The National Bureau of Economic Research released a pretty interesting study, “The Value of School Facilities: Evidence from a Dynamic Regression Discontinuity Design.”
The study “analyzes the impact of voter-approved school bond issues on school district balance sheets, local housing prices, and student achievement.” Researchers found that 1) sample school districts did, indeed, use bond funds to improve school facilities and the addition of bond funds had no effect on operating revenues and expenditures; 2) local housing prices increased substantially; and 3) there was a small, brief improvement in student achievement.
As far as student achievement findings, I think there is more to the story. Researchers used third grade test scores as a measure of student achievement. It takes approximately two years to renovate or build a new elementary school, so any increase in third grade test scores attributable to a renovated or new elementary school should have appeared in years two, three, and four after bond passage.
Instead, researchers found that, by year four, the school level effects were still at nearly zero standard deviations. Test scores began to increase in year five and peaked in year six, only to drop to zero school level standard deviations by year eight. We have no details about the district facilities plans or state testing regiment, so it is impossible to know these variables interact in the real world.
