Ten years after the 1996 Telecommunications Act, which was supposed to accelerate the introduction of high-speed communications systems, the U.S. has dropped from first to 15th in the world for the percentage of residents with high-speed Internet access. Increasingly, local governments are stepping in where the private sector and federal government have failed. Hundreds of cities are currently debating strategies to develop citywide broadband networks. They share common goals -- universal coverage, equitable access, increased competition, and more effective use of the new communications systems for municipal services, especially those related to public safety.
Their discussions often ignore or give short shrift to a crucial issue: who will own the information network? Ownership matters. As we will argue in this report, public ownership of the physical infrastructure may be the only way to guarantee future competition. It is clearly the only way that communities can influence the design of their future information systems. And public ownership can allow a community to tap into the growing exchange of information to generate significant revenues while enabling all households in the city to have affordable access.
As of mid-2006, more than 650 cities own telecommunications systems. These range from downtown fiber optic networks that connect public buildings and major businesses, to citywide Wi-Fi networks that offer retail service to all residences and businesses. These publicly owned networks have proved remarkably successful in meeting the community's need for advanced services at fair prices.
This first wave of public ownership largely occurred in cities that already owned their electricity networks. That ownership was born a century ago out of public frustration at privately owned utilities' refusal to extend service beyond larger cities. Today, municipal electric utilities are expanding into broadband telecommunications, born of a similar frustration at telecommunications companies' slow response to the needs of small and rural communities.
More recently, communities without municipal electric utilities have begun exploring a governmental role in accelerating the deployment of highspeed information networks. These urban and suburban communities already have some level of high-speed Internet access through cable and telephone company networks.
The incumbent suppliers vigorously oppose any municipal involvement, either through public ownership or by facilitating a competitive network. At the same time, companies that had been leasing space on incumbents' networks view municipal involvement as an opportunity to build their own networks, with public support. They offer cities what appear to be very attractive arrangements if the city grants them an exclusive contract.
Large cities -- Philadelphia, San Francisco, Minneapolis, Boston, Houston, Seattle, and others -- have become the front lines in the battle for affordable, high-speed information and communication networks.
So far, these larger cities have tended to choose privately owned, for-profit networks. They choose expedience over security. They choose the comfort of dependence rather than the risks and rewards of independence. They choose a small, guaranteed income via a franchise fee over the potentially large benefits, financial and otherwise, that stem from public ownership. We believe such a choice does a disservice to their households and businesses, as well as the local government itself.
What is Public Ownership?
Public ownership means ownership by citizens, customers, or the community. It comes in many different forms.
Municipal Networks are owned by a local government entity. This may be the city itself, as in Saint Cloud, Florida, or a municipal utility, as in Moorhead, Minnesota.
Cooperative Networks are customer-owned, as is the case with the Mountain Area Information Network in North Carolina.
Non-profit Networks often are a partnership between a number of public and non-profit entities. OneCommunity (formerly OneCleveland), for example, is owned by a non-profit organization established through a partnership between a number of public and non-profit entities.
Community Networks consist of individual users owning the hardware and voluntarily participating in an ad-hoc network. Some are sponsored by non-profit organizations. Typically these networks offer free access. SoCalFreenet, NYCWireless, Seattle Wireless, and Ile Sans Fil in Montreal are all community networks.
Hybrid Networks. Many networks are hybrids, building on the strengths of multiple partners. For example, REA-ALP Internet Services is a partnership between Runestone Electric Association, a rural electric cooperative, and Alexandria Light and Power, a municipal utility. The Urbana Project is a partnership between Champaign-Urbana Community Wireless Network and the City of Urbana. Austin Wireless is a community wireless network, but operates some portions of its network in cooperation with the City of Austin.
Telecommunications Regulation
Over the last two decades, both the regulation and structure of wired and wireless telecommunications changed dramatically. The definition of public interest has been severely curtailed, as has the authority of local, state and federal governments to assert the public interest.
In 1984, Congress limited local authority to enforce cable franchise agreements. In 1987, the FCC eliminated the fairness doctrine. In 1994, the FCC began auctioning spectrum to the highest bidder, and the Internet backbone was turned over to private companies. In 1996, the Telecommunications Act lifted many of the restrictions on the concentration of media ownership, deregulated cable prices, and substantially deregulated the Baby Bells. One of the few remaining areas of public involvement was the requirement that phone companies allow competing Internet service providers (ISPs) to connect to their customers via their networks.
The 1996 Telecommunications Act contained this requirement. Its intent was to spur broadband infrastructure deployment. Cable companies were not covered by the same rules. They were not required to offer equitable access to their networks to ISPs. As people began moving from dial-up to broadband, the different regulatory regimes for phone and cable companies became important. Local governments tried to rectify this inconsistent treatment of companies offering the same service -- high-speed, alwayson Internet access -- by requiring their cable franchisees to become like the phone companies, that is, common carriers, and to allow other service providers to use their copper lines at a fair price.
Courts consistently struck down these local efforts, even when cities made open access a condition for renewing a cable franchise. The courts agreed with the FCC's position that Congress had preempted local authority on this issue. The U.S. Supreme Court upheld the FCC's position in its 2005 Brand X decision.
Before the Supreme Court issued its decsion, in 2003, the FCC ruled that telephone companies did not have to share the fiber optic portions of their networks. It left to state governments to determine whether wholesale access rates for competing ISPs should be regulated. Most states chose not to regulate rates. Almost immediately after the Brand X decision, the FCC extended its exemption from common carrier requirements to phone companies' data networks as well.
Today, neither cable nor phone companies are required to allow competing Internet service providers to use their networks (though some choose to do so).
Meanwhile, technology is moving us into an era in which text, voice and video are carried over the same broadband networks. The incumbent communications companies are now trying to be everything to everyone with "triple play" packages. The FCC has used this as a further justification for deregulation, arguing that the existence of cable, phone, and satellite networks, and the emerging technologies of broadband transmission over power lines and terrestrial wireless, creates an adequate level of competition between network owners.
Such an argument is, at best premature. Approximately 98 percent of high-speed Internet connections come from cable or phone companies. For most households, even in larger cities, the market is dominated by one cable company and one phone company. Many neighborhoods do not even have two choices, since not all areas of phone company networks are equipped to offer DSL. If they do offer DSL, it is at speeds of 1.5 Mbps or less, compared to 3 to 6 Mbps from cable, and with no capacity to support video.
Some ten percent of households do not have access to broadband from any provider at any price. At the national level, the telecommunications industry is consolidating. Only slightly more competition exists in the telephone sector than in the days of Ma Bell. In 1984, AT&T was broken into eight regional "Baby Bells." Ensuing mergers and acquisitions have left us with just three: Verizon, AT&T, and the much smaller Qwest. The two largest cellular phone companies, Verizon Wireless (majority owned by Verizon), and Cingular (soon to be wholly owned by AT&T) currently command more than half the market. Two cable companies, Comcast and Time Warner, control 47 percent of the cable television market.
The lack of competition has slowed the expansion of the U.S. broadband market. We are 15th in the world in broadband penetration, according to the International Telecommunications Union, down from 4th in 2001. We perform even more poorly in the ITU's "digital opportunity" index, which considers price and capacity as well as other factors, coming in 21st. Broadband subscribers in the U.S. pay twice as much as customers in Asia and Europe for one-twentieth the speed.
The Internet was invented in the U.S., but other countries are now taking the lead. For example, the private companies that own the Internet backbone in the U.S. have resisted upgrading to a new version of the Internet address system (IPv6) for nearly a decade. IPv6 greatly expands the pool of Internet addresses, allowing everything from cars to thermostats to have unique addresses, and allows for increased network security.
This year, China converted to IPv6, and now the U.S. will have to follow its lead. American high-tech companies like Google are setting up research facilities in Asia, because U.S. broadband subscribers do not have the capacity to use new applications under development.
FCC Commissioner Michael J. Copps recently wrote in the Washington Post: America's record in expanding broadband communication is so poor that it should be viewed as an outrage by every consumer and business person in the country. Too few of us have broadband connections, and those who do pay too much for service that is too slow. It's hurting our economy, and things are only going to get worse if we don't do something about it.
Why Public Ownership?
The stakes are high. Local governments are stepping in where state and federal policies of privatization and deregulation have failed. Despite a brief backlash against municipal broadband projects, it is increasingly accepted that cities have the authority to develop telecommunications plans. In elaborating such plans, they must take into account many factors, but the one that will have the greatest effect on competition, equity, and public benefits is the decision about who will own the network.
We propose five arguments for public ownership.
1.High-speed information and communication networks are essential public infrastructure.
Much of the infrastructure of the country -- water, sewer, roads, airports, seaports -- is publicly owned. Indeed, virtually all economists and economic development experts believe that public infrastructure is essential for improving productivity and maintaining competitiveness. Just as high quality road systems are needed to transport people and goods, high quality wired and wireless networks are needed to transport information. Both networks allow individuals and businesses in a community to connect to each other and the outside world. For over 100 years, cities have successfully built and managed public infrastructure like roads and water and sewer systems. Information networks are new kinds of infrastructure, but they are not outside the competencies of local government.
Some apprehensiveness on the part of policy makers is understandable. Computers and related technology seem be evolving far faster than government can keep pace. But while technological change is constant, in this case it does not make today's technology obsolete. For example, the DSL technology available to most U.S. households has not changed for a decade, even though faster alternatives are in use in other countries. A futureproof technology is one for which the useful life exceeds the payback period.
The fundamentals of highspeed information networks are actually quite established. Optical fiber cables are to this century what copper wires were to the last, and their capacity is essentially unlimited. When the electronics that "light up" the fiber can no longer support the level of traffic on the network, they can be replaced without replacing the fiber.
Wireless has a shorter life-span, but also a shorter payback period. Rapid upgrades are part of both private and public wireless business plans. Public ownership of the physical network does not necessarily mean the city either manages the network or provides services. Benton Public Utility District in Washington State contracted for the construction of a fiber and wireless network, which it now manages as a wholesale only network. A halfdozen private companies sell retail services, including multiple ISPs and a home security company.
UTOPIA (the Utah Telecommunications Open Infrastructure Agency) is financed and owned by a consortium of cities that contracted with a private company to build and manage the network, and has several providers of video, voice and Internet services, including AT&T.
Cities own roads, but they do not operate freight companies or deliver pizzas. Modern information infrastructure easily allows the transport layer (the road, or in this case the network hardware) to be separated from the service layer (the pizza delivery, or in this case Internet access or video services).
A publicly owned network would not be a monopoly. Other networks would continue to exist. Indeed, as is explained in more detail below, the existence of publicly owned networks can raise the quality of services and the level of competition.
As Franklin D. Roosevelt said, "the very fact that a community can, by vote of the electorate, create a yardstick of its own, will, in most cases, guarantee good service and low rates to its population. I might call the right of the people to own and operate their own utility something like this: a 'birch rod' in the cupboard to be taken out and used only when the 'child' gets beyond the point where a mere scolding does no good."
2. Public ownership ensures competition.
Tens of thousands of miles of fiber optic backbone cable have been laid by the private sector, but there is little incentive for the private sector to bring highspeed connections the "last mile" to homes and businesses (sometimes called the "first mile", to emphasize the fact that users are creators of content as well as consumers of content).
Owners of existing cable and phone networks have strong incentives to make use of their existing infrastructure for as long as possible. What's more, consolidation in the industry means that companies serving hundreds of markets make choices based on what is most advantageous for the corporation as a whole rather than any individual community. Potential service providers seeking to compete with the incumbent cable and phone companies cannot use existing networks, or obtain access at rates that allow them to offer competitive services. Thus, to reach customers, they must build their own network infrastructure. But here they face a significant barrier to entry. An overbuilder faces the difficult challenge of having to simultaneously repay capital expenditures and compete for market share against incumbents that have already amortized their major capital investments.
A publicly owned, open access network could be open to all service providers on the same terms, thereby encouraging the entry of new service providers. It would allow competing service providers to lease capacity on the network in order to sell services to customers. Customers could then choose Internet service providers according to the combination of price, speed and service that fits their needs. Competition would ensure fair rates, and if any service provider restricted what could be done with its connection, customers could choose a different service provider.
Cities establishing new, privately owned citywide networks can require the owner to allow fair access. But it is unclear whether these contractual obligations will be enforceable in the future. History indicates they may not. Cities negotiated cable franchise agreements that were later preempted by state and federal laws. Thus public ownership, which allows the public to establish the rules for using that infrastructure, may be the only way to ensure a network will provide open, nondiscriminatory access in the future.
3. Publicly owned networks can generate significant revenue.
Telecommunications networks are different from traditional public works like roads because they can be self-financing both in terms of initial construction costs and ongoing upgrades. Indeed, a growing body of data suggests an information network can be a very profitable investment, for the city and for its households and businesses. Cities should welcome this prospect, given the strain on municipal budgets from increasing costs of public safety, health, welfare, and aging infrastructure.
Saint Louis Park, Minnesota's publicly owned fiber network that connects public buildings has a five year payback period because it dramatically lowered operating costs below what the City was paying for leased T-1 lines. In three years, the publicly owned wireless network of Buffalo, Minnesota (population 10,000) generated over $150,000 in profits from a $750,000 investment.
We offer the case of our hometown, Minneapolis, to demonstrate both the potential profitability of a publicly owned network and as a cautionary tale for cities tempted to use the public purse to allow private firms to capture those profits. In April 2005, the City of Minneapolis issued a request for bids for a citywide wireless network. The City ruled out public ownership from the outset, insisting that given its weakened financial state, it could not afford the capital investment. In September 2006, the City announced the winning bidder, a small local company with gross sales of around $10 million in 2005.
Since the company itself was far weaker than the City in terms of being able to finance a $10 million system, the City, under the terms of the 10-year franchise, agreed to purchase a minimum of $1.25 million in services each year, and likely much more. Part of this commitment, $2 million, will be prepaid before the network is launched.
The prepayment and the City's ongoing commitment to purchase services will enable the small, privately held company to finance the build-out. Indeed, the $2 million prepayment for services will cover about one-quarter of the cost of building the wireless network.
In other words, the City, which previously declared it lacked the financial wherewithal to finance the network, is financing the network. For the same amount of money the City could have owned the network, used subscriber revenues to pay operating expenses, and provided free services to itself. As part of the agreement, five percent of net pretax revenue (that is, revenue after operating and debt expenses but before taxes) will go into a digital inclusion fund. City officials expects $4000 in the first year of operation and $1.7 million in the tenth year. The total value of cash and in-kind donations (i.e. free Internet access for Community Technology Centers) is expected to be $11.5 million over ten years.
The company will receive 95 percent of the net income, over $32 million in pretax profits in the tenth year alone. The company will likely realize profits upwards of $130 million over ten years. A city-owned, wholesale access only network would not have been as profitable as a private retail network. Our analysis of the City's numbers concludes that a publicly owned network might earn the City about $51 million over ten years, or roughly four times what it will receive from the contractual agreement. Some cities see public ownership not as an opportunity to make money, but as a way to strengthen the local economy. They view the city not narrowly as a municipal corporation trying to balance its internal books, but as a public corporation trying to maximize the total benefit to its community owners.
Saint Cloud, Florida, for example, chose to invest in citywide wireless in part to keep more money in their citizens' pockets. The $3 million capital expenditure is just 7 percent of the city's outstanding debt; its $300,000 annual operating costs represent just 1.5 percent of the city's general fund expenditures in 2006. The free, city supported service is saving the average household $450 per year -- the amount they previously paid for broadband Internet access. That's more than the average household pays annually in local property taxes.
A publicly owned system can spur private competitors to lower their rates or improve their services, which will benefit all city households and businesses. The Clarksville, Tennessee Department of Electricity (CDE), for example, is asking local voters to approve a bond for $25 million to install a fiber to the home network. At launch, CDE's price will be lower than the current Internet provider, Charter. CDE fully expects that Charter will respond by lowering its rates, perhaps below that of the City's. And that's fine with the City. "That's not a bad thing," CDE General Manager Ken Spradlin says, "because not all our customers are going to choose to do business with us, but they are all our customers."
4. Public ownership ensures universal access
Society as a whole benefits when information and communication networks are accessible to everyone. More people on the network means more participants in online communities, and more customers for online products and services.
Private companies balance the price they charge against the number of households willing to subscribe at any given price. It makes no difference to the companies whether they generate $100,000 from 1000 people paying $100 per month, or 100,000 people paying $1.
Publicly owned road, water and sewer, and sidewalk networks connect all households without discrimination. All have access to the same services, though they may purchase different amounts of these services based on household economics and need. A publicly owned telecommunications network similarly can choose to make a basic level of access available to everyone at a low cost, or offer free or subsidized access to some households. Cities may be able to negotiate such requirements in initial contracts. But as pointed out above, federal and state intervention in cable franchises over the years demonstrates that local governments cannot count on retaining the authority to enforce these contracts.
Cities that choose private networks get one chance to set rules governing the network, in contract negotiations. After that, they rely on corporate good will.
Opponents of publicly owned information networks argue that the private sector is more responsive to customer demands than the public sector. Customers are not asking for high-capacity connections to the home right now, they argue, but once they do, the private sector will respond more quickly and efficiently than the public sector. Yet the evidence indicates that the even the most aggressive telecommunications companies do not intend to serve everyone. Lower income areas, whether urban or rural, and sparsely populated areas, regardless of income are not attractive places for new investments.
The claims that access speeds have increased and prices have dropped are true only if phone and cable offerings are considered separately. The average DSL speed of 1.5 Mbps available to most households has been the same for almost a decade. Phone companies have offered promotional prices, but often only to customers who buy home phone service, and they have created lower price tiers for 768 kbps or slower connections. Meanwhile, cable companies have kept prices high, but increased advertised download speeds.
When these companies do invest, the incentive is to do so in ways that provide a quick return on investment. Consumer rights advocate Bruce Kushnick points out that over the last decade, states gave the Baby Bells tax breaks and deregulated some prices in exchange for their commitment to deploy high capacity, high speed fiber optic networks.
They did so, but instead of making investments in very high speed, very high capacity networks , they made lesser investments to add DSL to their existing copper networks. DSL offers much slower speeds, but it is almost immediately profitable. Yet even here, while the phone network is universal, DSL is not.
The single most reliable predictor of whether a household has broadband connections is income. Broadband data signals can travel only a limited distance over existing copper-based phone and cable networks, and companies are unwilling to invest in upgrades where average revenue per household is low. In rural areas, expensive satellite (upwards of $50 per month, plus hundreds of dollars for the dish) is often the only alternative to dial-up. In urban areas, a large percentage of households have access to cable modem service, typically at rates of more than $40 per month, but not DSL service, which provides slower speeds at a lower price.
Universal broadband access will be a long, long time coming from private companies, if it comes at all. Consider the two highest profile projects currently underway. AT&T (formerly SBC) plans to run fiber to the streets of some 19 million homes in 13 states by 2009. The company will continue to use existing copper connections from the street to the home. Verizon is spending $6 billion to run fiber directly to about 6 million homes by the end of 2006, and another 9 to 14 million homes by the end of 2010. Combined, these deployments might reach about one-third of U.S. households in 2009, overwhelmingly located in communities of aboveaverage income.
Neither company plans to extend fiber to all their customers, ever, because "... there will be areas that are just not economic to offer fiber everywhere," says AT&T's Homezone managing director Ken Tysell.
Phone companies are making this investment primarily to be able to offer video, a market dominated by cable television companies (less than one-third of households that subscribe to paid television do so through satellite rather than cable). No cable companies have announced efforts either to connect fiber to subscribers' homes or to increase the capacity of their networks. Instead, they are packaging cable modem speeds that are slightly faster than current DSL offerings, along with video and voice over Internet protocol. A recent study from an industry supported research center indicates the capacity of these networks is strained as a result of these "triple play" packages.
Most cities included full build-out and anti-redlining provisions in their cable franchise agreements. Cable must be available everywhere in the community. But phone companies are now lobbying at the state and federal level to create new franchising systems that would bypass local authorities and eliminate antiredlining provisions. Companies would be allowed to build out infrastructure only in the areas with the highest profit potential, that is, densely populated neighborhoods with higher incomes. Moreover, some of these proposals would allow cable companies to exit existing franchise agreements if a phone company began offering video services in any portion of the local market.
We are already seeing the results of the move to statelevel franchising without build-out requirements. Communities with above average income have at least two competing providers of very high-speed networks -- capable of providing video -- while neighborhoods of lesser means are bypassed. Any infrastructure investments by the cable companies are in areas where they face competition from telephone companies. Lowerincome and rural areas, many of which already have lesser networks, are ignored.
For example, of the Pennsylvania communities in which telephone companies have filed plans to upgrade their networks, 85 percent are above the state median income. Meanwhile, Verizon is replacing its copper networks with fiber in certain Boston suburbs, but is reportedly trying to sell rather than upgrade its copper networks in Maine, New Hampshire and Vermont.
In New York, Syracuse is getting fiber to the home, but DSL customers in Queens, New York City are being told there is no more capacity in their area. "You can't wire everything for unlimited capacity," said a Verizon spokesman. "It's more effective to engineer capacity to be a fixed percentage above the average use in a given day."
5. Public ownership can ensure non-discriminatory networks.
Network neutrality is the term used to describe a network whose customers can use their broadband connections to access the content of their choosing, run the Internet applications of their choosing, and attach to their connection any devices of their choosing. This is possible because with Internet Protocol, bits are bits. Whatever you do with your internet connection -- listen to radio programs, post your work on a web site, send pictures to family, or talk to friends in Canada -- is broken down into little packets of data that move through the network in the same way.
With network neutrality, there can still be multiple tiers of service (i.e. $15 per month for 1 Mbps, $30 per month for 3 Mbps). Neutrality simply means that when customers pay for a connection with a certain level of service, they should be able to use that connection however they choose. The elimination of common carrier requirements and increased Internet traffic -- both as a result of more people online and more bandwidth intensive applications like video -- have brought a new urgency to the debate on network neutrality.
Cable and phone companies have begun insisting they need to manage traffic in order to ensure "quality of service." A typically cited example is that X-rays shouldn't get tied up in network traffic created by someone downloading a movie. But reasonable traffic management can be incorporated into a network without changing the nature of the Internet. Just as emergency vehicles, like ambulances, can take priority on the roads, so emergency pieces of information, like X-rays, can be given priority over information highways.
Private network owners argue that they need to charge differential rates in order to manage web traffic and provide quality of service. In reality they desire this ability to allow them to maximize their profits. Instead of offering faster or more affordable connections, they would charge you for what you do with your connection. For example, they can charge one rate to download video created by their own company, but a higher rate to download video from an independent filmmaker, and an even higher rate to post your own video for others to download. A digital book purchased from Amazon.com would download faster than the same book from your local bookstore or an independent author, just because the larger company can afford to pay for priority for its traffic. Those who own the network could make customer interaction with the Internet more like cable television. For example, AT&T's new service uses Internet technology, but won't allow users to browse just any content using the box on their televisions.
According to the Wall Street Journal, "While the Homezone set top-box will be connected to the Internet, users won't be able to surf to any Web Site. They will only be able to download content from providers who have made deals with AT&T." With publicly owned networks, customers can be sure that any traffic management mechanisms are necessary and not simply to improve profitability. Communities can insist on neutrality from any service provider that uses the network, a form of local regulation they could not enforce if they were relying on privately owned networks. Or, if the market is large enough to support multiple service providers, they can leave neutrality to the market, knowing that unhappy customers can easily change service providers.
