What have American schools been doing about the troubling trends in student performance?
For the last twenty years American schools have been trying in many and varied ways to improve student performance. Educational systems did not wait until the 1983 presidential report, A Nation At Risk, warned of a "rising tide of mediocrity" to begin seeking improvement. The national decline in test scores was apparent by the early 1970s, and efforts to turn that decline around – to boost student achievement – began in earnest at that time. For example, the 1970s saw a strong nationwide movement to hold schools accountable for student performance. The 1970s also brought innovations in curriculum, instruction, and special programs almost as numerous as the state and local systems of education that implemented them. To be sure, the 1980s saw stronger waves of reform sweep across the entire nation. But the important point is that America's school systems began introducing reforms attacking educational mediocrity for a long time and have continued to pursue such reforms over the last two decades. 
This experience should at least make reformers skeptical of new efforts to improve education through the existing school systems. While it would be premature to pass final judgment on the school reforms of the last five years, it is fair to say that America's school systems can provide little evidence that their last two decades of reform efforts have paid off. There is also ample evidence that the dominant approaches to reform that the schools have used, approaches that rely heavily on spending and regulation, have not been working out.
It may come as a surprise to some participants in the current educational debate, but public schools are increasingly well-funded institutions. From 1970 to 1987 per-pupil current expenditures increased 407 percent in nominal terms against an inflation rate of 177 percent.  That's a real increase of 83 percent. Total expenditures per student in daily attendance reached $4,300 by 1987.  For the sake of comparison, that amount is more than twice the cost of educating a student in a Catholic school, where research indicates the education is superior.  Even during the 1970s, the period of steepest decline in student achievement, per-pupil current expenditures in the public schools increased 44 percent in real terms. 
But what were these increased financial resources being used for? Some portion of the additional money was being used for two things that educational systems have long argued are vital to school improvement. Teacher salaries were increased and class sizes were decreased. From 1970 to 1987 the average teacher salary in the United States increased from $8,560 to $26,700, and the average ratio of pupils to teachers (a proxy for class size) fell from 22.3 to 17.6.  Moreover, the increases in teacher salaries purchased at least nominal gains in teacher quality. The percentage of teachers with master degrees doubled by 1983, reaching 53 percent. 
These "improvements" account for about half of the 83 percent real increase in per-pupil education spending over the period. Although the increase in average teacher salaries amounted to only 12 percent in real terms, the increase had to be paid to 27 percent more teachers per student. The cost of teacher salaries therefore rose 42 percent per student from 1970 to 1987. If the returns to this investment have been meager – and this apparently is so – there are several immediate reasons. One is that reductions in pupil-teacher ratios or class sizes of the magnitude achieved during the seventies and eighties simply may not produce systematic improvements in student achievement. Another is that these reductions were not achieved through careful efforts to increase student-teacher contact but rather as a by- product of efforts to minimize teacher layoffs during a period of declining student enrollments. A final reason is that the higher average salaries – actually 7 percent lower during the 1970s – were not being used to attract more talented teachers into the profession but to compensate those already in the aging teaching force for increasing experience and educational attainment.
If the increased investment in teacher salaries did not pay off as hoped, however, it had, and still has, a better prospect of improving education than most of the rest of the increase in public school spending that occurred from 1970 to the present. Schools are not spending more today than twenty years ago because of better books, materials, laboratories, equipment or other obvious improvements in instructional facilities. No, at least half of the 83 percent real increase in educational spending per pupil from 1970 to 1987 was consumed by such things as more expensive employee fringe benefits (which doubled their share of school system budgets); rising "fixed costs" such as rent, maintenance, and insurance; increasing use of "auxiliary teaching services "provided by aides and counselors; and last but not least, educational administration.  Indeed, after teacher salaries and fringe benefits, school bureaucracy may be the single largest beneficiary of the substantial increase in educational expenditures over the last two decades.
Because of problems of data availability and comparability it is impossible to estimate with confidence the size of the real increase in all administrative costs per pupil in America's public schools since 1970. But the data that are available describe very significant growth in public school bureaucracy over the last two decades. From 1977 to 1987, when the ratio of students to teachers nationwide fell 8.4 percent, the ratio of students to central office professional personnel dropped 21.0 percent.  In other words, administrative employment outside of schools was growing at two-and-a-half times the rate of instructional employment inside the schools. Between 1960 and 1980 local school spending on administration and other non-instructional functions grew by 1.07 percent in real terms, almost twice the rate of per-pupil instructional expenditures. 
More instructional matters were also being taken out of the classroom: between 1960 and 1984 the number of non-classroom instructional personnel in America's school systems grew 400 percent, nearly seven times the rate of growth in the number of classroom teachers.  In 1983, the last date for which such figures are available, full-time classroom teachers represented barely half (54 percent) of all local school employment; administrators represented 13 percent.  Whatever its precise magnitude, though, the recent growth in public school bureaucracy may have harmed more than helped the academic performance of schools. And, if the growth in bureaucracy was not generally beneficial, it partially explains why two decades of growth in school expenditures was not very effective.
In due course, we will provide many reasons for concern about school bureaucracy, but an immediate reason has to do with the role of bureaucracy in school reform. Two decades of school reform have substantially increased the regulation of public schools. The local public school is subject to far more regulations today than it was when the nation's educational slide began. Schools are more constrained in their use of personnel, their design of curriculum, their choice of instructional methods, their maintenance of discipline, and their provision of special programs. School reform is not solely responsible for this. Collective bargaining with increasingly powerful teacher unions has helped to constrain schools. And the school systems, for their own reasons, have seen fit to take authority out of schools and to centralize it in school headquarters.  Still, from the countless special programs of the federal government – for example, compensatory education – to the curriculum specifications of state departments of education, to the implementation of these innovations by district offices, school reform has increased the regulation of local schools.
During the 1980s, moreover, this tendency has picked up pace. The powerful waves of school reform that have swept the nation since 1983 have followed well-worn paths. First, educational spending has been increased more than anyone thought possible when A Nation At Risk called for a long list of expensive reforms. Between 1953 and 1988 aggregate spending on public elementary and secondary education increased by $56 billion – an amount greater than one percent of the Gross National Product.  Second, many new regulations have been written. Some of these regulations may be desirable (though research does not encourage optimism). For example, almost all states have imposed higher graduation requirements on high schools, and many states have required competency tests of new teachers. But much of the regulation – as we shall explain – has the prospect of backfiring. The increased regulation of student and teacher performance, now being widely implemented through evaluation and accountability systems composed of tests and a host of other "objective" criteria, could easily rob schools of vitality and undermine their performance. This is a fairly well-known danger, but it is a danger that educational systems, now so heavily dependent on central administration, are willing to accept. It is also a danger that politicians, ultimately responsible for these systems, can hardly avoid. With education organized as it is, politicians interested in improving school performance have little choice but to provide educational systems with more money and then try to regulate how those systems use it. Within the existing systems, reform options are limited.