E. Frank Stephenson at DOL writes on an attempt to allow Sunday beer sales at minor league baseball games in Georgia:
The legislative kerfuffle over Sunday beer for Gwinnett brought to mind my recent IJSF paper (with math colleague Ron Taylor and former student Andrew Chupp). Rome's 2004 referendum created a natural experiment for us to assess the effect of alcohol availability on attendance. We compared the Rome Braves 2003 and 2004 Sunday attendance (before beer sales were allowed) to their 2005-2006 Sunday attendance when beer was available. (We controlled for lots of other factors that might influence attendance--promotions, rehab appearances by Chipper Jones, ...) We found that allowing Sunday sales resulted in a small (2%) and statistically insignificant increase in attendance. Attendance does seem to be influenced by cheap beer--the team offers two for one beverages on Thursdays and draws about 8% more fans than on other weeknights. While Rome's experience may not carry over to other communities, our paper does call into question the conventional wisdom about beer and attendance.
This calls into mind Oi's "Disneyland Dilemma" QJE paper. If concession pricing is indeed part of a two-part pricing scheme aimed at capturing consumer surplus and the team priced tickets and beer "correctly", then allowing beer sales would not be expected to alter attendance. Or as Steven Landsburg could have written in his Armchair Economist: Economics & Everyday Life, why is beer so expensive at the ballpark?