The standard story told about Microsoft Corporation is that it became three different businesses in twenty-five years. There is some truth to this. The Microsoft of 2000 sold a single product, Windows, to a captive audience. The Microsoft of 2010 was a mature franchise with a faintly desperate Bing strategy and a deteriorating relationship with the consumer. The Microsoft of today is a cloud-computing utility with a sideline in office productivity that no one has ever quite replicated. The reinventions were not gentle. They were also, mostly, the work of one CEO.

What matters for the protocol is none of that. What matters is that Microsoft has, for the entirety of the past decade, generated more cash than it can spend on the businesses it already runs. It pays out the surplus in dividends and buybacks. The cash is real. The dividend coverage is more than three times. The growth in operating income for the trailing five years averages something just shy of twenty percent.

The cash is real. The dividend coverage is more than three times.

The protocol takes the five forward years of analyst-consensus EPS — which currently land between thirteen and eighteen percent annual growth — averages them, caps at twenty percent, and produces G1 of roughly nineteen-and-a-half percent. The terminal G2 is two-point-three, anchored to U.S. real GDP. The discount rate, with the technology sector premium of two percentage points layered on top of the 10-year Treasury, comes out around six-and-a-half. Owner Earnings of one hundred and two billion dollars compounds for ten years on this fade schedule and lands at an intrinsic-value-per-share number that is north of seven hundred.

That number is too high for our taste, frankly, and we will tell you why. The protocol's growth input is forward analyst consensus, and the forward analyst consensus on Microsoft is, as forward analyst consensus on Microsoft has consistently been, optimistic. We apply a two-percentage-point risk premium for technology already; we do not apply a second one for “analysts get this name wrong on the upside.” If you wanted to, you could. At a 9% discount instead of 6.5%, the intrinsic-value-per-share comes down to something in the high-three-hundreds, well within twenty percent of today's asking price.

The interesting question is which version of that protocol you believe. Ours says Mr. Market is offering Microsoft below value. A more conservative analyst — one who thinks the cloud-computing revenue line will start to slow in 2027 or 2028, which is a perfectly sensible thing to think — gets a flatter answer. We publish the optimistic one because Mr. Market's protocol, as written, says you take the analyst consensus and apply the sector premium and produce the number. The discipline is consistency, not nuance. Nuance is what comes next, on the subscriber side, where we publish our own discount-rate sensitivity work alongside the protocol's baseline.

For now, on Microsoft, today: Δ%, by the protocol, is meaningfully positive. By a more cautious read, less so. Either way, the business itself is doing the thing it has done for a decade — compounding cash at a rate that very few enterprises in history have managed. The patient owner does not need a verdict today. The patient owner needs to know what the business is doing. The business is, as ever, doing fine.

— The editors