It is good at providing access where the state does not have capacityApr 11th 2019Print edition | Special report
“WE WANTED TO pop champagne when they said they were opening Spark Soweto,” said Ntebogeleng Malevu. Before the new branch of a fast-growing chain of low-cost South African schools opened in the township on the outskirts of Johannesburg in January, Ms Malevu, a nurse, would wake her six-year-old daughter Qhawe at 4am to travel to another Spark school in the city’s northern suburbs. The transport cost nearly as much as the tuition (2,310 rand, or $158, a month for primary-school pupils).
Parents prefer private schools. A global survey in 2017 found that they were a lot more likely to give the teaching at their children’s school a positive rating if it was private than if it was public; parents in Chile have voted with their children’s feet in favour of the private sector.
Governments are often less keen on private education. Some of the reasons for their hostility are bad ones: a reluctance to cede power, the opportunity for patronage, the influence of teachers’ unions. But some are entirely in order: governments need to promote quality, access to schooling and equity. The private sector is good at some, but not all, of those.
The evidence on quality is ambiguous. Private institutions dominate the upper ranks of the global higher-education leagues. Seven of the top ten places in the Times Higher Education ranking are taken by American private non-profit universities and three by British institutions, which although regarded as public in Britain, are privately run and funded largely by user fees. The most highly rated clearly public institution, ETH Zurich, which is, ultimately, run by the state, is in 11th place.
But these rankings depend almost entirely on the universities’ research performance. The standard of education they deliver is hard to measure. The only useful proxy is earnings, but a study of alumni of America’s most selective colleges by Stacy Dale and Alan Krueger found that their higher earnings were explained by background and intellect. Top universities provided a boost only to blacks and Hispanics—presumably because they gained a useful network to which many white students already had access.
At the bottom of the market, America’s for-profit colleges—largely vocational outfits that take students who cannot get into the state system—do poorly. A study of their alumni’s employment history showed not just that they performed worse in the labour market than similar people who went to (much cheaper) public colleges, but also that they barely earned more than those who did not go to college at all. On average, in other words, the time and money that they spent on their education had been largely wasted.
In some countries private schools do exceptionally well. According to Varsity, Cambridge University’s student newspaper, Westminster got an average of 79 pupils a year into Oxford and Cambridge in 2006-16, more than any other school in the world. But Westminster is one of Britain’s most selective schools, attracts bright pupils from all over the world and spends four and a half times as much per child as the public sector does. Educating the world’s cleverest children with vast resources is not the biggest challenge in teaching.
A better test of schools is whether they add value—in other words, produce outcomes better than would be expected given where children started. In the OECD’s latest PISA test private-school pupils did a lot better than public-school ones in reading and science, but after controlling for economic background they did little better in reading and worse in science. An American study concluded that private schools added no value, a British one the opposite: the university and labour-market outcomes of two cohorts of people, born in 1958 and 1970, who attended private schools were considerably better than those of government-school alumni, even when ability and background were taken into account. The gap was bigger in the younger cohort, presumably because private schools have come to focus more on academic achievement.
In poor countries, the evidence tends to favour the private sector. Out of 21 studies in Africa and South Asia surveyed for Britain’s Department for International Development (DFID), 14 found that children learned more at private schools and seven found no difference. In none of the studies did government schools come out on top, but the private-schools’ margin was not overwhelming. In the most rigorous study, carried out in the Indian state of Andhra Pradesh, pupils’ scores in most subjects were the same in both types of schools, though they did better in Hindi at private schools. The maths scores of the private schools that taught in the local language, Telugu, were higher than those taught in English, suggesting that while private schools confer an advantage, being taught in English is a disadvantage.
Opponents of private schools often argue that they undermine public schools, but the evidence does not support that claim. A review of studies from America, Canada and Sweden concludes that virtually all of them showed that public schools do better when they are up against voucher schools; the few studies of the issue in the DFID review, from India, Pakistan and Kenya, found the same.
The qualitative differences between private and public schools are marginal, though. More strikingly, private schools cost less. Of seven studies in the DFID review, none found government schools to be cheaper. A study comparing the cost-effectiveness of public and private schools in eight Indian states found that the private sector did better in all of them; the differential ranged from 1.5 times in Bihar to 29 times in Uttar Pradesh.
The private sector’s efficiency is one reason why it does well at providing access to education. Another is its speed: in fast-growing cities, governments struggle to provide schools, but wherever there are people, schools spring up. A worldwide review of voucher schemes has shown that governments that cannot provide enough capacity can increase access by enabling children to attend private schools. “Kids are being born every day,” says Murad Raas, education minister in Pakistan’s Punjab province, where 2.6m children are in private schools on voucher schemes and 11m in government schools. “We don’t have the funds to accommodate them all. I’m very open to anything that can benefit them.”
The main reason for the private sector’s superior efficiency seems to be teachers, who are paid less and are more likely to turn up for work than at government schools. Politics has a lot to do with that. Teachers’ unions have huge bargaining power—in India, for instance, they man polling stations and have reserved seats in state assemblies—and can therefore protect their members from being held to account for poor performance. An Indian study found that in 3,000 government schools, only one principal had ever dismissed a teacher; among 600 private schools, 35 had.
The other explanation for better performance in the private sector could be competition, but it does not seem to be working all that well. One problem with the market is that parents often lack information. Researchers in Pakistan tried providing parents with cards showing their child’s test scores and the average scores of the schools in the village. Where this was done, children learned substantially more, fees were lower, enrolment went up and bad private schools were more likely to close down.
A second problem is parental priorities. Parents want their children to go to schools with the best outcomes, but those establishments may achieve good results not because they add more value but because their intake is richer and cleverer. A Chilean study showed that parents presented with data on both outcomes and value added were interested in the former, not the latter.
They are being perfectly rational. Parents want to ensure not just that their children get a good education, but also that their classmates come from “good” families, because the company they keep will shape their behaviour and provide their network. Employers, too, are likely to favour the products of schools with good exam results. But the incentive for parents to choose outcomes over value added limits the efficacy of parental choice as a mechanism for improving schools.
It also helps explain why the market tends to increase inequality. Whereas governments promote social integration, parents actively seek stratification. Furong Ren, a parent at Dehong, Dulwich College International’s bilingual sister school in Shanghai, explains that “when parents get together, all they talk about is how China is developing a class system, and they want their children to be on top.” Fees and selective admissions, which favour rich kids, encourage schools and families to sort themselves by income. For governments concerned about social mobility, that is a problem.
As private education grows, its strengths and weaknesses are becoming increasingly apparent. It is good at providing access where the state does not have capacity; in poor countries, the education it offers is slightly better than the government variety. But it also encourages inequality and discourages social mobility.
Many teachers’ unions and left-wing politicians favour getting rid of private schools. That would solve the equity problem, but there would be a cost in terms of both access and quality. Without the private sector, many children in fast-growing cities in the developing world would be in worse schools or on the streets.
Another approach is to regulate private education, by, for instance, setting stringent standards for facilities and teaching. That is a reasonable thing to do in countries where the state works well, but a state that cannot provide decent education is unlikely to be a good regulator. DFID looked at 19 studies to see whether developing-country governments were any good at regulating schools; 14 concluded that they were not, three that they were, and two were not sure. Bribery is a common problem.
A third approach is for governments to partner with the private sector, through vouchers or subsidies. The idea is to allow society to benefit from the private sector’s virtues while mitigating the inequality it fosters.
Such partnerships are spreading, but their performance so far has been mixed. They suffer from the same problem as regulation does: governments that cannot provide education are unlikely to be good at commissioning it. India’s reservation of 25% of private-school places for poorer children has not been a great success. The government has been slow to pay its bills, the initiative has got bogged down in legal action, ten years after launch only 16% of private schools are taking part, and a study in Karnataka found that most of the families taking up the vouchers had been sending their children to private school anyway. And where schools charge fees or set admissions tests, such partnerships can become vehicles for subsidies to the better-off and encourage stratification, as Chile’s original voucher system did.
Yet well-designed public-private partnerships can work. Two of the world’s best education systems—those of the Netherlands and Hong Kong—are based on them. In both places, schools get public funding, a lot of autonomy and hefty state regulation to raise standards and limit inequity. Chile’s voucher-based education system, despite its flaws, outperforms those of its neighbours. They are especially good for countries whose governments struggle to provide access to education: in Pakistan’s Punjab province 2.6m children go to school thanks to vouchers; a PPP in Uganda enrolled 400,000. Design and monitoring are crucial, says Harry Patrinos of the World Bank: “performance has to be measured, rigorously and often, and schemes adjusted accordingly.”
Above all, governments should stop regarding private education as an enemy. Its growth is the result of people’s deepest urge—to look after their children. Whether through buying expensive houses near the best government schools or by forking out for private-school fees, they will find a way to do that. The private-education boom may be fostering inequality but it is also causing unprecedented amounts of money and energy to be spent on improving humanity’s brains. Governments should encourage that—but spread the benefits as widely as possible.