LEGISLATIVE FISCAL ESTIMATE

[First Reprint]

SENATE COMMITTEE SUBSTITUTE FOR

SENATE, No. 1872

STATE OF NEW JERSEY

214th LEGISLATURE

DATED: MARCH 4, 2011

SUMMARY

Synopsis:	“Opportunity Scholarship Act”; establishes pilot program in Department of Treasury providing tax credits to entities contributing to scholarships for low-income students.
Type of Impact:	Revenue Decrease in General Fund; Expenditure Decrease in General Fund or Property Tax Relief Fund
Agencies Affected:	Department of the Treasury; Department of Education; Local School Districts

Office of Legislative Services Estimate
Fiscal Impact	Year 1	Year 2		Year 3	Year 4	Year 5
State Revenue Loss	$35.7 million		$71.4 million	$137.4 million	$229.0 million	$366.4 million
State Cost Savings	$35.7 million		$71.4 million	$137.4 million	$229.0 million	$366.4 million
State               Net Change	$0		$0	$0	$0	$0
Local Revenue Loss	$46.2 million		$93.7 million	$185.3 million	$321.4 million	$547.4 million
Local      Revenue Increase	$10.5 million		$22.3 million	$47.9 million	$92.4 million	$181.0 million
Local               Net Change	-$35.7 million		-$71.4 million	-$137.4 million	-$229.0 million	-$366.4 million
Local Cost Savings	Indeterminate – See comments below

· Under the assumption that corporations will make contributions to scholarship organizations in an amount sufficient to fund the maximum number of scholarships authorized by the legislation, the Office of Legislative Services (OLS) estimates that the State would collect $839.9 million less in the corporate business (CBT) tax over the five years of the pilot program.

· The OLS estimates that $1.194 billion in State school aid would initially be withheld from the 13 participant school districts during the course of the pilot program.  Of this amount, $839.9 million would be retained by the State to offset the CBT revenue loss, inclusive of the costs of scholarships awarded to students currently enrolled in nonpublic schools and the scholarship organizations’ administrative costs.[1]  The remaining $354.1 million would be redistributed to the participant districts.  The bill authorizes an additional withholding to offset the costs incurred by the State for administering the State assessments to scholarship students and two assessments, to be selected by the Commissioner of Education, to each first-year scholarship student.  The cost of these assessments, and the corresponding amount of State school aid that would be withheld is indeterminate; however, the OLS anticipates that the additional cost would be minimal in the context of the overall scholarship program.

· The revenue increase estimates for school districts are based on the assumption that the State provides school aid pursuant to the provisions of the “School Funding Reform Act of 2008” (SFRA), P.L.2007, c.260.  To the extent that future budgets provide State aid at amounts lower than what would be provided under SFRA, the amount of aid initially withheld and the subsequent revenue increase would be smaller; however, the net change in district revenue would continue to be equal to the amount of CBT revenue foregone and the cost of the required assessments.

· The process of withholding State school aid to offset the revenue loss and reimbursing any excess withholding results in some of the districts supporting a disproportionate share of the program’s cost relative to the share of scholarships awarded to the districts’ residents.

· There are three additional possible, albeit indeterminate, fiscal effects of S-1872 SCS (1R): 1) the bill may decrease State expenditures, and decrease district revenue, as a result of districts receiving less State aid as enrollment declines; 2) enrollment declines in the 13 districts may lead to decreased operational expenditures in the districts; and 3) the participant districts may incur increased transportation costs and increased transportation aid provided by the State.

· It has been suggested that the legislation would avoid cost increases in the school districts by averting the closure of nonpublic schools and the subsequent enrollment of those students in the public school system.  Within the 13 school districts affected by S-1872 SCS (1R), the OLS did not find a clear pattern of public school districts’ enrollments increasing when nonpublic schools closed.  As such, this possibility is not incorporated into the fiscal analysis.

BILL DESCRIPTION

      Senate Bill No. 1872 SCS (1R) of 2010 provides for the establishment of a five-year pilot program in which corporations are given a tax credit equal to 100 percent of the contributions made to scholarship organizations that are authorized to provide scholarships to low-income students who either attend a “chronically failing school” (CFS) or attend nonpublic school in the State, but reside in a school district in which a CFS is located.  A CFS is a public school, other than a charter school, in which, in each of the two most recent years for which State assessment data are available: 1) the percent of students scoring in the partially proficient range exceeded 40 percent for both language arts and mathematics; or 2) the percent of students scoring in the partially proficient range exceeded 65 percent for either language arts or mathematics.  Additionally, the school must be located in one of 13 specified school districts.[2]  S-1872 SCS (1R) of 2010 defines a low-income child as one who resides in a household where the income is no greater than 250 percent of the federal poverty threshold (this would equal $55,125 for a family of four based on the most recently published federal poverty guidelines).

      The maximum number of scholarships awarded under the pilot program would be 3,900 in the first year, 7,800 in the second year, 15,000 in the third year, 25,000 in the fourth year, and 40,000 in the final year; in each year, no more than 25 percent of the scholarships may be awarded to a student who was enrolled in a nonpublic school on the effective date of the bill (unless, as of August 1, some of the scholarships have not been awarded).  Three scholarship organizations, one in the north, central, and southern regions of the State, will receive an allotment of corporate contributions in direct proportion to the number of scholarships allocated to each.  Available scholarships are allocated across the regions in proportion to the share of the Statewide enrollment in CFS located in the region (for 75 percent of the available scholarships) and in proportion to the share of the Statewide enrollment in nonpublic schools located in a district in which a CFS is located (for 25 percent of the available scholarships).

      Revenue losses due to the corporate tax credits are offset by withholding State school aid from participant districts.  Initially, an amount equal to the aid per pupil in certain State aid categories[3] would be deducted from the participant districts’ State aid for each resident student  enrolled in a public school in the district who receives a scholarship.  To the extent that the amount withheld from all participant school districts exceeds the total tax credit authorized under the bill, the excess withholding would be distributed to the participant districts in proportion to the districts’ share of the total initial State school aid withholding.

      The participant school districts would be required to provide transportation services to all scholarship students on the same basis as the districts are required to provide transportation to students enrolled in nonpublic schools pursuant to N.J.S.18A:39-1.  If the district provides transportation to its own students who live more than two miles (2.5 miles for high school students) from the school of attendance, then the district must also provide transportation services to scholarship students who live a similar distance, but within 20 miles, of the school that the scholarship student attends.  Given that the resident school district is responsible for providing transportation services, transportation aid would not be deducted as part of the State school aid withholding described previously.

FISCAL ANALYSIS

EXECUTIVE BRANCH

      None received.

OFFICE OF LEGISLATIVE SERVICES

Identifying Chronically Failing Schools

      The OLS used assessment data reports available on the Department of Education’s website to determine which schools would meet the definition of CFS included in the legislation.  Using data for the 2008-2009 and 2009-2010 school years, the OLS calculated the weighted average percent of students rated as partially proficient on each test administered, with the number of valid test scores serving as the weight. This process identified 180 schools, approximately half of all schools located in the 13 districts, that meet the criteria for CFS.  The fiscal estimate assumes that the list of CFS will not change through the period of the pilot program.

Maximum Scholarship Amount

      S-1872 SCS (1R) stipulates that each scholarship will equal the lesser of: 1) the actual per pupil cost of the school attended by the scholarship student, or 2) for elementary school students, the greater of $8,000 or 40 percent of the average comparative cost per pupil in the participant districts; for high school students, the corresponding values are $11,000 and 59 percent of the average comparative cost per pupil.  The average comparative cost per pupil in the 13 districts during the 2008-2009 school year (the most recent school year for this audited expenditures are available from the Comparative Spending Guide) was $16,973.  Since the average comparative cost per pupil would yield scholarship amounts of $6,789 for elementary school students and $10,014 for high school students, this analysis estimated the cost of funding the scholarships using the higher figures of $8,000 and $11,000.  It should be noted, however, that the cost of the scholarship program would be lower if the actual educational costs of schools attended by scholarship recipients are lower than the amounts included in the analysis.[4]

Allocation of Scholarships Across Regions

      S-1872 SCS (1R) specifies that 75 percent of the available scholarships are to be allocated to each region based on the share of students enrolled in a CFS in the region relative to the total enrollment in a CFS Statewide.  Similarly, 25 percent of the available scholarships are to be allocated based on the region’s share of students enrolled in nonpublic schools located in a district in which a CFS is located relative to the Statewide enrollment in nonpublic schools located in a district in which a CFS is located.  Enrollment figures for CFS were obtained from the 2009-2010 fall survey data collected by the Department of Education, while nonpublic school enrollment data were obtained from the biennial Private School Survey (PSS) administered by the National Center for Education Statistics (the most recent available PSS data are from the 2007-2008 school year) .  Table 1 shows the regions in which the participant districts are located, while Table 2 provides the estimated allocation of scholarships across the regions in the fifth year of the scholarship program.

Table 1

Region Assigned to Participating Districts

North	Central	South
East Orange	Trenton	Camden
Newark	Perth Amboy	Lakewood
Orange	Asbury Park
Jersey City	Elizabeth
Passaic	Plainfield
Paterson

Table 2

Estimated Allocation of Scholarships Across Regions– Year Five

	Public	Nonpublic	Total
North	17,979	5,993	23,972
Central	6,383	2,128	8,511
South	5,638	1,879	7,517

Note: Figures may not total to scholarship limits included in legislation due to rounding.

Allocation of Scholarships Across Districts

      To allocate scholarships across school districts within each region, it is necessary to estimate the number of students in a CFS who live in a household in which the income is no greater than 250 percent of the federal poverty threshold.  Income-related data collected by the Department of Education only identify whether a student is eligible for the federal free or reduced-price lunch program, for which eligibility is limited to those students living in a household with income no greater than 185 percent of the federal poverty threshold.  As an alternative, the OLS estimated income eligibility for the scholarship program using data from the American Community Survey (ACS) collected by the Census Bureau.  Table 3 shows the estimated share of school-aged students in each municipality who live in a household with income less than 250 percent of the federal poverty threshold.[5]

      The analysis estimates the number of eligible public school students by multiplying the enrollment in each CFS by the eligibility rate shown in Table 3.  Additionally, scholarships for students currently enrolled in nonpublic schools are allocated by multiplying the eligibility rates in Table 3 by the number of students residing in the district who attend a nonpublic school, as estimated in the ACS.  The available scholarships are apportioned among participant districts in proportion to the number of eligible students in each district; this method yields an allocation

Table 3

Estimated Share of Income-Eligible Students

County	District	Estimated Percent Eligible
Camden	Camden	88.1%
Essex	East Orange	71.3%
Essex	Newark	72.7%
Essex	Orange	62.3%
Hudson	Jersey City	64.0%
Mercer	Trenton	72.7%
Middlesex	Perth Amboy	63.8%
Monmouth	Asbury Park	81.8%
Ocean	Lakewood	75.9%
Passaic	Passaic	78.3%
Passaic	Paterson	78.7%
Union	Elizabeth	65.3%
Union	Plainfield	55.5%

that approximates what would occur if scholarships are awarded through a lottery and eligible students across all districts are equally likely to apply for a scholarship.  Table 4 displays the resulting allocation of scholarships in the fifth year of the program.

Table 4

Estimated District-Level Allocation of Scholarships – Year Five

County	District	Elementary - Public	High School - Public	Elementary - Nonpublic	High School - Nonpublic
Camden	Camden	4,033	594	203	94
Essex	East Orange	555	532	227	167
Essex	Newark	4,473	1,818	1,218	686
Essex	Orange	311	266	198	74
Hudson	Jersey City	2,158	484	1,002	470
Mercer	Trenton	2,042	678	491	242
Middlesex	Perth Amboy	662	0	133	44
Monmouth	Asbury Park	455	142	107	44
Ocean	Lakewood	572	439	1,261	321
Passaic	Passaic	2,135	0	684	263
Passaic	Paterson	4,123	1,123	606	398
Union	Elizabeth	1,542	0	406	221
Union	Plainfield	732	331	319	121

Note: Figures may not total to scholarship limits included in legislation due to rounding.

Estimated Tax Credit and State School Aid Withholding

      The total number of scholarships awarded would be determined by the availability of revenue provided through corporate contributions.  Since the legislation provides for a tax credit equal to the amount contributed, corporations can make donations at no cost to themselves.  Therefore, it appears to be reasonable to assume that sufficient contributions would be made to fund the maximum number of scholarships authorized under the bill.

      Using this assumption, estimating the total amount of corporate tax credits to be taken under the bill, and the resulting reduction in General Fund revenue, is straightforward.  The estimated number of scholarships to be awarded to elementary school students is multiplied by $8,000, while the number of scholarships to be distributed to high school students is multiplied by $11,000.  The resulting figures are summed and increased by five percent to account for the portion scholarship organizations are allowed to retain to administer the program.  This yields a total of $839.9 million in tax credits over five years.[6]

      S-1872 SCS (1R) provides for the withholding of State school aid from a school district for each of its students who receives a scholarship.  The OLS estimated the participant districts’ total State school aid over the five years of the pilot program pursuant to the provisions of the “School Funding Reform Act of 2008” (SFRA).  For simplicity, the aid estimates assume that the only relevant factor that changes over time is the districts’ enrollment, which will decline as a result of students receiving scholarships.  The resulting State aid per pupil amounts are multiplied by the estimated number of public school students receiving a scholarship.  This results in a cumulative State aid withholding of $1.194 billion over five years.

      To the extent that the total amount withheld exceeds the total amount of tax credits necessary to provide the maximum number of scholarships (and pay for the administrative costs incurred by the scholarship organizations and the costs of assessments to be administered) the difference, $354.1 million over five years, would be redistributed back to the participant districts in proportion to the districts’ share of the total initial State school aid withholding.  This funding mechanism results in some districts supporting a disproportionate share of the program’s costs relative to the number of scholarships that would be received by its students.  Table 5 shows the share of scholarships, both for students enrolled in a CFS and those enrolled in a nonpublic school on the effective date of the bill, and the share of the Statewide aid withholding, net of any amount returned, for each district.[7]

Table 5

Share of Scholarships Received and Contribution to Total Aid Withholding

County	District	Share of Scholarships	Share of Total Aid Withholding
Camden	Camden	12.3%	21.0%
Essex	East Orange	3.7%	3.9%
Essex	Newark	20.5%	20.6%
Essex	Orange	2.1%	1.6%
Hudson	Jersey City	10.3%	7.4%
Mercer	Trenton	8.1%	8.7%
Middlesex	Perth Amboy	2.1%	1.7%
Monmouth	Asbury Park	1.9%	3.1%
Ocean	Lakewood	6.5%	0.8%
Passaic	Passaic	7.7%	6.4%
Passaic	Paterson	15.6%	17.8%
Union	Elizabeth	5.4%	4.0%
Union	Plainfield	3.8%	3.0%

Other Potential Fiscal Implications

      S-1872 SCS (1R) may have at least three other possible, albeit indeterminate, fiscal implications.  First, the legislation may lead to a decrease in State expenditures, as well as a decrease in revenue to the participating school districts, through a decrease in State school aid resulting from enrollment declines.  Under the provisions of SFRA, adjustment aid is provided to school districts to ensure that districts continue to receive at least as much State aid as it received in the 2008-2009 school year.  However, the law provides that, beginning in the 2011-2012 school year, adjustment aid would be decreased for any district in which enrollment declines by more than five percent relative to the enrollment in the 2008-2009 school year.  It is plausible that the enrollment declines caused by students receiving scholarships under this program will cause districts to reach the five percent threshold (or having larger enrollment declines beyond five percent and, therefore, having larger State aid reductions than what would have otherwise occurred).

      Among the participant districts that do not receive adjustment aid, it is possible that S-1872 SCS (1R) will have an impact on the total amount of State school aid received.  A reduction in the districts’ enrollment would lead to somewhat smaller adequacy budgets and, all other factors being held constant, may lead to less equalization aid being awarded.[8]  Enrollment declines may also lead to decreases in the amount of special education and security categorical aids received by the districts.[9]

      These possibilities rely on the assumption that State school aid is provided pursuant to the provisions of the SFRA.  If aid is provided through another mechanism specified in the annual appropriations act, then the interaction between S-1872 SCS (1R) and education finance policy would be indeterminate.  It should be noted that any such interactions would be directly related to the degree to which State aid is contingent on districts’ enrollment changes.

      Second, the resulting enrollment declines may lead to decreased operational expenditures in the participant districts; however, the degree to which such possible savings will be realized is indeterminate.  The largest enrollment declines in the participant districts would occur in the final years of the pilot program.  The enrollment declines that occur in the earlier years of the pilot program may not be sufficient to allow for systematic changes, such as consolidating classrooms.  In the later years, when the enrollment declines might be sufficient, a district may be reluctant to make such changes given the scholarship program’s temporary nature.

      Third, the bill may lead to increased transportation aid provided by the State and transportation costs incurred by school districts.  The legislation requires the participant school districts to provide transportation services to scholarship students in a manner consistent with the services provided to nonpublic school students pursuant to N.J.S.18A:39-1.  It is possible that a scholarship student would require transportation services that were not necessary when the student was enrolled in the CFS, leading to increased transportation costs incurred by the district and transportation aid provided by the State.

Preventing Nonpublic School Closures

      It has been suggested that S-1872 SCS (1R) may avert cost increases in the participant school districts by preventing the closure of nonpublic schools and the subsequent enrollment of those students in public school districts.  This analysis does not incorporate such a possibility for two reasons.  First, there is no systematic method for determining which nonpublic schools are currently at risk of closing, the number of students necessary for the school to avoid closure, or the propensity for scholarship students to enroll in such schools (as opposed to those not on the brink of closure).  Second, as shown in Tables 6A through 6C, there is no clear pattern of nonpublic school closures leading to enrollment spikes in the public school districts in which the nonpublic schools are located.

      Using data from the PSS for three time periods, the OLS identified nonpublic schools that closed[10] in the 13 participant districts (if a participant district is not shown, then the analysis did not identify any nonpublic school closures during the time period).  The tables show the number of nonpublic schools that closed during the time period, the total enrollment in the nonpublic schools prior to closure, and the change in the public school districts’ enrollment during the same time period.  In each time period, there is no apparent enrollment increase associated with the closure of nonpublic schools; in the majority of cases, the public school district experienced a decrease in enrollment.  While the available data are not sufficient to determine the reason that this occurs, there are at least two possible explanations.  First, it is possible that, upon closure, most nonpublic school students enroll in a different nonpublic.  Second, other population dynamics may have led to decreases in the public school districts’ enrollment that more than offset any new enrollment caused by a nonpublic school closure.  Either scenario would suggest, at least during the time period analyzed, that nonpublic school closures do not necessarily lead to a net increase in public school enrollments (although, it is possible that they prevent greater enrollment declines that would have led to expenditure decreases in the public schools).

Table 6A

Nonpublic School Closures and Change in Public School Enrollment

2001-2002 to 2003-2004

County	Districts	# Nonpublic School Closures	Nonpublic School Enrollment Prior to Closure	Change in Public School Enrollment
Essex	East Orange	2	223	-561
Essex	Newark	2	478	69
Essex	Orange	1	171	-36
Hudson	Jersey City	5	897	-1,170
Mercer	Trenton	2	654	93
Middlesex	Perth Amboy	1	62	-94
Ocean	Lakewood	1	45	-7
Passaic	Paterson	2	300	166
Union	Elizabeth	1	188	-315

Table 6B

Nonpublic School Closures and Change in Public School Enrollment

2003-2004 to 2005-2006

County	Districts	# Nonpublic School Closures	Nonpublic School Enrollment Prior to Closure	Change in Public School Enrollment
Essex	East Orange	1	32	-1,026
Essex	Newark	9	1,080	-88
Hudson	Jersey City	3	205	-812
Mercer	Trenton	5	767	-986
Middlesex	Perth Amboy	2	262	151
Ocean	Lakewood	3	383	67
Union	Elizabeth	3	536	-274
Union	Plainfield	2	332	-441

Table 6C

Nonpublic School Closures and Change in Public School Enrollment

2005-2006 to 2007-2008

County	Districts	# Nonpublic School Closures	Nonpublic School Enrollment Prior to Closure	Change in Public School Enrollment
Essex	Newark	2	387	29
Hudson	Jersey City	2	417	-998
Mercer	Trenton	1	276	-234
Monmouth	Asbury Park	1	30	-284
Ocean	Lakewood	3	1,834	-76
Passaic	Passaic	1	125	237
Passaic	Paterson	1	185	-643

Section:	Education
Analyst:	Allen T. Dupree Senior Fiscal Analyst
Approved:	David J. Rosen Legislative Budget and Finance Officer

This legislative fiscal estimate has been produced by the Office of Legislative Services due to the failure of the Executive Branch to respond to our request for a fiscal note.

This fiscal estimate has been prepared pursuant to P.L.1980, c.67 (C.52:13B-6 et seq.).

---

[1] The OLS notes that depending on taxpayer behavior with respect to the timing of certain tax payments, the decrease in CBT collections may be greater or less than the school aid reduction in any given year; however, over the life of the program, the two would be equal.  For simplicity, this analysis assumes that the two will be equal in each year.

[2] The districts included in the legislation are Asbury Park, Camden, East Orange, Elizabeth, Jersey City, Lakewood, Newark, Orange, Passaic, Paterson, Perth Amboy, Plainfield, and Trenton.

[3] The categories include equalization aid, special education and security categorical aids, and adjustment aid. 

[4] One should note the distinction between a nonpublic school’s tuition rate and its actual cost per pupil.  As discussed in Private Schooling in the U. S.: Expenditures, Supply, and Policy Implications, (Baker, 2009), tuition is typically lower than actual costs, as the educational costs are sometimes supported with other revenue sources in addition to tuition.  Cost estimates included in the report, particularly in Figure 11, suggest a possibility that the actual costs of nonpublic schools attended by scholarship recipients will be greater than the amounts included in the legislation, meaning that scholarships awarded would equal $8,000 for elementary school students and $11,000 for high school students.

[5] The ACS data used in this analysis is the five-year summary data for years 2005-2009.  These are the only ACS data that provide estimates for all municipalities in the State.  The ACS data do not include an income category that ends at 250 percent of poverty.  Alternatively, this analysis calculates the number of children living in households with income less than 200 percent of poverty relative to all children except those living in households with income between 200 and 299 percent of poverty.  This method makes the assumption that the distribution within the category of 200 to 299 percent of poverty mirrors the distribution at lower and higher income levels.

[6] This estimate assumes that the elementary and high school level scholarships are distributed in proportion to the grade level enrollment figures of the CFS or nonpublic school.  If a greater share of scholarships were to be awarded to elementary school students, the total cost would be lower.

[7] The table shows figures for the fifth year of the program.  The pattern is similar in the earlier years.

[8] This, however, would likely not lead to a decrease in State expenditures.  Rather, the equalization aid would more likely be reallocated to other districts.  It should be noted that any such impact would be marginal.

[9] Whether or not these factors actually lead to a decrease in total aid awarded to the districts would depend on whether or not the district has reached the State aid growth limit established under SFRA.  If the district has already exceeded its maximum State aid increase for the year, these fluctuations beyond the limit would not have any actual impact on the districts’ revenues or the State’s expenditures.

[10] A nonpublic school is identified as having closed if it appears in one iteration of the PSS, but is not included in any subsequent survey.  This approach may overstate the number of school closures if a nonpublic school remained open, but simply stopped responding to the PSS.  On the other hand, this method may understate the number of closures if a nonpublic school that closed failed to respond to the PSS even when it was operational.  The PSS response rates exceeded 90 percent in the years included in this analysis; as such, nonresponse likely had little impact on the results.