Syndicate content

Rights and Welfare Economics

Shanta Devarajan's picture

ML028S19 World Bank Some 135 countries have constitutional provisions for free and nondiscriminatory education for all. Seventy-three countries guarantee the right to medical services. And 41 countries have either enshrined the right to water in their constitutions or have framed the right in national legislation.  All of these actions are aimed at protecting the rights of poor people. 

Yet, it is poor people who are losing out on access to these services.  In Mali, whereas almost everyone has access to a primary school, and 67 percent from the richest quintile complete primary school, only 23 percent from the poorest quintile do.  The percentage completing higher levels of education is in the single digits. In rural India, in the period since the Right to Education act was passed, student learning outcomes in public schools have been declining.  Equatorial Guinea, with a per-capita income of $20,000, has a child mortality rate of 118 per 1,000 births, comparable to that of Togo with a much lower per-capita income.  As a result of intermittent (or nonexistent) water supply through networks, poor people in South Asia and Africa have to buy water from vendors at 5-16 times the meter rate.

What’s going on?  Too often, the rights to education, health and water have been interpreted as the government’s financing and providing these services.  But governments have limited resources.  With limited resources, welfare economics tells us that governments should spend on: (i) those things that the private sector will not finance—public goods and externalities; and (ii) redistribution.  By neglecting these principles and spending on private goods, governments have shortchanged poor people—the very people these rights were intended to protect.
For example, the lion’s share of public spending on health goes towards hospitals—largely a private good.  In the absence of insurance markets, rich and poor people need hospitals for catastrophic care—to treat heart attacks, strokes, etc.  Since the rich have greater political power, they make sure these hospitals are located near where they live—in urban areas.  The result is that, in India, about 33 percent of public health spending accrues to the richest quintile, while less than 10 percent goes to the poorest quintile (Figure).

Similarly, governments provide quality higher education for free or at a low price on grounds that poor people should not be denied a university education because of cost. Given excess demand, governments ration entry to these universities by an entrance exam.  But free high-quality education is a huge benefit to anybody who can get it.  So the rich send their children to the best secondary schools in order that they have the best chance of passing the exam and entering university.  The result is that most of the students at public universities come from the richer quintiles.  And higher education is expensive.  In Africa, about 43 percent of public education resources go to the richest 10 percent of the population. 

Governments subsidize water (another private good) in order to enable poor people to get access to it.  But a subsidy means that politicians control the water utility that receives it.  They send the water to neighborhoods where their political clients live.  Those left out, frequently the poor and marginalized, have to buy water from vendors at much higher prices.

In light of all this evidence, it is tempting to suggest that governments stop financing private goods, like hospitals, higher education and water, and refocus public spending on public goods and externalities, such as sanitation.  Unfortunately, this is almost impossible politically.  The non-poor, who have captured the rents from government financing of private goods, will resist every attempt at having them cut—witness the protests against raising university tuition fees or water tariffs in most countries.  Furthermore, since politicians can exercise control over the allocation of these rents, they too will resist. 

What can be done?  One approach would be to invoke another right, the right to information.  Poor people can be a powerful political force for change—if they are informed about how much they are losing by government’s continuing to spend on private goods that are captured by the non-poor.  The experience with information campaigns on public expenditure leakages and community-based monitoring is promising.  But we need to better understand how information influences citizens’ behavior before we can scale up these efforts.

A second approach harks back to the second role of government, namely, redistribution.  Public spending in the form of subsidies or free services is often justified on this objective, although as we have seen, the redistribution is frequently in the wrong direction.  To achieve real redistribution, therefore, the most efficient method is to give cash to poor people.  Rather than subsidies that risk being captured by the non-poor, giving cash to poor people to purchase private goods in the market has a better chance of public resources’ going to those for whom they were intended.  To be sure, this approach will also be difficult politically, but it could serve as a useful benchmark against which alternative proposals can be measured.


Submitted by Imran on

I agree that Governments are highly constrained in terms of the resources they have at their disposal, and therefore have to make difficult decisions about exactly what to finance and what to leave to others to finance. However, I think the way in which you categorise - bluntly and without much explanation - hospitals, higher education, and water as 'private goods' is incorrect, especially when many of the economically and socially successful countries of the world use public finds to finance these goods. Your argument is more about where and how Governments should provide these goods, which makes sense, but to just dismiss them as private goods is unrealistic of the social realities faced in the world, and frankly, lazy 'Economics'.

Submitted by Shanta on

Imran: Thanks for your comment. By "private goods," I was referring to goods and services where there was no clear-cut market failure (public goods or externalities) associated with them. Sanitation, for example, does have an externality (people without latrines still benefit from their neighbor's having a latrine, as I described in my earlier post). By contrast, with water, there is no externality (I don't benefit from your drinking a glass of water), nor is it a public good in the sense of being non-rival and non-excludable (when you consume water, I can't consume that same glass of water). There is therefore a clear rationale for government's subsidizing of sanitation. With water, the only rationale is that of equity--we would like poor people to be able to get access to water. But, as I showed, subsidizing water does not guarantee that poor people will be able to get it at the subsidized rate because the non-poor, who can also get the subsidized water, are able to capture it through their political clout. That is why it's better to give poor people cash, and let them buy water at market prices, the same prices that the non-poor pay. By referring to water as a private good, I was not "dismissing it." Rather, I was attempting to identify the precise rationale for public financing of the good and service, something that is critical when governments have limited resources. Shanta

Submitted by Imran on

Thanks for your explanation Shanta - that is helpful. Though interesting that Governments of the world are, by implication of your argument, wasting enormous resources by looking into how to provide hospitals and tertiary education if, as you suggest, there are no clear-cut market failures. And with water, who owns water? Should a company have the right to own water which ordinary citizens use? Or does water belong to the community/society/country where it falls? I totally agree in the need for a market-based system to ensure efficient distribution and allocation of water, but to make profits from this system can be perverse, especially in cultures where water is treated as a particularly special good. (No need to reply - and I'm largely just playing devil's advocate here). Thanks.

Submitted by Shanta on

Imran: I know you said no need to reply, but I want to clarify two points. First, the reason governments are wasting resources on hospitals and tertiary education is that the people who benefit from them--the non-poor--are politically powerful, and are able to resist attempts at reform. Second, on water, it is precisely because water belongs to society that we should charge for it. If we don't charge for it, there will be no water left (as we are finding out in many countries). What you do with the revenues from water is a different issue, on which I don't think welfare economics has a particular view. The revenues could go to the government, or to a private company (or both). This is not about privatization. It is about charging marginal costs for water, whether it is a public utility or private. Shanta

Submitted by Rob Yates on

Looking at specifically at the issue of purchasing health services: can you provide any evidence that "giving cash to poor people to purchase goods in the market has a better chance of public resources' going to those for whom they were intended"?

Sure poor people may temporarily end-up with additional cash but are you suggesting that individual private purchasing of health services is more efficient and equitable than public purchasing of services? I thought that since Arrow highlighted the massive market failures in health (due to huge information asymmetries) that the consensus was that the state should be heavily involved in purchasing healthcare services. This certainly was the view of the recent Lancet Commission investing in health:

Submitted by Shanta on

Rob: From the graph in the post you can see that, if the Indian government took the amount it currently spends on hospitals and gave it out equally to all citizens (i.e., without any targeting), it would be a more equitable distribution than the current one. That was what I meant by "giving cash...has a better chance of public resources' going to whom they were intended." On your second point about information asymmetry, Arrow's original paper and the literature since then has shown that government should regulate health, and more specifically, health insurance markets. It doesn't necessarily follow that government should finance all of health care. There may be a case for government to be the single purchaser on cost-efficiency grounds, but even here, it doesn't follow that government should turn around and provide the health services for free to everybody, including the rich. Shanta

Submitted by Rob Yates on

I completely take you point on service provision Shanta. However I would suggest that all over the world countries have learnt that it is more efficient and equitable for the government to be heavily involved in all 3 of the main financing functions - resource generation, pooling AND purchasing. Only a tiny number of rich countries try and regulate a market of private health insurances and the most famous of these, the US, is notoriously inefficient and inequitable. It's also clear that the trend in the US is towards more of a socialized health financing system.

Submitted by Anders Zeijlon on

Shanta gives a friendly but critical nudge to the rights-approach to service provision implying that including such rights in the law or constitution is pretty useless – and perhaps even harmful as it may be interpreted as committing the state to both producing the service and providing it free of charge. However, he will be pleased to learn that the human rights covenants and associated bodies fully agree that putting the service provision in law in only a first step. In fact, state parties to the relevant covenant (here the Covenant on Economic, Social and Cultural Rights) undertake to take steps, individually and through international assistance and co-operation, especially economic and technical, to the maximum of its available resources, with a view to achieving progressively the full realization of the rights recognized in the covenant by all appropriate means (public provision not a must) including particularly the adoption of legislative measures. States also undertake to guarantee that the rights will be exercised without discrimination of any kind as to race, color, sex, language, religion, political or other opinion, national or social origin, property, birth or other status. How can you argue with that?

Submitted by Shanta on

Anders: Thanks for weighing in. I can't argue with what is said in the Covenant, but I might argue with what is left unsaid. When state parties undertake to achieve the full realization of rights, do they ever take into account government budget constraints? If they did, they would then have to state explicitly that government financing should be for those services where there is either a market failure (public goods and externalities) or redistribution motive. By not making this point explicit, the rights approach has inadvertently allowed public resources to be captured by the non-poor, precisely the opposite outcome from what the approach was trying to achieve. Shanta

Submitted by Anders Zeijlon on

Well argued, as always. But I would say that the Covenant is better off not getting into HOW the countries should realize the respective rights, not least because it should be appropriate for all countries, rich, poor, western, socialist, etc. Just think about all the OECD countries with public doctors, hospitals and pharmacies who would be in breach with the Covenant if it said that public production and provision is only allowed when there are externalities or public goods! It is enough that it binds the counties to the goal of realization and to being serious about it. Then, cleaver economists and others can argue about implementation. Another, also interesting, problem is the extent to which the legal system should play a role in health provision. Many countries have tons of lawsuits where hopeful patients seek a court decision to get treatment, and Varun and others are researching if the court decisions get enforced when thry grant treatment. In Scandinavia the courts don't touch cases like that.

Submitted by Shanta on

Anders: I take your point about the covenant not mandating how rights should be enforced. But to make sure these rights are in fact realized, we need to have a dialogue between the welfare economics community and the human rights community which, I guess we are having now!

Submitted by Anders Zeijlon on

I think human rights and welfare economics live quite weel together, and typically pull in the same direction - especially after the human rights folks adopted the concept of progressive realization of rights (a state is not violating the right as long as there are serious efforts in place to realize it). But one difficult area is that of compensation and distribution. When I took welfare economics we were taught that moving from situation A to B is preferred and justified if the winners' befefits are greater than the losers' costs, i.e. if there is a net gain and the winners could compensate the losters and still be better off than before. Human rights practiciners will say that this is not enough but that compensation must actually take place for A to be preferred to B. On another note, I fully agree with your point about the importance of right to information - the by now well known example of the school budget that was made puplic in a village in Uganda and led to public pressure for funds to be used as intended is a good illustration.

Submitted by Shanta on

Anders, this is turning out to be an interesting discussion. First, in monitoring whether the state is making serious efforts to realize rights (the progressive realization of rights), do they monitor whether there is progress in the poor getting access to these rights? If they did, they would realize that countries like India, Tanzania, and others are retrogressing. Second, the welfare economics concept you refer to, whether the change leads to a potential Pareto improvement, is precisely the logic behind the statement that government should concentrate on solving market failures. For when government finances a public good, society is better off even when that good is financed by taxation (assuming the marginal cost of taxation is not too high), or by cutting back on spending on private goods. The losers in the latter case are those who have been enjoying these private goods for free or at subsidized rates. Typically, they are the non-poor. So if human rights practitioners wish to compensate the losers in a Pareto improving change, they should be aware that they are compensating who were not the ones that the rights were designed to protect. Finally, I'm glad you agree on the right to information, but has it ever been articulated that this right is more important than the others in that it can help the others achieve their intended outcome? Shanta

Submitted by Sri-Ram Aiyer on

Excellent piece Shanta, and so true about public goods. The point on redistribution has been shown empirically to be correct, judging by the Bolsa Familia program in Brazil initiated under President Lula, and the conditional cash transfer (CCT) programs in Colombia (although first attempted in Mexico's poorest states, during the Salinas presidency). But these too need a good degree of enforcement of basic conditions, e.g., school attendance, visits to primary health clinics for vaccination etc.

The right to information is not common even in countries with ostensibly reasonably open news media, most often since the sources of info are often government agencies.

The underlying problem seems to be efficiency of public expenditures, whereby spending on subsidies such as for food, fuel and fertilizer is, more often than not, abused by intermediaries who pocket most of the funds--witness India. The other is the political pressures in all types of regimes to cater to particular vested interests--of many stripes, religious, political, tribal etc. The media may again offer a solution as more information becomes available from a wide range of sources.

Submitted by Andrea Planter-Otter, MD, MPH on

Thank you for the clear distinctions among rationales for government spending (and taxation). You say that governments adopt goals like "free health care" and "free water", which are mainly private goods, for essentially populist reasons and then too often the benefits are captured mainly by the rich. Am I correct in my understanding?

If so, what role should World Bank and international technocrats play? In particular, just last month the World Bank has adopted universal health coverage as its one and only goal for all its aid programmes in all developing countries. Would better health of the poor be a more appropriate goal, as I believe? For example through sanitation, reducing pollution and other externalities? Or do you think it makes no difference what the World Bank-endorsed goal for health of the poor is?

If it has influence and funds government spending (taxpayers have to repay the loans), then it should perhaps avoid exacerbating the perverse redistributive impacts you have highlighted.

Submitted by Andrea Planter-Otter, MD, MPH on

Essentially, government is an insurance fund with an army.

This paraphrases Paul Krugman, refering to the US federal budget. It's less valid in countries where, unlike in the US, public goods like primary education, public transport, disease control and prevention, sanitation and police are provided mainly through the central government budget, but it still aligns with your public goods and redistribution categories. Redistribution is often like insurance that provides compensation for and recovery from adverse shocks, some lasting a long time.

Submitted by Rajan on

This issue of capture of public goods is sadly too common. It was interesting to hear about low cost private schooling offered by Bridge International academies in Kenya and how the poor see this as the path forward for their children rather than poorly performing government schools. Cash transfers to the poor (school vouchers) would increase the competition, and one hopes make government schools better performing. Link to the NPR story is here

A functioning public information and disclosure would no doubt lead to increased transparency, especially where scarce public resources like public housing are purported to be given to poor etc. In the past, whenever a government agency made such allocations, a good number always went to the well connected.

Submitted by Gautam on

Dear Shanta,

Two things: one, I'm not sure about the private goods logic, especially when it comes to education. Lots of scholars argue that because the social benefits of a good healthcare system and a good educational system reach beyond the immediate benefits to the transacting parties, the market will always underproduce (this argument is very prominent in the fair use/education literature at the moment).

Two, I'm unconvinced by the empirical claim that there is no evidence of market failure in cases such as healthcare and education. I think that orphan drugs are an excellent example of market failure in healthcare, and the Marrakesh Treaty was ratified precisely because there was market failure in providing educational materials to visually impaired persons.

Submitted by Shanta on

Gautam: Thanks for the perceptive comments. On education, I know lots of scholars argue that there are externalities, but the empirical evidence is rather thin. And even if there were some externalities (especially in primary education, for which there is some evidence), it is still the case that there is a huge private benefit to education as well. So the government should subsidize education only to the extent of the difference between the social and private benefit, not the whole thing.
On orphan drugs, you are right that there may be a market failure, but this is in the research and production of the drugs. There is then a case for government subsidizing the research, as I believe they do. It doesn't follow that government should also subsidize the provision of these drugs to rich and poor alike. Shanta

Submitted by keshab on

Thanks for very thoughtful analysis. Nepal is most appropriate place to prove your argument where a small section of urban elites are pushing the agenda for subsidized education and subsidized water and oil and rest of rural poor are dying.

Add new comment