Public Policy Considerations Favor Fiber, But So Far Industry Isn’t Buying, Study Finds

Cisco forecasts that global Internet traffic will grow by 30 percent per year to 1.36 zettabytes by 2016, Koboldt said. That means that the gigabyte equivalent of every movie ever made will cross the Internet every three seconds, he said. Cloud computing and the “next big thing to come” will bring consumer and social welfare benefits, but will need high bandwidth and low latency, he said. Those are areas where fiber networks excel but they need to be deployed to the home, he said. There’s a big question mark over the ability of existing copper infrastructure to meet the digital agenda targets, he said.

There are several reasons why fiber isn’t taking off in Europe, Koboldt said. Rollout costs are substantial and largely “sunk,” exposing investors to the uncertainty of future demand. There’s the chicken-and-egg problem that services need infrastructure and infrastructure needs services, he said. There’s no point in financing services with high bandwidth and low latency if there’s no demand for them, but new services are key to making ultrafast broadband interesting to users, he said. In addition, the future regulatory environment is uncertain. At this point, the “wait and see” approach is quite attractive and has few downsides, he said.

Investment in fiber isn’t just suffering because of the prospect of regulation, Koboldt said. The threat of having to provide cost-based access to fiber matters, and promising to allow high returns on FTTH investment is unlikely to be enough, he said. Potentially very large returns in successful cases are needed to compensate for the risks of buildout, he said. Regulators may not want to concede returns of 30-50 percent in successful cases, he said. And permitting higher charges for fiber access may not be relevant if consumers aren’t willing to pay more for fiber, he said.

This is where the competing copper access charge comes into play, Koboldt said. Copper and fiber compete, not just for funding but also for end-user demand, he said. Incumbents are likely to invest in fiber to get higher access fees, but new entrants are more likely to favor cheaper access to existing copper services, he said. Lower retail prices for copper-based access products limit return on investment on fiber, he said. The fiber premium is potentially very limiting because consumers don’t appreciate that what they get from fiber isn’t necessarily what they expected, he said. There’s evidence that low price drives fiber take-up, not the technology’s higher bandwidth and low latency, Koboldt said. The reason fiber has succeeded in some countries is that it’s cheaper, not that it’s better, he said.

Addressing those problems will probably require a mix of cooperative public-private infrastructure agreements, said DotEcon associate John Gunnigan. In addition, some regulation is good for promoting fiber, he said. Consumers should be given better information about service quality, as has been done in the mobile telephony arena. Challenges in the fiber business case should be addressed by considering differentiated access prices and co-investment from service providers; and there should be more certainty and clarity about specific regulation of fiber networks in the future, he said. During the transition phase, fiber investment could be boosted by measures preventing access price differentials from feeding through to retail prices, he said. For example, there could be a “tax” on copper access to drive a wedge between copper wholesale and retail prices, or, in a more radical approach, incumbents could be allowed to withdraw copper-based access products as soon as their fiber networks are available, he said. Or there could be higher access charges on fiber and copper networks.

There are four possible solutions to spurring fiber access networks, Gunnigan said. One is direct public sector involvement through public-private partnership, state aid, and so forth. There is also full government funding, where the government pays for and owns a network intended to be wholesale, open access only, he said.

There are also two “blue sky” options, Gunnigan said. One is to make regulators promote investment in fiber by allowing or disallowing certain investment, as is done in airport regulation. But this leads to picking winners, does away with technology neutrality and rules out all copper upgrades, he said. The most far-out option would be a “fiber switchover” policy, he said. But it wouldn’t offer a dividend like digital transition did, and it’s unclear how such a policy could be coordinated across the EU, he said. Still, despite huge legal issues surrounding operators’ freedom to use their assets, and serious practical problems, it could be a last resort if fiber deployment doesn’t take off, he said.

Asked if they advocate moving away from technology neutrality, Gunnigan said the issue is whether Europe will see the fiber rollout that the digital agenda demands under current rules. If not, a different model may be needed, which could lead to choosing technologies, he said. The question must be addressed, he said.

Technology neutrality runs the risk of becoming a “mantra,” said Koboldt. Allowing the market to decide which technology is used is fine as long as there’s effective competition, he said. If there isn’t, it’s quite possible that there will be market failure in terms of the market-driven technology, he said. That doesn’t mean regulators should pick winners, because their track record isn’t good, but that if market-based technologies aren’t working, they will have to consider another tack, he said. At some point, either technology neutrality or the digital agenda targets will have to give in order to meet those goals, Gunnigan said.