Sunday, 3 November 2013

A flawed case for social engineering

Will compelling private schools to ensure that 25 percent of their students must be from the economically weaker groups affect the quality of education in these schools? It most certainly will, private schools have always maintained. In the absence of hard evidence, they haven’t been able to back up their claims. But then, nor have those who rubbished these gloomy prophecies as elitist claptrap been able to irrefutably disprove them.
Well, there’s good news for the latter and, therefore, bad news for the former. A recent study by a University of California San Diego professor, Karthik Muralidharan, junks the theory that the reservation proviso could bring down the grades of general category students in private schools. This is alarmist misinformation by a small fraction of snobby schools in the major metros, Muralidharan scoffed at a talk he gave in the capital last week. He feels the much-maligned Clause 12 of the Right to Education Act – which mandates the reservation – is actually a “rare example” of a policy that improves equity and efficiency. It could also be the biggest school inclusion programme in the world, he believes, given the huge numbers involved in India. 
Muralidharan’s assertion isn’t an ideological one; it is based on evidence from the Andhra Pradesh School Choice project. The project was initiated to find answers to two questions. One, are private schools more or less effective than government schools regardless of the social and economic background of the students? (It is often argued that private schools are able to show better student performance because these students come from relatively better off homes than those who go to government schools). Two, how will the intake of economically weaker students under Clause 12 affect the students who are already in the private schools? 
The four-year project, which started in 2008, was conducted in 180 villages in five districts of Andhra Pradesh by the Azim Premji Foundation, as part of a memorandum of understanding between the state government and the World Bank. In all these villages, students in government schools were offered vouchers which would allow them to study in a private school of their choice in the village. 
The vouchers covered school fees, books and stationery as well as uniforms and shoes but not the cost of transport to private schools that were not in the village. Nor did it compensate for the loss of mid-day meals that are provided in government schools. Then, a lottery first chose 90 villages that were to be treatment or voucher villages. The other 90 villages that did not get selected for vouchers became the control villages. Within the voucher villages, another lottery chose students who had applied for vouchers. A total of 1,980 households out of the 3,097 that applied for the voucher got it. Of these, 1,210 accepted the vouchers and enrolled their children in private schools at the start of the project. Private schools that participated in the programme were not allowed to cherry-pick students; they had to accept the students who won vouchers and chose to study there. The rest of the households who had applied for but did not get the vouchers were the control group, whose children continued in government schools. Independent tests then looked at learning outcomes of students after two years and four years. Since both sets of students were from identical socio-economic backgrounds, any differences in learning outcomes, the thesis went, would be because of the change in schooling. 
Most of the results were predictable, of course. The voucher students who went to private schools did better than the non-voucher students. The study reaches this conclusion in a somewhat roundabout way. On the face of it, there was not much difference in scores on two main subjects – Telugu and maths. So it did appear that private schools were not more effective, and that differences were because of family background. 
But then it was found that private schools had less qualified, less experienced and lower paid teachers than government schools. They, however, had better attendance than their government school counterparts. Private schools also had longer days and years and more teachers. And though they spent lesser time teaching Telugu and maths, they spent more time in teaching English, social studies and Hindi (this was taught as a third language, which government schools did not). What’s more, the similar scores in Telugu and maths were achieved despite private schools spending less time teaching them than government schools did. So clearly, private schools were giving more bang for the buck, since the per-child cost is one-third that of government schools. 
The survey, however, knocks the bottom out of the argument of many private schools that Clause 12 will adversely affect the performance of students already in private schools because of lower-performing scholarship students. The study found no effect at all. It then looked at whether the number of scholarship students was a factor – the more the number of such students, the greater the negative effect. That too drew a blank. 
Going by these findings, Muralidharan – who thinks Clause 12 is not a bad idea – suggests that schools should not be allowed to cherry-pick from students hailing from lower income backgrounds. This, he argues, could happen if compliance with the proviso is seen as something that schools should be doing at their individual level. Instead, he says, it should become a `system-level’ issue, with the provision being implemented in a coordinated way at the city, block or district level. He suggests a system of private schools providing audited enrolment and fee data to the government; low income parents listing their preference for private or government schools; and a lottery system to allocate schools to children from low income families. 
Muralidharan could be treading on thin ice here. The method he suggests could be quite complicated when it comes to actual implementation, given the huge numbers – of students and schools – involved. It also places too much faith in the efficiency of government systems. Efficiency levels vary widely from the centre to the state to local governments, between states, within states and between local bodies as well. 
But the problem with his thesis is that he thinks Clause 12 is the right way to ensure equity in education. That is just not true. Forcing equity like this may work in a few individual cases; it cannot be a national-level government policy. It is not just an issue of whether elite schools want to take children from lower income backgrounds. Parents who pay a bomb to send their children to these schools do so because of their exclusivity. It may sound snobbish but people are entitled to their snootiness. The state cannot force social engineering; it will backfire. In any case, the elite schools will simply bribe their way out of any lottery or other system that is put in place. They won’t find it difficult to do so; the very people who will be implementing or overseeing this system – the politicians and bureaucrats – send their children to these elite schools because they do not want them to mix with the hoi polloi. 
So, instead of looking at how to fix a few elite schools which cater to a very small minority of school going children, why not look at expanding the education market? If the idea is to ensure that all children get access to a basic minimum quality of education, a better way would be to make it easier for private schools to come up, within a broad regulatory framework with reasonable rules. Also, it might be better to design a proper school voucher programme which will allow parents to choose between government and private schools. This could go hand-in-hand with changing the way government schools function so that they compete with private schools for the voucher students. There will be no need for complicated lotteries which will invariably be rigged, given the kind of cronyism that prevails in India today.

No comments: