When Eric Hanushek's aunt graduated as valedictorian of her college class decades ago, she decided to become a teacher - a choice bright young women often made in those days. She eventually became a public school principal.
"Today valedictorians don't go into teaching," says Mr. Hanushek, an economist at the University of Rochester in New York State. They are more likely to go into law or business, or become university professors. Teaching in elementary or secondary schools (junior high or senior high schools) has lost considerable status as a profession. Teachers' wages have also risen slower than the pay of nonteachers. That is especially true for women.
Yet, notes Hanushek, public education costs have steadily risen. Real expenditures per student - after deducting inflation - increased 3.5 percent per year on average over the period from 1890 to 1990. Real public expenditures on primary and secondary education rose from $2 billion to more than $187 billion in constant 1990 dollars. Even if student population had remained constant in those 100 years, expenditures would still have risen by 25 times. Public school spending has increased from less than 1 percent of national output in 1890 to 3.4 percent in 1990.
Despite such growth, relatively little attention has been devoted to the costs of education, Hanushek holds. "Nobody tries to put it into the larger perspective." That is, not until recently, when he and Steven Rivkin, an economist at Amherst College in Amherst, Mass., analyzed 100 years of statistics on student enrollment, instructional staff per pupil and the school-year length, the price of instructional staff, and other spending. Their key findings, outlined in a paper for the National Bureau of Economic Research in Cambridge, Mass., include:
*Most of the rising cost of public school education stems from higher teacher salaries, rising as wages increased in general. Teacher wages climbed from $34 per day in 1890 (in 1990 dollars) to more than $177 per day in 1990.
*A second key cost factor has been a decline in the number of pupils per teacher. The number was 20.5 in 1970 and 15.4 in 1990, a change accounting for 85 percent of the $25 billion hike in instructional staff spending in those 20 years.
*Costs other than teacher pay have gone from one-fourth of total school spending in 1890 to 54 percent in 1990. But some teacher benefits and instructional costs are included in this.
*The education of students classified as having mental and physical handicaps has been a less significant but fast-rising cost. Under a 1975 federal law, these "special" students have first call on available funds. Such pupils cost about 2.3 times more than regular students.
The decline in the teacher-pupil ratio has been welcomed by teachers. "Teachers, like everybody else, appreciate better working conditions," Hanushek says. Though students may get more individual attention, he finds "no consistent evidence" that a drop in the number of students per classroom in the 15- to 40-pupil range results in better student performance. Of 300 studies he reviewed, 15 percent found achievement improved significantly as the ratio fell; 13 percent showed significantly lower achievement.
Most teachers work hard, Hanushek says. But they don't carry out their duties much differently when the number of students in a class is reduced. Good teachers remain good teachers; bad teachers remain bad teachers.
Hanushek and Rivkin are concerned that in the years ahead, American school systems will have a hard time maintaining the quality of teachers. In the 1980s, they note, a falling student enrollment allowed per-student expenditure to rise faster than total spending. Class sizes decreased, soaking up any potential savings. But this fortuitous situation has ended and reversed in the 1990s. The number of students is growing again. The two economists suspect this will put pressure on school-district budgets, and that taxpayers will become more concerned about costs and the educational returns on their tax dollars.
What's to be done? Hanushek suggests school systems weed out poor teachers and pay good teachers salaries that are more competitive with other jobs. Since unions and tenure rules often make firing a teacher difficult, school officials may end up seeking nonfinancial incentives for teacher performance - the adult equivalent of the stars teachers use to recognize good work.