Icon Our Work

The debate in the World Bank on the Proxy Means Test hots up: or does it?


The World Bank HQ

It has been great to see the response so far to the debate on the proxy means test (PMT); and, as you’ll know from a recent blog by Nick Freeland, the World Bank itself is beginning to question the value and worth of the PMT. However, we don’t yet know the depth of this questioning. It’s only very recently that members of the Social Protection and Labor Group were claiming that the PMT was accurate (despite all the evidence to the contrary) and we see lots of evidence that they’re continuing to promote PMTs around the world.

In response to Nick’s blog, Martin Ravallion – one of the authors of the Bank’s critique of the PMT – left the following comment: ‘If the Bank was the monolithic, sluggish, institution so resistant to criticism and change that you paint in your blog post, why did the Bank support this research and allow the paper to be published in its own working paper series?’

Of course, Martin is quite right. The Bank is not a monolithic structure: it’s highly heterogeneous. Its staff have different academic backgrounds, ideologies, genders, capabilities, etc. And, it’s also an institution with different departments. It’s noticeable that the critique of the PMT came from the Bank’s Development Research Group and not from the Social Protection and Labor Group. Nor is this the first time that the Bank has given conflicting views within a very short period of time: in 2010, Dethier, Pesthier and Ali produced a paper in which they argued that universal pensions would be a great tool for poverty reduction in Latin America. Yet, only a year later, three members of the Social Protection and Labor group – Acosta, Leite and Rigolini – argued exactly the opposite, strongly advocating for poverty targeted pensions (we produced a paper at the time that offered a robust critique of Acosta, Leite and Rigolini’s analysis).

So, the big question on everyone’s lips (yes, we know, not really) is whether the Social Protection and Labor Group is changing its view on the proxy means test.

Nick has sent us the following cartoon – drawn by his wife, Pippa – which, we fear, encapsulates the current debate in the World Bank on the proxy means test. It’d be great to hear from the Social Protection and Labor Group if they believe we’re wrong.




Stephen Kidd is Senior Social Policy Specialist at Development Pathways who has worked as a consultant and advisor on social development and social protection for the past 30 years.

Follow Stephen on Twitter: @JustKIDDing_DP



  • I’m a civil servant who has had the opportunity to implement various targeting mechanisms. I think it is time we heard from the implementer s and policy makers on this issue. Hopeful i will find time together with my colleagues to issue a paper in favor of PMT. So before you give up on PMT ,Wait a minute. PMT is not that bad. Actually the fact is that no targeting mechanism is perfect or bad. The same applies to universalism. It is more of about how it is implemented and designed. Those who also support universal schemes deliberately ignore the fact that providing social protection is sometimes dictated more by political consideration than the technical requirements of a program. They project a universal approach as the alternative to targeting which is not true. Coming to PMT, It is important to note that very few programs use PMT as a single targeting mechanism. If this was the case then i would agree with most of the criticisms against it. But the reality is that PMT is often used as one of the many elements in the selection of beneficiaries overcoming the shortfalls that have been advanced so far in this debate. Most programs use a combination of mechanisms such as self targeting, community based targeting, PMT itself, categorical targeting etc. So the debate so far against PMT is not balanced. So wait a minute before you discredit PMT.
    The value of any approach or process should be determined by the beneficiaries not by how nice it is written on paper. Good ideas that can not be implemented are worthless.


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