Are broadband Internet connections a “public good?”

That is what economists would ask when approaching an issue like the one that St. Paul’s City Council took on Wednesday. Economists also would want to know if private market forces are failing to produce broadband access in optimal quantities. Note that economists’ use of the term “public good” is more specific than “some good or service that happens to be supplied by government.” Como Park has a miniature golf course. It is not clear, however, that the public would suffer greatly from the lack of mini golf courses if that one on public property did not exist. In fairness, this golf course is not owned by the city or run by city employees. It’s a private business operated on a concession basis. To economists, public goods are goods or services that private markets will not produce in optimal quantities because markets “fail” for some reason. Commonly it is because they involve benefits that cannot be limited to specific private purchasers, but rather benefit society as a whole. The alert from a tornado warning system with radar and warning sirens cannot be restricted practically to paying customers. Public goods usually are “non-rival” and “non-divisible.” Non-rival means that my hearing a warning siren does not keep any of my neighbors from hearing it. Non-divisible means it is impossible to specify what part of the warning benefits me and which separate part benefits each neighbor. National defense, public safety and navigational aids such as lighthouses meet this definition most completely. Other goods may not, but still are commonly provided by government because the spillover benefits are substantial even if the goods are somewhat rival or divisible. Public health and education are examples. A shot given to me cannot be given to someone else. A school desk occupied by my child is not available to another student. The shot keeps me from getting sick. Instruction makes my child more skilled, hence more productive and more likely to have higher earnings. Nevertheless, the benefits of public health and education are not limited to the individuals directly involved. If many people are vaccinated, the risk of unvaccinated people getting sick is reduced. An educated citizenry makes for a more productive society as a whole. So governments typically provide at least some level of these services even if they don’t meet the strict criteria of “public goods.” Broadband Internet access clearly is neither non-rival nor non-divisible and hence not a public good in the strict sense. The question is whether it has enough spillover benefits so that private firms will not produce it in socially optimal quantities. The comparison with water, electricity, gas and telephone service is obvious. These are vital services for households, but with the exception of the reduction in infectious disease afforded by safe water and sewers, households directly capture most of their benefits. How we produce these goods depends more on historical happenstance and pragmatism than any clear principles. In St. Paul, the city provides water and sewers, but private firms provide gas, electricity, telephone and cable-TV service. The same is true for most municipalities, but not all. Many cities, including Rochester and Austin in Minnesota, have municipal electric power systems. Some cities, such as Des Moines, Iowa, still have private water systems. Some smaller towns have municipal telephone and cable TV. In Europe, government ownership of telephone and electrical systems was the rule for decades. But in France, where the state has taken a large economic role since the days of Louis XIV, water supply has been almost entirely private. There is no clear evidence which approach is best. Advocates of municipal broadband systems argue that there are large economic spillovers. Universal access to broadband will make businesses and households more efficient, they say. Moreover, broadband access for children will increase learning and help children in poor households achieve the same educational levels as those from higher-income families who already buy broadband. Opponents argue that, as with cable TV, there are few spillovers benefiting society as a whole. The efficacy of broadband access in improving productivity or educational outcomes is unproved at best. We don’t need public provision of Internet-based video games any more than we need public provision of “Sex and the City.” Moreover, there is little evidence that municipal ownership uses resources more efficiently than existing private firms. If broadband is a public good, then subsidizing it with general tax dollars may be justifiable. But cities implementing such systems generally are establishing them as self-financing public entities. Many find that the capital costs are higher than projected, and public sign-ups lower. Advocates respond with comparisons to the 1920s, when municipal electric and gas systems were established because investor-owned utility firms scorned investing in small systems. We now see a clear market failure from lack of information about how profitable serving smaller cities might be. There is no clear right or wrong answer on this issue. © 2005 Edward Lotterman Chanarambie Consulting, Inc.