Bill Gates

Bill Gates is an American software entrepreneur, born in Seattle in 1955, who co-founded Microsoft with Paul Allen in 1975 and led it as chief executive for twenty-five years. His significance for an investor is not the scale of the company but the clarity of the economics underneath it. Software has almost no marginal cost, which means the second copy sold is nearly pure margin, and Microsoft is the clearest demonstration of what happens when that property is combined with a position everyone else has to build on top of.

The decision that made the rest possible was contractual rather than technical. When IBM came to Microsoft in 1980 for an operating system for its forthcoming personal computer, Microsoft did not have one. It licensed, and later bought outright, 86-DOS from Seattle Computer Products, and delivered it to IBM as PC DOS for a one-time fee. Crucially, the arrangement was not exclusive: Microsoft kept the right to license the same system to other manufacturers. When the IBM PC design was cloned, every clone maker needed an operating system, and Microsoft owned the one that ran the software people already had.

That is the platform pattern in its purest form. Application developers wrote for the system with the most users, and buyers chose the system with the most applications, so each reinforced the other and the position became progressively harder to attack. Microsoft went public in 1986 and spent the following decade extending the same logic from the operating system into applications, where Office repeated the pattern at the document-format level.

The position also brought the most significant antitrust case of the era. The Department of Justice and a group of state attorneys general sued in May 1998 over the bundling of Internet Explorer with Windows. The district court found monopolisation violations in April 2000 and ordered a breakup that June; the appeals court overturned the breakup remedy in June 2001 while leaving significant findings intact, and a settlement was reached in November 2001 and approved a year later. Gates had already handed the chief executive role to Steve Ballmer in 2000, becoming chief software architect until 2008 and stepping down as chairman in 2014. He left Microsoft’s board, and Berkshire Hathaway’s, in March 2020.

The organising insight is that software’s cost structure is unlike almost anything else an investor analyses. Producing the first copy is expensive and producing every copy after it costs approximately nothing, so beyond breakeven almost all incremental revenue reaches the bottom line. That property alone does not build a business, because a competitor can price at the same near-zero marginal cost. What converts it into a durable position is a reason for buyers to stay, and that is what the platform structure supplies.

The platform argument is about whose problem you are solving. Gates positioned Microsoft to serve the developers writing applications rather than only the people buying computers, on the reasoning that buyers follow software. Every additional application made the system more valuable to buyers, every additional buyer made it more attractive to developers, and the loop compounds without the platform owner having to win each individual purchase decision on merit. An investor recognises this as a network effect and a switching cost operating together.

The licensing decision of 1980 is where the philosophy becomes concrete, and it is worth being precise about what happened. Microsoft did not out-engineer anyone; it sold IBM a system it did not yet own, for a modest one-time fee, while keeping the right to sell the same thing to everyone else. IBM captured the hardware margin and watched it compete away as clones arrived. Microsoft kept the layer that did not commoditise. The transferable lesson is about where in a value chain the durable economics sit, which is a question that survives long after the specific technology has gone.

His approach to competition was famously aggressive, and the antitrust record is the cost of that. The pattern the courts examined was bundling: using a strong position in one product to establish another, in a way that made it hard for buyers to choose otherwise. That is an efficient strategy from inside the company and an anticompetitive one from outside it, and the case established that at sufficient scale the second description carries legal weight. Investors analysing dominant platforms today are working with the framework that case produced.

The final phase is a different discipline. Gates moved from operating a company to allocating a foundation’s capital, and the 2010 Giving Pledge, launched with Warren Buffett and Melinda French Gates, committed signatories to give away the majority of their wealth. Whether one regards large private philanthropy as an appropriate mechanism is a live argument, but the transition itself is instructive: the skill of running an operating business and the skill of deploying capital are related without being the same, and he is one of the few people to have done both publicly at scale.

Microsoft made platform strategy legible. The vocabulary investors now use about network effects, switching costs and standards ownership was largely assembled by people trying to explain why one software company captured the economics of an industry that thousands of manufacturers competed in.

The antitrust case shaped how dominant technology companies are analysed and regulated. Every subsequent argument about bundling, default settings and self-preferencing is conducted against the framework the 1998 to 2002 proceedings produced.

The Giving Pledge, launched with Warren Buffett, restated in modern form an argument Andrew Carnegie made in 1889: that large fortunes should be distributed by their holders during their lifetimes rather than left to accumulate. Whether that is the right mechanism remains contested, but it has become the reference point for the debate.