Should Content Providers be Allowed to Contribute to the Cost of Mobile Bandwidth?

Posted By George Ford On March 2, 2012 @ 2:28 PM In "Over the Top" Services,Federal Communications Commission,Mobile Broadband,Net Neutrality,Network Agnostic Devices,Spectrum,Spectrum Caps,Spectrum Exhaust,Wireless | Comments Disabled

A recent article in the Wall Street Journal caught my attention, and I’m sure the attention of many others.  The article—AT&T May Try Billing App Makers [1] (February 28, 2012)—reported that AT&T and content providers were discussing ways in which the providers of mobile content, like video streaming, could pay for (in whole or part) the cost of the data traffic on behalf of the end user.  According to the article, the interest in a content-payer system is being encouraged by content developers that “could use the feature to drum up new business from customers wary of using data-heavy services like mobile video” in a world of usage-based pricing.

Not unexpectedly, the political interest groups went berserk at the news, claiming that these were exactly the type of arrangements the network neutrality rules were designed to block.  Could it be, however, that such pricing arrangements are good for consumers, good for content developers, and good for the network operators?  A win-win-win?  The answer is YES.

We asked this very question five years ago and provided the answer in a Policy Paper entitled Network Neutrality and Foreclosing Market Exchange: A Transaction Cost Analysis [2], first released in March 2007 and then academically-published in 2009 [3].  In that paper, we modeled the issue from a multi-sided market, transactions cost perspective.  For the set up, we had many content providers providing services to many consumers via a broadband intermediary—a two sided market with the broadband provider in the middle.  We wanted to evaluate the effects of a rule that prohibited the broadband providers from charging a positive price to the content providers—a rule often associated with network neutrality and, in fact, a rule included in the FCC’s Open Internet Order.  To create a relevant scenario, we assumed that in addition to the broadband provider selling connections to the end-users, it could also sell an enhanced “bump” in service to permit the sale of content needed prioritization or enhanced speed (say a video or ball game).  A key feature of the model is the assumption that the costs of setting up the enhancement are lower for the content provider relative to the end user.  For example, it is cheaper for Amazon.com to set up shipping for its sales than it would be for the consumer to do so on their own.  This modeling choice can be interpreted quite generally and encompasses a wide range of potential arrangements.

Using this setup, we derive the equilibrium prices for content and broadband services both in an unregulated state and under a rule where only the end-user could be charged for the enhancement (that is, the Open Internet Order’s rule).  Our model reveals that rules that prohibit efficient commercial transactions between content and broadband service providers could, in fact, be bad for all participants.  To begin with, under the prohibition consumers would pay higher prices for both broadband service and content.  These higher prices result in less broadband adoption, and the price affects all broadband customers, not just those that want to purchase the enhancement.  Under the law of demand, higher broadband prices mean lower broadband adoption, which is plainly contrary to the stated policy goals of the U.S. government.  Also, higher content prices mean fewer content purchases.  Fewer purchases, in turn, are likely to lead to fewer content providers, given the nature of costs in that business.  Furthermore, if the broadband provider is also a content provider, the regulation shifts sales to the broadband provider’s affiliate.  Sounds like a nightmare—but one induced not by free market exchange, but by regulations aimed at stopping it.

Opponents of innovation in the pricing of broadband capacity claim that such innovation harms consumers and reduces content, but our analysis shows this need not be correct.  Indeed, such prohibitions may in fact be the root cause of undesirable outcomes—higher prices to consumers, fewer content providers, and the shifting of business to the content affiliate of a broadband provider.

The network neutrality regulations recently set by the FCC prohibit for wireline networks the arrangements now being proposed by content firms for mobile networks.   Now the political interests groups will certainly step up their already existing efforts to extend such regulations to the mobile networks.   As we all know, the content firms were instrumental in the development of the Internet regulations, and it appears they now may be defending themselves against those very rules.  After writing our paper, I was convinced that eventually it would be the very content providers that supported network neutrality regulations that would become the most vocal opponents of the regulations.  It appears that I’m about to find out if that’s true.

 


Article printed from @lawandeconomics: https://www.phoenix-center.org/blog

URL to article: https://www.phoenix-center.org/blog/archives/475

URLs in this post:

[1] AT&T May Try Billing App Makers: http://online.wsj.com/article/SB10001424052970204653604577249080966030276.html?KEYWORDS=ATT+May+Try+Billing+App+Makers

[2] Network Neutrality and Foreclosing Market Exchange: A Transaction Cost Analysis: https://www.phoenix-center.org/pcpp/PCPP28Final.pdf

[3] academically-published in 2009: http://www.inderscience.com/search/index.php?action=record&rec_id=24782

[4] : https://www.phoenix-center.org/blog/archives/475?share=twitter

[5] : https://www.phoenix-center.org/blog/archives/475?share=facebook

[6] : https://www.phoenix-center.org/blog/archives/475?share=reddit

[7] More: #

[8] : https://www.phoenix-center.org/blog/archives/475?share=linkedin

[9] : https://www.phoenix-center.org/blog/archives/475?share=tumblr

[10] : https://www.phoenix-center.org/blog/archives/475?share=pinterest

[11] : https://www.phoenix-center.org/blog/archives/475?share=pocket

Copyright © 2020 @lawandeconomics. All rights reserved.