Lessons from the Browser Wars
The first-mover advantage is well chronicled, but it didn't help Netscape when Microsoft launched Internet Explorer. What drives technology adoption, and do browser upstarts such as Firefox stand a chance? A Q&A with professor Pai-Ling Yin.
In a famous example of how first movers can lose their advantage, second-mover Microsoft won the Web browser wars from Netscape and continues to dominate the market today. But that competition was the subject of another "war," this one among researchers who study how technology is diffused into the market.
The debate was this: Did Microsoft win because its Internet Explorer was the technologically superior product to Netscape Navigator, or was Microsoft just more successful at the distribution end by convincing most PC companies, some argue by anticompetitive tactics, to include IE on every PC shipped in the late 1990s? Researchers line up on both sides of the argument.
A recent working paper by Harvard Business School professor Pai-Ling Yin and Stanford professor Timothy F. Bresnahan offers an answer. Looking at both the pace of adoption of new versions of the browsers and the brand choice made by users, "distribution played a larger role than did technical progress in determining the market outcomes," the scholars conclude in the paper "Economic and Technical Drivers of Technology Choice: Browsers."
The implications are significant. Browsers simplified access to the online world, transforming the Internet from a communications vehicle for academics into a mass consumer phenomenon. At the same time, the growing usefulness of the Internet drove sales of personal computers off the chart—the installed base of PCs doubled to 213 million computers between 1995 and 1999. Understanding the complex interaction between the two technologies and how a second mover was able to unseat the incumbent market champion will help innovators of all stripes learn how new technology gets diffused into mass markets.
In this interview, Yin discusses the results and implications for new browser upstarts such as Firefox and Camino. Do they stand a chance against Internet Explorer? Yin says Microsoft's entrenchment in corporate IT, the maturation of the PC market, and the technical difficulty many consumers have in switching to a new browser make it hard soil for new challengers to take root.
Sara Grant: Why did you choose to study Web browsers to look at the idea of diffusion of innovation in a market?
Pai-Ling Yin: Web browsers were the turning point in mass commercializing the Internet. They were the easy-to-use user interface that permitted the average person to access information on servers around the world.
As a tool for exploring how standards are set when new technologies hit the market, the browser wars exhibit many features we like to study: competition between two viable alternatives, rapidly improving technologies, the ability of firms to use strategic levers such as market power and channels of distribution, growth in demand leading to diffusion of the new technology through the population, and uncertainty. Thus, this is one example from which we can generalize lessons regarding the outcome of diffusion of innovation into a market.
Q: In your paper, you explain that your analysis of the "classical concern" in the diffusion of new technologies is based on technological progress versus economic resources. Can you explain this?
A: The classical debate in economics has been whether the market always produces the "best" outcome. Best can be defined in different ways. In one stream of the debate, "best" has been defined as "technologically superior" (again, a term that can be defined differently). So, does the market always lead the "technologically superior" outcome, or can economic actors take actions to influence the market outcome so that a "technologically inferior" outcome arises?
We conclude that while both technological progress (measured by releases of newer and better versions of browsers: version 1, version 1.1, version 2, etc.) and strategic actions (distribution browsers with PC purchases) increase the rate of diffusion of browsers into the population, the strategic actions (distribution or restrictions on distribution in the case of Netscape) are twice as important as technical progress.
Q: Your findings show that at the height of the competition between Netscape and Microsoft during the 1990s, Microsoft's Internet Explorer became the de facto browser standard. What does this say in regards to first-mover versus second-mover advantage, and about future browser battles?
A: What is interesting are the lessons we can learn about how a fast second mover can upset the normally strong barriers to entry that a first-mover advantage in a network setting can create. In short, the big lesson learned is that a window of opportunity exists for a second mover to challenge a first mover in this setting early on when the new technology has not yet diffused through the entire population—the second mover can try to influence new users rather than get the small installed base to switch over.
The second mover has to have some sort of asymmetric advantage, such as control over the distribution of a complement (in this case, the PC), in order to slow the build-up of network effects around the first mover and ensure that the second mover's product begins to build up a critical mass. A number of smaller strategic elements converge to generate this window of opportunity, but I encourage people to read the paper as well as a chapter on this phenomenon which is forthcoming in Shane M. Greenstein & Victor Stango (Eds.), Standards and Public Policy, Cambridge University Press.
As for the future of Web browsers, the standard has been set, and it will be very hard to displace IE. Although Firefox is touted as the new challenger, its share of end-users is only estimated at 10 percent, and end-users are less of a barrier to further market share than are Webmasters.
The big lesson learned is that a window of opportunity exists for a second-mover to challenge a first-mover.
A: Firefox and Camino claim only a small market share, and those users are on the tech-savvy end of the user spectrum. In particular for Mac-centric browsers, we still live in a PC-dominated world. Until IT managers in large enterprises are willing to support these browsers, they will not hit the mainstream in any major fashion.
Why won't an IT manager support these versions? The biggest headache with these browsers is that the majority of Web sites are optimized on IE. Try going to some of the major commercial airline sites, or to the Web sites of older and smaller companies. If you use Firefox, sometimes you will find missing menus, missing pictures, links that don't work, etc. Even Netscape doesn't work with all of our Harvard Business School applications.
But it's Webmasters who are the real barrier to a late-to-the-game second mover in the browser market. Because different browsers require slightly different code to be viewable, it is costly for Webmasters to write for different types of browsers. They will tend to pick the browser that is most used by the majority of end-users. Thus, the source of network effects in this market is indirect. While end-users don't know which browser other users are using, the developers of the content that make the browser so useful do care that everyone is using a similar browser.
Q: So Firefox and other new browsers, no matter that they have new features and refinements that IE lacks, remain at a competitive disadvantage?
A: Game over. Firefox and the others have to get the installed base of IE users to switch to their browser, a much harder proposition than IE faced in the '90s when all it had to do was get new users to pick IE rather than Netscape as their first browser.
Until IT managers in large enterprises are willing to support these browsers, they will not hit the mainstream in any major fashion.
Either the innovations would have to be huge improvements over what IE can currently do, or a huge problem would have to arise with current IE use to create an opening for such a late second mover to make headway and lead people to go through the hassle of switching their browser.
Even with the security issues that plague IE, Microsoft has a huge amount of cash. If any browser really became a threat, Microsoft could easily imitate their innovation, or fix the IE problem with a patch. Indeed, IE successfully caught up to Netscape's quality by throwing a lot of money and manpower into IE browser development.
Q: What are you working on next?
A: Most closely related to the browser wars is research with professor Estelle Cantillon on how a second mover tipped the derivatives exchange market in the German Bund.
These works contribute to a more general theme in my research on how firms can change industry structure. The market tipping work addresses one part of that question: How can second movers break incumbent control? Casework and early-stage research attack a second aspect of that general theme: Can technology drive changes in demand? In a world of growing information resources, what are the drivers of search? How do people search, really? What would the answers to these questions imply for industry structure and business models in the next twenty years?