectations Investing (book site)
   https://www.expectationsinvesting.com/
2. The reverse DCF engine and the hurdle rate — read from this repository: vss/valuation.py; reference/FRAMEWORK-EDITS.md E28, E29

The slow way, made survivable. Built 2026-09-13 from the repository as it stood that day. NAME A and NAME B are fictional; no company, price or holding appears on this site. Not financial advice: one private investor's method, shared as it is.