A number of people who communicated with the Property Tax Commission, whether through correspondence or at public meetings, expressed support for tax restructuring to shift the local portion of education funding to statewide revenue sources. In addition to being cited by tax restructuring proponents as a model for New Jersey, Michigan’s education funding reform, called Proposal A, represents one of the few broad statewide tax restructuring efforts in decades. Because of this, Property Tax Commission staff developed this discussion paper about the Michigan model and its applicability to New Jersey. This discussion paper is being made available to inform more fully the public debate on the merits and likely consequences of a Michigan-style tax restructuring in New Jersey.

With the basic elements of taxation put in place in most states by the end of the 1960s, broad scale tax restructurings are now relatively rare in the United States. In a majority of states today, personal and corporate income taxes, sales and use taxes, and property taxes provide the primary tax revenue sources for state and/or local governments. A handful of states do not tax income, a few others do not tax sales, and New Hampshire taxes neither.

In recent years, voters or legislators in several states have been presented with sweeping tax restructuring proposals aimed at reducing reliance on property taxation. These restructuring plans were typically rejected. The most notable recent exception to these rejections was in Michigan, where Proposal A was approved by the electorate in March, 1994.

Briefly, Proposal A sought to shift the burden of public education funding from locally-levied property taxes to broader state taxes by establishing a statewide education property tax, raising various other state taxes, and limiting the ability of local school districts to raise property taxes in addition to the state school tax. For a more detailed description of the program, please refer to the attached summary of Proposal A prepared by the Citizens Research Council of Michigan.

Proposal A’s approval by the voters is unusual in a number of respects. First, between 1972 and 1994 Michigan voters rejected 13 of 14 ballot questions concerning either property taxation, school finance, or both. These questions were placed on the ballot either by the legislature or through the state’s initiative petition process. Each of the 13 proposals defeated during this period would have reduced Michigan’s reliance on the local property tax. Five of the proposals would have raised the sales tax rate.

Like most fiscal restructuring ballot measures, Proposal A was complex. This complexity can be seen in the attached summary of the proposal given to the Property Tax Commission. The political and fiscal dynamic surrounding Proposal A’s passage was also complex. The 1993 enactment of a statute eliminating property taxation to support school operations, for example, did not determine a source of replacement revenue. That replacement source was determined in a subsequent bill, passed with little public discussion on Christmas Eve, 1993, which would have statutorily increased the personal income tax and business tax rates unless voters approved Proposal A at the polls. The revenue from these tax increases would have been used to replace the property tax revenues formerly used to fund school operations. Thus, whether or not the electorate approved Proposal A, there would have been a major shift to the state in the sources of school funding.

Proposal A significantly changed Michigan’s tax structure. Reliance on the property tax diminished, but the percentage of total property taxes paid by owners of commercial, industrial, and utility properties increased from 28% to 43%. Meanwhile, the proportion of total state and local tax revenue derived from the sales tax increased from 16% to 25%. All told, the FY1998 net tax reduction from Proposal A was estimated to be $1.087 billion.

While "hold harmless" provisions perpetuated significant spending disparities between some districts, Proposal A made important changes in education funding. The super-majority legislative vote required by the proposal to increase school operating taxes beyond those in place in February 1994, for example, has been a key factor in limiting spending increases. Greater control of education policy shifted from communities to the state level. A July 5, 1998, Detroit News article (attached) is indicative of rising public concern over this increased state control.

Is the Michigan Proposal a model for New Jersey? Given New Jersey’s uniquely high education costs, a tax system able to fund schools in Michigan may either be inadequate for New Jersey or may require excessively high tax rates. To consider this concern, some comparisons of Michigan and New Jersey are in order.

This chart presents data to compare Michigan and New Jersey in several categories relevant to the school funding discussion.

1 Average daily public school attendance, 1994-95. (The first year in which Proposal A was in effect.)

2 State and local combined K-12 public school revenues, 1994-95.

3 Total K-12 public school expenditures per pupil in average daily attendance,1994-95.

4 Classroom instruction portion of total K-12 public school expenditures, 1994-95.

5 Per pupil general administration expenditures (which excludes school building based administrative costs, such as those associated with principals).

As indicated by the chart above, Michigan and New Jersey are comparable with respect to a number of factors relevant to the school funding discussion, including population and number of school districts. Despite this comparability, however, New Jersey not only spends more money per pupil than Michigan, but also more than any other state in the nation and more than any nation in the world. It is, therefore, plausible that New Jersey has an education expenditure problem rather than simply an over-reliance on the property tax.

The hypothesis that New Jersey has a cost problem is supported by the existence of two areas of dramatic difference in expenditure patterns between Michigan and New Jersey. First, with 84 more school districts than New Jersey, per pupil general administration expenditures (which excludes school building based administrative costs such as those associated with principals) were $145 per pupil in Michigan and $303 per pupil in the Garden State. Second, with more school districts, vastly larger land area, and a much sparser population distribution, pupil transportation costs were $297 per student in Michigan and $492 per student in New Jersey. This figure does not factor in the number of students actually transported.

To further illustrate the differences between New Jersey and Michigan, consider the contrast between Michigan’s and New Jersey’s sales tax rates. In Michigan, the sales tax rate was 4% prior to the passage of Proposal A and 6% after. This rate increase made Michigan’s sales and use tax comparable to that of neighboring states. New Jersey's current sales tax rate is 6%, as is Pennsylvania's. Delaware does not tax sales. New York has a state sales tax rate of 4%, with an additional amount levied locally. If New Jersey’s sales tax rate were to be increased so that, like Michigan, the increase generated about half of the new revenues used to replace the local property tax for school operating purposes, New Jersey would require a sales tax rate of 10.25%. This would be not only the highest sales tax rate in our region but also the highest rate in the country.

An additional point to consider when examining a possible transfer of funding from the property tax to the sales tax is that property tax payments are deductible for federal income tax purposes, for those who itemize, but sales tax payments are not. While this deductibility issue was not extensively discussed in Michigan, it would likely be an issue for New Jerseyans to consider. This is especially true given the well documented small return on taxes paid to the federal government by New Jersey residents.

Another tax provision of the Michigan model increased imposts on tobacco products. In view of a recent increase in tobacco taxation in New Jersey, an additional increase in the taxation of these products may not be likely in the near term.

Finally, while Proposal A did provide a measure of property tax relief to Michigan taxpayers, the erosion of local influence over education matters that it prompted would likely be an important factor in evaluating the appropriateness of the Michigan approach for New Jersey, with its affinity for local control over educational decisions. For example, New Jersey school districts already spending well above the state average would, under the Michigan model, lose almost all local ability to determine their local education spending level.

In summary, there are significant differences between Michigan at the time of Proposal A’s adoption and New Jersey in 1998.

Attached:

This summary of the elements of Proposal A was prepared by the Citizens Research Council of Michigan and given to the Property Tax Commission. CRCM is an independent policy research organization based in a Detroit suburb.

Proposal A, as adopted, consists of both constitutional changes, which may be changed only with voter approval, and statutory changes, which may be changed by the legislature or by the voters through initiative.

No direct changes were made in the property tax rate limitations in the Michigan Constitution.

By far, the bulk of the Michigan plan was, and continues to be, accomplished statutorily. The statutes are complex, in part because of transitional provisions. The statutory component of the plan continues to evolve and will likely change further over time.

The basic foundation allowance, which was $5,000 in 1994-95, is indexed each year to the change in school aid fluid revenue and the change in pupil count. The foundation allowance of a given school district will be adjusted by the dollar amount of the basic allowance.

The state payment to each district equals the difference between the foundation allowance and the local revenue per pupil, calculated by assuming that the district levies the lesser of 18 mills on non-homestead property (see below) or the number of mills it levied in 1993, regardless of how many mills are actually levied.

approval, may levy additional millage on homestead property up to 18 mills in order to maintain existing per pupil revenue. If 18 mills is still not enough to prevent a decline, further millage may be levied uniformly on all property until a decline is prevented. A relatively small number of districts (52 out of a total of 556) were affected by this so-called "hold harmless" provision.