Innovation Spillovers and the “Dirt Road” Fallacy: The Intellectual Bankruptcy of Banning Optional Transactions for Enhanced Delivery Over the Internet

AbstractJ. Gregory Sidak & David J. Teece

In October 2009, the Federal Communications Commission proposed “net neutrality” regulations, including a new rule that would have the effect of banning optional business-to-business transactions between broadband Internet service providers (ISPs) and content providers for enhanced delivery of packets over the Internet. The proposed “nondiscrimination” rule would have the ironic effect of actively discriminating against any kind of content or application that is differentiated by its need for greater assurance of higher quality transmission across the Internet (known as quality of service, or QoS) than undifferentiated best-effort delivery can offer. This result not only would reduce static efficiency by encouraging higher consumer prices, but also would reduce dynamic efficiency by retarding innovation.

The proposed rule manifests an inverse relationship between means and ends, for it would actively thwart the Commission’s stated purpose of promoting innovation both in and at the edges of the network. These economic considerations set the bar very high for those who claim that the new regulation is needed to prevent theoretical harms that have not materialized in more than a decade of real-world experience.

By now, the economic arguments in favor of network neutrality regulation have coalesced around three principal theories. The first is the theory that, if permitted to charge suppliers of content or applications for optional higher quality delivery, network operators will ignore positive spillover effects and set charges at higher than socially optimal levels. The second is the theory that vertically integrated network operators will foreclose independent providers of Internet content and applications. A third and less clearly articulated theory is that the broadband ISP will degrade the quality of best-effort delivery of Internet packets – reducing the quality of best-effort delivery to that of a “dirt road” – as a means of coercing suppliers of content or applications into purchasing superior QoS.

We show that none of these three theories of harm is plausible. Certainly, none justifies the proposed across-the-board ban on optional business-to-business QoS transactions between ISPs and content providers – transactions that could prove particularly valuable to smaller content providers looking to differentiate their offerings from and compete with larger content rivals that have the scale and resources to meet their QoS needs with third-party or self-deployed content delivery networks.

Download as PDF