Note: This document is the executive summary of a longer paper. For a copy of the full work, please contact FARU.
Analysis of Local Effort in New York State School Districts
The State Aid Work Group
Policy Implications for the Regents Proposal
School funding in New York State is a reflection of a State and local partnership. Over the past forty years State funding has constituted less than half of school revenues and the State share has ranged from 38 to 48 percent statewide. Thus, in New York State, the majority of education funding has traditionally come from local sources.
With the exception of the fiscally dependent Big Five City School Districts, the primary means of raising revenue locally for education has been the property tax. There is no minimum requirement for property tax effort in New York State and school districts have the discretion to tax (and spend) as they see fit. Thus, the local contribution to total education revenue for a district is largely a result of local decision making.
The local contribution to education also depends on local fiscal capacity. The local share is affected by the extent to which State Aid supplements local spending. State Aid is allocated to school districts using roughly thirty different aid formulas which distribute funds based largely on school district fiscal capacity. Wealthy districts receive less State Aid than poorer districts. Thus, local effort is most appropriately viewed as a function of both district ability to pay and district willingness to pay.
A State and local partnership is necessary to ensure that adequate resources are available to help students meet high standards – particularly when needs are high and resources are limited. Towards this goal, the Regents continue to emphasize the importance of local effort. This analysis provides a framework from which a broader policy discussion can ensue. In this paper, we examined the key drivers of local effort –changes in property values, pupil enrollments, income wealth, revenue, etc. The study includes both changes over time (1993-94 to 1999-2000) for categories of districts of various need/resource types as well as a single-year snapshot (2000-01) which provides a more current view of the local effort problem.
Local Effort Continues to be a Concern in New York State
Three measures of tax effort are used to describe the tax effort problem: a) "lost levy" – which refers to the amount of local tax revenue that districts "lost" in 2000-01 by taxing themselves below the statewide median tax rate of $17.75 per $1,000 actual value; b) "effective lost levy" – which refers to that portion of the unexpended lost levy that would have to be raised in order to bring a district up to the median2 statewide spending level of $11,550 per pupil. This second criterion enables us to recognize more clearly that many districts with high property wealth can generate substantial local levies per pupil at relatively low tax rates, i.e., while their lost levy problem is substantial, their spending level/pupil is still typically far greater than the median spending level per pupil seen statewide. In short, these types of districts would not be characterized as having a true, or "effective" lost levy problem and their high spending levels would almost without exception, remove them from this type of policy concern.
The third criterion used to define the local effort problem was student performance itself. Some school districts may tax themselves below a statewide median tax rate, and fall below the median spending level, yet nevertheless meet high standards for student achievement. Relatively low taxing and spending behavior appears less problematic for high performing districts than for low performing ones. In order to identify districts that are low taxing, low spending, and needing to improve their academic performance, the same effective lost levy analysis was applied in conjunction with average student performance on the New York State 4th and 8th grade English Language Arts and Math examinations. For the purposes of this study, any district with an average score below the level three cut-point on any two or more of the four exams was considered to be in need of improvement.
Lost Levy (taxing below the state median). The total statewide levy lost for districts taxing below the state median for 2000-01 was $1.8 billion. Downstate suburban districts had a lost levy of $862 million, which accounted for 47 percent of the total. While the lost levy suggests a potential policy concern, without additional analysis it does not permit us to draw definitive conclusions about local effort because lost levy does not take into account the spending level attained by districts. The fact that high wealth districts can raise substantial local revenue with low tax rates requires the incorporation of a target expenditure level into the discussion of local effort. Nonetheless, it is clear that while districts such as New York City are struggling to improve student achievement, while at the same time educating substantial proportions of pupils in poverty, the fact that taxing at the median would yield an additional $610 million in local revenue (as seen in Chart 1) should not be minimized.
Effective Lost Levy (taxing and spending below the State median). When considering only those districts that were low taxing and spending below the median expenditure per pupil of $11,550, the total effective lost levy was $187.9 million. There were 161 districts identified as both low taxing and low spending, which placed them in the effective lost levy category.
Effective Lost Levy and Low Performance. Low taxing and low spending is of greatest concern in cases where student achievement is also low. If we further narrow the definition of effective lost levy districts to only those with low academic performance, 75 districts remain in this category. For these 75 districts, the total effective lost levy was $108 million, with New York City accounting for 45 percent or $49 million of the total effective lost levy. Rural districts were disproportionately represented in the effective lost levy category. While Rural districts account for 26.3 percent of all districts in the state, they comprise almost half (49.3 percent) of the 75 districts with effective lost levy and poor performance.
Local Effort and Revenue Per Pupil are Related
When examining local effort rates of the four need/resource categories (excluding the Big Five) from 1993-94 to 1999-2000, patterns over time were similar across need/resource categories. For all four groups, local effort increased during the first four years and decreased during the final two years. Urban/suburban high need districts exerted the greatest local effort compared to districts in other need categories (excluding the Big Five). Average need districts exerted effort slightly higher than the median in all six years of this analysis. Low need districts had the lowest effort rates compared to districts in other categories. Rural high need districts were found to have higher effort rates when compared to low need districts, but remained below the median.
As districts reduced local effort from 1997-98 to 1999-2000, revenue per pupil from the State3 was increasing. Chart 1-2 shows State revenue per pupil (dashed line) and local revenue per pupil (solid line) for each need/resource category. As shown in the chart, local revenue has decreased while State revenue has increased from 1997-98 to 1999-2000. The largest State aid increases during this period have been in the high need rural districts and the high need urban/suburban districts. These increases were met with reductions in local effort rates as well as reductions in local revenue per pupil.
Local Effort is a Critical Concern in the Fiscally Dependent Big Five
Local effort in the Big Five cities is especially important both because of the large number of pupils served as well as the fiscal dependence of these districts on the municipalities in which they reside. Unlike other districts in the State, the Big Five districts do not levy school taxes but are dependent on the municipality for their budgets.4 Chart 1-3 shows changes in State and local revenue per pupil for each of the large cities, followed by a detailed overview of local effort changes from 1993-94 to 1999-2000 for each of the Big Five districts.
New York City. On a per pupil basis, New York City was the only Big Five district to increase local revenue from 1993-94 to 1999-2000 (by $747/pupil). State revenue per pupil increased by $1,057 during this same time period. New York City’s property effort rates have steadily increased each year, placing the City close to the median effort rate at the end of the study period. In terms of income, this additional effort represented a smaller fraction of adjusted gross income in 1999-2000 than it did in 1993-94. As in the other Big Five districts, local education revenue as a percentage of income was substantially below the median.
Buffalo. While revenue per pupil from the State increased by $2,927 over the period studied, local revenue per pupil decreased by $68. Effort rates in Buffalo have been increasing and were above the median in 1999-2000. However, the district has struggled with declining property values (-15.7 percent from 1993-94 to 1999-2000) and income levels that have not kept pace with the rest of the State. Thus, the fiscal capacity of the Buffalo school district has declined to the point where its 12.47 percent increase in property effort rates over six years has been more than offset by these declining property values.
Rochester. Rochester is one of the highest effort districts in the State and the property effort rates increased by 9.02 percent in the years covered by this analysis. However, like Buffalo, Rochester has experienced a substantial decline in property value (-13.8 percent) and lagging personal income. Additionally, Rochester has had a 12.56 percent increase in pupils. Thus, despite increasing effort rates, both declining fiscal capacity and increasing enrollments have led to a decrease of $693/pupil in local revenue. Revenue from the State has increased by $1,840 per pupil during this same period.
Syracuse. Property effort rates declined substantially from 1993-94 to 1996-97 then increased while remaining below the median from 1996-97 to 1999-2000. Property value and income have increased, but have not kept pace with statewide trends. Additionally, Syracuse has experienced a 6.01 percent increase in enrollments. While the actual value effort rate has decreased by 21.39 percent over the time of this analysis (reducing its local contribution by $634 per pupil), State revenue per pupil has increased by $1,219 per pupil.
Yonkers. Yonkers exerted above-median local effort from 1993-94 to 1997-98, after which the district substantially reduced local effort on a variety of measures. This reduction in effort is most likely due to the addition of $81.8 million for an educational improvement program for desegregation purposes. Despite lagging property values and income and increasing enrollments (19.03 percent), Yonkers reduced its effort by 11.38 percent which resulted in a $1,693 decrease in local revenue per pupil. During this same time period, State revenue increased by $3,621 per pupil – largely due to the educational improvement program funds that resulted from a desegregation court order.
Local Effort is Fundamentally Linked to Fiscal Capacity and this Relationship is Non-linear
A strong relationship exists between a district’s need relative to fiscal capacity and the tendency toward low taxing and low spending behavior. As the level of district need increases, districts are more likely to fall into the effective lost levy category. In the five lowest need/fiscal capacity deciles (337 districts), only 13.1 percent of the districts were found to be low taxing and low spending. However, in the five highest need/fiscal capacity deciles (340 districts), 34.1 percent of the districts were identified as effective lost levy districts.
High Need Districts Have Less Incentive to Increase Effort
This can be attributed, in part, to the fact that as property wealth increases, districts will enjoy a greater per pupil increase in local levy for a given increase in their effective tax effort rate. As seen in Chart 1-4, as the Combined Wealth Ratio of a district increases, so does the levy per pupil at a standard level of effort (one mill).5 Therefore low wealth districts have less of an incentive to increase their tax effort than high wealth districts. The line on the chart shows the percentage of low taxing districts. When the levy associated with a standard level of tax effort is low – such as in the low wealth deciles, a greater percentage of districts tax below the median. As the property value per pupil increases, and therefore the associated levy per pupil increases, the likelihood that a district will be found to be low taxing decreases. This relationship holds up until the eighth decile, after which the percentage of districts found to be low taxing begins to increase due to the substantial resources generated at low tax effort levels in high wealth districts. Thus we find that there is a strong, non-linear relationship between wealth and local effort with very wealthy districts and very poor districts having a greater tendency toward low tax effort.
Policy Implications for the Regents State Aid Proposal: