Major customers such as governments and large corporations are critical if Linux is ever to knock out the Windows monopoly, according to an upcoming economic study authored by two Harvard faculty members.

The academic paper, "Dynamic Mixed Duopoly: A Model Motivated by Linux vs. Windows," was written by assistant professor Ramon Casadesus-Masanell and professor Pankaj Ghemawat of the Harvard Business School, and will be published in a special issue of Management Science.

The two based their research on a simplified economic model attempting to recreate the dynamics of Windows' competition with Linux, where Windows has market share and profitability on its side, while Linux benefits from a faster development cycle and lower cost.

Casadesus-Masanell and Ghemawat found, to their surprise, that Linux's advantages by themselves didn't mean Linux would ultimately oust Windows, because of Windows' initially dominant market share.

"This result holds true regardless of the strength of Linux's demand-side learning. Furthermore, the result persists regardless of the intrinsically better design and potential differential value of Linux," they said in an interview with Working Knowledge, a Harvard Business School publication. "In other words, harnessing demand-side learning more efficiently is not sufficient for Linux to win the competitive battle against Windows."

Demand-side learning is the researchers' term for the open source development model's ability to take advantage of suggestions and contributions from users.

One factor that could allow Linux to win is "strategic buyers" - large organisations such as governments and large companies, who commit to buying Linux because they feel more comfortable having access to source code. "This may be one main reason why Microsoft has been providing chunks of Windows' source code to governments," the researchers said.

Another surprise was that the spread of improperly licensed copies of Windows can actually help Microsoft, as long as it's mostly going to people who wouldn't ordinarily be paying for the software. That's because the more people using the software, the greater its network effects, which make the software more valuable and allow Microsoft to charge more.

Windows piracy also helps to dampen enthusiasm for Linux, they said. "We found that in countries where piracy is highest, Linux has the lowest penetration rate," said Casadesus-Masanell and Ghemawat.

The third finding that surprised the two researchers was that Linux doesn't necessarily imply better "social welfare" effects than Windows. Whatever positive effects open source may offer, Linux's popularity means that developers aren't working to full efficiency, since they're dividing their efforts between two operating systems.

"With a monopoly, the efforts to develop new software and improve the platform are directed towards one system only and this may turn out to be better from a social welfare perspective," the researchers said.

The findings suggest particular ways Microsoft can hobble Linux, which largely are things Microsoft is already doing - such as attempting to increase its own "demand-side learning," as well as making it as difficult as possible for Windows applications and Microsoft Office documents to run on Linux.

Microsoft can't expect it to be easy to win against Linux, however. "Given that Linux was born in 1992 in an industry already dominated by Microsoft, and given that the financial motive is secondary, it will be very hard for anyone to oust Linux," the authors said.

The researchers admit that the formal economic model they've constructed is highly simplified, and may not reflect important factors that exist in the real world.

One issue they're planning to research in the future, for example, is the mysterious incentive to contribute to open source in the first place. "The economic approach may shed light on why supposedly rational individuals are willing to spend valuable time and effort without extrinsic, financial incentives," the researchers said.

Only an economist could ponder why people might do something for anything other than money.