1. Although our constitution does not require attendance in primary or secondary schools, state law has long compelled school attendance. State laws regulate almost every aspect of these schools, including the number of days in the school year. Private schools are also regulated by the Nonpublic School Act. A law passed in 1995 provides for a gradual increase in the number of school days per year up to 195 by the year 2011-2012.

  2. Milton Friedman, Capitalism and Freedom. Chicago: University of Chicago Press, 1962. I term this the "modern" school choice debate, since schools were privately funded for centuries before governments took over the responsibility. For an acknowledgment of his influence, see The New York Times editorial notebook article quoted below, in the discussion on addressing common myths.

  3. James F. Hirni, "Introduction," School Choice Programs: What’s Happening in the States. Washington, D.C.: Heritage Foundation, 1996.

  4. In Vermont, local school boards in certain towns will pay tuition at any public or approved private ("independent") secondary school, in or out of the state. The school boards are required to pay an amount equivalent to the average tuition charge of the state’s union high school districts, with the parent paying the remaining amount. See John McClaughry, "Educational Choice in Vermont." Concord, VT: Ethan Allen Institute, n.d.; available at the Ethan Allen Institute’s World Wide Web site, http://www.ethanallen.org/. McClaughry, the president of the Ethan Allen Institute, was the vice chair of the state’s Senate Education Committee in 1991-1992.

  5. William Allen and Eugenia Toma, A New Framework for Public Education in Michigan. East Lansing, Michigan: Michigan State University, 1996, pp. i-x.

  6. Proposal A added a two-percent sales tax and a six-mill state property tax, both dedicated to the aid of government schools. Other taxes are also used to support the government school system. See the section on fiscal analysis for a more complete discussion of the taxpayer cost of government schools.

  7. See the discussion on choice improving the whole system, below, for an interesting example of advertising by public schools when parents gain choices.

  8. Daniel McGroarty (in Break These Chains. Rocklin, CA: Prima Publishing, 1996, p. 193) reviews researcher John F. Witte’s assessment of parental involvement in the Milwaukee Choice Program. In that program, parents making choices about their children’s schools had "high parental involvement coming into the schools and even higher involvement once there." According to Witte, parents responding to surveys felt that in their choice schools they found "what they professed they were looking for when they entered the program—increased learning and discipline."

  9. John E. Chubb and Terry M. Moe, Politics, Markets, and America’s Schools. Washington, D.C.: The Brookings Institute, 1990, p.4. Chubb and Moe note that David B. Tyack’s book One Best System: A History of American Urban Education made the term "one best system" popular, although his work does not hold that one system is best for all.

  10. Chubb and Moe, Politics, Markets, and America’s Schools, p. 189.

  11. Gary Becker, "Human Capital and the Family," (available on cassette tape) Grand Rapids, Michigan: Acton Institute, 1996. Becker’s work, Human Capital (1970) is the seminal work on the subject. See also his essay "Human Capital" in David Henderson, The Fortune Encyclopedia of Economics: New York: Warner Books, 1993.

  12. Becker notes a small difference of opinion with his colleague Milton Friedman on this issue. Becker believes a voucher system should be targeted to the poor, because they suffer the most from the current system. Friedman favors a more open system. This study recognizes both these views and advocates a tuition tax credit that has the same maximum value for all, but is somewhat more accessible to low-income parents.

  13. Scott Gordon, executive director of CEO Michigan, chronicles the impressive expansion of the scholarship program: In the first scholarship year, 1993-94, 3 scholarships were granted, followed by 10 scholarships in 1994-95. During the 1995-96 school year, 162 scholarships were awarded, a number that doubled to 330 during the 1996-97 school year. CEO Michigan currently assists 462 students with scholarships.

  14. Carol Stevens, "Detroit’s Cornerstone Schools Get Rave Reviews on Capitol Hill," The Detroit News, May 15, 1997, p. A-1.

  15. A variety of these programs exist at both the state and federal level. (For a discussion of Michigan’s tuition tax credit for college tuition, see the financial analysis section of this study.) However, a handful of colleges, led by Michigan’s Hillsdale College, refuse to accept any government funding because they do not wish to accept the government controls that accompany it.

  16. Although the new inter-district options are promising, stronger legislation is necessary. Because the School Aid Act is examined every year during the appropriations process, the inter-district choice option could be deleted or altered in future budget years.

  17. "Sending Child To New Public School Costs Parents," Detroit Free Press, November 19,1991, p. 3A.

  18. The statute that originally provided for public school academies, P.A. 362 of 1993, was ruled to be unconstitutional by the Ingham County Circuit Court in November of 1994. The court held that public school academies were not public schools under state law because they were not under the exclusive control of the state. As a result of that ruling, another statute, P.A. 416 of 1994, was enacted to address the constitutional questions and to replace the previous statute.

  19. Chester E. Finn, Jr., "Cheating the Child to Save the ‘System’," The New York Times, May 17, 1997.

  20. See "Teacher Unions: Are They Helping or Hurting Education?," Mackinac Center for Public Policy, http://www.mackinac.org/9399.

  21. http://www.mackinac.org/9399

  22. Mike Antonucci, "Who’s an Extremist?" The Wall Street Journal, May 20, 1997. Antonucci is the managing editor of the California-based Western Journalism Center.

  23. McGroarty, Break These Chains, p. 188.

  24. Chester E. Finn, Jr., Bruno V. Manno, and Louann A. Bierlein, "Executive Summary," Charter Schools in Action: What Have We Learned?. Indianapolis, IN: Hudson Institute, 1996, p.1.

  25. Public schools receive a basic per pupil foundation grant for each student attending their schools. The Lansing School District foundation grant for the 1996-97 school year is $5,912. When the district loses students, it loses the grant for those students. It is important to note, however, that the amount of the per pupil grant flowing from the state to the school district remained constant.

  26. See the discussion on public and private school costs, in Appendix I, which details how private school tuition is typically supplemented by substantial other private funds.

  27. It is interesting to note the parallels between the actions of those who fought competition in the auto industry and those who fight it in the education industry. Many auto union leaders supported protectionist trade barriers to prevent foreign manufacturers from selling in the US market. Similarly, many teacher union officials want to preserve protectionist barriers preventing private schools from selling their services in an open market.

  28. Mark Mayes, "Lansing Schools to Unleash Marketing Push," Lansing State Journal, May 5, 1997. The advertising even extends to placemats in local restaurants touting a "parade of schools."

  29. Data from Michigan Department of Education, 1996-97 year, indicate 17,851 students enrolled during the year. Thus, the movement of about 1 in 24 students was all it took to force significant change in the Lansing district.

  30. Brent Staples, "Showdown in Milwaukee/How Choice Changes Public Schools," The New York Times, May 15, 1997.

  31. McGroarty, Break These Chains, p. 192.

  32. From "The New Colossus," a sonnet by poet and essayist Emma Lazarus, which is inscribed on the Statue of Liberty.

  33. "A Lesson From Milwaukee," editorial, The New York Times, April 25, 1997.

  34. Hugh Calkins, "School Vouchers—A Middle View," Education Week, Sept. 11, 1996, p. 35.

  35. See the example of the Lansing School District, above.

  36. Bill Johnson, "Powerful Baptist Council Leans Toward Favoring School Vouchers," The Detroit News, April 20, 1997, p .5B

  37. Johnson, "Powerful Baptist Council Leans Toward Favoring School Vouchers," p. 6B.

  38. "Michiganders choose vouchers to help kids in failing schools," Lansing, Michigan: Teach Michigan Education Fund, February 24, 1997; Johnson, "Favoring School Vouchers," p. 5B. The poll also asked a separate question for the Detroit Free Press about a proposal that the state take over failing public school districts. The public opposed the plan by a 53% majority.

  39. The pollsters indicate a margin of error of four percentage points.

  40. The founding works of the public choice school are those by Gordon Tullock and James Buchanan, who co-authored The Calculus of Consent in 1962. Buchanan was awarded the 1986 Nobel Prize in Economics for his work. The Society of Public Choice was started in 1965 by these two men, and is loosely affiliated with a journal devoted to the field. The Society has a Web site at http://sunsite.unc.edu/public-choice.

  41. Nineteen centuries before the emergence of "public choice" as a school of economics, the Roman lawyer Pliny the Younger made a direct observation on this point. He noted that he was less likely to send children to a school maintained by public funds than to those supported by private tuition, because the teachers at the taxpayer-funded school paid more attention to the concerns of politics than to the desires of the parents who placed their children in their care. Letters and Panegyricus, "Letters," Book IV, XIII (Louden, William Heineman, 1969). Quoted in Andrew Coulson, "Markets versus Monopolies in Education: The Historical Evidence," Education Policy Analysis Archives, vol. 4 number 9. (June 1996) College of Education; Arizona State University. Available at http://olam.ed.asu.edu/epaa/v4n9.html.

  42. The fiscal analysis section, in Appendix I, cites a study showing the likely voting behavior of teachers and parents in a Michigan school millage election. Their behavior was entirely consistent with the "public choice" theory discussed here.

  43. Chubb and Moe present a thorough discussion of public choice and politics in their chapter entitled "An Institutional Perspective on Schools," in Politics, Markets and America’s Schools, pp.26-68.

  44. Allen and Toma, A New Framework, pp. 129-131.

  45. Edward M. Gramlich, Distinguished Lecture on Economics in Government, "Setting National Priorities: 1992," in Journal of Economic Perspectives, vol 6 No. 2 (Spring 1992), p. 7-8; John Chubb and Eric Hanushek, "Reforming Educational Reform," in Henry J. Aaron, ed., Setting National Priorities: Policies for the Nineties. Washington, D.C: Brookings Institution, 1990, pp. 213-47.

  46. See, for example, in Michigan, Patrick L. Anderson, Proposal A: An analysis of the June 2, 1993 Statewide Ballot Question. Midland, Michigan: Mackinac Center for Public Policy, 1993. Anderson summarizes analyses, including earlier work by Hanushek, showing no connection between government school spending and performance.

  47. Allen and Toma, A New Framework pp. 102-103. Allen and Toma use regression analysis to isolate the contributions of various factors to better performance, measured by graduation rate and dropout rates in various Michigan public school districts. Public expenditures per student was not statistically significant (at the .05 level) in explaining either graduation or dropout rates, and was significant at the .10 level for dropout rate alone. In layman’s terms, this evidence indicates there is less than a one-in-twenty chance that public expenditures per student on their own cause a change in graduation or dropout rates.

  48. Lewis J. Perelman, School’s Out. New York: William Morrow, 1992, p. 184.

  49. Allen and Toma, A New Framework, p. 1.

  50. Constitution of the State of Michigan of 1963, Article VIII, Section 1.

  51. This is dated from the court decisions outlawing school prayer which occurred in the late 1950s and early 1960s, but there has been constant tension over the balance between government and religion since the founding of the nation. Cases on schools and religion have reached the U. S. Supreme Court on several occasions, and these cases mostly determine the outlines of allowable activities. The seminal supreme court cases in the last few decades have been Everson v Board of Education (1947), Lemon v Kurtzman (1971), Committee on Public Education v Nyquist (1973), Mueller v Allen (1982), and Rosenberger v Rector (1995).

  52. Rosenberger v Rector and Visitors of Univ. of Va. 515 US ____ , 132 L Ed 2d 700, 115 Sct 2510 (1995), (slip op. at 18). The court’s discussion here even encompasses direct aid, as long as the aid is for activities that are neutral toward religion. On the tax preference issue in particular, Justice Thomas in his concurrence in Rosenberger v Rector, p. 9, gives an extensive history of government tax exemptions for religious institutions.

  53. See "State Ballot Issue C: Prohibit Use of Public Money for Support of Non-Public Schools," Council Comment, No. 833, Detroit, Michigan: Citizens Research Council of Michigan, October 14, 1970.

  54. Traverse City School District v Attorney General, 384 Mich 390 at 407 (1971), note 2; and 411, note 3. "Shared time" generally means the education of children primarily attending nonpublic schools at public school or other taxpayer-funded facilities, where secular subjects are taught.

  55. Compiled Law 7379 of 1929; Council Comment, October 1970.

  56. The law allowed up to 50 percent of the salaries of certified lay teachers to be reimbursed by the state. Council Comments, October 1970.

  57. Traverse City School District v Attorney General, 384 Mich 390 at 406 (1971), note 1.

  58. Advisory Opinion re Constitutionality of 1970 PA 100, 384 Mich 82 (1970); as cited by the same court in Traverse City School District v Attorney General, 384 Mich 390 at 406 (1971).

  59. Council Comments, October 1970.

  60. The Supreme Court later noted the "voter was barraged by contradictory statements," including those made by prominent supporters and public officials. Traverse City School District v Attorney General, 384 Mich 390 at 407-409 (1971), note 2. See also Council Comment, October 1970.

  61. Constitution of the State of Michigan of 1963, Article VIII, Section 2; added by amendment in 1970.

  62. Traverse City School District v Attorney General, 384 Mich. 390, 185 NW 2d 9 (1971).

  63. Traverse City School District v Attorney General.

  64. The state Individual Income tax is a tax levied on the adjusted income of all Michigan residents, currently at a rate of 4.4 percent.

  65. The Single Business Tax is value-added tax levied on Michigan businesses with gross sales in excess of $250,000. It is calculated using a formula that considers gross sales, payroll expenses, and the value of capital.

  66. The state education property tax is the six-mill property tax levied on all owners of real property in Michigan. It is important to note that revenues from this tax accrue to the state of Michigan school aid fund, not local units of government. However, the tax is collected by local units of government.

  67. Executive Budget Tax Expenditure Appendix, Fiscal Year 1996-1997, Lansing, Michigan Department of Treasury.

  68. This includes revenues of $4.33 billion from the Individual Income tax, $2 billion from the Single Business Tax, and $1.14 billion from the state education property tax. FY 1996-97 Appropriations Report, Senate Fiscal Agency, July 1996, p. 140.

  69. For fiscal year 1996-97, the benchmarks are basic foundation allowance, $5,308; average of school districts, $5,646; average of pupils, $5,876. See Senate Fiscal Agency, 1996 Statistical Yearbook, p. 61.

  70. The Executive Budget 1998, p. C2, calls for 79 percent of all students to be at or above the foundation allowance in fiscal year 1998.

  71. As discussed above, the alternative school tuition figures used by the model are those projected for private schools. If the high-end data for tuition-paid public schools were included, the overall weighted average alternative school tuition would change by $5, or 2/10 of 1%, in the first year. Our estimates of private school tuition are not that precise, and a change of that magnitude would not affect the dynamics of the system.

  72. State of Michigan Department of Management and Budget, Executive Budget 1998, February 1997. The FY 1998 Budget provides funding for the 1997-98 school year.

  73. Under the new school finance system, all school districts will have at least the foundation guarantee allowance to spend per pupil. As of the passage of Proposal A, the foundation allowance was set significantly above the average expenditure, so that the new system creates an increasing expenditure for the majority of districts. However, special rules limit the annual increases, so that a fraction of the districts by 1998 will not receive the full allowance. At the same time, a number of very high spending school districts were "held harmless" and guaranteed their current expenditure plus additional increases. These additional increases will be smaller in magnitude than for the majority of districts, so the variation in spending is declining. As of 1997, DMB estimates showed the "average" per pupil expenditure to be about 6 percent above the "formula guarantee."

  74. "Categorical" expenditures include special education, special education transportation, at-risk programs, vocational education, and early childhood programs. See Senate Fiscal Agency, 1996 Statistical Yearbook, p. 57; or annual school aid appropriation bills.

  75. Much of the acquisition and construction of school facilities is financed through debt millage, which is levied in addition to operating millage. Public employees themselves do pay state income tax, and local income taxes in those Michigan cities that levy them.

  76. Religious schools receive tax exemptions as well. It is interesting to note that schools and other local units of government sometimes must pay special IRS taxes for "arbitrage," when they borrow at tax-exempt rates and then invest the proceeds in taxable securities.

  77. There is no consistent data source for all alternative schools. This estimate emerged from discussions with private school representatives. Typically, elementary school tuition runs around $1,000 to $2,000 per year, and high-school tuition runs between $3,000 to $4,000 per year. Tuition at religious schools is usually less than that at nonreligious private schools. Tuition at elite private academies often exceeds $10,000, which is near the per-pupil expenditures of the local public school districts.

  78. According to a 1996 Cato Institute study, the average tuition for all private schools, elementary and secondary, is $3,116; less than half of the average cost per pupil in the public school, $6,857. In 1994, 67 percent of all private schools charged $2,500 or less for tuition; and more than 5,000 private schools charged under $1,000. David Boaz and R. Morris Barrett, What Would a School Voucher Buy? The Real Cost of Private Schools. Washington, D.C.: Cato Institute, March 26, 1996.

  79. Public schools also receive some of these services, but the requirement in the Constitution for a "free" education prevents government schools from charging for any basic service.

  80. This number is the "basic foundation allowance" established in the Executive Budget 1998, which was published by the Department of Management and Budget in February 1997.

  81. In the context of school finance reform, in which one of the goals was to reduce the disparity among school districts’ spending, this is known as "averaging up."

  82. Confusion sometimes arises when the budget for school aid is discussed because the school calendar is different from the state’s fiscal year. The state’s FY 1998 basic foundation allowance grant covers the 1997-98 school year; the 1998-1999 school year will be provided for in the FY 1999 budget. To make matters more confusing, the state sometimes front-loads or back-loads the expenditures, and occasionally provides advance payments to assist school districts with their cash flow.

  83. Conversation between Joseph P. Overton of the Mackinac Center for Public Policy and Mr. Powell of the Saginaw Public Schools, October 8, 1997.

  84. This inference is drawn by using the percentage of students paying tuition in these districts and applying it to the statewide number of public school students. The tuition is calculated as a weighted average of tuition paid by students in the sample.

  85. There are a number of reasons to expect these "unreleased" students to decline, although to never reach zero. First, once school choice is opened up by constitutional amendment, there would appear to be little to be gained by punishing parents who exercise it within the public school system. Second, more and more school districts are awakening—however reluctantly—to school choice. The example of Lansing, discussed in Section I, is instructive. Many of these districts are adopting cooperative agreements with surrounding districts. Third, there will always be some students who, for a variety of reasons, fall outside the boundaries of expected circumstances. Our survey discovered at least one student from out of the country, and another who missed a deadline for enrollment in a schools of choice program. This means the number of students in this category will probably never be zero. Fourth, as part of legislative implementation of a tuition tax credit program, such practices as charging tuition in public schools will probably become more restricted. Finally, under the current and proposed Article VIII section 2 of our Constitution, the state promises "free public elementary and secondary schools." Thus, this practice, which is already suspect, could never get too widespread.

  86. The constitutional amendment specifies the amount by referring to the average state and local per-pupil expenditure for operating purposes. This will not be exactly the same as the guarantee amount. The analysis uses the guarantee amount to predict both the government system costs and the private credit costs so any difference between the "average" and "guarantee" amount would affect both systems similarly. This keeps the comparison of costs consistent, although it somewhat underestimates the cost savings when students migrate from traditional public schools to alternative schools.

  87. David Henderson, "Demand," in David Henderson, ed., The Fortune Encyclopedia of Economics. New York: Warner Books, 1993.

  88. Marshall also pioneered the ubiquitous price-and-quantity charts, with their demand and supply curves, which are part and parcel of any economics course. His seminal work is Principles of Economics, 8th Edition. London: Macmillan, 1920.

  89. In mathematical terms, price elasticity of demand = ;
    where P and Q denote price and quantity, and the delta symbol denotes a change in the variable. This measure is equivalent to ; which is more intuitively the "percentage change in quantity given a percentage change in price."

  90. Technically, a one-to-one relationship such as that indicated in the example produces a price elasticity of negative 1.0, because the increase in demand is associated with a decrease in price. Since almost all goods have a negative price elasticity (meaning that a lower price encourages a higher demand for the good), many discussions simply drop the sign and talk only about the magnitude of the elasticity. The rare cases of goods with a price elasticity of demand that is positive (meaning that people buy more when the price goes up) are termed "Giffen goods," after Sir Richard Giffen, who noted that this occurs with staple foods among the very poor. See, for example, Intriligator, Econometric Models, Applications, and Methods. Englewood Cliffs, NJ: Prentice Hall, 1978; p. 216, citing Marshall’s Principles of Economics.

  91. For example, Intriligator cites: Goldberger and Gamaletsos, "A Cross-Country Comparison of Consumer Expenditure Patterns," European Economic Review, 1: 357-400, 1970, who studied consumer behavior in the US, Canada, the UK and other developed countries; Houthaker, "New Evidence on Demand Elasticities," Econometrics 33: 797-801, who estimated income and price elasticities for food, rent, clothing, and consumer durables in Western Europe and North America; and many others. See also the discussion following.

  92. See, for example, Pindyck and Rubenfeld, Econometric Models and Economic Forecasts, 2d ed.. New York: McGraw Hill, 1981, chapter 13. Pindyck and Rubenfeld discuss the estimation of elasticities, particularly the difficulty in using time-series data to estimate elasticities at a point in time. They include examples, including a seminal work on automobile price elasticities done by Saul Hymans of the University of Michigan, who estimated the short-term price elasticity of demand for automobiles at –1.07 and a long-run elasticity of –0.36. Saul Hymans, "Consumer Durable Spending: Explanation and Prediction," Brookings Papers on Economic Activity, No. 2. pp. 173-199, 1970. Intriligator, in Econometric Models, contains an extensive discussion of the various estimation methods for elasticities, and numerous citations.

  93. James Gwartney and Richard Stroup, Economics: Private and Public Choice, 8th edition. New York: Dryden, 1997. Used with permission.

  94. In addition to the general theory and citations above, Pindyck and Rubenfeld summarize an analysis done in Troy, Michigan, on the voting behavior of 425 individuals in a school millage election. They report a price elasticity of demand of –1.21 for public school services.

    Interestingly, the researcher included a variable in the equation for attendance at private schools, expecting to find that having children in a private school would be associated strongly with voting "no" on public school millage. However, the estimation did not show a statistically significant difference in behavior for parents currently sending their children to public and private schools. School employees, however, were much more likely to vote "yes" on millages, as were the parents of multiple children in the public school system. Daniel Rubenfeld, "Voting in a School Election: A Micro Analysis," Review of Economics and Statistics, vol. 59, no. 1. pp. 30-42, February 1977. For a discussion of "public choice" economics and political support for public school spending, see A Public Choice Analysis of the Current System in section I, February 1977, pp. 30-42.

  95. The two best examples are Cornerstone Schools in Detroit, and Children’s Educational Opportunity (CEO) in Grand Rapids. Both offer scholarships to parents who want to send their children to private schools, or in the case of CEO, to a private school or out-of-district public school. Both have scrambled to raise funds to support the scholarship program, which has met with significant demand. Since many parents in these programs have below-average income, their willingness to support the additional expenses not covered by the scholarship underlines the strong desire of families to place their children in better schools, even at financial sacrifice.

  96. See the discussion on Private School Costs and Public School Costs, above.

  97. This would occur because charitable and religious organizations, which are chronically short of resources, would find it possible to fulfill their educational mission with less of an internal subsidy to an affiliated school.

  98. With charter schools starting at 15,000 students compared with the traditional public school total of over 1.6 million, including that sector in this fiscal analysis would make almost no change. Home-schoolers, whose numbers are hard to estimate in any case, are likely to be even more resistant to moving and have an even smaller overall fiscal impact. This is not to underestimate the role they play, only to observe that the fiscal effect of including them would be small. See the discussion beginning on page 56, "Public, Private, and Alternative Schools," regarding public school tuition students.

  99. If there were any change in this amount, which includes intermediate school district funding, certain categorical programs, some federal aid, and other local factors, it would likely favor the UTTC plan. If private schools start to take over some of these functions, and they are somewhat more efficient than the public sector, then this would add to the savings that taxpayers would see in the UTTC plan.

  100. Michigan citizens pay numerous taxes whose revenues are dedicated partially to the School Aid Fund, including sales, tobacco products, cigarettes, industrial facilities, liquor excise, real estate transfer, state education, and use taxes. Annual Report of the State Treasurer, 1995-96, Lansing, Michigan: Michigan Department of Treasury, table 14.

  101. Clearinghouse organizations would have to keep separate accounts for administrative expenses to avoid mixing creditable tuition payments with deductible administrative expenses. Such separate accounting is already necessary for charities that engage in certain profit-making enterprises. Separate accounts are also kept by numerous nonprofit, for-profit, government, and voluntary organizations to meet numerous accounting, disclosure, legal, regulatory, and prudential requirements.