Federal, state, and local expenditures for public education amounted to $184 billion, or $4,538 per pupil, in 1987-88 (Lieberman, 1989: 29). At the same time, educational achievement in the United States, whether measured in terms of student performance on standardized tests, literacy rates, or other dimensions of learning, has been stagnant for a decade or more (National Commission on Excellence in Education, 1983). As a result, public concern with public education has reached a level not seen since the Soviet Union launched its Sputnik satellite in 1957. Calls for educational reform are rampant in many states, with proposals being advanced for increasing teacher salaries, implementing merit pay for teachers, adopting more stringent teacher training and certification requiremens, and introducing various other initiatives designed to improve the quality of education delivered by the public schools. Experiments with introducing choice into the public education process are also beginning to be undertaken. In 1987, for example, Minnesota implemented a plan allowing students to enroll at any public school within the state. Similarly, public school officials in Boston are considering a proposal that would divide the city into three autonomous zones and allow elementary and middle school students to attend any school within their zone of residence (Kelly, 1990). All such experiments are designed to improve the quality of public education by weakening the monopoly power of local school districts. The premise of this approach to educational reform is that by allowing greater freedom of