Daniel Akst, guest-blogging at Marginal Revolution, discusses an intriguing paper which estimates the effect of an NFL franchise on property values in US cities. The authors, Gerald Carlino and Edward Coulson, find that the presence of an NFL franchise raises the average household rent in a city by about 7-10%. They offer this as evidence that sports teams generate significant external benefits that are not captured by team owners. Akst compares this "existence value" to the value some folks put on old growth forests.
There is no question that such benefits exist, and that they are of considerable magnitude. Indeed, a more direct indicator of external benefits from commercial sport are sports sections in newspapers. No other industry has a section of the paper devoted to it on a daily basis. People love sport, and enjoy following their teams, whether they purchase tickets or television rights or not. Newspapers capture some of this value. Landowners in cities with sport franchises might capture some as well.
Carlino and Coulson's paper is an interesting attempt to measure this effect. But something has never sat right with me about the study. I find the magnitude of the rent increase implausibly large. Eight percent of housing expenditure? On average across all the citizens of a city as large as New York? That would generate a significant increase in property values, and thus property tax collections. The paper puts this figure at $7.1 billion in present value for NYC. That's enormous (although so is NYC).
One problem with the analysis is that factors which make a city attractive to the NFL also make the city an attractive place to live. The NFL is not looking to locate where people ain't. Carlino and Couslon do what they can to control for these factors, and note this issue in the paper's conclusion. But they can't be sure that that the NFL variable is uncorrelated with unmeasured attributes of NFL cities that drive both the NFL to locate there, and independently, people too. If there is any correlation of that sort, their estimate of "the NFL effect" on rents is biased upwards, and I suspect that to be the case.
Another measure of the external value of a sports team was constructed by Johnson, Groothuis, and Whitehead. Their paper is the source of the $5.57 figure in Akst's post - the amount the typical citizen in Pittsburgh is willing to pay (in taxes) to ensure the Penguins stay there to play hockey. The figure comes from a survey taken when the Pittsburgh hockey franchise was a relocation risk as it was going bankrupt in 1998. As Akst states in his post, this figure translates into $66 million in value over a 30 year horizon. That might not pay for an entire hockey rink, but it's not small potaoes either. For comparison, Carlino and Carlson obtain a property tax revenue gain of $128 million for the Pittsburgh Steelers. The Johnson et al. figure is an upper bound, so these figures are not wildly at variance with each other.
I think everyone understands that these "existence values" are not trivial sums. But whether they are in the tens or hundreds of millions of dollars is beside the point. What "obstructionist" economists object to is the use of studies projecting a large economic impact on jobs and income to promote stadium subsidies, when all evidence suggests the contrary. As Tom Kirkendall argues, the subsidy pitch can be made without recourse to junk science. We've been saying this for quite some time, but the argument keeps falling on deaf ears (save Tom's).
Notes: 1. Monopoly control over franchises is critical for the subsidy game to work. Economists have objections to the monopoly aspect of this issue as well. But one can accept that, and the consequence that subsidies will follow as owners of artificially scarce franchises play off one location against another. Junk science is another matter.
2. You can obtain an earlier version of the Johnson et al. paper here, or from the Journal of Sports Economics if you have access. Similarly for Carlino and Coulson's working paper, and the recently published version at the Journal of Urban Economics. I'm not sold on the numbers, but both papers are decent attempts to quantify an elusive concept. And their numbers will rule until someone comes up with something better.
3. The budget woes of Pittsburgh are problematic for the Carlino-Coulson calculations. Pittsburgh's stadium subsidies do not look like investments that are paying for themselves.