Even as the piece probably will induce significant mashing of teeth among network neutrality advocates, I strongly recommend a recent article by Greg Sidak entitled A Consumer-Welfare Approach to Network Neutrality Regulation of the Internet; available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=928582. It’s a comprehensive document with many compelling and legitimate points.
I am not completely opposed to “access tiering” if an ISP can offer complete end-to-end enhanced routing without violating Service Level Agreements and without deliberately dropping packets and degrading service to “regular” peering and transit users. Greg makes some fair points about the right of a network operator to discriminate on price and QOS both downstream to end users and upstream to other ISPs and content providers. I’m not sure how this can be done in light of existing peering and transit agreements, which are largely best efforts, but the potential for a complete end-to-end, “better than best efforts” exists.
He and I part company on a number of issues including his conclusion that the broadband access marketplace is robustly competitive and therefore there is no risk of price squeezes, anticompetitive pricing of access tiering, collusion, etc. Greg sees a competitive marketplace, based on the inclusion of dial up Internet access and based on the FCC’s broadband penetration statistics. I don’t agree that VoIP easily routes via POTS Internet access, particularly because Vonage and others state their service requires broadband access. The FCC’s use of zip codes is a very flawed measure for determining whether consumers have competitive options. The zip code measure attributes access in terms of numbers of ISPs if even one potential subscriber exists, regardless of price. So just about every zip code area has satellite access despite the lack of significant cross-elasticities between a $100 a month service and a $50 service. BTW Greg reports an average price of $25 for broadband access! I’m paying $60, but then again I’m one of the extraordinarily rare U.S. consumers in one of those rare rural zip codes that has one and only one option, cable modem service, unless you include DBS.
Greg reiterates much of the private property common law taking arguments to justify the premise that network operators should have total control over their networks including the right to deny interconnection, refuse to carry specific content or applications, including unaffiliated ventures’ VoIP traffic, and the option of vertically integrating and favoring corporate affiliate’s traffic. He does not give much credence to the rights of DSL/cable modem subscribers to expect an unfettered bitstream pathway to the Internet cloud.
I was confused by Greg’s extensive examination of the Madison River case and his apparent endorsement of the lawfulness in a network operator’s option of blocking downstream carriage of specific bitstreams. He does not acknowledge that the FCC could engage in an enforcement action only because Madison River falls under Title II of the Communications Act, as a telecommunications service provider and not an ISP. Madison River refused to terminate telephony traffic. I don’t think Title I would have the same force as Title II if an ISP refused to terminate traffic, particularly if the traffic is deemed part of an information service.
Lastly I did not see much discussion of whether a network service provider has the ability and inclination to drop packets and create congestion. Absent robust inter-modal competition an ISP could accrue monetary benefits by punishing ventures that do not agree to take more expensive routing options and who trigger robust demand for their services by end user DSL and cable modem customers.
I am not completely opposed to “access tiering” if an ISP can offer complete end-to-end enhanced routing without violating Service Level Agreements and without deliberately dropping packets and degrading service to “regular” peering and transit users. Greg makes some fair points about the right of a network operator to discriminate on price and QOS both downstream to end users and upstream to other ISPs and content providers. I’m not sure how this can be done in light of existing peering and transit agreements, which are largely best efforts, but the potential for a complete end-to-end, “better than best efforts” exists.
He and I part company on a number of issues including his conclusion that the broadband access marketplace is robustly competitive and therefore there is no risk of price squeezes, anticompetitive pricing of access tiering, collusion, etc. Greg sees a competitive marketplace, based on the inclusion of dial up Internet access and based on the FCC’s broadband penetration statistics. I don’t agree that VoIP easily routes via POTS Internet access, particularly because Vonage and others state their service requires broadband access. The FCC’s use of zip codes is a very flawed measure for determining whether consumers have competitive options. The zip code measure attributes access in terms of numbers of ISPs if even one potential subscriber exists, regardless of price. So just about every zip code area has satellite access despite the lack of significant cross-elasticities between a $100 a month service and a $50 service. BTW Greg reports an average price of $25 for broadband access! I’m paying $60, but then again I’m one of the extraordinarily rare U.S. consumers in one of those rare rural zip codes that has one and only one option, cable modem service, unless you include DBS.
Greg reiterates much of the private property common law taking arguments to justify the premise that network operators should have total control over their networks including the right to deny interconnection, refuse to carry specific content or applications, including unaffiliated ventures’ VoIP traffic, and the option of vertically integrating and favoring corporate affiliate’s traffic. He does not give much credence to the rights of DSL/cable modem subscribers to expect an unfettered bitstream pathway to the Internet cloud.
I was confused by Greg’s extensive examination of the Madison River case and his apparent endorsement of the lawfulness in a network operator’s option of blocking downstream carriage of specific bitstreams. He does not acknowledge that the FCC could engage in an enforcement action only because Madison River falls under Title II of the Communications Act, as a telecommunications service provider and not an ISP. Madison River refused to terminate telephony traffic. I don’t think Title I would have the same force as Title II if an ISP refused to terminate traffic, particularly if the traffic is deemed part of an information service.
Lastly I did not see much discussion of whether a network service provider has the ability and inclination to drop packets and create congestion. Absent robust inter-modal competition an ISP could accrue monetary benefits by punishing ventures that do not agree to take more expensive routing options and who trigger robust demand for their services by end user DSL and cable modem customers.