Participants of the Global Meeting

In late November 2009, the nine partners in the CGAP-Ford Foundation Graduation Program met in Bangladesh, during a workshop hosted by BRAC Development Institute. The first part of the meeting was kept to pilot implementers only—it proved to be an incredible forum for candid discussions about the challenges of implementing such a demanding model.

There was a particularly lively discussion on targeting. The graduation model is intended for the poorest—food insecure households, those at the bottom of the economic ladder, and often excluded from regular microfinance. Targeting, to ensure both that the poorest are reached and that better off households are excluded, is therefore an extremely important element of the program. And because the program is based on household-level economic activities, only people who are physically or mentally able to manage enterprises can join.

The discussion between the nine pilots showed how much the targeting criteria is dependent on local context; for example, food insecurity seems to be a solid indication of poverty in Ethiopia and Haiti, but much less so in Peru and Yemen. Most pilots try to select households that do not own any productive assets. But this is actually often much easier said than done: in Pakistan some households were excluded at first because they were perceived as owning a cow, whereas it turned out that they were only taking care of the animal for a small fee. Lack of access to productive land might be a reasonable indicator of poverty in South Asia, but definitely not in the highlands of Peru where all villagers have access to communal grounds for cultivation and grazing. 

There is a lot more to effective targeting than  inclusion and exclusion indicators—process is at least as equally important.

Most pilots in the program start with a geographic targeting of poorest villages, usually using national poverty maps, or based on their local knowledge of the area. Once they have selected program sites, they need to find out who are the poorest people in the communities. Pilots tend to combine participatory rural appraisals led by the communities with more traditional household surveys conducted by their staff—where possible, using tools such as the PPI. In addition, CGAP expert Syed Hashemi, has insisted that all households are visited by senior managers to minimize mis-selection at the pilot stage.

Targeting Efficiency – How well can we identify the poor? by IFMR looks at Bandhan’s Ultra Poor Program in West Bengal and the report shows that participatory  rural appraisals are “reasonably good indicators of economic well-being” and could serve as a decent basis for targeting. But among the nine pilots, there was a general consensus the combination of methods had further improved the targeting…while making selection process more time consuming and costly. As they start scaling up, how will programs deal with the trade-off between multiple levels of targeting and the higher cost of doing so?