The overwhelming majority of students received a relatively small amount of schools' additional spending.

We are often told that per-pupil spending has doubled in real terms over the last 25 years but the investment hasn't produced academic gains. So why should we keep throwing good money after bad? The negative equation -- more spending equals no greater student achievement -- fits right in with the current anti- government mood. Why should taxpayers play Santa Claus? But the math is wrong, as researchers from the Economic Policy Institute (EPI) demonstrate in a new study.

The study, "Where's the Money Gone?: Changes in the Level and Composition of Education Spending," shows that while spending has gone up since the 1960's much of the growth was soaked up by price inflation in the goods and services schools purchased. Furthermore, most of the remaining dollars went for programs schools had never provided before. Most of the budget used to go straight into the core academic program, but schools began directing and increasing share to new responsibilities the courts and Congress mandated for students with special needs. Many were children with severe disabilities who would not even have been in school in an earlier era.

To get a clearer picture of what happened to education costs, researchers Richard Rothstein and Karen Hawley Miles looked in detail at how a representative group of nine school districts (large, small; rich, poor; fast-growing, slow-growing; rural, suburban, urban) spent their money between 1967 and 1991. They analyzed expenditures receipt by receipt. This enabled them to pinpoint how inflation worked on each district's budget. They found that the Consumer Price Index ( CPI) -- the usual gauge of education inflation -- didn't match what was happening to costs. The districts were having to spend more just to stay even.

It's not that critics deliberately misstate the case. But even though schools received a 100 percent funding increase over 25 years based on the CPI, they can't buy twice the amount of teaching, supervision, textbooks, and support services that they bought before. The study show that, when inflation is accounted for, per-pupil spending actually grew by 61 percent. Schools can buy 61 percent more now than they could 25 years ago -- still a substantial increase, but not double the amount.

What did schools do with the additional money? Here's what the researchers found when they sorted expenditures in the nine districts by program: The overall share of per-pupil spending going to "regular" education -- the instructional program for most students -- dropped from 80 percent to 59 percent, while the share for special education grew from 4 percent to 17 percent. By 1991, special education and other targeted programs (including bilingual, vocational, and compensatory education; desegregation; dropout prevention; school lunch; and alternative education) had received about 60 percent of all new school funding. Special education alone accounted for 3 8 percent; regular classrooms received only 28 percent.

This means that the overwhelming majority of students in public schools received a relatively small amount -- 1 percent per year -- of the schools' additional spending over a long period of time. Disabled students, students in poverty, and student with special problems got more money, and there have been increases in their achievement levels. Although there's still a long way to go, test scores of poor and minority students have risen steadily relative to those of middle-class students. High school graduation rates of minority and white students are converging. And educational possibilities unimaginable only two decades ago have been opened for millions of youngsters with disabilities.

These gains haven't come cheap. New York City, for example, spends $5,200 per pupil in regular education but $23,600 per pupil in special education. Districts are sometimes required to spend $50,000-$100,000 on a single student. The urban districts in the EPI study poured almost all of their new resources into special programs rather than into regular education; one put 18 percent into feeding programs alone. The suburban districts showed the same trend, but they had enough money to increase spending on regular education as well.

What about teacher salaries? They represented a relatively stable share of spending in the nine districts. Schools hired many new teachers trained in working with students with special needs. But growth in the budget share going to salaries didn't come from staffing; it was just that over time there were more senior teachers than junior teachers being paid, so the average salary level went up. In reality, the salaries of most teachers barely kept pace with the CPI and some fell behind.

The large sums being spent on special education undoubtedly are a response to the fact that nothing at all was spent on these youngsters before. It may be time to pause and reevaluate special education's role in driving school costs. But we aren't getting any help from the critics. As long as they keep promoting the false equation of higher spending with poor achievement, we'll never get to the real, much more difficult, problem: How should we reshape our education priorities to make America's schools second to none in academic excellence?