Another new study has refuted the case that more competition and choice between schools leads to higher student results. The paper reviewed research evidence in several countries and concluded that it is “mixed and modest”. It also found that choice and competition leads to greater social stratification between schools.
These results suggest that the policies followed by the Howard, Rudd and Gillard governments over the last 15 years have taken Australian education in the wrong direction by promoting more choice and competition. These policies have failed to improve school results in Australia over this period but have increased social stratification. The results add to the case for education policies to focus directly on improving the results of low income and other disadvantaged students instead of relying on market mechanisms.
The paper reviewed research studies on the relative performance of public and private schools and whether more competition between schools improves student results in public schools. The authors are academics from Columbia University in New York and the US National Bureau of Economic Research. The paper was published by the Institute for the Study of Labor in Bonn, Germany, in December.
In the United States, studies comparing the results of attending Catholic schools relative to attending a public school have not produced any evidence of a consistent and substantial advantage of private over public schools. The paper cites one recent review which concluded that:
Measured solely by achievement and attainment effects, existing evidence does not support the view that private schools are generally superior to public schools in all settings.
Another study concluded that:
[t]he best research to date finds relatively small achievement gains for students offered education vouchers, most of which are not statistically different from zero.
Studies of an experiment in New York City that randomly allocated school choice vouchers to low income students have found that winning a voucher to attend private a school had a modest, statistically insignificant impact on student learning.
The paper also reviewed some sophisticated studies of private and public school performance in Colombia, India and Pakistan.
The Columbian studies tend to support the hypothesis that vouchers to attend private schools enhance student performance. However, the paper points out that they cannot distinguish between three plausible explanations for the results: increased competition/choice, external effects on student incentives, and an increase in resources allocated to education.
The Indian study reviewed found no difference in private/public student performance. A study of private schools operating in Pakistan without government subsidies suggests that an advantage persists for private schools even after controlling for child, household, and community characteristics. However, this result relies on less solid empirical identification than the studies on Colombia.
In the light of these results, the paper concludes that there is little evidence of superior performance by private schools:
….the literature on whether there is a private test productivity advantage produces mixed results; the estimated effects do not seem to be of a regularity or magnitude such that transferring students into private schools would by itself substantially and reliably raise achievement. [p. 8]
It says that this finding is consistent with a broader literature that attending a higher achieving school or class has little to no effect on academic performance. Some of these find positive effects but no uniform pattern emerges.
The paper also reviews US and developing country evidence of the impact of greater private school competition on the results of public schools. The market theory predicts that it will improve public school results. However, the paper concludes that the US studies show mixed results rather than a distinct finding that greater choice raises test outcomes. They also show increased social stratification between schools.
Chile’s long experiment with vouchers for private schools since 1981 provides a crucial test case of the impact of a large-scale introduction of competition into an educational market. It has attracted a number of studies.
One major study found that municipalities that experienced faster growth in private sector market share did not display higher growth in test scores or years of schooling. It did find increasing stratification between school sectors, with the higher income students in the public sector moving to private schools. These findings suggest competition has little effect on test results.
Other studies of the Chile experience have found that the larger private sector has led to higher student achievement. However, the review paper suggests that there are issues related to the validity of the variables used in these studies.
The paper also suggests that if there is a private school advantage it would be expected that Chile’s national and international test results to have improved over the long period in which large numbers of children were transferred into the private sector. Instead, national test scores in Chile have been largely stagnant even as educational spending has increased substantially. The paper says that this indicates that “privatization has thus been associated with a decline in school test productivity” [p. 10].
Overall, the paper makes the following conclusion from its review of studies of the long experience in Chile with competition from a large private school sector:
…the research on the effect of large scale school choice reforms on average outcomes in Chile produces mixed results. In contrast, there is robust evidence that competition resulted in increased stratification by family background and ability. These results are generally consistent with the consensus within Chile (as articulated recently by the country’s conservative President Piñera) that the school choice system “perpetuates inequality”, and has not done enough to raise learning. [p.11]
The paper attributes the lack of success of markets in education to a gap between the economic theory of education markets and the reality of education service provision. It says that education services do not come close to satisfying basic assumptions of economic theory of market efficiency.
Education consists of a variety of services that cannot be separately priced and parents cannot always make their preferred choice because high achieving schools ration places. It is costly for students to switch schools and is not feasible to do so for many parents. It is not possible to contract for specific outcomes and some key outcomes such as long-term employment prospects can only be observed after a student has finished school. Moreover, school results are extremely difficult to disentangle from their student composition and so school reputations can be durably built around student composition. Unlike policy makers, parents and students do not necessarily care about test scores per se, instead many care more about the opportunities that attending a particular school will offer them in the future.
Given these characteristics of education services, the paper says it is “not surprising” that the predicted results of the theory are not realised in the empirical findings of research studies on competition and choice [p. 15]. Education is a complex service and competition is “not sufficient” to improve school results [p. 29].
Trevor Cobbold
W. Bentley MacLeod & Miguel Urquiola 2012, Competition and Educational Productivity: Incentives Writ Large, Discussion Paper No. 7063, Institute for the Study of Labor, Bonn, December.