«THE EFFECT OF SCHOOL FINANCE REFORMS ON THE DISTRIBUTION OF SPENDING, ACADEMIC ACHIEVEMENT, AND ADULT OUTCOMES C. Kirabo Jackson Rucker Johnson ...»
NBER WORKING PAPER SERIES
THE EFFECT OF SCHOOL FINANCE REFORMS ON THE DISTRIBUTION OF
SPENDING, ACADEMIC ACHIEVEMENT, AND ADULT OUTCOMES
C. Kirabo Jackson
Working Paper 20118
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue Cambridge, MA 02138 May 2014 We wish to thank the PSID staff for access to the confidential restricted-use PSID geocode data. We are grateful for helpful comments received from seminar participants at UC-Berkeley and Harvard University. This project was supported by funding from the National Science Foundation. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
NBER working papers are circulated for discussion and comment purposes. They have not been peer- reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.
© 2014 by C. Kirabo Jackson, Rucker Johnson, and Claudia Persico. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source.
The Effect of School Finance Reforms on the Distribution of Spending, Academic Achievement, and Adult Outcomes C. Kirabo Jackson, Rucker Johnson, Claudia Persico NBER Working Paper No. 20118 May 2014 JEL No. H0,H52,H71,H72,I0,I24,I3,J0
ABSTRACTThe school finance reforms (SFRs) that began in the early 1970s and accelerated in the 1980s caused some of the most dramatic changes in the structure of K–12 education spending in U.S. history. We analyze the effects of these reforms on the level and distribution of school district spending, as well as their effects on subsequent educational and economic outcomes.
In Part One, using a newly compiled database of school finance reforms and a recently available long panel of annual school district data on per-pupil spending that spans 1967–2010, we present an event-study analysis of the effects of different types of school finance reforms on per-pupil spending in low- and high-income school districts. We find that SFRs have been instrumental in equalizing school spending between low- and high-income districts and many reforms do so by increasing spending for poor districts. While all reforms reduce spending inequality, there are important differences by reform type: adequacy-based court-ordered reforms increase overall school spending, while equity-based court-ordered reforms reduce the variance of spending with little effect on overall levels; reforms that entail high tax prices (the amount of taxes a district must raise to increase spending by one dollar) reduce long-run spending for all districts, and those that entail low tax prices lead to increased spending growth, particularly for low-income districts.
In Part Two, we link the spending and reform data to detailed, nationally-representative data on children born between 1955 and 1985 and followed through 2011 (the Panel Study of Income Dynamics) to study the effect of the reform-induced changes in school spending on long-run adult outcomes. These birth cohorts straddle the period in which most of the major school finance reform litigation accelerated, and thus the cohorts were differentially exposed, depending on place and year of birth. We use the timing of the passage of court-mandated reforms as an exogenous shifter of school spending across cohorts within the same district. Event-study and instrumental variable models reveal that a 20 percent increase in per-pupil spending each year for all 12 years of public school for children from poor families leads to about 0.9 more completed years of education, 25 percent higher earnings, and a 20 percentage-point reduction in the annual incidence of adult poverty; we find no effects for children from non-poor families. The magnitudes of these effects are sufficiently large to eliminate between two-thirds and all of the gaps in these adult outcomes between those raised in poor families and those raised in non-poor families. We present several pieces of evidence to support a causal interpretation of the estimates.
C. Kirabo Jackson Claudia Persico Northwestern University Northwestern University School of Education and Social Policy Institute for Policy Research 2040 Sheridan Road 2040 Sheridan Road Evanston, IL 60208 Evanston, IL 60208 and NBER email@example.com firstname.lastname@example.org Rucker Johnson Goldman School of Public Policy University of California, Berkeley 2607 Hearst Avenue Berkeley, CA 94720-7320 and NBER email@example.com I. INTRODUCTION Ensuring equal educational opportunities for all children has long been an American ideal (Strickland, 1991; Browning and Long, 1974). However, the rules that determine school funding have not necessarily lived up to this ideal. In most states, prior to the 1970s, the vast majority of resources spent on K–12 schooling was raised at the local level, primarily through local property taxes (Howell and Miller, 1997; Hoxby, 1996). Because the local property tax base is generally higher in areas with higher home values, and there were persistently high levels of residential segregation by socioeconomic status, heavy reliance on local financing contributed to wealthier districts’ ability to spend more per student.1 In response to large within-state differences in perpupil spending across wealthy and poor districts, state supreme courts overturned school finance systems in 28 states between 1971 and 2010, and many states have implemented legislative reforms that led to important changes in public education funding.2 These school finance reforms (SFRs) caused some of the largest changes in the structure of K–12 education spending in United States history.3 Existing research indicates that SFRs have led to greater equalization of school spending within states in the short run (Card and Payne, 2002; Murray, Evans, and Schwab, 1998).
However, there are four important unresolved questions that remain:
(1) Do existing studies suffer from biases associated with low-quality data? Previous national studies rely on data that were only available every five years starting in 1972.4 The low frequency of the data both precluded detailed analyses of outcomes surrounding the timing of reforms and rendered authors unable to rule out the possibility that the effects were driven by
1 Note that many low-income urban districts raise local funding from commercial property, so although low-income students typically receive lower levels of funding on average, this is not always the case (Hoxby, 1996).
2 The first of these cases was the well-known California case, Serrano v. Priest, decided in 1971.
3 Furthermore, nine states are currently reforming their school finance rules, and ten states are in legal battles regarding school financing. States that are reforming their school finance rules: Alaska (Moore v. State of Alaska), Indiana (Hamilton Southeastern Schools v. Daniels), Maine, Michigan (Gov. Rick Snyder’s proposal), Minnesota (statewide task force), New Jersey (Abbott v. Burke), Ohio, Rhode Island (Woonsocket School Committee v.
Carcieri), and Washington (McCleary v. State of Washington). States that are currently in legal battles: Alabama (Lynch v. State of Alabama), California (Robles-Wong v. State of California), Colorado (Lobato v. State of Colorado), Connecticut (Connecticut Coalition for Justice in Education Funding v. Rell), Florida (Citizens for Strong Schools v. Haridopolos), Kansas (Shawnee Mission School District v. State of Kansas; Ganon v. State of Kansas), New Hampshire, New York (Hussein v. State of New York), South Carolina (Abbeville Co. Sch. Dist. v.
State of South Carolina), and Texas (Texas Taxpayer & Student Fairness Coalition v. Scott).
4 Arizona, California,* Idaho, Kansas,* New York, New Jersey,* Washington, and Wisconsin* all had important court decisions that either overturned or upheld the state school finance system before the second possible data point in 1977. States with an asterisk (*) are states in which the status quo was deemed unconstitutional.
pre-existing trend differences between reform and non-reform states.
(2) Do SFRs lead to enduring spending changes? Researchers have found that SFRs may affect marginal income tax rates (McGuire and Anderson, 2011), residential sorting (Tiebout, 1956), and shifting of income sources for school spending (Brunner and Sonstelie, 2003); be capitalized into housing prices (Epple and Ferreyra, 2008); and lead to loopholes or subsequent reforms to undo the effects of SFRs (Imazeki and Reschovsky, 2004). Accordingly, the effects of SFRs on school spending in the short run might be quite different from those in the long run.
(3) How do different kinds of reforms affect the distribution of school spending in the short and long run? There is substantial variation in how different states implement SFRs (Hoxby, 2001). Because policy-makers must choose not only whether to implement reforms but also what kinds of reforms to implement, it is important to know how different kinds of reforms affect the distribution of school spending in both the short and long run.
(4) How do changes in school spending caused by SFRs affect the long-run outcomes of affected children? The motivation behind SFRs was to reduce gaps in educational opportunity and subsequent socioeconomic well-being between children from poor and affluent families.
However, the extent to which improvements in outcomes for low-income children was achieved is unclear. Hoxby (2001) finds mixed evidence on the effect of increased per-pupil spending associated with SFRs on high-school dropout rates. Card and Payne (2002) find that courtmandated SFRs that reduce inequality in spending are associated with reduced gaps in SAT scores between students from low- and high-income families.5 In contrast, Downes and Figlio (1998) find that reforms in response to court mandates do not result in significant changes in the distribution of test scores.6 In addition to the fact that the evidence on student achievement is mixed, there is mounting evidence that focusing on effects on test scores may miss important effects on longer-run outcomes (Heckman, Pinto, & Savelyev, forthcoming; Jackson, 2012).
Accordingly, the effect that SFRs may have on long-run outcomes remains unknown.
This paper tackles these four questions through an analysis of the effects of SFRs on the level and distribution of school spending, as well as on subsequent educational and economic attainment outcomes in adulthood. The analysis proceeds in two parts.
In Part One, covered in Sections II and III, we tackle the first three questions and
5 As acknowledged by the authors, the data used in this study may suffer from selection to SAT taking.
6 However, Downes and Figlio (1998) find that plans that impose tax or expenditure limits on local governments reduce overall student performance on standardized tests.
investigate the effects of SFRs on district spending, both in terms of absolute levels and in equalizing spending between districts in a state. We address these questions using newly released panel data on per-pupil spending at the school district level going back to 1967, five years before the first reforms, and available annually from 1970 through 2010. We compile a comprehensive inventory of the timing of school finance litigation and legislative changes in state aid formulas that occurred between 1970 and 2010. We also codify reforms into several types, based on the ways the reform influenced the school funding formulas. With the higher-frequency, districtlevel data (previous studies used data points five or ten years apart), we conduct a detailed analysis of the timing of changes in outcomes in relation to the timing of reforms and assess the degree of pre-existing trends in spending leading up to the enactment of reforms.
Using the longest district-level panel on school spending that has ever been used to analyze these issues, we document the effects of SFRs on spending up to 20 years after reforms.
Because many states implemented different aspects of reforms at different times, the highfrequency annual data allow us to distinguish the effects of different types of reforms on school spending. We analyze the effects of different kinds of court-mandated and legislative reforms on school-spending disparities between rich and poor districts and on the overall level of per-pupil spending. To document the evolution of school spending before and after reforms, we present a flexible semi-parametric Difference-in-Difference (DiD) event-study analysis. That is, we show how the year-to-year change in outcomes for districts in reform states differed from those for districts in other states over the same time period. We present estimates both for several years before and several years after reforms, and we document the effect of reforms on districts by their percentile of the state income distribution prior to the reforms.