Questions about news and research: gsb_info
@gsb.stanford.edu
(650) 723-3157

Pricing in a Networked World

January 2004

STANFORD GRADUATE SCHOOL OF BUSINESS — At the time of the U.S. Justice Department's landmark antitrust case against Microsoft, the software firm was selling its Windows operating system to computer manufacturers for an extremely low price given the company's dominance in the market. At the same time, it was charging significantly more for its Microsoft Office application suite. Given that both products had comparable market shares, why did Microsoft charge only about $60 for Windows — its "base" product — instead of the $1,800 that many estimate it could have demanded, and why did it choose to price Office — the "complementary" good — at nearly four times as much?

The question was key during the trial for understanding Microsoft's competitive position. Never answered to the satisfaction of many observers, the query launched Brian Viard, assistant professor of strategic management, and Nicholas Economides, professor of economics at New York University's Stern School of Business and executive director of the NET Institute, to look for an answer. They examined pricing, compatibility, and product quality decisions for both primary, or "base," goods and the secondary, or "complementary," goods that rely on the base product to function.

The researchers argue that firms such as Microsoft that enjoy a monopoly on base but not complementary goods have an incentive to set low prices on base items to attract customers who use complementary products, even if they come from other vendors. There is also an incentive to improve the quality of both kinds of goods. Strong sales of applications that run on Windows, for example, even those manufactured by other firms, will benefit Microsoft by increasing demand for Windows.

The issue is particularly important for high-tech industries where consumers create virtual networks of demand for specific products such as PC or Macintosh computers, specific types of music players, or Nintendo game consoles. As more consumers choose one line of products over another, that particular line inspires the creation of more and more complementary products and support services and can create enough support for a primary product to completely squeeze out its competition. "The greater number of complementary goods created for a product, the more people will buy the product," explains Viard. This leads to what economists term the "network effect."

"One of the implications of our model is that if a firm produces both the base good and complementary goods, that firm has more of a profit motive for investing in improving the complementary goods than does an independent company," said Viard. "So, for example, Microsoft has more of an incentive to improve its word-processing package than another firm that produces such a package but does not produce an operating system. This is because of the feedback effect between the two kinds of goods: If I improve the complementary good, it helps sales of the base good, and vice versa."

It follows that firms that acquire complementary goods-producing companies end up with more of an incentive to improve the quality of such products. Similarly, a firm has more incentive to make base and complementary goods compatible if it owns both than if another company owns the complementary good. "And there are lots of decisions that firms make about how compatible to make their products," Viard points out.

So why did Microsoft price Windows so low relative to Office at the time of the trial — a practice it continues today? One possible reason, says Viard, is that "they don't want to crank up the price for Windows, the base good, because they don't want to choke off the positive feedback effect with those complementary goods that Microsoft does not produce, such as statistical packages and so forth. They control the price of Office, but they don't control the price of other firms' complementary goods. So they prefer to crank up the price of Office, because people value it very highly and are willing to pay a lot for it."

Prior to Viard and Economides' study, theorists measured the network effect in two mutually exclusive ways: either by looking at the detailed underlying microeconomics or by taking a more global view that assesses value as a function of how many other people buy the product. The researchers' study is the first to unite these two models by using firms' profits as a baseline measurement.

The model, says Viard, has implications for future antitrust cases that may involve base and complementary goods. It also may prove useful to producers of base goods who are deciding whether or not to own complementary goods. Finally, it may provide information on how to price goods in cases in which a monopolist owns the base good but not all of the complementary goods.

by Marguerite Rigoglioso

Pricing of Complementary Goods and Network Effects
Nicholas Economides and V. Brian Viard
Stanford Research Paper #1812, 2003

"The Microsoft Antitrust Case," Nicholas Economides, Journal of Industry, Competition and Trade: From Theory to Policy, 2001, Volume 1, pp. 7-39

Competition Policy in Network Industries: An Introduction
Nicholas Economides

Information Rules: A Strategic Guide to the Network Economy, Carl Shapiro and Hal R. Varian, Chapter 7 ("Networks and Positive Feedback"), Harvard Business School Press, 1998, pp. 173-225

More Research Papers
More Cases

...more research news