Wilkie, Aronson, Bar
Towards Neutral Ground on Net Neutrality, Simon Wilkie, Jonathan Aronson, Francois Bar. But Wilkie wants to talk about what's going on at Annenberg (USC).
He notes that on Feb. 10 the USC Annenberg Center hosted a closed conference on Net Neutrality. Included content providers, network operators, equipment manufacturers, academics, Wall Street analysts, public interest. They developed a broadly supported list of principles. Paper explains that process and proposal.
Long and extensive dialogue. Emerged with a “split consensus.” Almost a consensus, with a few holdouts. More agreement than disagreement.
Annenberg Principles (see annenberg.edu). This is the first in a series of processes Annenberg wants to have/host.
First point of consensus: Net Neutrality is a horrible term. Too value laden and politicized. Need better terms. Ambiguity re what broadband meant. Speed, QoS? fixed and/or wireless. We're going to take 1.25Mbps to the home as the standard.
NN components: End to end principle — any end user can access any other content provider or end user, attach any non interfering CPE, point of interconnection.
Nondiscrimination: two types — customer level (based on speed, access, quality and quantity of service); transport level (in terms of interconnection, many types).
Four categories for transport level discrim:
DS1 — “neutral” eg a posted tariff; any firm can attain a priority level by paying posted price.
DS2 — “non-neutral” exclusive deal eg a preferred or affiliated content provider receives access to end users on exclusive terms
F1 foreclosure Type 1: network operator forecloses access to its subscribers from certain other content providers sites/users
F2 foreclosure Type2: network operator offers affiliated exclusive content only to its own subs.
In the discussion, even consumer advocates thought tiering okay — charging more for higher capacity fine. Neutral transport price discrimination D1 may be efficiency enhancing. But discrimination of D2 and F1 is likely harmful. Discrimination F2 is ambiguous — take a “wait and see” approach.
Modified end to end principle: we looked at what's available to consumers now. If there are deviations now, consumers shouldn't lose those if they are useful. Speeds of 1.25Mbps ubiquitous today. Allows delivery of VHS quality video. So let's modify the principle. They should have this 1.25 available in the marketplace. But be able to pay more for more.
Economic theory behind this: costs a lot to build a network. Vertical integration can be good. We've got mix and match components, so we can have multiple equilibria. Equilibrium does not imply efficiency. So bottom line is that we need better information on the market parameters.
Sunk costs: investments in network upgrades are sunk. Low marginal costs of moving bits. Competition in 'dumb' pipes' drives prices to marginal cost. Fixed costs not recouped — therefore people won't invest. There's a need for some type of price discrimination.
Internet is really a two sided market. Needs an intermediary to make the market. Multiple sources of revenue for network providers increases incentive to invest. There's a hold up problem: if intermediary can extract rents then that discourages innovation at the edge (content providers and end users).
So - what to do? Market reality is a duopoly likely for the short term. Some potential entrants — aws, sprint, clearwire, power line. But long legacy of entrant failures. Notes that incremental value of adding a MVPD customer is around $2000. Can't justify the investment, so has to be another revenue sources.
Case study 1 — videotext market. Look at minitel v. Bildschrimtext (DT). Nascent information services. Technology developed in UK in 1970s. Minitel deployed by France Telecom in 1982. Bildschrimtext deployed by DT. Minitel still functioning, Bildschrimtext is dead. Minitel was open to any information providers. Neutral but not open. Non discriminatory access (like DS1). Created private information services as “kiosks”. Huge initial success. Limited by use of only FT services. eclipsed by Internet. DT was closed — controlled content, limited both innovationa and usage. Killed off in 2001. [pretty gross case study!]
Market falure - case study 2 Australian cable TV market, similar cultrue to US, similar suburban population. Duopoly policy. No program access rules. Each franchise gained exclusive access to movie studio libraries. Take rate about 20% compared with 85% in US. Implies significant welfare losses.
These are real issues. We're concerned about high prices because of lack of competition and we're also concerned about low quality.
Principles:
Both operators and customers should win.
Any regulation should be national and uniform and with a light touch.
There should be basic access to broadband — meaningful, neutral internet connectivity. 1.25Mbps. Beyond this, all bets are off. Baseline should be reviewed every four years.
Customers should receive clear, understandable terms and conditions of service re preferred deals/content discrimination. FTC should enforce.
Comments
One Response to “Wilkie, Aronson, Bar”
Got something to say?
I don't know if he'll admit to being this old, but I took a class from Jonathan Aronson in Fall, 1981 in USC's Thematic Option program for freshmen. He was absolutely terrific. Ask him if he remembers the guy who got the only A he gave that semester.