Why Do Households Without Children Support Local Public Schools? Linking House Price Capitalization to School Spending

Abstract

While residents receive similar benefits from many local public expenditures, only about one-third of all households have children in the public schools. In this paper we argue that capitalization of school spending into house prices can encourage residents to support spending on schools, even if the residents themselves will never have children in the schools. To examine this hypothesis, we take advantage of differences across communities in the extent of house price capitalization based on the availability of land or population density. We show that fiscal variables and amenities are capitalized to a much greater extent in Massachusetts cities and towns with little available land and that these localities also spend more on schools. Next, we use data from school districts in 49 states to show that per pupil spending is positively related to population density, a proxy for the availability of land. Consistent with a model tying house price capitalization to school spending, we show that the positive correlation between density and spending persists only in locations with high homeownership rates. Communities with a higher percentage of residents above 65 years old have increased school expenditures only in places with high population densities, and this correlation grows for the percentage of elderly above 75 or 85 years old who have a shorter expected duration in their house. The positive relationship between percentage elderly and school spending is confined to central cities and suburbs of large metropolitan areas and does not exist in places where land for new construction may be easier to obtain. These results support models in which house price capitalization encourages more efficient provision of public services and provide an explanation for why some elderly residents might support local spending on schools. According to the 1990 U.S. Census, 36 percent of all households have children below 18 years. Furthermore, the National Household Education Survey (NHES) of the U. S. Department of Education documents that for 1993, 91.2 percent of students in grades 3-12 are enrolled in public schools, while 8.8 percent of the students are enrolled in private schools.

The capitalization of school quality into house values: A review

“…Instead of student characteristics, some studies include population demographics for the neighborhood such as shares of African Americans (or whites) (Bayer et al, 2007;Clauretie and Neill, 2000;Kane et al, 2006), and levels of education (Black, 1999;Brasington and Haurin, 2006;Reback, 2005;Barrow and Rouse, 2004). Some studies include median or mean household income as a proxy for residential housing demand or as a neighborhood amenity (Bayer et al, 2007;Davidoff and Leigh, 2008;Hilber and Mayer, 2009;Kane et al, 2003). As earlier discussed in greater detail, we believe that income does influence housing demand and thus is unlikely to be exogenous in the bid-function envelope, and that it is unlikely to be a good proxy for actual measures of neighborhood characteristics.…”

“…The occurrence and timings of these inspections are therefore independent of unobserved neighborhood factors. Hilber andMayer (2009) andBrunner et al (2002) include both test scores and school spending in their regressions but instrument only for school spending, perhaps because school spending is of primary interest in these studies. This approach appears to be a step in the right direction, but it is unlikely to fully eliminate endogeneity bias.…”

“…Property tax rates generally receive less attention in the studies we review (see column 5 of Table 2). A few studies (e.g., Clapp et al, 2008;Hilber and Mayer, 2009) incorporate effective property tax rates; several studies, however, employ nominal tax rates. While some of the latter studies claim that these nominal rates are identical or at least close to effective tax rates as a result of assessment requirements (Crone, 2006;Mitchell, 2000), other make no mention of the link to effective rates (Gravel et al, 2006;Mathur, 2008).…”

“…Per-pupil expenditures have a positive and significant relationship with house prices in 16 of Brasington and Haurin's (2006) 21 regressions. Dhar and Ross (2010), Hilber andMayer (2009), andBrunner et al (2002) find the positive capitalization of school expenditures into property values. Particularly, Brunner et al (2002) present a large effect of reform-induced spending per pupil in which a one-dollar rise in spending per pupil results approximately in a sixdollar increase in the price of a home, all else equal.…”

House prices and the provision of local public services: capitalization under school choice programs

“…Furthermore, the presence of an academic agreement in the baseline year is plausibly exogenous to the percentage change in residential property values in the district. 14 Table 5 shows estimates for both equations of a Heckman selection model estimating house price effects, where the main equation is analogous to the model estimated in column (5) of Table 3. The estimated coefficients of the transfer rate variables in the main equation are very close in magnitude to the OLS estimates.…”