4. The three parts of a valuation, and which weighs

Level 1: The price is a forecast

Open any discounted cash flow valuation and take it apart, and you will always find the same three pieces. Knowing which of them carries the weight of the price completely changes where you should concentrate your skepticism, and explains why arguing about next quarter is almost always arguing about crumbs.

The three pieces a price breaks into

Every valuation of a company is made up of:

• The explicit period: the present value of the cash flows projected year by year, usually between five and ten years.
• The residual value: the present value of everything that comes after the last projected year, condensed into a single figure.
• Net cash: the cash and financial investments the company holds, less the debt it owes.

The first two add up to what is called the value of the business. Add the third and you arrive at the value belonging to shareholders. And the split between the three pieces is far less even than almost anyone suspects.

Part one: the explicit period, the only one anybody argues about

It is the visible part of the work: the sheet where somebody writes down sales, margins and investment year by year. It takes up ninety percent of an analyst's effort and of the argument on any forum.

Let us see how much it weighs. Take a business that will generate 105 next year, growing 5% a year for five years, discounted at 8%. Adding up the five flows brought back to today, the explicit period is worth around 460.

Part two: the residual, where almost everything lives

At the end of the fifth year the company does not vanish. Suppose that from there on it grows 2.5% a year for ever. The value of that tail, sitting at year five, is the year-six cash flow divided by the difference between 8% and 2.5%: some 2,378. Brought back to today, around 1,619.

Total: some 2,079. The residual accounts for 78% of the value. In other words, nearly four fifths of what you pay depends on assumptions about what happens after the fifth year, territory in which nobody has any information and about which almost nobody argues.

The immediate temptation is to lengthen the explicit period. Let us do it: ten years instead of five. Now the explicit period is worth some 859 and the residual some 1,406, out of a total of 2,265. The residual drops to 62%. It looks like an improvement, and it is not: the years six to ten you have added are pure forecast, every bit as made up as the residual they replace. You have not reduced your dependence on the distant future; you have relabeled it.

For a mature S&P 500 company it is normal to find between 65% and 85% of the value in the residual on a five-year horizon; in growth companies with no profits today, the figure comfortably clears 90%. Damodaran has spent decades repeating that this is where valuations are won and lost, and where the abuses hide: type in perpetual growth of 6% when the nominal economy grows at 4%, and the company ends up, mathematically, larger than the world that contains it. Perpetual growth cannot sustainably exceed the nominal growth of the economy, and that limit is the first check to run on anybody else's valuation.

Part three: net cash, the only part you can check

Against two parts made of assumptions, the third is read straight off the balance sheet. It is money that is either there or not, and debt that is either owed or not. Subtracting it or adding it changes the analysis more than it looks, because the multiple everybody quotes is worked out on the total market cap, mixing the business up with the piggy bank.

And there is a second-order effect worth registering: debt amplifies the expectations gap. If the business is worth 2,079 and there is 1,000 of debt, shareholders are left with 1,079. A 10% error in the valuation of the business, some 208, becomes a 19% error for the shareholder. The same surprise, twice the consequences.

Real case: Apple in 2013

At the close of its 2013 financial year, Apple reported around $147 billion in cash and investments, against some $17 billion of debt issued that same year. Net cash: around $130 billion. Its market cap was then in the region of $450 billion and its net profit for the year some $37 billion.

The headline said Apple traded at about 12 times earnings. The correct sum is a different one: strip out the net cash and the operating business cost some $320 billion, that is, less than 9 times its profit, while the index traded well above that. Translated into expectations: the price was not discounting growth, it was discounting deterioration.

That was the real debate of 2013, and it is where the public battle between the company and several activist investors over what to do with the piggy bank was fought. None of it was visible from the published multiple.

What you take away from this level

• A price is a closed list of assumptions about sales, margins, capital and discounting.
• Getting the quarter right gives you no edge: the consensus is already inside the price.
• Return comes from the gap between what happens and what was discounted.
• Between 60% and 90% of the price lives after the last year anybody bothers to project.
• Net cash is the only piece you can verify, and it changes the multiple everybody repeats.