Research Report October 2002

Amendment 9: Florida's Ballot Initiative to Reduce Class Size Amendment 9 -- scheduled for the November 2002 ballot -- constitutionally mandates legislative funding, beginning in fiscal year 2003-2004, to reduce the number of students in each classroom. If Florida voters approve this amendment, the numbers of students per class would be reduced by at least two students per year. By the year 2010-2011, classrooms would be capped at the following limits: 18 students in grades pre-kindergarten through 3; 22 students for grades 4 through 8; and 25 students for grades 9 through 12.

For Florida voters to approve Amendment 9 would be a blunder of major fiscal proportions. Florida would have made a similar costly mistake in 1995, had the court not stepped in to overturn the STOP (Stop Turning Out Prisoners, Inc.) constitutional ballot amendment that would have constitutionally required prisoners to serve 85 percent of their sentences.

At the time, it was erroneously assumed that a large number of new prisons would have to be built to accommodate a huge projected growth in prisoners. But a significant, unanticipated change in prison demographics-a major drop in future crime rates-lowered rather than increased the need for more prisons. Had the STOP Amendment succeeded, Floridians would have been strapped with a huge tax bill (an additional one cent sales tax) to house additional prisoners that never showed up.

Amendment 9 is more fanciful, if not deceptive, than the STOP Amendment, seeking to place in constitutional concrete an unproven mandate costing tens of billions of dollars (see text box on next page). It would do more harm than good. If approved, this will be the broadest, most expensive class size reduction effort in the nation, and it would be permanent-regardless of future demographics-unless reversed by another constitutional amendment.

Amendment 9 is a deceptive, scatter-shot-from-the-hip approach to reducing class size. Rather than targeting the student populations most in need and providing them useful assistance, this initiative would apply arbitrarily to all public students, in all grades, and to all subjects-even when empirical evidence shows it would make little or no difference. There is no precedent nationally whereby a state has sought through constitutional amendment to reduce class size across-the-board as would Amendment 9. If Florida knows something that other states fail to grasp, it universally is the best-kept empirical secret in "town and gown."

A Question of Costs vs. Benefits

What should Florida taxpayers expect in the way of funding and results should they approve Amendment 9?

Estimated Costs State economists estimate the initiative's cumulative cost over the next 8 years at $20 billion to $27.5 billion. This is $9.3 billion for classroom construction and $18 billion for operating costs, mostly for the hiring more teachers. The average annual cost is estimated to be $2.5 billion-$3.4 billion for the first eight years and $2.5 billion annually (in today's dollars) thereafter.

Accomplishing this class size initiative within the existing tax structure will require a major overhaul of state spending patterns that would threaten to reduce or eliminate funding for the many competing needs within Florida's education system. The intransigent list of potential casualties is alarming: higher salaries for good, qualified teachers; textbooks and instructional materials for all students; supplemental academic instruction for low-performing students; professional development for teachers; expanded community college programs to address Florida's workforce development needs; and university need-based financial aid. Amendment 9 additionally could, eliminate or reduce government services outside of education. Health care and human services needs likely will be candidates for substantial spending cuts.

Even without passage of Amendment 9, Florida will face a budget challenge in excess of $2 billion in Fiscal Year 2003-2004. Education and Medicaid will require $900 million in additional funding to keep up with growth in school enrollment and the elderly poor. To compound the fiscal problem, as Florida TaxWatch revealed in a May 2002 Budget Watch report, lawmakers last year unwisely spent more than $1 billion dollars of nonrecurring revenues on recurring expenses. Additionally, significant increases in remote sales over the Internet (estimated to be upwards to $1 billion in 2001-2002) and revenue decreases due to the accelerated phase out by the Federal government of the Federal estate tax further erodes the state's revenue sources. Florida's debt tripled in the past decade, as well. Florida clearly cannot fund Amendment 9 without substantial spending cuts and/or funding from alternative revenue sources. The text box on the next page puts in perspective potential sources of the new revenues required and the scope of the fiscal challenge ahead should Amendment 9 be approved.

Will smaller class sizes raise student achievement as the proponents of Amendment 9 claim?

Over 1000 studies have attempted to answer this question, yet have failed to resolve the issue. Some research (see text box below) provides limited evidence that class size reduction can be beneficial when focused in some grades, with some students, and when it is part of a larger program of school improvement strategies. But Amendment 9 is a sweeping approach that would apply, across the board, to all grades, all students, and all classes.

Rather than conclusively proving that student achievement improves writ-large when class sizes are reduced, as Amendment 9 proponents would have voters believe, there is mounting Florida-based evidence showing that class size reduction is not the silver bullet its supporters claim.

Putting Things In Perspective Over eight years, the estimated $27.5 billion price tag averages out to $3.4 billion annually.* Based on FY 2004 population, this means an average cost of $201 per Floridian annually or $507 per household for the first eight years. Over the eight years, the total cost would be $1,343 per capita and $3,382 per household. Even assuming that tourists and other out-of-state taxpayers pay 20% of the cost burden, the total investment by Floridians would be $1,074 per capita and $2,705 per household. The state's corporate income tax yields approximately $1.1 billion. Tripling this tax rate could generate an additional $3.3 billion annually. A 1-penny increase to the state's 6-cent sales tax would raise $2.9 billion annually (based on FY 2004 estimated collections). It would take a $1.17 increase to raise $3.4 billion annually. It would take a personal income tax of 6.68 percent of federal tax liability to equal $3.4 billion. A state personal income tax (currently prohibited by the State Constitution) of 1 percent of Floridians' federal taxable income would raise approximately $2.4 billion annually. It takes 3.31 mills of property tax (total state taxable value) to raise $3.4 billion.** Eliminating a variety of sales tax exemptions and exclusions, including services, equals over $23 billion annually. A significant portion of these exemptions are necessities groceries, medical, residential utilities, and rents, etc. Currently, the state spends approximately $16 billion annually on education. An annual increase of $2.5 billion is a 12.5 percent increase. The total General Revenue Fund for the current fiscal year is just under $20 billion. The lottery provides less than $900 million annually to education. The gross receipts tax on utilities, which funds public school construction, raises less than $800 million annually. Four major tax sources combined (gross receipts, corporate income, beverage, and tobacco taxes) are expected to bring in $2.97 billion in FY 2004. Therefore, a doubling of these four tax rates would not raise sufficient revenue. *This is the state economists' estimate. The low end of their estimated cost is $20 billion over eight years. They estimate that Amendment 9 would cost $2.5 billion annually after the first eight years. **The Amendment requires the funding to come from the Legislature, not the local school districts. The property tax is a local tax.

As early as 1995 Florida began experimenting with reduced class sizes. Since that time, Florida has provided over $2 billion to districts for initiatives such as class size reductions that target improved student performance. In 1999 the Federal government began providing funds for class size reduction in districts with the highest concentrations of children in poverty and the highest overall enrollments. Florida has received nearly $200 million in Federal funds. Schools that hired teachers with these funds were able to reduce the average class size from about 23 students to 18 in the targeted grades. There is, however, scant, if any, empirical evidence that these reductions in class size significantly improved student performance.

The Dubious Relationship Between Class Size and Student Performance* The best-designed study available, the Project STAR experiment in Tennessee, found that smaller classes at best matter in kindergarten and that minority and inner city children gain the most from class size reduction. In Wisconsin, a class size initiative (SAGE) targeted elementary schools with a high concentration of low-income students. Classes were limited to 15 students and the class size reduction initiative was accompanied by other program initiatives to implement a rigorous curriculum. Although one study of the Wisconsin initiative found statistically higher student performance of SAGE students in language arts and math in 1st, 2nd and 3rd grades, another study concluded (1) that smaller classes in the 2nd and 3rd grades had minimal impact and, in some instances, no additional impact on student achievement; (2) minority students can benefit from very small classes in the early grades, but how large and long-lasting the benefit will be is unclear; and (3) smaller classes had different effects on achievement, depending on subject areas. Some research suggests that positive effects are less likely if teachers do not change their instructional methods and classroom procedures. One report suggests that the most efficient and effective use of education dollars to raise student achievement involves targeted assistance for minority and low-income students with funding for lower pupil-teacher ratios, more widespread kindergarten efforts, and more adequate teaching resources. *A bibliography of research sources on the issue of class size reduction can be found on the Florida TaxWatch Website at http://www.floridataxwatch.org.

Class size is most likely just one of many variables in a complex mix that may or may not, to an unascertained degree, positively affect student performance: This year's Florida elementary schools earning "As," "Bs," "Cs," "Ds," and "Fs" had median class sizes of 24.1, 23.2, 22.6, 21.9 and 19.9 respectively. Of course, for such evidence to be conclusive, the extent to which "lower-graded schools" on average had lower median class sizes than higher-graded schools (or conversely), needs to be measured for its statistical validity and reliability and other possible intervening variables evaluated.

Such a review of information regarding the impact of class size on student achievement illuminates the overriding problems posed by Amendment 9-the available data are incomplete and inconclusive, thereby raising more questions than answers.

A Question of Having Enough
Qualified Teachers and What Will Work

Research by University of Tennessee statistician, Dr. William Sanders, underscores the shortsightedness and inherent dangers in applying a simplistic solution, as Amendment 9 does, to solving the complex issue of student achievement. After evaluating six million student records and the performance of more than 30,000 elementary school teachers, Dr. Sanders concludes that "teacher effectiveness is 10 to 20 times as significant as the effects of other things," including class size.

California's class size reduction initiative provides an example of a state initiative that failed to provide conclusive evidence linking achievement gains to class size reduction. California's experience also provides an example of the serious implementation problems such initiatives bring to bear. In that state, preexisting problems involving teacher shortages, quality, and distribution took on crisis proportions. The percent of teachers not fully credentialed increased from 1.8% to 12.5%. Of 18,000 new teachers hired, almost one-fourth had no teaching credentials. A proliferation of emergency-permit teachers hit high poverty areas when teachers fled to meet the demand for increased personnel in more affluent districts.

Florida voters take note: Amendment 9 is far more sweeping than was California's effort to reduce class size. Florida's K-12 education problems, therefore, will likely be more sweeping and problematic than California's if Amendment 9 passes.

Even without Amendment 9 Florida's preexisting teacher shortage is expected to reach 162,000 in the next decade. Part of the problem is due to the unprecedented numbers of teachers estimated to retire in the decade following 2006. State economists estimate that an additional 31,000 new teachers will be required over the next eight years to implement the massive class size reduction required by Amendment 9. If Florida voters pass this amendment, the state will be required to attract nearly 200,000 new teachers to avoid a crisis in the classroom.

The shortage of teachers is not limited to Florida. It's a national crisis, made more acute by the passage of the Federal "No Child Left Behind Act" that requires, as a condition to receive federal education dollars, that there be a "highly qualified" teacher in every classroom by the year 2006. Competition for qualified teachers is intensive and already is beginning to escalate. This month the New York Times reported that a ten-year veteran teacher left her job in Tampa, earning $32,000 a year, for a teaching job in Fulton County, Georgia. She is earning $51,000 this year and, with additional education and experience, can increase her salary to $75,000.

How do Florida teacher salaries compare? Their average teacher salary is $5,000 below the national average of $43,250. The implications are clear and disturbing. Attracting first-year teachers to Florida to meet the current and projected shortage is problematic unless their pay scale becomes competitive. Just to the north of Florida's borders, the salary of first-year teachers in Fulton County is $36,500. This is $7,500 above the national average of beginning teacher salaries-$28,986-and $10,714 more than the average beginning teacher's salary in Florida-$25,786. It thus is highly unlikely, if Florida voters approve Amendment 9, that 31,000 out-of-state teachers (in addition to the current projection of 162,000 needed) will flock to Florida in pursuit of substandard pay.

According to the Florida Department of Education, in 2000-2001, teachers not appropriately certified filled 6.9 percent of elementary education vacancies. This is almost triple the number hired in 1990. Dr. Sanders' finding underscores another chilling prospect should Amendment 9 become constitutional law. Florida will be forced to hire even more uncertified teachers than it does currently to resolve a pending teacher shortage epidemic.

There are many approaches to teaching being tried and tested in classrooms across the nation that appear to have more promise for advancing student achievement than does Amendment 9. For example, Florida TaxWatch, for the past three years has been evaluating one such program-the Institute for School Innovation's CHILD Program.* CHILD-an innovative management process approach to teaching children-has demonstrated its students' ability, in a majority of classes tested over the first two years of evaluation, to significantly outscore students from traditional school classes-grades one through five-despite class size differences. CHILD students, from two of three schools in the first year of evaluation and from four of five schools in the second year, excelled despite the fact that their class sizes were approximately 30 students, whereas the non-CHILD students were in classes of approximately 20 students. In the third year of evaluation, as many of larger class-size CHILD schools significantly performed as well as the non-CHILD schools, despite the increased intensity of standardized testing and mounting pressures to excel in all schools and classes.

A Question about Kids and the Future of Florida's Economy

Despite the many drawbacks of Amendment 9, one poll shows as this report goes to press that 71 percent of the electorate supports class size reduction as a matter of principle. Only 49 percent, however, supports reducing class size once they learn that taxes must be raised to pay for it. Aside from the exorbitant price tag to reduce class size, there is another equally unacceptable price to pay if this initiative becomes part of Florida's constitution: More students being taught by poorly qualified teachers. Florida voters need to be fully aware of what they are voting for before they vote on Amendment 9.

It is ironic that Amendment 9 begins with language promoting such a seemingly innocuous stated purpose: "To assure that children attending public schools obtain a high quality education...." Its proponents must be counting on most voters not thoroughly studying the issue and opposing what appears at first blush to be a pristine, well intentioned, and even noble, cause. But CAVEAT EMPTOR must be the watchword as Florida voters go to the polls. Amendment 9 is anything but what it appears to be. If voters carefully and dispassionately examine its consequences, they can only conclude that it threatens, not just the fiscal stability of state finance and budgeting, but the education of Florida's kids-the seed corn of Florida's workforce.

* The first two years of the CHILD evaluation results by Florida TaxWatch-Phases I and II-may be found at http://www.floridataxwatch.org or http://www.ifsi.org. The third year of evaluation-Phase III, currently in progress-will be available on both Websites upon completion.