TCS Daily

"Separating 'the last mile' of the regional Bells' operations from their other services would promote competition in broadband."

By James K. Glassman - October 9, 2001 12:00 AM

Separating "the last mile" of the regional Bells' operations from their other services would promote competition in broadband, Robert Hall, noted Stanford University economist, told Tech Central Station Host James K. Glassman. A new study by the Hoover Institution scholar and Columbia University economist William Lehr says "conflicts of interest" keep the Bells from fully cooperating with competitors. Structural separation overcomes that problem, as was proven by the break up of the old Bell monopoly in 1984. "Long distance became a competitive industry and the local phone companies cooperated willingly with all long-distance carriers, " Hall said. The Bells' alternative of regulatory relief from requirements to open new local network elements to competitors would lead to a " big regulated phone company again," he says.

James Glassman: Robert Hall, there's a good deal of controversy these days about how to deploy broadband more broadly. First, do you think there is a crisis surrounding the deployment of broadband?

Robert Hall: I'm not sure I'd call it a crisis. But there are certainly major structural problems that relate to the fact that the entities that are most directly in control of the deployment of broadband have conflicts of interests and limited incentives to push it as hard as they could. The result is that although broadband is moving rapidly, it's not moving as rapidly as it should, and we remain in a situation where only a few percent of the American public has access to broadband.

Glassman: When you say conflicts of interests, you're talking about something that you describe in your paper as the difficulties of having one firm be cooperative and competitive at the same time, is that right?

Hall: That's exactly right.

Glassman: Could you explain that?

Hall: In the case of the phone companies, which represent one group for broadband through the existing copper wires that they have to connect to homes, they are supposed to provide the use of those wires to other sellers of phone services. But since they're also providing the same services themselves, they are being asked to be a supplier to someone who's also a competitor. The result is that they tend to be a very reluctant supplier and to stall, basically preventing the rapid expansion of the use of those wires by a competitive industry that could grow rapidly.

Glassman: Now, that particular set-up was the result of the 1996 Telecommunications Act. Do you think there's any way that the Telecom Act of '96, if aggressively enforced, could bring about the kind of competition that would bring us broadband at high-quality, low-cost wide deployment?

Hall: It's hard for me to answer that question. The approach that the Telecom Act took was not to try to deal with the competition versus cooperation issue in some clear, direct way, but rather to put it in the hands of regulators, whose job it was to get telephone companies to cooperate. But legislating cooperation and trying to enforce it through the regulatory system has proven to be frustrating to the consumer, who's still waiting for many of the benefits and alternatives. The other way to go, not embodied in the Telecom Act, might have done a better job.

Glassman: Let's talk about the other way to go. You refer in your paper back to 1984 when there was a structural separation of the Bell system that certainly seemed to produce vigorous competition. Could you just describe the history and then tell us why you think it might work going forward?

Hall: Sure. Effective in 1984 the long-distance business was separated from local telephone service. This is important, because the long distance industry depends on the local phone companies to provide their connections both to the person making the call and to the person receiving the call. As a result, the industry absolutely requires the cooperation of local phone companies. Before 1984, when the same company was the local phone company and the long-distance company, as in the old Bell system, the local phone company was very careful not to cooperate with other long-distance carriers, such as MCI. Then came the structural separation, which said the local phone company was no longer involved in long distance. Long distance became a competitive industry and the local phone companies cooperated willingly with all long-distance carriers, both with what became AT&T and companies such as MCI and Sprint, which developed their own long-distance businesses. It was a very successful application of this idea of structural separation.

Glassman: Wasn't there a certain amount of cooperation, however, that was necessary because AT&T did lease its lines to its competitors? In other words, you had a long distance company that also was forced to cooperate with competitors. Is that right?

Hall: It was, but it was in an area where the competitors could build their own networks. And that's certainly what they did. So, it was easier to get that type of cooperation from AT&T because there was a meaningful alternative. And the presence of a meaningful alternative meant that if AT&T didn't cooperate then they just wouldn't have the business. That gave them a strong incentive to cooperate. Unfortunately, there isn't a similar incentive when you turn to similar issues now rising with respect to local telephone service.

Glassman: Is that because at this point the last mile involves so much of a sunk cost investment that it would be very hard for competitors to say, "Well, if you don't cooperate with us in leasing lines, we'll just start our own"?

Hall: That's right. "We'll just tear up streets and put in new wires," that's just not going to happen.

Glassman: How would structural separation work to solve the problem of bringing competition to local service and then, the important link, to broadband?

Hall: Well, the basic idea would be to create a new entity out of each local phone company. It would be minimal in the sense that it would just own the last mile and limited parts of the rest of the network -- the parts that are difficult for rivals to reproduce. That part would separate and become a separately owned company. It would be regulated, and it would then serve customers that actually provided phone service and broadband service over those wires. Those would be the competitive companies. They would include not only the other parts of the current local phone companies, but also the companies that would like to compete in local service, including companies such as AT&T. The idea is to harness this principle of structural separation at a level closer to the part of the phone system that is difficult to reproduce, mainly the local network.

Glassman: So, you have one company that would own the last mile and then would lease access to that to all comers?

Hall: Exactly.

Glassman: And that would necessitate, I expect, that the prices at which it does this leasing would have to be regulated because it would be a monopoly.

Hall: It would, exactly. And it should be a monopoly, because it's inefficient to try to have more than one set of wires going to homes. It is desirable, since one set of wires is plenty of capacity for all the communication that a family wants to do. It doesn't really make economic sense to have more than one set of wires. It's not an area where the idea of competition through separate physical facilities really makes any sense.

Glassman: I asked Michael Powell (the new chairman of the Federal Communications Commission) a few months ago, when he was talking about how the spectrum was owned by the public, "Well, if the Spectrum is owned by the public, is the last mile owned by the public in the sense that enormous subsidies, government protection, nurturing were involved in developing this facility?" Is there an argument that when you split the company, if you split the Bells in two, could the last mile be eventually a publicly owned utility?

Hall: Public ownership of facilities has a mixed history. Generally speaking the American principle has been to not to push for public ownership. On the other hand, some cities are building their own communications facilities. The City of Palo Alto, for example, has fiber facilities. In fact, you can get fiber to your own home in Palo Alto through a municipally owned system that's very efficient, and people are happy with it. I think if you do go for a government solution, though, it should be one based on local competence and not on trying to make public ownership be a universal feature. There are some governments that you really wouldn't want to put in charge of something as important as communication.

Glassman: You are both a professor of economics at Stanford and you're also a fellow at the Hoover Institution, which is part of Stanford. And Hoover has a reputation as a conservative think tank.

Hall: I would change that slightly. It's very dedicated to the principle of the free market. Whether it's conservative or not actually varies quite a bit, but the common thread of thinking here is the belief in harnessing the principles of the free market.

Glassman: I've pointed out the same thing about the American Enterprise Institute, with which I am a affiliated. But the point is that you and your colleagues have a great deal of respect for property rights and for free-market competition. Some of the people who say that structural separation is not a good idea argue that it's essentially a seizure of property that's owned by a private corporation. And just as these people have objected to the break-up of Microsoft, they're saying that a company, such as Verizon, should not be broken up by law. How do you respond to that?

Hall: Well, first of all, there's no threat to the principle of property rights. Just to go back again to the history. William Baxter, the architect of structural separation, was one of the most rigorous free-market thinkers you could imagine and was always regarded as a strict free-market thinker. He was brought in to handle antitrust issues by the Reagan administration, which was generally devoted to the free market. So there's really no conflict between free-market thinking and the principle of structural separation. It is true that the way that structural separation is accomplished should give full regard to property rights; that is, the shareholders of the phone companies under a regulatory regime have put their money into building these facilities and the facilities should not be taken from them. The Fifth Amendment to the Constitution forbids that. But, just as we handled that problem in 1984, we can do it again.

Glassman: By setting up a separate company, which the current shareholders of the Bells would own?

Hall: Right. And also the regulation of that must give them regard. I mean, you can regulate value out of existence by setting prices too low, and that's not what should be done. The prices should be realistic with respect to the costs involved in creating the network.

Glassman: The Bell companies argue that there's nothing wrong with the Telecommunications Act of '96 that a few changes would not fix. One of the major changes would be to forbid companies -- their competitors -- from tapping into or connecting with improvements that they make in their system ...

Hall: Relieving them of the obligation to provide the services of approved parts of the network.

Glassman: Right. That is the basis of the Tauzin-Dingell bill, which in its original form, anyway, also allowed the Bells immediately into the data part of long distance without meeting requirements that were in the Telecom Act. What do you think of that approach?

Hall: Well, it's getting into a very hybrid solution. We've always had the possibility that we should go back to the pre-1984 system in which we have a unified phone company that does everything and doesn't cooperate with any other entity. Instead of using the principles of competition in the free market, we could have a big regulated phone company again. A lot of these ideas -- these improvements -- would help reduce competition. They would eliminate the 1984 restrictions, and the idea of allowing transported data is one of those. Those all would be moving back toward the pre-1984 model, and that model didn't work very well. Those of us who believe that a competitive solution to these problems is a superior solution are hoping for a different type of restructuring of the '96 Telecom Act, which there's pretty wide agreement needs improvement.

Glassman: So you believe that if a bill like Tauzin-Dingell were passed, it would lead to more regulation and less competition than a solution like structural separation?

Hall: Exactly. I think it's fairly clear. The whole idea of what the proponents are saying is let's make the local phone companies bigger entities that can do more things. Let's relieve them of their obligation under the '96 Telecom Act to share their facilities, particularly if the new facilities don't have to be shared. Well, ultimately, all facilities are going to be new facilities, and it would mean a return to the old model. It worked, but it didn't work as well as what we could achieve through a more competitive approach.

Glassman: Well, thank you very much, Professor Hall. I want to make sure that everyone knows that your paper with William Lehr is available through our website,

Hall: Great.

Glassman: Thank you very much.

TCS Daily Archives