When Canadian schools suffer from poor track records and students fall behind, provincial governments always favour the same response: throw more money at the problem. New crises prompt political actors to request further rounds of new investments, reinvestments, refinancing, improved financing — the slogans change but the strategy is always to try to secure improvements in educational outcomes by increasing the quantity of inputs used.
But this sidesteps the crucial issue of how resources are used in the first place. The reality is that by downsizing ministries of education, making school funding follow the student and introducing greater school autonomy, outcomes for students could improve while actually reducing government spending on education.
There is a large empirical literature documenting the weak relationship between input levels and educational performance. To take just one example, Quebec data show that performance on PISA (Programme for International Student Achievement) tests for mathematics has been quite stable, declining only 1.5 per cent since 2006, even though spending per pupil has increased 18 per cent.
What does help? The literature clearly shows that systems that decentralize management to the local level, introduce choice and exit options for parents and create local feedback mechanisms (such as participating in school associations) heighten the efficiency of any given spending level. In such systems, the state generally disengages from producing the service and concentrates solely on financing it — in ways tied to parental choices.
This makes sense for a variety of reasons. First, as a rule, “one-size-fits-all” policies tend to yield disappointing outcomes for heterogenous populations. Second, parental involvement tends to be higher in decentralized systems, and this creates a positive feedback loop between school administrators and local populations. Third, tying funding to parental choices gives parents an exit option, which in turn generates strong incentives for schools to provide higher-quality customization.
The benefits of this customization are considerable and improve both cognitive outcomes (namely, performance on test scores) and non-cognitive outcomes.
On cognitive outcomes the literature is quite robust, as it relies on randomized control trials (RCTs). In most RCT testing, increases in parental choice find a clear positive effect on cognitive outcomes. Moreover, the few that find adverse effects involve cases where the gains in school autonomy and parental choice were small compared with other cases studied in the literature. The bottom line is that meaningful increases in parental choice and school autonomy tend to yield positive outcomes in terms of performance.
Yet cognitive benefits are not the lion’s share of the benefits of parental choice and school autonomy. For parents, schooling is not only about scores on standardized tests for reading and math. They also consider the social environment in which their children will learn and whether it will be good for their mental well-being. In fact, the literature on parental choice in schooling shows a strong association with improvements in students’ mental health. One recent study concentrating on American teenagers between 15 and 19 years of age found a 10 per cent reduction in suicide rates following the introduction of school choice, among other similar beneficial effects.
Though the implications of improved test scores are largely intuitive the benefits of non-cognitive improvements may be harder to grasp. Yet they probably rival cognitive development in importance, as mental health in early adolescence is a strong predictor of incomes later in life, as well as adaptability to different work environments.
Not surprisingly, being bullied between the ages of 13 and 16 significantly reduces educational achievements and earnings by age 25. As school choice is tied to less bullying, the benefits of this policy will be considerable in the long run.
Everyone agrees that improving both cognitive and non-cognitive outcomes for children is important and that educational policy is a key influence on outcomes. But the idea that increasing government expenditures on education will secure those improvements is misleading. How money is spent weighs more heavily than the level of spending.
The bulk of the empirical literature in the economics of education suggests that policies that improve parental choice and school autonomy provide better ways to spend. The only question is how to adapt school choice and autonomy to each Canadian province’s circumstances for the benefit of parents and students across the country.
Vincent Geloso est économiste senior à l’IEDM et l’auteur de « Amélioration des résultats en éducation: il faut plus de choix, et non plus de dépenses ». Il signe ce texte à titre personnel.