The "Great" Debate

To kick off the second day of Freedom to Connect, Charlie Firestone, executive director of the Aspen Institute, moderated a debate about how to create a "better" regulatory regime. Participants in the debate included Rick Whitt, MCI's Senior Director of Global Policy and Planning; Tim Wu, a visiting professor at the University of Chicago Law School; Randy May, a senior fellow and director of communications policy studies at the Progress & Freedom Foundation; and James Gattuso, a research fellow at the Heritage Foundation. What follows is a partial transcript of their discussion:

Charlie Firestone: This is the great debate over regulation. This is what Washington's about. Companies spend billions on it. Careers are built on it. Thousands of children are sent to school on it. What is the proper regime for regulation? If we can find fair advantage, that's the ballgame. What's the paradigm for regulation that will move forward consumer interest and public welfare?

Each new media has had a regulatory scheme based on the past. The beginning of radio was called the wireless. Now we have these stovepipe vertical regulatory regimes. Yet, as we know, in the home, we see different ways of doing the same thing. Should we regulate each platform individually, or should we have the same thing?

Resolve that the communications acts' stovepipe vertical regulatory regime should be replaced with a horizontal regulatory regime. To debate this, we have two teams.

Rick Whitt: Most of you here are familiar with what's behind a horizontal approach: the ubiquity of network layering, the agnostic nature of Internet protocols, and the end-to-end principles that make up the Internet. We now have a system that's broken up and makes no sense. It assumes clear, categorizable characteristics. The result is that we have an inflexible approach.

Service categories are examined by regulators in isolation. Regulators also tend to concentrate on the services provided to end users rather than look at the networks overall. This artifice fails to reflect the reailty of merged markets and networks. Services themselves can't be defined in any way that matches up with these silos. A packet is a packet, regardless of whether it's a data or video packet.

Forcing consistency stifles creativity. The problem is over-regulation, as well as under-regulation. We should build our laws around the Net, rather than the other way around. The layers knock over the silos. It also looks at the way layers interact with each other and gives you more granularity. Kevin Werbach just wrote a paper called Breaking the Ice that I think is a great piece of work.

The key benefits of a layered approach? It's a flexible conceptual tool. The concept of layering has been around since the '60s, even before the adoption of TCP/IP. The deregulatory nature is aimed at policy makers. It also helps to clarify the issues and give you a framework to get to the answers. Should we be using antitrust law? That fits nicely together with a horizontal view. Layering analysis helps uncover the critical roles at the edges of the network. It looks at what the consumer is doing rather than what happens in the background.

Randy May: Rick and I go back a long way as colleagues. Nevertheless, Rick spent a lot of time explaining why the vertical stovepipes we have now are no longer an appropriate regulatory framework. We all agree on that. And at the very end, he said the horizontal approach is better. That might possibly be true as well. But the real issue is whether the horizontal layers approach is really the place we want to be in terms of a policy framework. I don't think it is.

The reason comes down to this: It would be a mistake to take an approach that everyone acknowledges is based on techno-functional characteristics and replace it with a framework that is based on another set of techno-functional characteristics. We're in an environment now in which technology changes rapidly. There are always changes. It's difficult to anticipate how technology's going to change. You don't want to lock in a framework that's based on technology.

It would be difficult now or in the future to agree on how to define the layers. I have an article by him right here, and he refers to several layering models. They're all different. This is problematic. It's not worth directly overturning the stovepipes to get to this place. We need to get to a better place.

What is that better place? It's a regime in which we look at market power. When you get cable TV and Internet at home, is that one or two products? We need to look at what consumers think. That gets you away from these distinctions based on technology and functional characteristics.

Tim Wu: I want to relieve you of any doubt that the layer model is too difficult to understand. The layered model need not be complicated. It need not be any form of replication of the seven-layer OSI model. It can be as simple as a single classification and a single rule. The layered model we're talking about is no different than the model that's already in the head of the smartest FCC regulators. It's the difference between transport infrastructure and applications.

All a layered model would have to do is separate the issues about infrastructure and the issues about applications. They're different. The vertical silos obscure that control of the physical infrastructure is too often used by large telcos to block market entry.

Why is there so much opposition to Rick's proposal? It's the first proposal in awhile that would open competition.

James Gattuso: I don't think we have as much to disagree on as we've played out. No one likes the current regulatory regime. Left to its own devices, the government would screw that up trying to find a solution. We've heard two different descriptions of the layered model. That's the problem. We don't know what it says. It's a Gertrude Stein theory of regulation. It can be a helpful way to look at the world. Layers exist. But should it drive public policy?

The key factors are competition and choice. Layers can help you identify what the markets might be, but not always. Just because someone's in a layer doesn't mean it's competitive. There are other considerations, as well. Public policy needs to look at the problems we're facing directly, not as a secondary consideration.

Add New Comment