This is a segment from The Breakdown newsletter. To read full editions, subscribe.
“The value of any stock, bond or businesstoday is determined by the cash inflows and outflows … that can be expected to occur during the remaining life of the asset.”
— Warren Buffett
When Warren Buffett explained that the value of every stock, bond or business is a function of its future cash flows, he was paraphrasing John William Burr, who formally articulated the “discounted cash flow” (DCF) method of valuation in his 1938 book The Theory of Investment Value.
Burr didn’t invent the ideas behind his calculations, however.
The first recorded instance of discounting future cash flows to present value dates to 1773, when an engineer, Edward Smith, estimated the “present worth” of a coal mining operation based on its future stream of income.
In fact, the logic of present value has been used even longer than that.
With the publication of his Liber Abaci in 1202, Leonardo Pisano (better known as Fibonacci) is credited asthe “first to develop present value analysis for comparing the economic value of alternative contractual cash flows.”
823 years later, that’s still the best way to do it: Discounted cash flow is as close as you can get to a fundamental truth in the art of financial valuation.
It rarely feels that way, however, which is confusing — the best argument against DCF is that it doesn’t match our lived experience as investors.
Stocks, for example, trade up, down and all around for reasons that are entirely unrelated to future cash flows — even in hindsight, it’s impossible to discern any cash-flow logic in the price charts.
John Maynard Keynes is right to call stock market investing a “beauty contest” that’s unrelated to fundamentals — and there’s no disputing Robert Shiller’s findings that stock prices are far more volatile than can be explained by subsequent changes in dividend payments.
But I draw the line at Austrian economist Carl Menger’s conclusion that “value does not exist outside the consciousness of men.”
Inferring value from price movements (and human behavior) is like concluding the stars revolve around Earth — it only looks that way from where we stand.
It’s true that DCF can’t be proven in a mathematical sense — it isn’t a physical law of the universe.
But it’s as close to a law as we can get in finance, however unrelated it is to how asset prices move in the real world.
We can pseudo-prove this with a thought experiment.
Imagine, for example, you were asked to bid for two assets: Asset A, which is guaranteed to pay holders $10 a year for the next 10 years, and Asset B, which may or may not pay you an unknowable amount of money at some unknowable time.
Which would you offer money for?
Asset A, of course, because you could calculate how much to offer by putting its predictable stream of future cash flows into an excel model and discounting them back to a present value.
You could do that for Asset B, too, except that you’d have to guess at all the model’s inputs, which makes it much less useful in practice.
In theory, however, it’s still the mental model to use for understanding why financial assets have value.
Asset A in the above example is a bond, of course, and Asset B is a stock — which is why people rarely use DCF models for stocks.
DCF models require too many guesses about things that will happen too far in the future (out to 10 years, typically, and even that’s too short to really know).
Instead, we like to look one, two or three years out, guess at just one thing — say, earnings — and pick a somewhat arbitrary multiple of that guess to arrive at a price target.
But a price target is a forecast of where a stock will trade, not what it’s worth.
Valuation metrics like price-to-earnings, price-to-sales, price-to-EBITDA — these are all stand-ins for the unknowable discounted cash flow valuation of a stock.
However you want to think about it, what any stock is worth remains the present value of its future cash flows, which are annoyingly unknowable.
This is why hardly anyone actually values stocks this way — not even Warren Buffett builds DCF models.
“Warren often talks about these discounted cash flows,” Charlie Munger once noted, “but I’ve never seen him do one.”
That’s because Munger and Buffett both worried that the “false precision” of a DCF model was likely to lead them astray — if the value of an investment “doesn’t just scream out at you,” Buffett says, you shouldn’t invest.
In other words, you should know that the value of a financial asset is a function of its future cash flows — but you shouldn’t try to calculate it.
For investors, DCF is less of a practical playbook and more of a state of mind.
But crypto is different, right?
On the most recent episode of the Empire podcast, crypto investor Tom Dunleavy predicted that the fees people pay to use layer-1 blockchains like Ethereum “are effectively going to zero.”
That made me double-check the title of the podcast — “The Bull Case for Ethereum” — because that sounds like the ultimate bear case for a cash flow-minded investor like myself.
But Dunleavy thinks I’m doing it wrong: When valuing a layer-1 token like ETH, “revenue and cash flow is the exact wrong way to think about it.”
Instead, L1s should be valued “as a percentage of assets that will come onchain at some point in the future.” (“Theoretically,” he adds.)
He might be right about that, I don’t know — and he might also be right that the price of ETH will be correlated to the amount of assets Ethereum hosts (although that remains speculative).
But to argue that the value of ether is based on anything other than its future cash flows is to argue that ether is exempt from Buffett’s dictum about stocks, bonds and businesses.
Lots of people think it should be.
The “Ethereum investor” Ryan Berckmans, for example, argues the case for ether by refuting the idea that it’s a business selling blockspace: “What Ethereum sells is freedom and global prosperity.”
Maybe so — but how else would you measure an investment in freedom and prosperity if not by forecasting how much people will pay to access it?
Berckmans may have implicitly conceded the point later in the podcast: “Value accrual will happen, yes, through fees in the long term.”
That about sums things up for me: Buy a crypto token for any reason, sure — freedom, prosperity, vibes — but the only way to invest in a token is on the assumption that its current price is justified by its future cash flows.
