Is opera good for growth?

TODAY'S "Oh, really?" moment comes courtesy of Jack Ewing at Economix:Modern Germans may still be harvesting significant economic benefits from extravagant opera houses built by spendthrift Baroque princes, according to a study published this month by the Ifo Institute for Economic Research in Munich.The economists behind the study, Oliver Falck, Michael Fritsch and Stephan Heblich, argued that Baroque opera houses attract well-educated workers who prefer to live near cultural amenities. Proximity to an opera house can increase regional growth by as much as 2 percentage points, they wrote.They concluded that political leaders should think twice before reducing culture spending. Oh, really?The study by Mr. Falck and the other economists examined 29 opera houses built before 1800 or just afterward. By limiting themselves to venues constructed before the advent of the industrial revolution, the authors sought to eliminate the possibility that opera houses were a result, rather than a cause, of regional economic growth.The study corrected for other factors that might explain higher growth, like the presence of a university or seaport. Some opera venues were in major cities like Berlin, Munich and Hamburg, but others were in smaller cities like Bautzen, Passau and Stralsund.The authors also looked at regions with similar characteristics, minus the opera house. Ok, maybe there's something to this. Maybe most smart people really like opera enough to move to cities that have opera houses, thereby making those cities more productive. But colour me sceptical.The authors of the paper go to great lengths to control for other factors that might bias their results. They only look at opera houses built before or immediately after 1800, that is, prior to the industrial revolution. And they control for local economic conditions at the time of construction of the opera house. But does that necessarily mean that it's the present impact of the opera houses that's driving growth?Here's my alternative explanation: education levels are persistent. It's quite possible that courts built opera houses in the 18th century for reasons of prestige, and those opera houses attracted skilled musicians and music lovers in the decades thereafter. And it's those 19th century concentrations of educated individuals that are responsible for high skill levels and growth now. Not the opera.Seem strange? Have a look at this:As you can see, there's a tight relationship between school enrolment in 1900 and income a century later. Ed Glaeser wrote on this back in 2009:One reason that historical education levels have such predictive power is that educational investment is extremely persistent. School enrollments in 1900 can explain more than 72 percent of the variation in years of schooling across countries today, as measured by data collected by Robert J. Barro and Jong-Wha Lee; a similar degree of persistence occurs across United States cities.Educated parents and teachers produce educated children; societies that invested in schooling a century ago still generally do so today. Moreover, education has an extraordinarily high contemporaneous relationship with national income levels.Before I sign off on job-creating opera-construction initiatives, I'd like to see an effort to control for education levels a century ago.