«Introduction This paper examines the fiscal condition of school districts in Nebraska. Our methodology is a new one that has been applied to ...»
In other words, we use our analysis of school costs to calculate the cost of providing an average-quality education in a district with the share of handicapped students, share of students eligible for transportation, enrollment scale, and class of the hypothetical consolidated district. A large-scale consolidation plan might increase total transportation costs by cutting the number of school buildings.
On the other hand, some current school districts are widely scattered not contiguous, so that consolidation might lower transportation costs in some places. We are unable to determine the impact of our consolidation experiments on total transportation costs.
The county-wide districts are treated as either class 2 districts or class 3 districts, depending on their population. This experiment accounts for potential economies of scale to be achieved through consolidation, such as spreading the costs of administration, libraries, and other system-wide functions over a larger number of students, and it accounts for gains or losses due to cost differences among classes of school districts. In particular, this consolidation experiment eliminates class 1 districts and therefore eliminates their special cost advantages. As explained earlier, we assume that interdistrict spending differences that are left over after controlling for all observable factors are cost differences, not differences in service quality. If this assumption is incorrect, then our calculations understate the gains from consolidation. Because these left-over spending differences are large, almost $2,500 per pupil between class 1 and class 3 districts, the degree of understatement could be substantial.
The results of this experiment are dramatic. First, we find that consolidation greatly lowers the fiscal disparities across school districts. The standard deviation in the need-capacity gap, a measure of the extent to which the gap varies across districts, is cut more than in half, from $2,621 in the current system to $1,107 with consolidation. Furthermore, the range in gaps also drops substantially, from $-27,159 to $4,364 under the current system (see Table 8) to $-$3,774 to $3,191 with consolidation. Thus without any new state aid or other resources, consolidation can greatly lower the unfair fiscal advantages enjoyed by some districts and lessen the fiscal disadvantages experienced by others.
Second, the decline in costs made possible by consolidation, largely due to economies of scale, lowers the need-capacity gap in the district serving the average student to -$1,864, a drop of $555 per student below the current system. Because our calculations hold educational service quality constant, this drop corresponds to an average cost savings of $555 for educating every student in this state. Thus, this consolidation experiment can be said to cut the cost of educating students in Nebraska by $555 multiplied by 266,000 students or over $147 million. This result should not be interpreted as an estimate of the cost savings from an actual consolidation plan; instead, it dramatizes the potential cost savings from consolidation, even with conservative assumptions.
Our analysis reveals two important facts about the fiscal conditions of school districts in Nebraska.
First, and most important, school districts vary greatly in their revenue-raising capacities, in their expenditure needs, and in their need-capacity gaps. This variation exists across all classes and all sizes of districts. Differences in income account for a large part of this variation, but other factors, including differences in the ability of districts to export taxes and in the costs of providing education (due to the presence of handicapped students, economies of scale, and other factors) all contribute to the current fiscal disparities across the state's schools.
Second, the largest and the smallest districts are in much better condition, on average, than the districts with enrollments between 100 and 1000 students. These medium-sized districts, which are concentrated in classes 2 and 3, do not have the high per student income that the smallest districts have, nor can they take full advantage of economies of scale. As a result, these medium-sized districts have, on average, both a relatively low capacity to generate revenue and relatively higher expenditure need.
Despite the existence of an equalizing aid program, overall state aid in Nebraska does not offset existing fiscal disparities across school districts--and indeed even exacerbates it to a small degree. The foundation grant program, unlike foundation grants in other states, does not provide more aid to districts with low revenue-raising capacity, and the incentive aid program rewards the districts that are in the best fiscal condition. Moreover, the equalizing aid program does not consider many key aspects of a district's fiscal condition.
The predominance of tiny school districts in Nebraska has many fiscal consequences, perhaps the most important of which is the existence of enormous fiscal disparities across districts. Taxpayers in some districts must either accept much higher tax burdens or much lower educational quality than taxpayers in other districts. A major consolidation plan would greatly lessen these fiscal disparities and might save taxpayers in the state a great deal of money by creating districts that can take advantage of economies of scale.
AppendixThe Determinant of School District Spending
As explained in the text, we estimate the cost of education by examining the relationship between spending and various cost factors, holding constant the variables that influence school quality. This appendix presents the results of our regression analysis of school district spending in Nebraska, which draws on the large literature on local public spending (see Inman, 1979, or Rubinfeld, 1985). Detailed variable definitions and regression results are presented in Tables A1 and A2. Data sources are listed in Ratcliff, Riddle, and Yinger (1988).
The sample for our regressions consists of 865 school districts. We eliminated a few districts for which we had insufficient information as well as districts that contract all of their students. In addition, we excluded 4 districts with expenditures per student vastly greater than the average (in excess of $10,000 per student). These districts are all very small and including them yields highly misleading results.
Voters' demand for public services increases with their income and decreases with the "price" of those services. The "price" of public services is the amount a voter must pay in taxes for another unit of services. This so-called tax price is inversely related to the property tax base in the school district; the greater the tax base, the lower the taxes each voter must pay to raise a given amount of revenue.
We find that both income and tax price have a statistically significant impact on school district spending.
Our measure of income, as mentioned previously, is the aggregate household income of a district divided by the number of students in that district. We observe that as income rises, expenditure on education also rises. When income is small say, $10,000 per student, a $1,000 rise in income per student (a 10 percent gain) will create a $23 increase in spending per student. When income per student in a district is much higher, say $50,000 per student, income must rise by $5,000 (also a 10 percent gain) in order to obtain the same $23 increase in education spending per student. We also find that a district with a tax price one standard deviation above the average spends about $244 less per student than a district with an average tax price. In other words, for every 10 percent increase in tax price, a district will spend about $40 less per student.
We also found that districts with a larger ratio of residents to students tend to spend more per student, presumably because they can spread the cost of educating each student over more adults. To be precise, spending per student increases approximately $24 for every 100 percent increase in this ratio.
For example, if two districts both have 10 students but one has a total population of 30 whereas the other has a total population of 60, the second district will spend $24 more per student than the first district, all else equal.
Certain social and economic characteristics that affect the cost of providing education also vary among districts. We estimate the impact of these characteristics on the cost of education by determining their impact on school district appending in Nebraska, controlling for other factors. The cost factors that prove to be statistically significant in Nebraska are the number of special education students, the number of students eligible for transportation, the overall size of the student population, the proportion of students in secondary school, and the class of district. As explained in the text these results were used to calculate each district's expenditure need.
We also looked at the impact of enrollment in private schools on spending for public education. We found that for every 1 percent increase in the number of children attending private schools in a county, all the school districts in that county spend $17 less per student for public education. This result has two possible explanations. First, voters in some districts may simply have a preference for sending their children to private schools and may therefore not support spending for public education. Second, the public schools in some districts may be providing a poor-quality education, so parents decide to enroll their students in private schools.
Aid from the state and from the federal government tends to stimulate spending by school districts. For every dollar of direct aid given by the state (not federal pass-through aid) school district spending increases by ~$1.26. For every dollar of federal aid, school district spending increases by $1.06. The response to aid varies from school district to school district, but these results imply that, on average, school districts are inclined not to reduce the tax burden when aid is received, but instead spend the aid on education. These spending impacts are not significantly different from $1.00, so we do not conclude that districts match a small portion of their aid with additional spending generated from their own revenue. When aid is received from other local governments, the school district spends $0.89 of every dollar in local aid received and passes the remaining $0.11 on as tax savings.
We also find that some interschool receipts and payments for tuition, transportation, and special education have large impacts on spending. For example, $1.00 of special education payments to another district cuts own spending by $0.89. In contrast, our analysis also finds no systematic variation in spending between districts that participate in an Educational Service Unit (ESU) and districts that do not. ESUs collect fees from participating districts but also receive separately collected property taxes and federal and state aid. The revenue that ESUs collect in addition to the school payments averages $50 for every student enrolled in districts that participate in ESUs. One might have expected that the districts that participate in the ESU system would spend, all else equal, $50 less per student than the schools providing ESU-type services on their own. The evidence does not support this expectation.
The services provided by ESUs do not appear to be substitutes for services previously provided by school district themselves.
Footnotes The authors are, respectively, Governor's Fellow, Department of Natural Resources, State of New Jersey; Academic Computing Specialist, Metropolitan Studies Program, Maxwell School Syracuse University; and Professor of Economics and Public Administration and Senior Research Associate in the Metropolitan Studies Program, Maxwell School, Syracuse University. This paper is based on a report for the Nebraska Comprehensive Tax Study (Ratcliffe, Riddle, and Yinger, 1988), which was commissioned by the Nebraska Legislature. The authors thank Tim Kemper at the Bureau of Business Research, University of Nebraska at Lincoln; John Clark and Mert Smith, Nebraska Department of Education; Deborah Thomas, the Revenue Committee Council for the State Legislature; and Eric Will, a legislative aide to Senator Vard Johnson.
References Advisory Commission on Intergovernmental Relations. 1986.
Measuring State Fiscal Capacity: Alternative Methods and Their Uses Washington, D.C.: By the Author.
Bradbury, K. L. 1983. "Structural Fiscal Distress in Cities--Causes and Consequences." New England Economic Review (January/February): 33-44.
Bradbury, K. L., H. F. Ladd, M. Perrault, A. Reschovsky, and J. Yinger. 1984. "State Aid to Offset Fiscal Disparities across Communities." National Tax Journal 37 (June):151-70.
Bradbury, K. L. and H. F. Ladd. 1985. "Changes in the Revenue-Raising Capacity of U.S. Cities, 1970-1982," New England Economic Review(March/April): 20-37 Cohen, Matthew C. 1983. "Tax Capacity, Tax Effort, and State Funding of Schools in Ohio Journal of Education Finance 9 (Fall): 141-56.
Cohn, Elchanan. Forthcoming. "School Management and Efficiency." In International Encyclopedia of Education: Research and Studies, Supplementary Volume 2. Oxford: Pergamon Press.
Hanushek, Eric A. 1979. "Conceptual and Empirical Issues in the Estimation of Educational Production Functions." Journal of Human Resources 14 (Summer): 351-88.
Hanushek, Eric A. 1986. "The Economics of Schooling." Journal of Economic Literature 24 (September): 1141-75.
Henderson, V., P. Mieszkowski, and Y. Sauvageau. 1978. "Peer Group Effects and Educational Production Functions." Journal of Public Economics 10 (August): 97-106.
Inman, R. P. 1979. "The Fiscal Performance of Local Governments: An Interpretive Review." In P.
Mieszkowski and M. Straszheim, eds., Current Issues in Urban Economics, pp. 270-321. Baltimore:
Johns Hopkins University Press.
Kenny, Lawrence W. 1982. "Economies of Scale in Schooling." Economics of Education Review2 (Winter): 1-24.
Ladd, H. F., and J. Yinger. 1989. America's Ailing Cities: Fiscal Health and the Design of Urban Policy. Baltimore: Johns Hopkins University Press.
Ratcliff, K. B., Riddle, and J. Yinger. 1988. "The Fiscal Condition of School Districts in Nebraska: Is Small Beautiful." In M. Wasylenko and J. Yinger, eds., Final Report of the Nebraska Comprehensive Tax Study, ch 11. Syracuse University, Maxwell School, Metropolitan Studies Program.
Rubinfeld, D. L. 1985. "The Economics of the Local Public Sector." In A. Auerbach and M.
Feldstein, eds., Handbook of Public Economics, pp. 87-161. Amsterdam: North-Holland.
Summers, A. A., and B. L. Wolfe. 1978. "Do Schools Make a Difference?" American Economic Review 67 (September): 639-52.
Yinger, J. 1988a. "The Fiscal Condition of County Governments in Nebraska." In M. Wasylenko and J. Yinger, eds., Final Report of the Nebraska Comprehensive Tax Study, ch 9. Syracuse University, Maxwell School, Metropolitan Studies Program.