Yesterday, in a panel discussion at Policy Exchange – coinciding with their release of a report advocating ‘social enterprise’ schools, which would be forced to reinvest 50% of surplus – I made the case that publicly funded for-profit schools should be allowed.
Why is the profit motive important? Here, I agree entirely with the report’s conclusions. Firstly, profit provides strong incentives for schools to grow and capitalise on scale economies. Non-profit schools generally do not have these incentives. Secondly, non-profit schools have a more difficult time finding up-front capital because they cannot target investors and obtain funds in exchange for future potential profits. Thirdly, even when non-profit schools grow, they do so because of philanthropy, and there are few incentives for philanthropists to back the best schools, as evidence from California shows.
In essence, in order to produce a well-functioning education market in the UK, we need a supply-side dynamic that gives parents and pupils more than just a theoretical right to choose schools. Such a supply-side dynamic requires for-profit actors.
But doesn’t the profit motive drive down quality? The short answer is no. On the whole, research from America, Chile and Sweden displays either positive or null effects vis-à-vis public and non-profit schools. One study from Sweden displays a small negative effect in upper-secondary education but as I’ve argued here, the results are likely to be driven by other factors than the profit motive.
I would also like to emphasise that null effects are important since we are more interested in the overall systemic-level effects from competition than in specific ‘school effects’. While evidence on competition’s effects from various reforms around the world is mixed – which could be due to the fact that competitive incentives have rarely increased more than marginally in most cases, and also that it takes time before competition matures – cross-national research analysing PISA scores and focusing on long-run, systemic effects, is not: the best estimation strategies find relatively large positive effects of specifically private-school competition on PISA scores, for pupils in both private and public schools, and also that it drives down costs. In other words, private school competition gives us more bang for our buck.
However, I also think it is important to discuss the evidence from Chile, since there is no mention of it in the Policy Exchange report. Chile is the only country apart from Sweden that allows for-profit schools to participate in a universal voucher scheme. And just as in Sweden, many have blamed these schools for producing more segregation rather than increasing quality. Is this true? No. The most recent research demonstrates that for-profit schools that do not charge top-up fees enrol more disadvantaged pupils than non-profit and public schools. What about performance? Here the evidence is a bit more complicated: for-profit chain schools perform better than public and Protestant schools, while there are no systematic differences between for-profit chain schools, Catholic schools and non-sectarian, non-profit schools. Among pupils in eighth grade, however, for-profit chain schools also outcompete non-sectarian, non-profit schools. Small, independent for-profit schools, on the other hand, perform on a par with public and Protestant ones, but lower than the other types of for-profit and non-profit schools. Yet, the evidence is still clear: nothing indicates that the profit motive per se is dangerous. If anything, the opposite is true. While there have been, and continue to be, many flaws in the Chilean school choice model, the profit motive is not one of them.
What about the Social Enterprise model? I argue that while it is a step in the right direction, it is not the best policy option available. A duty to reinvest 50% of surplus would make it more difficult to attract investors to support new schools, or expansions of existing ones, since it means much higher risk taking. The corollary is that the sought-after supply-side dynamic might not materialise or at the very least be significantly diluted. This is not a risk we should take.
Secondly, I understand that the social enterprise model is supposed to be a political solution, but I wonder whether critics of profits would be satisfied since they do not claim that it is okay if 50% of profits go to what they argue are exploitative capitalists. Instead, they argue that all money should be reinvested. I therefore question whether the proposal makes for-profit schools more politically palatable.
But aren’t there dangers in producing a more competitive framework in education, which for-profit schools are supposed to do? Yes, there certainly are. Evidence from reforms all over the world indicates that while it is relatively easy to make schools compete, it is more difficult to make them compete by increasing quality. An example is Sweden, which combines an extremely decentralised grading system – where individual teachers set the grades – with a heavily centralised admissions system that depends almost entirely on those grades. This has produced perverse incentives among rational parents and pupils to seek out the schools and teachers who give the highest grades for the least effort. A more relevant example for the UK is the danger of undifferentiated funding. Since the real cost of educating a pupil is a function of his/her background and ability, funding should be systematically differentiated based on pupil background and prior ability. If it is not, there is a danger that schools compete by attracting richer and more high-performing pupils rather than by raising quality. While the pupil premium is a step forward in this respect, it is not sufficient.
The upshot is that if schools are given opportunities to compete by other means than quality, they are likely to do so. When we allow commercial interests to enter the publicly funded education sector, it is therefore crucial that we also promote a healthy framework that harnesses incentives and channels them towards improving quality.
In conclusion, I agree with Eric Hanushek, one of the world’s foremost education economists, that the key problem in today’s education systems is that there are no incentives to improve pupil performance. For-profit schools are not sufficient to improve the incentive structure. Many other complementary reforms are necessary, and I discuss these in detail in an upcoming paper for the Institute of Directors.
Instead, I argue that for-profit schools are but one important element in a coherent reform package designed to transform the current education system into an education market. I’m certainly aware of the political difficulties involved in allowing businesses to participate in a publicly funded education sector. However, if there is anything the past 30 years of school choice research has shown, it is that half-hearted reforms are unlikely to generate more than meagre gains.