The Internet as “Spillover-Rich” Infrastructure


April 16th, 2016

A key source that informs my perspective on special access policy—and telecom policy in general—is Brett Frischmann’s 2012 book, Infrastructure: The Social Value of Shared Resources.  Selected chapters from the book can be found here, and Frischmann provides a good overview of the book, including its Internet-related sections, in this talk at Harvard’s Berkman Center).  Frischmann is a professor and co-Director of the Intellectual Property and Information Law program at Cardozo Law School in New York City.

On page five of the book, Frischmann notes that “most economists recognize that infrastructure resources are important to society precisely because infrastructure resources give rise to large social gains.”  But he also notes that infrastructure’s “externalities are sufficiently difficult to observe or measure quantitatively, much less capture in economic transactions, and the benefits may be diffuse and sufficiently small in magnitude to escape the attention of individual beneficiaries.”  On page 6 he cites the “comedy of the commons” concept developed by Carol Rose which, he explains, “arises where open access to a resource leads to scale returns—greater social value with greater use of the resource.”

The book includes a chapter focused specifically on the Internet, which Frischmann describes on page 345 as “a spillover-rich environment because of the basic user capabilities it provides and the incredibly wide variety of user activities that generate and share public and social goods.”

But, as with other infrastructure, it is not easy to quantify the net value of Internet-supported social goods, or even identify what all of those social goods are.  As Frischmann explains on pg. 347, Internet connectivity involves “a very high degree of social value uncertainty.”

 It is impossible to predict with any degree of confidence who or what will be the sources of social value in the future.  Accordingly, there is no reason to defer to private firms in this context. First, there is no reason to believe that firms are better informed, capable of maximizing social value, or likely to resist the pressure to discriminate, prioritize, or optimize the infrastructure based on foreseeable and appropriable private returns. Second, there is no reason to trust that markets will correct misallocations…[F]irms may be strongly biased in their estimation of the future market value to favor services that they currently offer or expect to offer, sponsor, or otherwise control, and to disfavor those that they do not.

One recent example of the tendency of private access network owners with market power to “discriminate, prioritize, or optimize the infrastructure based on…appropriable private returns” is Comcast’s recently launched Stream TV service.  The element of discrimination is clear in that Stream TV, unlike competing services like Netflix, does not count against the data caps that Comcast customers are or will be subject to.

On page 346 Frischmann points to two demand-driven problems that need to be considered by Internet-related policies:

 The Internet infrastructure is a mixed infrastructure, and as such, it faces the two types of demand-driven problems discussed throughout this book: First, it faces concerns about undersupply and underuse of infrastructure to produce infrastructure-dependent public and social goods, which leads to underproduction of those goods. Second, it faces concerns that infrastructure development may be skewed in socially undesirable directions. For example, if private infrastructure owners prematurely optimize infrastructure for uses that they expect will maximize their private returns, and in doing so choose a path that forecloses production of various public or social goods that would yield greater net social returns, the social option value of the Internet is reduced. This latter concern may involve dynamic shifts in the nature of the Internet infrastructure, such as optimizing networks in a manner that shifts from mixed infrastructure toward commercial infrastructure.

In key respects, these two problems correspond to the social harms of primary concern to Singer (undersupply by companies subject to regulation) and Cooper (actions by private infrastructure owners to “maximize their financial returns” in ways “that foreclose production of various public or social goods that would yield greater net social returns”).

Frischmann’s analysis of infrastructure and Internet dynamics leads him to conclude that the abundant but difficult to predict and quantify social benefits of the Internet are best supported via commons management.  On page 7 he describes commons management as “the situation in which a resource is accessible to all members of a community on nondiscriminatory terms, meaning terms that do not depend on the users’ identity or intended use.”  This, he says, “can be implemented through a variety of public and private institutions, including open access, common property, and other resource management or governance regimes.

Frischmann acknowledges that “grouping ‘open access’ and ‘commons’ under the ‘commons management’ umbrella will be troublesome to some property scholars:”

Open access typically implies no ownership or property rights. No entity possesses the right to exclude others from the resource; all who want access can get access, typically for free. Commons typically involves some form of communal ownership (community property rights, public property rights, joint ownership rights), such that members of the relevant community obtain access “under rules that may range from ‘anything goes’ to quite crisply articulated formal rules that are effectively enforced” and nonmembers can be excluded.

 There are at least three dimensions of distinction between open access and commons as traditionally understood: first, ownership (none vs. communal/group); second, the definition of community (public at large vs. a more narrowly defined and circumscribed group with some boundary between members and nonmembers); and third, the degree of exclusion (none vs. exclusion of nonmembers). These distinctions are important, especially for understanding different institutions and how social arrangements operate at different scales.

In two other posts (see here and here) I consider key issues related to infrastructure ownership in more detail, from the perspective of the “generative” vs. “extractive” ownership framework developed by Marjorie Kelly in her book, Owning our Future.


Tags: , , ,