by Felipe González and Pablo Muñoz: Guest contributors
Two months ago, The Washington Post published an article about the educational system in Chile and Betsy DeVos, currently the 11th United States Secretary of Education. DeVos is a well-known advocate of school choice, a policy that General Augusto Pinochet implemented in Chile in the 1980s. This article narrates the Chilean experience.
The Chilean educational system
The 1973 coup d’état that overthrew the democratic and self-described socialist government of Salvador Allende was a critical juncture in Chilean history. In the following seventeen years a brutal dictatorship led by army general Augusto Pinochet assassinated thousands of citizens and implemented a package of market-based policy reforms that radically shook the country. In education, the Pinochet regime followed the recommendations of economists trained at the University of Chicago. These policies pushed Chile to transition from universal free education to a market-based system in which parents choose schools for their children. After twenty seven years and six democratic governments, the educational system remains fundamentally unchanged.
We emphasize two reforms that exemplify the spirit of the changes made in the 1980s. First, by taking the administration of public education out of the Ministry of Education in the central government and placing it on the hands of mayors as local administrators, the Pinochet regime decentralized the educational system. The objective was to increase the allocative efficiency of resources by relying on private agents in the market. Second, the regime introduced a voucher scheme that transferred money from the central government to private school owners and local administrators. Overall, this privatization process eliminated teachers’ historical contracts and destroyed their unions.
As a direct consequence of the introduction of a market-based educational system in 1981, the percentage of students that attended public schools declined from 80 percent in 1980 to 40 percent in 2016. Nowadays, students in private voucher schools – equivalent to U.S. charter schools – represent 50 percent of total enrollment. The remaining 10 percent of students are enrolled in private schools that work outside of the voucher system. Given the incentives placed by the reforms of the regime, it is not surprising that the former best public schools “migrated” to the private sector (Hsieh and Urquiola, 2006).
An invisible hand in education?
The motivation behind Pinochet’s educational reform was simple. First, give resources to students and let them and their families choose among schools. Second, schools will compete in quality to attract students, who choose schools of better quality. Third, quality of education will increase and learning will improve. This logic is nothing more than the well-known efficiency argument first posed by Adam Smith and popularly known as the “invisible hand.” Less popular among non-economists is the monumental work by Arrow and Debreu (1954), who proved the necessary conditions for a market-based system to maximize allocative efficiency. It will not surprise the reader that these conditions are hardly met in real life and that is why a significant amount of current research in economics deviates from the view of perfectly competitive markets and focuses on imperfections. Education is not an exception and its imperfections are more than enough for us to be skeptical about “invisible hand” type of arguments to support a free market in education.
The first set of deviations from perfectly competitive markets come from the “demand side,” i.e. how parents choose schools for their children. Researchers have shown that this decision is influenced by school fees, school quality, and the geographic distance between home and school (e.g. Gallego and Hernando 2009 and Cuesta et al. 2017, among others). The imperfections arise for several reasons. The unequal distribution of income in a society implies that children can only choose certain schools. In addition, school quality is hard to measure and it is confounded by students who attend that school. Because parents care about classmates, students will tend to sort across schools by ability and/or socioeconomic background. Moreover, the geographic segregation of households combined with transportation costs make it impossible for children to have the same opportunities at the moment of choosing a school.
The second set of deviations come from the “supply side,” i.e. the behavior of schools. Research has pointed to many undesirable responses of schools to improve their position in the market. For example, schools tend to manipulate the pool of students that take standardized tests to improve their school-level test scores and seem as of better quality to the eyes of the public and the government (Cuesta et al. 2017). This behavior affects parents choice of schools and cause the government to incorrectly assign teacher bonuses. Schools are also strategic when considering where to locate geographically, who to accept as students, and how to set their school fees. Moreover, a market oriented educational system might exhibits large turnover with direct consequences for the students. School closure increase school dropouts and grade retention (Grau et al. 2017).
The claim of improved efficiency with a free market system involves many assumptions about the behavior of parents when choosing schools and about schools when they compete in the market. Research has shown that these assumptions are unlikely to hold in the real world. Therefore, the conclusions derived from these assumptions need to be reassessed under more realistic conditions. Moreover, the “invisible hand” is silent regarding inequality and socioeconomic factors which have a huge influence on student outcomes (Mizala and Romanguera, 2000). A market-based educational system actually allows the influence of these factors to grow over time (Muñoz and Redondo, 2013).
Even though the framework of perfectly competitive markets is inappropriate to evaluate a voucher policy in education, this does not necessarily implies that vouchers are a wrong policy per se. Indeed, depending on the market imperfections that we think are prevalent in education, one could argue that the introduction of vouchers could increase, decrease, or have no effects on the outcome of interest. For example, Neilson (2013) shows that the increase in the amount of the voucher in Chile led to an increase in test scores that seems to be explained by more competition among schools in the market. Feigenberg et al (2017), however, raises doubts about the previous findings and argues that strategic responses by schools explain part of the apparent increase in test scores. Abdulkadiroglu et al (2017) study the Louisiana Scholarship Program and find that vouchers actually decreased academic outcomes.
Research can be complicated and we simply need more evidence to understand what are the consequences of introducing vouchers in education. Until now, however, the accumulation of knowledge suggests that the effects of vouchers drastically depend on country-specific institutional features. Indeed, Epple, Romano, and Urquiola (2017) provide the most comprehensive study of vouchers and conclude that these might work in some cases and not in others. Small and large scale programs seem to have different effects and more work is needed to understand why is this case.
The Chilean experience with vouchers in the last 30 years provides many lessons that need to be considered before introducing a voucher scheme. Vouchers do not guarantee an improvement in educational outcomes.
Abdulkadiroglu, A., Pathak, P., Walters, C. (2017). “Free to choose? Can school choice reduce student achievement?.” Working Paper.
Arrow, K., Debreu, G. (1954). “Existence of an equilibrium for a competitive economy.” Econometrica, vol. 22 No. 3, p. 265-290.
Cuesta, J.I., González, F., Larroulet, C. (2017). “Distorted quality signals in school markets.” Working Paper.
Epple, D., Romano, R., Urquiola, M. (2017). “School vouchers: a survey of the economics literature.” Journal of Economic Literature.
Feigenberg, B. Rivkin, S., Yan, R. (2017). “Illusory gains from Chile’s targeted school voucher experiment.” NBER Working Paper No. 23178.
Gallego, F. and Hernando, A. (2009). “School choice in Chile: looking at the demand side.” Pontificia Universidad Católica de Chile, Documento de Trabajo #356.
Grau, N. Hojman, D and Mizala, A. (2017) “School Closure and Educational Attainment: Evidence from a Market-based System. Working paper.
Hsieh, C. and Urquiola, M. (2006). “The effects of generalized school choice on achievement and stratification: evidence from Chile’s voucher program.” Journal of Public Economics, vol. 90, p. 1477-1503.
Mizala, A. and P. Romaguera (2000), “Determinación de factores explicativos de los resultados escolares en educación media en Chile”, Serie Economía, No. 85, Santiago, Chile, Centre for Applied Economics, Department of Industrial Engineering, Faculty of Physical and Mathematical Sciences, University of Chile.
Muñoz, P. and Redondo, A. “Inequality and academic achievement in Chile.” CEPAL REVIEW, 109 (2013): 107-123.
Neilson, C. (2014). “Targeted vouchers, competition among schools, and the academic achievement of poor students.” Princeton University, Working Paper.