In June of 2008, William Gates Jr. will step down as chief software architect at Microsoft and move more of his time to the foundation named for him and his wife. It is a quasi retirement from Microsoft, not a full retirement. Gates intends to come back now and again to help. His June 2006 announcement made headlines, even though most of us—except those who happen to work there—have only a passing interest in the internal machinations of Microsoft. It gained attention because of who said it.
Bill Gates’ life has a unique arc. Gates founded and then managed a company that provides software for personal computers, and, in the process, became unbelievably wealthy, to the tune of $50 billion at last count. Many others got wealthy along with him, Steve Ballmer and Paul Allen, most notably, but Gates’ wealth tops them all, making him comparable to only a small number of other business figures in history—‘‘robber barons’’ in popular speak—such as Carnegie, Rockefeller, Bell, and Ford.
Gates does not always seem certain how to manage his unique status in society and history. To be sure, a large team of public-relations experts at Microsoft try to build, manage, or modify the message about Bill’s talents and views. Yet, he goes off message every so often.
On occasion, one of these public gaffes illuminates the professional side of Gates’ life. My favorite example occurs in volume III of Bob Cringely’s movie, Nerds 2.0.1. Cringely conducted the interview featured in this part of the film before the Federal Antitrust case had been filed, so Gates is not defensive. Cringely asks him if Microsoft is losing money by giving away its browser, and Gates is relaxed as he answers.
He pauses, smiles, and then, speaking with just a hint of dry humor, says, ‘‘Well, nobody ever accused Microsoft of not knowing how to make money.’’ The smiling Gates then exhales in a little involuntary laugh, a response to his own joke. There is nothing suave or planned about it: just a cross between a guffaw and a boyish giggle. After another pause, Gates provides an explanation of how Internet Explorer helps the Windows business. It’s interesting, but it seems to dissolve in the revelation of the earlier moment: Gates knows how to make money, and he knows that he knows.
That is a long way of motivating today’s question: As Gates’ quasi retirement from Microsoft approaches, what judgment should history render about how Bill Gates acquired his wealth?
To be sure, any path to this much wealth involves a combination of considerable luck, skill, and effort. Yet, that platitude is unsatisfying for explaining Gates’ wealth. He deserves to be measured by the yardstick applied to history’s great robber barons. What does that yardstick gauge? It examines the difference between fair and foul. A fair path to wealth involves entrepreneurial effort, creating new value while capturing some fraction of it. A path is foul if it involves some something unethical or illegal—that is, if the acquisition of wealth involves breaking rules, either legal ones or important business principles.
Both fair and foul apply to some degree to any individual. The open question concerns how much. In particular, which mattered most in Bill Gates’ career, and why? To answer this question, I will recall many stories from his career and interpret the long arc with praise and harshness.
Did Gates make money by creating value? The answer to this question is not as straightforward as it looks. To be sure, Gates helped found and manage a company that eventually grew into a giant. He helped take it public in 1987, and he has always owned a significant fraction of that company, down to less than 10 percent most recently. His personal wealth appreciated along with the stock.
Why is this not straightforward? Gates did not make money through commercializing one big invention, the way Alexander Graham Bell did, for example. Rather, Microsoft made money by employing Gates’ uncommon savvy in the art of software commercialization. Gates made a gazillion little decisions. The strategic and financial benefits accumulated over years, occasionally becoming manifest in a few well-known events, such as the rollout of the Office suite and Windows95.
It is hard to characterize any career after three decades, particularly when it manifests in many decisions, but let me try to touch on some of Gates’ most salient managerial traits. Gates repeatedly approached situations with the intent to out-think and out-execute others. In pursuit of that approach, Gates became a devoted student of computer industry business models and lessons from other markets. Using those lessons, he developed strategies that moved beyond the barbershop wisdom of the most sophisticated Silicon Valley salons. Perhaps his greatest inventions in this sense were his business models for software platforms. He invented them before many consultants and business analysts.
His best analyses could be breathtaking in insight and depth. For example, ‘‘The Internet Tidal Wave,’’ the May 1995 memo that altered Microsoft’s strategy just prior to the browser wars, is an extraordinarily insightful document. The vast majority of CEOs on the planet would have been incapable of writing it if they had been in the same situation. (I do mean this: I circulate that memo in my class to give MBAs an example of what an exemplary executive memo looks like, and to help them appreciate why awake executives work hard to understand markets, even out of school.)
Three predilections lie behind Gates’ savvy. First, Gates likes technical topics for their own sake—he taught himself to program and debug DEC’s computers when he was a teenager, for example. In addition, Gates draws on an almost unique mix of drive, persistence, and patience when pursuing his goals. For example, he furiously drives himself to learn things he believes he needs to know, and he will negotiate incessantly until he wins on the points he wants, but he will wait eons for results from investments.
Third, and less appreciated, Gates uses his authority inside his company to teach and to conduct lengthy investigations— that is, to convey a view, address questions and consider them in light of the available facts, and, in many cases, argue back and forth with employees. It seems to be no accident that every longtime Microsoft executive I have ever met can talk intelligently about their market and, more remarkably, about markets around them. It is as if they have taken regular pop quizzes on a wide variety of subjects. The CEO conducts the quizzes, and every manager has passed these quizzes many times.
Gates’ preferences manifest in Microsoft’s actions in a variety of ways, but one became especially evident early and has reappeared many times: Microsoft’s executives learned from experience and, more interestingly, arranged so that they did. For example, many observers recognize that Microsoft often has pursued a ‘‘three-version strategy’’—that is, its first version of a product experiments, the second responds to feedback, and the third puts together the lessons for a hit. Very few CEOs would authorize that strategy, or could succeed in executing it properly, but Microsoft has done it repeatedly.
All this seemed to meld into a coherent whole in the early 1990s, as several stunning market events unfolded. IBM effectively lost leadership of the PC industry (a long story). Yet, the market did not drift long without a software leader. Out popped Windows 3.0. That was followed by the Office Suite and Windows 3.1, then the announcement and big push for what would become Windows 95. In the interim, Microsoft brought out Encarta, which had a spectacular launch. It is rare to see any company have such a string of commercial successes. Gates had built something unprecedented and impressive. Indeed, at about that moment in time, it looked as if Microsoft could (and would) take over virtually any or all software markets it set its mind to address.
I should offer some balance to go with the compliments. Gates is not a flawless executive by any stretch. Gates’ greatest strategic weakness arose from his strength—the insistence that Microsoft act ‘‘smarter’’ than others. Companywide tactical actions, no matter how well they are designed, can be executed and coordinated only with centralized strategic decision making. At his own insistence, Gates was at the center of all those decisions, and that came with an obvious and widely reported drawback: He can be difficult to work with on a daily basis.
Over several decades, this yielded predictable results. Gates drove many talented employees to quit—out of frustration with his verbal tirades, over disagreements with some of his policy preferences, and because many of them could not see the point of working so hard after their stock options had appreciated to several hundred million dollars. Microsoft has also managed to keep many of its quality executives too, so the company did all right despite the departures. But, frankly, Gates was simply lucky the personality clashes never came back to bite him in kiss-and-tell scandalous books, or, what would be worse, in the guise of former employees managing competing firms. A few tried (for example, Rob Glazer at Real) and, but for circumstances, might have succeeded in displacing some of Microsoft’s businesses.
Centralization also has two broad and well-known drawbacks that do not matter until circumstances expose them: It requires a nearly omniscient strategy team, and it can become a bottleneck on new initiatives. Both of these were potential issues at Microsoft. After all, Gates and his team were only human, and they had no special crystal ball. They also were just a little too eager to judge every internal initiative, even after Microsoft had grown awash with money and they could afford to give their talented employees some discretion.
Both weaknesses came home to roost when the commercial Internet diffused, for example. Gates had not authorized any experimental projects in the browsers prior to April 1995, although employees had wanted to undertake some initiatives a year earlier. Gates did not see the value in those initiatives, and the habit of mind inside the firm favored Gates’ judgment over anyone else’s.
In addition, Gates’ strategy team erred in its reading of the technical and commercial trends. So did many others in the commercial computing industry, to be sure, but most corrected course in the fall of 1994. Microsoft did not alter its approach until April of 1995, because the company had to wait for its strategy team to agree to change course—that is, for Bill to change his mind (see, again, ‘‘The Internet Tidal Wave’’).
Let’s pause and summarize. On the whole, Gates got some of his wealth using fair means. It was not flawless, but it was not substantially foul by any reasonable definition. As an economist and student of the technology business, I cannot be too critical of an entrepreneur who gets rich by building an organization that executes well. Indeed, if that was all Gates had done, this would be the end of the essay. But it is not.
Gates got into foul territory because of his insistence that Microsoft attempt to remain self-serving in all its actions. That might sound like a small error until it is stated in proper context: Gates’ organization not only got big, but it also took on leadership role in the industry. Yet, after assuming that role, Gates was unwilling to become less self-serving, which is what leading a whole industry requires. His seeming tone-deafness to principles of industry leadership got him into legal trouble, and it lost him admirers.
Taking it from others
Did Gates get part of his wealth by taking it from others? The answer is certainly yes, but so what? Many forms of market competition involve market share moving from one firm to another. The harder question is whether Gates’ firm used appropriate tactics or not.
There has been debate on this question for decades, so I do not expect to settle it in one essay. Here is my big theme: Gates’ conniving and self-serving thoughtfulness suited him for entrepreneurial settings, not necessarily every setting. When his company took a leadership position, the situation required him to start taking actions appropriate for an industry leader. Yet, he did not change. Although Gates had shown tactical flexibility and intellectual resiliency in the first two decades of his career, his actions indicate that he did not perceive such change as a priority.
To develop this theme, I need to interpret the long arc of Gates’ strategic actions. As I said, this will touch on arguments that have simmered for many years.
Let’s start with his early years. Many technologists despise Gates for being a commercial opportunist from the outset.
For example, in the fall of 1980, IBM approached Microsoft to supply BASIC for the IBM PC. Then, after failing to secure an operating system from the first company it approached, IBM returned to Gates with that task as well. He agreed to supply the operating system, although he did not bother to tell IBM that he did not have one and he would get it from someone else in the Seattle area—which implies, of course, that Gates simply could have sent IBM to that other person. Instead, he bought the operating system, known as the Quick and Dirty Operating System, or QDOS, retooled it a bit, and resold it to IBM as the Disk Operating System, or DOS. Microsoft did not make much profit on the resale, but Gates did the deal because he seems to have had a strategic sense, even then, that the contract would seed a favorable market position.
Methinks the technologists do protest too much. This was ethically ambiguous behavior—no Boy Scout could earn a badge this way. More precisely, a little white lie and convenient silence supported personal gain in the face of a lucky opportunity. At the same time, extenuating circumstances make it easy to explain and partially excuse the behavior. Many others would have muffed the same opportunity had they gotten it (indeed, the provider of the CP/ M operating system did, which is why IBM asked Gates).
In addition, Gates was a reasonable choice for supplier. To be sure, his company had only several dozen employees and he was young and uncertified (for example, lacking a college degree), but, lucky for Gates, that was not unusual in the PC market at the time. He also already had extraordinary technical competence at that age, and IBM’s people had put him through enough tests to see it. In retrospect, he deserves some respect for recognizing the opportunity and being foresighted enough to take advantage of it. That is how entrepreneurial business works.
Many technologists also despise Gates for imitating Apple’s look and feel in Microsoft’s products. To this, I initially give the same answer as above: It was ethically ambiguous behavior, not a clear foul. Apple had an opportunity to dominate the industry and muffed it (a long story). In contrast, Microsoft implemented many of the good ideas in commercially relevant ways. If that was not enough, there was also a lawsuit about whether the imitation was illegal, and the courts sided with Gates. Once again, we should give Gates his due for building an organization that executed so well when others did not.
From these stories, it is a short leap to the thing most technologists really resent—that is, Microsoft imitated ideas other companies had pioneered and implemented them in ways that were more commercially successful, but the pioneers never got paid for their contributions. For example, the Office Suite, comprising Excel, Word, and Power- Point, borrowed lessons and design innovations from others, such as Lotus, WordPerfect, or Harvard Graphics, not to mention some applications for the Macintosh. The same is true of Access, Outlook, and several other modules.
To this, I give almost the same answer as above. Being commercially successful in this way is not a clear foul. It is a common occurrence in young markets— albeit, it is rare to see it at this scale. It would be a foul in the Netscape case, but it was more than the imitation alone that was the issue. (A full explanation is coming.)
Moreover, it is hard to recall against the haze of whining that followed later, but the vast majority of developers rather liked what Microsoft was doing for much of the early 1990s. Users did, too. That was when Microsoft implemented another one of its most valuable innovations, the application programming interface, or API, in Windows 3.0. It helped turn the third version of Windows into the kind of blockbuster platform that Apple or IBM would have had if either of them had implemented the same feature and developed an organization to support it (another really long story).
Was the idea for APIs new? Of course not. It had been discussed in computer science for ages, and others had tried related things. Windows 2.0 also experimented with a simple attempt (a long story). The real challenge involved going from blackboard to implementation in a functioning organization where revenue needed to exceed cost. This was precisely the type of activity Gates was good at leading. Did Microsoft make APIs work better than anybody had ever forecast? Yes, in part because in 1989 the company was a competent provider of software tools, and in part because the company implemented APIs in a platform that carefully respected backward compatibility with investments made by users and developers, building a de facto standard into a core piece of a platform on which the whole industry could move forward. We should give Microsoft its due for doing things well.
Was there anything illegal about APIs? No, not until the company put APIs on steroids, supporting an even greater range of applications in Windows 95.
While that implementation would be valuable for many others in the industry, the company also started to cross the line by initiating actions that set up too many controlling quid pro quos for the use of their APIs.
You did read correctly. The problems started to emerge at the height of Microsoft’s commercial success. Strategic continuity caused part of the problem. Some of what might have been ethically ambiguous when Gates was an entrepreneur was a foul coming from an industry leader. Yet, Gates did not act as if he worried about it. He did not act as if he appreciated that industry leadership required any major adjustments.
That will take another column to fully explain.