School Finance Alternatives for West Virginia

Qingshui Zhou and Dale Colyer

    School finance has been a contentious issue in West Virginia for many years.  Fourteen years ago,Judge Arthur Recht ruled in Pauley v. Bailey (1982) that students in the state's poorest countieswere not being provided the thorough and efficient education guaranteed by the West Virginia Constitution because school funding was heavily dependent on local property taxes.  School districts located in the state's poorest counties simply did not have the fiscal capacity necessary to meet the state constitution's mandate.  Since then, the state has created a School Building Authority to assist school districts in the construction and renovation of schools, provided additional funds to increase and equalize teacher's salaries, changed the state's school funding formula to provide greater fiscal assistance to the state's poorer school districts, and twice put before the state's voters a statewide excess property tax levy that would have increased and equalized school funding across the state if it had been approved by the voters.  Despite these efforts, funding disparities among the state's fifty-five school districts continue.  Presently, the courts are once again evaluating the state's efforts to see if they are sufficient to meet the state constitution's mandate to provide a thorough and efficient education for all students in the state (Blackford 1996).

    West Virginia is not the only state wrestling with school finance issues.  The historic reliance on property taxes as the primary means of financing public education at the elementary and secondary levels has come under increasing criticism in many states.  A crisis in school funding in Michigan, for example, led to a radical reform - the abolishment of the use of local property taxes as a primary funding mechanism for elementary and secondary education (Fisher and Wassmer 1995).  In its place, Michigan put together a blend of other tax sources, including a statewide property tax and an increase in the sales tax.  It further established an entirely new system for distributing revenues aimed at reducing disparities between communities in per-pupil spending (Ballew et al. 1994).  Following Michigan's lead several other states are currently investigating whether the local property tax can, or should, be replaced with other revenue sources.  They are primarily interested in how these changes impact the amount of revenue available for schools.  One recent analysis, for example, concluded that an increased reliance on the personal income tax is the best tool for increasing school revenue, followed by the sales tax (Gold 1996).  Another important issue that often gets overlooked is how school finance reform impacts tax equity (i.e., do the changes increase or decrease the amount of taxes paid by the poor). This article evaluates what would happen to tax equity if West Virginia decided to follow Michigan's lead and replaced the property tax as the primary means of financing elementary and secondary education with either an increase in the general sales tax or the personal income tax.

West Virginia Taxes Today

    West Virginia's state and local tax system consists of three major taxes (income, sales and property) plus several less important ones (severance, business and occupation, etc.).  The personal income tax ($621 million in FY 1993) and corporate net income tax ($107 million in FY 1993) collectively account for about 35 percent of the state's general revenue.  Sales taxes (the 6 percent general sales tax and selective excise taxes, such as those on gasoline, tobacco, and alcoholic beverages) account for about 32 percent of the state's general revenue ($652 million in FY 1993).  Property taxes account for less than 1 percent of state general revenue but represent a major revenue source for local governments.  In FY 1993, property taxes in West Virginia generated $670.6 million, with about two thirds of that amount going to the state's fifty-five school districts (Bureau of Business Research 1995; West Virginia Department of Tax and Revenue 1995).

Estimating Tax Incidence Impacts

    A general tax burden model was developed to determine how school finance reform would affect tax equity in West Virginia.  In simplest terms, the model uses regression analysis to answer the question, "who gains and who loses" if West Virginia reformed its tax system? Readers interested in the assumptions, data sources, and estimation procedures used in the model should contact the author for further details.  Importantly, the model incorporates the 1996 changes that exempted those with income below $10,000 from paying state personal income taxes and uses assumptions concerning consumer behavior and tax elasticities that are accepted in the professional literature as being reasonable (Pechman and Okner 1974; Phares 1980; Pechman 1985; Zhou 1995).

Tax Policy Changes

    School districts received about $449 million in FY 1993 from property taxes.  The state's general sales tax would have to be increased from 6 percent to 10.1 percent to generate that amount of revenue (i.e., the general sales tax generated $652 million in FY 1993 and would have to be increased by 69 percent - 449/652 - to equal $1.101 billion).  The state's personal income tax rates would have to be increased by 72 percent to generate that amount (i.e., the personal income tax generated $621 million in FY 1993 and would have to be increased by 72 percent - 449/621 - to generate $1.07 billion).  This could be accomplished by increasing the rates for each income bracket by 72 percent equally across-the-board (from 4 percent to 6.88 percent for the $10,000-$25,000 income bracket, from 4.5 percent to 7.44 percent for the $25,000 to $40,000 income bracket, from 6 percent to 10.42 percent for the $40,000 to $60,000 income bracket, and from 7.5 percent to 12.9 percent for the over $60,000 income bracket).  These changes are used for the simulations described in the following sections.  The state could, of course, generate the $449 million in other ways.  For example, they could increase the general sales tax but exempt food from the tax increase, a situation that would mean a somewhat higher tax rate on other items but one which would be less regressive than an increase that includes food.  They could increase the personal income tax, but increase the rates for some income brackets more than others to make the system either more progressive or more regressive.  Each such change would produce a slightly different tax incidence pattern than the one presented here, but these changes could be analyzed with the general tax burden model used for this paper, provided that the basic data needed are available or could be obtained expeditiously.

Tax Burdens

    The general tax burden model's results are presented in Table 1.  It shows the income categories used for the tax burden estimates and 1)  the effective tax rates under the current system, 2) when the reliance on property taxes is replaced with a general sales tax increase to 10.1 percent, and 3) when the reliance on property taxes is replaced with an across-the-board increase in personal income taxes.

    The results indicate that when the school's property tax revenue is replaced by an increase in the general sales tax the tax burden on lower income groups increases and the tax burden on higher income groups decreases, making West Virginia's tax system more regressive.  The opposite occurs when the school's property tax revenue is replaced by an increase in personal property taxes,  the tax burden on lower income groups decreases while the tax burden on higher income groups increases, making West Virginia's tax system more progressive (see Table 1).

Conclusions

    As Table 1 demonstrates, the current school finance system is regressive (i.e., it taxes a higher percentage of the income of lower income groups than it does of higher income groups).  Replacing the school's property tax revenue with an increase in the general sales tax would make the school finance system even more regressive.  However, when an increase in the graduated personal income tax is used to replace the school's property tax revenue, the school finance system is made more progressive.  Also, as noted earlier, the state could reduce the regressive nature of the sales tax if it exempted food since persons with lower incomes tend to spend a higher proportion of their income on food than other income groups.

    An additional consideration in revising the school finance system is the impact that the revisions would have on federal income tax obligations.  Both property and income taxes are deductible for taxpayers who itemize their deductions, while the sales tax is not.  Thus, shifting to a higher sales tax would not only make the school finance system more regressive but it would also increase many West Virginians' federal income tax obligations.  On the other hand, the federal government would, in essence, at least partially subsidize the additional costs of shifting to a higher income tax, especially for those in the highest income groups since they are the most likely to itemize on their federal income tax form and to be in the highest federal income tax brackets.

Endnotes

1.    Traditionally, tax burdens have been estimated by making assumptions about the incidence of taxes and then using 'allocators' - a statistical series used as a proxy to allocate tax receipts among income groups (Pechman and Okner 1974; Phares 1980; Pechman 1985; Harmon 1989).  Another approach is the development of computable general equilibrium (CGE) models, in which an input-output matrix is first constructed to determine equilibrium prices, and then a consumer matrix multiplier is used to transform the tax effects into income class tax burdens (Keller 1980; Ballard et al 1985;  Hertel and Tsigas 1988;  Boyd and Newman 1991).  The attractiveness of CGE models is their ability to simulate both direct and indirect effects of tax policies on a variety of industries and consumer income classes which are closer to actual situations.  However, it is expensive and time-consuming to build and run a CGE model.  The general tax burden model used here requires assumptions similar to those of the allocator approach since a key component of the model is the elasticities of tax rates.

2.    The elasticity estimates were -.389407 for the sales tax, -.310451 for the property tax and  .589971 for the personal income tax.  All estimated coefficients were significant at the .0001 level.

3.    Since there is some debate in the professional literature concerning tax rate elasticities (the response of consumers to the new tax environments), the sensitivity of the results presented here were tested by varying the elasticity estimates.  For these tests, the sales tax elasticity was varied from 0.20 to 0.60 and the property tax elasticity was varied from 0.10 to 0.50 (absolute values).  The general tax burden model were then re-estimated to determine the potential impacts of replacing the school's portion of the property tax by increases in the general sales tax and personal income tax.  The results indicated that the net tax burden changes for each income group are only modestly affected by the changes in sales and property tax.

References

Ballew, Paul D., Richard H. Mattoon, and William A. Testa. 1994. "School Reform and Tax
    Reform: A Successful Marriage?" Government Finance Review 10: 32, 34.

Ballard, Charles L., Don Fullerton, John B. Shoven, and John Whalley. 1985.  A General
    Equilibrium Model for Tax Policy Evaluation. Chicago: University of Chicago Press.

Blackford, Linda. 1996. "Judge Delays School Decision." Charleston Gazette, June 28, p. 1A,
    9A.

Boyd, Roy, and David H. Newman. 1991 "Tax Reform and Land-Using Sectors in the U.S.
    Economy: A General Equilibrium Analysis." American Journal of Agricultural Economics 73:
    399-409.

Bureau of Business Research. 1995. West Virginia Statistical Abstract 1995-96. Morgantown,
    WV: West Virginia University.

Fisher, Ronald C. and Robert W. Wassmer.  1995. "Centralizing Educational Responsibility in
    Michigan and Other States: New Constraints on States and Localities." National Tax Journal
    47:3: 417-428.

Gold, Steven D. 1996.  Separating Wheat from the Chaff: How Much Do Schools Really Benefit
    When States Raise Taxes on Their Behalf? (West Haven, CT: National Education Association
    Professional Library), cited in "Do Tax Increases Really Result in More School Funding?" 1996.
    State Legislatures (July/August): 11.

Harmon, Oskar R. 1989. "A New View of the Incidence of the Property Tax: The Case of New
    Jersey." Public Finance Quarterly 17: 323-348.

Hertel, Thomas W., and Marinos E. Tsigas. 1988. "Tax Policy and U.S. Agriculture: A General
    Equilibrium Analysis." American Journal of Agricultural Economics 70: 289-303.

Keller, Wouter J. 1980. Tax Incidence: A General Equilibrium Approach. Amsterdam:
    North-Holland Publishing.

Pechman, Joseph A. 1985. Who Paid the Taxes, 1966-1985?  Washington, DC: The Brookings
    Institution.

Pechman, Joseph A. and Benjamin A. Okner. 1974. Who Bears the Tax Burden? Washington,
    DC:  The Brookings Institution.

Phares, Donald. 1980. Who Pays States and Local Taxes? Cambridge, MA: Oelgeschlegar, Gunn
    and Hann.

West Virginia Department of Tax and Revenue. 1995. Classified Assessed Valuations Taxes
    Levied, 1994 Tax Year. Charleston, WV: Property Tax Division.

Zhou, Qingshui. 1995. "Impacts of Alternatives to the Property Tax for School Financing in West
    Virginia." M.S. Thesis. West Virginia University, Morgantown, West Virginia. Table 1 Effective Tax Rates for the Current and Two Alternative Tax Systems