"Inequality at birth is neither just nor unjust. What's just and unjust is the way institutions deal with it" - John Rawls
"I always had a certain dislike for general principles and abstract prescriptions. I think it's necessary to have an "empirical lantern" or a "visit with the patient" before being able to understand what is wrong with him. It is crucial to understand the peculiarity, the specificity, and also the unusual aspects of the case" - Albert O. Hirschman
Teacher voucher policy as a compromise between government run schools and student vouchers
Learning outcomes and cost effectiveness of budget private schools (BPS) don’t make a strong case to shift focus from government schools to BPS.
However, if the current situation of neglect of government schools continue, two another parameters — financial stress on families and plausible improvement in current state of BPS due to additional funds from vouchers — makes a case to shift focus from government schools to BPS.
Since student voucher system may not be feasible in short-term due to financial constraints, a voucher policy for BPS teachers to help them receive better training is a good compromise policy.
There are two common perceptions regarding budget private schools (BPS)
BPS perform better than government schools, even after accounting for differences in student characteristics.
The learning per unit cost (cost effectiveness) is higher for government schools as compared to BPS.
These two arguments are cited as basis to move towards a voucher system where students get vouchers from government and they are free to choose the school they like, over the present system of government run schools where students’ choice is restricted. For instance, this Mint editorial makes the above two arguments to advocate for school voucher system.
Though these two arguments seem intuitively correct, careful research suggests that they aren’t necessarily true.
On government-private comparison
The simple comparison of scores of BPS and government schools show that BPS performs better than government schools. The common objection to this is that students of different background attend government schools (GS)and the lower scores are just a reflection of that. There are also other studies that say that BPS performs better than GS even after controlling for student characteristics. Hence, there is a common perception that BPS perform better than GS.
However, the perception that BPS is better than GS is NOT necessarily true. There are limitations to such studies saying BPS is better than GS because you can’t account for all student characteristics and self selection unless it’s an experimental study.
Hence an experimental controlled study was done to address the limitations of such studies. The controlled study found a negligible difference in outcomes between government and private unaided schools.
For the argument sake, let’s assume that all the difference of scores between government schools and budget private schools (BPS) remains, even after adjusting for student characteristics.
Even then, it doesn’t make a strong case for BPS, because we are talking of a difference of mere 1.7% (18.6% -government and 20.3%-private). The difference may be significant statistically but not from an education point of view, in a broader picture.
When the learning levels are at an abysmal low of <20%, quibbling over 1.7% to argue that one is definitely better than other and advocating for far reaching reform based on this is inappropriate on many levels.
Another study corroborates the negligible difference between government and BPS. The following graph shows the distribution of scores of children from across 6 states, for three categories — government schools, budget private schools, and high free private schools — with 15,000 students for each category.
Source: Report on Budget Private Schools, CCS Delhi (page 54)
As one can see, there is a major or complete overlap between the distribution of government and budget private schools (red and green), demonstrating equivalent quality in both the schools. Particularly, note the distribution of mathematics scores of class 7 (bottom right). It’s almost a complete overlap between government and BPS, again demonstrating that BPS are no better than government schools.
In summary, the difference in scores between government and BPS is not of real world significance, the difference that can especially make a strong case to see BPS in its current state as an alternative to government schools.
The second common argument in support of BPS is that learning per unit cost is higher for private unaided schools.
But, cost effectiveness of learning is an inappropriate to use when the learning levels are abysmal.
Following the cost-effectiveness logic, all parents should send their kids to government schools because the cost (directly paid from your pocket) is zero. But they don’t! It’s because, if absolute levels are abysmal, it doesn’t matter if it is free of cost. It hides more than what it reveals. Same is the case, when we apply the cost effectiveness reasoning to government expenditure.
When the learning levels are abysmal, cost effectiveness metric doesn’t make a strong case to shift focus from government schools to BPS in its current state.
As discussed above, the two arguments in support of BPS — higher learning outcomes and cost effectiveness — aren’t on strong grounds. Hence, arguments that cite these two factors as basis to shift to a voucher based system are also not on strong grounds.
The arguments of higher learning levels in BPS and cost effectiveness don’t make a strong case for government to shift its policy from running government schools to merely funding education through vouchers and leaving the operations to the BPS in its current state.
Normally, if it was a few years back, I would have ended the article here, arguing for spending efforts on government schools, while non interfering with private schools. But, the inaction of governments in the face of giant learning crisis is making it difficult to defend the public school system as being the primary anchor to deliver education.
The inaction of governments is increasingly making a strong case to shift from government schools if we use two other parameters in our decision of government vs. private schools — “financial stress on the families” due to education expenditure and “possibility of improvement of BPS” with additional funds to vouchers.
With increasing shift of students to private schools, school fee is placing a huge financial stress on families. With massive shift of students from government to private schools, reaching 50% share in some cities, it is a huge burden on a signifiant number of people. The delay in reforming government schools means a delay in reverse shift from private to government, which means a prolonged burden on families.
If we provide education vouchers — in the worst case, even if the learning levels don’t improve, it will at least reduce the economic stress on families. In the best case, private schools may actually out perform the government schools with the additional money they get from voucher students!
The latter point regarding plausible improvement of BPS due to additional funds through vouchers is important. Let’s look at the scores’ distribution graph once again.
Source: Report on Budget Private Schools, CCS Delhi (page 54)
The blue line represents distribution of high fee private schools (HFP). If BPS receives funds through vouchers, it attains the spending capability similar to that of the HFP schools shown in the graph. The distribution of HFP (blue line) may not be totally due to difference in financial capability, it is also a reflection of the kind of students who attend those schools. But, if students (of profiles) who attend government schools and BPS are placed in these HFP, they might perform better than as they would other wise have if they were in government schools and BPS.
The controlled study that we talked of earlier only considered a comparison of only government and BPS. It didn’t consider the case of HFP. But, there is a reason that students of disadvantaged backgrounds might do better in HFP setting, as compared to government school or BPS merely because of presence of teacher, better teachers and so on. As of now, there’s anecdotal evidence of such effect in case of 25% students who are admitted to HFP, who otherwise would have attended a BPS or government school. It might actually happen on large scale through education vouchers.
There are three common objections cited against such voucher policy. One objection is a variant of arguments of “education is provided by governments world wide” and “education can’t be left to private sector”. The other objection is that we should wait for government to reform. The final objection is that private might not reach all places.
On the point of governments providing education, it is becoming a matter of ideology. There’s no hard rule that only governments can and should provide education. Also, someone did it doesn’t mean everyone else should do it.
In reality, the decision of government vs. private actually is about whether the government has better capacity to provide education than the private, and in case if it is not, is there a hope for revival in near future. If the government’s capacity is weak and there is no hope of its revival in near future, private might actually be a better system. By restricting such transition from government to private might actually end up being harmful.
For instance, in a war torn country, with an extremely weak capacity of government to provide public education, an alternative of PPP or voucher system might be better in relative terms, considering the time it is required for public systems there to be up and running. We can’t wait till the public systems are set up here on the ideological basis of “education should only be provided by the government”.
The long history of negligence and no signs of any plausible attempts for serious comprehensive reform is putting India on similar ground as the war torn countries with weak capacity and no hopes for revival in near future. Hence, not shifting from government school system to voucher system may actually be harmful, if things are to continue as they are today.
Regarding the next objection that we should in fact wait for governments to get its act together OR that governments in India are trying and that we need to be patient — to such arguments, I respectfully submit that any government at any given point of time can give a list of schemes that it’s running, claiming that it’s “working”. A mere repetition of schemes doesn’t mean anything. All that the governments (except the Delhi government) are doing currently is “try to appear to be working, and not actually working with a serious intent to reform, at the scale that it’s required to make a substantial impact”. Even if we put the best of our efforts, it will still take 5 years to bring the system out of coma and another 5 years to make it work. At our current pace, there is no hope!
Regarding the argument of “private not reaching all places” — at least let it reach places where it can! Why stop something to work at one place because it doesn’t work everywhere else?
Learning outcomes and cost effectiveness of BPS don’t make a strong case to shift focus from government schools to BPS.
If the current situation of neglect of government schools continue, two another parameters — financial stress on families and plausible improvement in current state of BPS due to additional funds from vouchers — makes a case to shift focus from government schools to BPS.
Either reform government schools on an urgent basis (not piece meal band-aid reforms) or give vouchers to children. Even in worst case scenario, it will reduce financial stress on families and in the best case might end up being helpful.
The Middle-Path — teacher training vouchers to BPS teachers
School vouchers may seem better policy but there are practical constraints in initiating it. Around 80% of total expenditure in education goes towards teacher salaries. This is a non-flexible part of the education budget because it isn’t an easy task to fire all the existing government teachers. It is thus not possible to re-allocate all the money spent on government schools to school vouchers. Also, some argue that existence of a public system is necessary to ensure a good default, so that private players remain competitive.
In this context, a good compromise policy would be one that addresses the constraints of BPS, while not requiring steps like firing government teachers or permanently close schools removing default option.
Given that the attendance and presence in class and other governance aspects are ensured in BPS, the binding constraint is then the capacity of teacher. Usually, teachers in BPS are untrained and are under-paid. Given the thin profit margins, schools also don’t invest on their teachers’ training. Further, there is also a fear that teachers might shift to better schools, after availing training.
A “teacher-voucher” policy would address both these issue of teacher capacity and weak incentives for school administration. As per this policy, anyone can get a “teacher-voucher” which can be availed in a teacher training institute that provides “feedback sessions on live classes” or “equips them with theoretical tools”. Schools can encourage their teachers to take up such training programmes. Since it is provided by government, school management won’t lose even if teachers end up shifting schools.
This might not relieve the financial stress on parents sending their kids to BPS but it will help such students by providing better quality education to them.
In totality, a well-designed “teacher training voucher” policy, would help students of BPS and address the teacher capacity issue in BPS. It is a good compromise between the current state of “namesake focus on government schools” and the alternative of “student vouchers”.