The following is an excerpt from my Reply Comments on FCC’s NPRM on reclassification of broadband ISP services under Title II of the Telecom Act. A link to the full filing will be provided when available.
Proponents of Title II reclassification have cited a variety of economic analyses that suggest that the Internet’s openness is a key driver of its value, and that without rules that clearly prevent blocking and discrimination, there would be less incentive to produce online content or develop new applications, thus reducing the overall worth of the network. Other economic studies have found that non-neutral conditions in the broadband market might maximize profits for ISPs but would ultimately minimize consumer welfare.
The first argument, that applications innovation will be harmed without rules against blocking and discrimination, has been made since the debate over network optimization and prioritization began ten years ago. Whatever the academic analyses have predicted, in practice, innovation has done nothing but explode over that time. At one end, there are the well-known applications such as Facebook, Twitter and Netflix, which have become Internet power players over a relatively short time period. More telling is what has happened at the other end. Wireless apps, which basically are Internet interfaces, are so easy to design that Mom-and-Pop stores, Boy Scout troops and small churches can create and disseminate them for free. Today’s Internet is as functional for a multinational corporation as it is for a suburban dentist. All this has occurred without the stringent regulation that critics say is necessary. Instead, the FCC’s current guidelines have served the industry well because they wisely to run in tandem with market forces, not against them.
This brings us to the second contention, which raises the false dichotomy between maximizing ISP profits and protecting consumer welfare. This argument applies the idea that low prices are beneficial for consumers. That is true, as long as prices are not set at an arbitrarily low price that does not reflect the market value of the product or service and the cost of bringing it to market. In the long run, these types of price controls work against consumers. For example, rent control was once believed to be a fair way to make housing affordable for people with modest incomes. In the end, it benefited a minority of renters who were lucky enough to maintain years-long residency in rent-controlled housing, often by skirting occupation rules, while creating a scarcity of apartments for all other consumers. Even for those lucky enough to live in a rent-controlled apartment, their landlords had no incentive to maintain their properties beyond minimal standards because their low return would never cover their investment. Rent control laws were popular in the 1960s. Faced with the untended outcomes decades later, most cities have since done away with them.
Title II regulation, which would open the Internet to all manner of price controls, stands to create just such a situation in the online world.
In practice, the economics of network interconnection require more flexibility, which in turn allows all parties—service provider, content provider and consumer alike—to derive value from the Internet. The economics begin right at the point of interconnection, and the beauty is that the market has worked. Many Internet peering and transit relations are symmetrical: the amount of data carriers transfer to each other over the long term is balanced. Carriers agree not to charge each other because, when all is settled, it’s a wash. Any small difference in transit and connection is not worth the transaction and accounting costs that would be incurred to recover it. Everyone saves time, money and hassle. But that doesn’t mean it’s a good idea to make it a universal rule. For the equation changes when the data stream becomes asymmetrical: when one network begins transmitting more data than another, and that other network must bear the cost of managing and delivering that data with no offsetting reciprocity. At that point, it is reasonable for the network responsible for handling the greater amount of traffic to ask for compensation. The agreement between Comcast and Netflix is an example. Although Netflix complains about paying a service provider for handling the last mile, it is a justifiable demand based on the cost Comcast must incur. Should the FCC regulate or prohibit such negotiated arrangements, it will depress network value, not increase it. The FCC traveled a similar road with UNE-P requirements in the late 1990s and early 2000s. The Commission sought to regulate the prices incumbent local exchange carriers (ILECs) could charge competitive local exchange carriers (CLECs) for access to their infrastructure. The result was that ILECs, faced with giving away their facilities at low-market rates, slowed investment in broadband. Meanwhile, the CLECs based their business models on cheap access to third-party networks and abandoned their own construction plans. When the Supreme Court struck down UNE-P, CLECs found themselves in an untenable business situation. A policy with goal of increasing broadband investment and competition, in the end, set back both.
There is a high risk that the FCC, again by attempting to use regulation to distort the pricing and value of network access, will derail Internet investment growth and value in similar ways.
 Christopher Yoo, Testimony before U.S. Senate Judiciary Committee hearing on the Comcast-Time Warner Cable Merger and the Impact on Consumers, April 9, 2014, pp. 5-7, available at http://www.judiciary.senate.gov/imo/media/doc/04-09-14YooTestimony.pdf.
 Larry Downes, “The Full Cost of Net Neutrality,” Forbes, Sept. 26, 2011, available at http://www.forbes.com/sites/larrydownes/2011/09/26/the-true-cost-of-net-neutrality/.
 MHz-Pop is a measure of spectrum value. Price per MHz-Pop = Sales Price / (MHz of license x population covered). For example: A license is for 15.0 MHz, covers 1 million in population (pops); sale price is $11.4 million; price per MHz-Pop = $11,400,000 / (1,000,000 x 15.0), or $0.76. More available at http://www.srr.com/article/spectrum-licenses-valuation-intricacies#sthash.BXZJo7Tv.dpuf.