Aspen Institute Fellow Recommends Big Changes In USF, Intercarrier Compensation, Use of Satellite Broadband

Blair Levin, Aspen Institute Fellow, says $10 billion, spent over 10 years, is enough to provide a minimum 4 Mbps downstream service for Americans in rural and isolated areas.

He proposes that the money be gotten by revamping the Universal Service Fund, including reducing or freezing funds currently allocated under the Interstate Access Support and Interstate Common Line Support funds, steps that would have immediate impact on many rural telcos and rural mobile providers.

Levin points out that there are about seven million housing units (about five percent of the total) without access to the 4 Mbps downstream and 1 Mbps upstream services the Federal Communications Commission now believes is a minimum.

The FCC has estimated the cost to provide such service with wired broadband at $32.4 billion, with a revenue projection of only $8.9 billion, leaving a $23.5 billion gap.

But Levin maintains that the costs are so high because of costs to build wired infrastructure to just 250,000 homes. Reaching those 250,000 homes would cost about $13.4 billion. Levin does not appear to believe that is a wise investment. So he suggests using satellite to reach the most-isolated, high-cost homes, instead. That would free up enough money to build out facilities to the roughly 6.75 million other rural homes.

In 2010, the federal fund (USF) is projected to make total outlays of $8.7 billion, but not specifically to support broadband access.

Some $4.6 billion is set aside for deployment of networks to high-cost areas, where population density or other factors would cause the price of services to consumers to be at a level that would not reasonably compare to urban areas (this is in addition to the 21 states that have similar high-cost funds that distribute a total of over $1.5 billion).

About $1.2 billion is allocated to provide discounts to make basic telephone service available
and affordable to low-income consumers (in addition, 33 states have similar programs).

Another $2.7 billion is reserved for subsidizing telecommunications services, Internet access and
internal connections to enable schools and libraries to connect to the Internet (in addition, nine states have similar programs).

Making better use of existing funding should be the first priority in any reform effort, Levin says. The universal service contribution factor—an assessment on interstate and international charges that usually appears as a surcharge on consumers’ phone bills—is already at about 15 percent (having risen dramatically in the last decade), he notes.

Further increases would create both political and policy problems, he suggests.

"More ambitious goals in terms of network speeds, at this time, would cause such an increase in the assessment on the current system that it could backfire in terms of driving America’s use of broadband," Levin argues. "For example, the FCC calculates that going from 4 Mbps to 6 Mbps would increase the investment gap by more than 100 percent."

The rational approach would be to avoid building fixed-line networks to serve a quarter million homes, at a cost of $13.4 billion, using satellite broadband. That would free up nearly all of the available funds to build fixed-line networks for 6.75 million rural households.

There are a number of problems with the current Universal Service Fund, Levin suggests. "Among these are that the fund is targeted to support analog voice requirements, rather than data networks; that the fund does not target unserved areas but rather funds particular kinds of companies; that the fund provides incentives for inefficient build outs; that there is no accountability for actually using the funds for their intended purposes; and that the support programs are not coordinated to
leverage the funds to maximize broader policy objectives," says Levin.

Though rural telcos might not like the idea, there are a number of current programs within the Universal Service Fund that need to be changed.

About $4 billion could be redireted to broadband support, over 10 years, by reductions in USF payments to wireless providers.

Interstate Access Support (IAS) payments could be reoriented to broadband, adding approximately $4 billion over 10 years.

Freezing Interstate Common Line Support (ICLS) would limit the growth of the existing high-cost fund and result in savings of about $1.8 billion over 10 years. Those funds also could be redirected to broadband support.

To accomplish this, the FCC would have to require that rate-of-return carriers move to incentive regulation.

Phasing out remaining legacy high-cost support for competitive carriers (wireless, primarily) would yield up to an additional $5.8 billion over the coming decade.

Together these actions would result in between $15 and 16 billion in savings from the existing high-cost program that could be used to support broadband facilities construction.

As logical as the changes might be, there will be resistance from any number of firms that currently rely on the current mechanisms for significant portions of their current revenue, including but not limited to, rural telcos.

Why Wireless Might be the Best Way to Serve the "Unserved"

A new study produced by the Federal Communications Commission might be interpreted as arguing for a wireless approach to bringing broadband to many unserved locations, said by the FCC to number seven million homes, adequate for service at 4 Mbps downstream and 1 Mbps upstream.

The most rural 250,000 housing units account for $13.4 billion of the total $23.5 billion investment required. In fact, as cost varies inversely with density and distance from a central office or cable headend, the cost curve is a reverse Pareto distribution (a reverse "long tail").

The FCC says wireless, such as a fourth-generation wireless network, is the lowest-cost technology in 90 percent of cases. The point is that population density generally is inversely related to access cost.

Broadband Stimulus: "Access" is Only 1 of 3 Problems

The "broadband stimulus" programs established by Congress in the American Recovery and Reinvestment Act, making $7.2 billion available to improve broadband service in rural, un-served and underserved areas, actually must address three different sets of problems, says the National Cable & Telecommunications Association.

The NCTA notes that "unserved areas" represent about nine to 10 million households, typically in rural areas. NCTA refers to areas without "wired" access, as satellite broadband is available in those areas, though in some cases not technically accessible because of obstructions (mountains are the biggest issue, though foliage and trees or other structures in the line of site can be an issue in some cases). NCTA thinks projects extending broadband access to these households should be given highest priority.

"Underserved populations" represent a different problem. About 35 million households already have access to wired and satellite broadband, but do not currently use the services. The problem there is not "access," but "affordability" or "perceived value." Many of these potential users simply do not use PCs or the Internet, even if affordability is not the key problem. Simply building more access networks will not solve these problems. Instead, "demand stimulation" is the problem to be addressed in the "underserved" bucket of potential users.

Demand-side stimulus is what is needed here, and that might include training, equipment subsidies or usage subsidies.

Separately, there are households in underserved areas which have broadband access, but not at speeds generally available throughout the rest of the country. In these areas, the government should proceed with caution, NCTA says.

The need for subsidy in these underserved areas is not as great as in unserved areas or for underserved populations, and subsidizing infrastructure in these areas runs the risk of subverting the commercial deployment already taking place, NCTA argues.

Subsidies to these underserved areas should therefore be carefully structured so as not to favor one technology over another, one provider over another, the public sector over the private sector, or otherwise upset marketplace dynamics, NCTA argues.

NCTA notes that more than 92 percent of U.S. households actually have wired broadband access available to them. Satellite providers would argue that the eight percent without wired access are precisely the segment best served by satellite services.

About 18 states are least represented by wired access facilities, where households unable to buy a wired broadband service are less than 94 percent of all homes.

Among unserved geographic areas, subsidies therefore should be targeted first to areas in which service would not otherwise be provided and that could support the ongoing costs of providing broadband service if government funded the costs of the underlying infrastructure, NCTA argues. These are areas where infrastructure cost prevents commercial payback, but where an on-going business case can be made, if infrastructure deployment is not an issue.

New mapping of broadband facilities will be finished only after the funds have been disbursed, so "mapping" does not help either NTIA or RUS make its awards.

Merely providing broadband access does not necessarily mean that customers will subscribe to it, NCTA and other policy advocates note. The larger problem is that many consumers fail to subscribe to broadband service even when it is available. That's a "demand" problem, not a "supply" problem. "Researchers studying broadband access have concluded that 'lack of interest' in broadband is the main reason that people do not purchase the service.

Indeed, about onequarter of adult Americans do not use the Internet at all; these individuals are disproportionately lower-income and older than average Internet users, NCTA notes.

Grants should be used to provide targeted subsidies to make broadband services more affordable, NCTA suggests.

Access isn't the only problem the "broadband stimulus" investments must tackle.