How Privatization and Competition Freed the Web and Made the Modern World Possible

The historical importance of the National Science Foundation's decision to surrender control of the internet


How the Internet Became Commercial: Innovation, Privatization, and the Birth of a New Network, by Shane Greenstein, Princeton University Press, 488 pages, $35 

In 1991, the internet was one data communications protocol among many competitors. It had taken hold in a collection of research and education institutions serving less than 2 percent of the population. A small, government-funded backbone that tried to exclude all commercial use held its participating networks together.

Twenty-five years later, the internet protocol is the lingua franca of the global digital economy. Nearly 4 billion people worldwide depend on it. It is the platform for the business giants of the 21st century—Google, Apple, Facebook, Ali Baba—and the site of monumental battles over regulation and governance. An entire digital ecosystem has grown up around the internet as it evolved from a government-sponsored research community to a commercial economy.

Dozens of books, academic papers, and magazine articles describe different aspects of that change. What we've long needed is a comprehensive history that synthesizes all those elements into a single narrative. How the Internet Became Commercial, a new book by the Harvard economist Shane Greenstein, is an imperfect but noteworthy attempt to fill that gap.

Greenstein asks: How did economic forces, government policies, and prevailing norms and institutions interact to encourage or discourage the decentralized innovation that we have come to associate with the internet economy? He devotes most of his attention to the economic part of this triad, with institutional analysis coming in a distant third.

The narrative begins with the privatization of the National Science Foundation (NSF) backbone—the larger network connecting all the local and regional networks—from 1992 to 1995. This opened up what had been a fairly closed network for education and research institutions to commercial use, and it replaced a single government backbone contractor with multiple competing private connectivity providers. We see the emergence of private internet service providers (ISPs) and an unregulated, decentralized market for interconnection among them. We also see the importance of dial-up Bulletin Board Systems (BBSes) as entrants driving competition in the ISP market. Making the jump from BBS to ISP was relatively easy in terms of the capital investment and expertise required. As demand for internet access grew, what had been a small, localized BBS market for computer nerds exchanging files and messages became a mass market for web access. There is a useful, if stylistically labored, comparison of the early internet boom to the California gold rush.

Greenstein is particularly good on the emergence of the World Wide Web and the commercialization of the web browser, including the early browser war between Netscape and Microsoft. Developed in research institutes while internet use was still primarily noncommercial, the web protocols for linking documents and other resources on the network, coupled with the browser's graphical user interface, pushed computer networking into mass adoption from 1992 to 1995. As internet content and applications increased and became richer, Greenstein shows the deepening of capital investment in the telecommunications infrastructure due to user demand for faster upload and download speeds. Dial-up modems went from 2.4 to 56 kilobits per second, and from there to broadband cable modems. Equally important, established firms such as IBM and brick-and-mortar enterprises adapted to the internet, which facilitated organizational efficiencies through more extensive access to relevant information about supply chains, work teams, accounting, and so on.

An illuminating chapter is devoted to Google's origins in an NSF-funded research project at Stanford, and to the powerful connections between search and advertising that were subsequently forged. The book also explores the rise of wireless internet access in the early 2000s. The period of analysis stops around 2005; Facebook does not enter the stage.

Greenstein documents not just successes but follies and dead ends. He has some fun with the "Internet exceptionalists"—investors, analysts, and would-be prophets, such as George Gilder, who believed that the rules of established businesses were outmoded and destined to fail. The exceptionalists came to reject the conventional benchmarks of financial success, such as earnings and price-earnings ratios, leading to a speculative bubble and the famous dot-com crash of 2000. Greenstein argues that most investment analysts had strong incentives to remain silent about these errors even when they knew better, because their immediate gains from encouraging investment based on those premises—and the social costs of contradicting one's colleagues—trumped their better judgment.

One of the book's most welcome strands is its exploration of the ways privatization and commercialization allowed the internet to reach more people and pushed it to generate valuable new services and capabilities at a large scale. Greenstein correctly anchors the web's rise in a global shift toward privatizing and demonopolizing telecommunications and information services around the world. Too often, the accomplishments of market-driven growth and innovation—and the public policies that facilitated them—are taken for granted (or ignored) while attention becomes fixated on how to control, regulate, surveil, or even block e-commerce. Calls for control come from left and right, and can involve the rhetoric of fighting big corporations, protecting national security, protecting children, ameliorating inequality, or resisting U.S. hegemony. All overlook the extent to which the internet capabilities we take for granted rely on market forces and globalized exchange.

The main message that emerges from Greenstein's story is the importance of decentralized decision making. Competition is important, he argues, not simply because it leads to lower prices but because it fosters a "diversity of innovative viewpoints"—a wide range of business models and technical platforms—that ultimately lead to more innovation and better services. While this often leads him to Hayekian conclusions, Greenstein also believes that antitrust can foster this diversity. From the AT&T breakup to the Microsoft browser wars to WorldCom's attempts at consolidating ownership of telecom infrastructure, he argues that antitrust interventions (and the threat of them) helped maintain healthy levels of decentralized decision making.

There are some gaps in this argument, particularly with regard to the Microsoft case. Greenstein documents Microsoft's aggressive actions to crowd Netscape out of the Windows platform, and he convincingly argues that Bill Gates' company limited consumers' options at the time. But his narrative overlooks three things.

First, Netscape's failure was partly attributable to the inferiority of its product. By late 1997, its software was bloated and crash-prone and was beginning to alienate many users. Second, even among those who loathed Microsoft, most of the remedies bruited by the 2000 antitrust actions, such as splitting the company into separate operating system and applications firms, did not seem likely to increase competition much but did threaten to raise compatibility issues. Third, and most important, it's clear that Google later succeeded at what Netscape was trying to do: create an alternative platform to Windows with a broad range of functionalities. You might argue that Google succeeded because a chastened Microsoft was no longer willing to use the same aggressive tactics. But you could also argue that Google was simply a successful innovator and that its ability to leverage its search dominance to create a new platform could not have been stopped.

Greenstein emphasizes the role of NSF funding in the rise of both the internet itself and businesses like Google, but he also makes it clear to any would-be industrial policy advocates that the NSF did not intend to build particular businesses or industries. One of the NSF's greatest successes, indeed, was to know when to pass off control of the internet backbone to the private sector, and to do so in a way that (despite some stumbling) led to a self-sustaining, growing, and innovation-friendly internet infrastructure. He also attacks the myth, especially popular in Europe and Asia, that "governments can financially support technological development for cheap." In the United States, he notes, "the cost of the internet or its future economic benefits did not shape the aspirations of the government. The NSF invested in developing Internet technologies to meet its agency mission, not with the intent of producing large economic gains."

Greenstein's book has its drawbacks. When it comes to broader policy conclusions, he often seems to be pulling his punches. It's hard to tell what he thinks about internet tax exemptions. Decisions that set the foundation for the net neutrality debate were made in the period he covers, but the book does not address them directly. He makes a strong case that common carrier regulation of the local telephone monopolies played a critical role in fostering the rise of ISPs in the early years, but he says nary a word about the 2002 decision to classify cable modem providers as an unregulated "information service."

Furthermore, Greenstein's writing style can be repetitive, and his book's structure leaves a lot to be desired: Instead of being organized conceptually, it relies on a roughly chronological structure anchored to concrete anecdotes, such as Al Gore's claim to have "took the initiative in creating the internet." As a result, key ideas are sprinkled throughout the narrative, in some cases repeated too often and in other cases inadequately developed. It is not always clear whether the author intends to make a serious contribution to economic and technological history or wants to popularize the economics with human interest stories.

The book's most important limitation is its almost exclusive focus on the United States. Yes, the internet had its origins in the U.S., and America's processes of privatization and commercialization had a global impact. But the success of Transmission Control Protocol/Internet Protocol in the standards war over data communication was a transnational phenomenon. The internet's success relied on globalized standards and standards-making processes, and on global software and equipment markets. Greenstein correctly roots his account of the rise of the internet in earlier moves in telecommunications policy, in particular the breakup of the Bell system. But the abandonment of the monopoly paradigm in telecommunications took place all over the world during the same period, and freer trade in both equipment and telecom services had an important impact on the internet's diffusion and innovative potential. A 1988 change in the International Telecommunication Regulations that liberalized international leased line usage, the 1994 North American Free Trade Agreement, and a 1996 World Trade Organization agreement on basic telecommunications services laid many of the foundations for the global commercialization of the internet, but they are not even mentioned in the book.

A worldwide perspective, though obviously more difficult to execute, would add depth to Greenstein's emphasis on decentralization. The commercial internet both reflected and reinforced a globalizing economy and society. In an age of backlash against that trend, it's important to understand how progressive and innovative decentralized market forces can be.