A $1.00 increase in per pupil state aid increases aggregate per pupil housingvalues by about $20.00, indicating that potential residents value education expenditure."
An enduring question in education policy is whether spending additional public resourcesimproves schools. Those who argue against strategies to increase public spending argue that currentspending levels are too high and that public schools spend their current revenues inefficiently.
In Using Market Valuation to Assess Public School Spending (NBER Working Paper No. 9054), co-authors Lisa Barrow and Cecilia Rouse first examine whether an additional dollar ofpublic money spent on schools increases residential property values. If it does, then consumerspresumably value the additional spending because it increases school quality. Further, the authorsevaluate whether the current level of spending is too high.
School expenditures are financed with both state and local contributions. Each state has afinancing formula for determining the state aid to each district, and these formulas are revisedperiodically. The state portion of the contribution often is negatively related to property valuesbecause states try to equalize expenditures across rich and poor districts. Thus, it is difficult to assessdirectly the effects of an increase in state aid on school quality as it may reflect a worsening of otherconditions in the local district that also affect school quality. To overcome this difficulty, the authorsexamine 1980-to-1990 changes in property values resulting from changes in state aid for schools thatarise solely from changes in state financing formulas -- they control for an extensive list of districtand county characteristics that may also affect property values.
Their results suggest that, overall, a $1.00 increase in per pupil state aid increases aggregateper pupil housing values by about $20.00, indicating that potential residents value educationexpenditure. Further, their results suggest that some of the increase in value reflects lower local taxburdens, but most reflects increases in total per pupil district expenditures. Finally, they concludethat there is no evidence that school districts are overspending, on net.
However, this overall result may mask important differences, because some school districtsmay operate more efficiently than others. Specifically, the authors note that because households withgreater income can afford to consider a wider range of schooling and housing options, schoolspending in districts with wealthier residents may be more efficient. Similarly, the degree of externalcompetition that a school district faces (from having many neighboring districts) or the district's sizemay also affect the efficiency of school spending.
To test whether wealthier and more educated school districts spend their revenues moreefficiently, the authors categorize school districts by average household income and education level ofthe adult population, as well as by the degree of competition faced by the district, and the district'ssize. Although potential district residents on average value additional state revenues, the authors findthat "large school districts, and those areas with fewer homeowners and in areas in which residentsare poor or less educated" are more likely to overspend.
-- Linda Gorman