Charging the poor for services

Tim Harford has an interesting article in this weekend’s Financial Times about private health and education in developing countries:

Imagine that your daily earnings were less than the price of this newspaper. Would you consider buying private education and private healthcare?

Before you make up your mind, here are a few considerations: government healthcare and primary education are free; the private-sector doctors are ignorant quacks and the teachers are poorly qualified; the private schools are cramped and often illegal. It doesn’t sound like a tough decision. Yet millions of very poor people around the world are taking the private-sector option. And, when you look a little closer at the choice, it’s not so hard to see why.

Now there is a dilemma here.

On the one hand, we know that charging even a very small amount massively reduces the take-up and impact of services such as health and education. (This survey by Holla and Kremer summarises the evidence.)  So charges excludes many people from access, and it seems likely that the poorest and most vulnerable will be excluded most of all.

On the other hand, we know that public services in developing countries are often poorly managed and badly delivered. That’s why, as Tim points out in his FT article, many of the very poorest people choose to go private instead.

Apologies if this is anecdotal, but I see this dilemma in practice every day. My partner works for Marie Stopes International, which operates 21 clinics for women (providing contraception and abortion) here in Ethiopia.  They charge their clients for services – a small amount which is just enough to pay for the cost of running the clinics.   The result is that they are very focused on delivering services that will bring their clients into the clinics every day – that is, services that they actually need, at a price they can afford.  My feeling is that, as a result, they are more focused on their customers than most public services in developing countries, and indeed in some developed countries, whether financed by aid or by taxation.

So how can we disentagle ourselves from the horns of this dilemma?  Here are three thoughts:

  • First, we should take seriously Tim’s observation that “a little accountability goes a long way” and think  much harder about how we can make public services more acountable.  You have probably heard about the way more funding reached Ugandan schools as a result of greater transparency (though the details have been disputed (pdf)). The work of my team on aid transparency is a modest contribution to this effort.
     
  • Second, we should not be ideological about whether the public or private sector actually provides services, as long as the government takes steps to ensure that there is universal access. For example, governments (with the support of donors) might issue vouchers to the poorest, enabling them to choose for themselves whether to use public or private services.
     
  • Third, in the long run this problem will be reduced if and when there is equitably shared economic growth which gives people sufficient incomes for these kinds of choices to be more reasonable.

2 thoughts on “Charging the poor for services”

  1. On the other hand, we know that public services in developing countries are often poorly managed and badly delivered. That’s why, as Tim points out in his FT article, many of the very poorest people choose to go private instead.

    Apologies if this is anecdotal, but I see this dilemma in practice every day. My partner works for Marie Stopes International, which operates 21 clinics for women (providing contraception and abortion) here in Ethiopia. They charge their clients for services – a small amount which is just enough to pay for the cost of running the clinics. The result is that they are very focused on delivering services that will bring their clients into the clinics every day – that is, services that they actually need, at a price they can afford.

    Presumably, the same shift towards user-pays, that provided data for the studies in the Kremer review ought to also provide evidence on whether fee-recovery systems lead to more responsive services or not? I.e. was there any improvement in service quality in countries that started to charge for state operated services?

    For what it’s worth, I think that poor government services aren’t the result of the absence of fee-related accountability but rather stem from deeper institutional issues, at work at many different levels of government.

    Or, to put it another way, do you think that if Marie Stopes suddenly stopped charging (income now coming from another source) everyone working at those clinics would suddenly start caring less about their patients?

  2. The research indicating there are sizable differences in demand between zero and non-zero prices almost exclusively assesses demand for merit goods (like bednets and condoms). I’ve never seen any evidence showing this phenomenon for normal goods/ services like malaria treatment. It’s important to distinguish – since the studies cannot be interpreted to shed light on pricing for many health goods and services.

    Most of the studies also look only at demand. They demonstrate that for some goods, like bednets, the small change from a low price to a zero price for such merit goods is likely to lead to a significant increase in demand. In order to know the impact on USE, you need to take into account the impact of the price change on supply also (as Owen discusses in the case of MSI). In many cases, public provision diminishes when you reduce the price, that is, when you abolish user fees. (The providers were being “incentivized” by the payments…). So, you simply don’t know the net (forgive me, not intended as a pun) impact.

    The Center for Global Development published a review recently of experiences applying performance-based incentives for health services and goods. Many of the case studies in the book show that a positive price received is an important stimulus to provider effort in the health sector, even if the amount of the incentive is quite small relative to the total reimbursement package.
    Here is a link to the book
    http://www.cgdev.org/content/publications/detail/1422178

Leave a Reply

Your email address will not be published. Required fields are marked *