DFID is measuring the wrong thing on poverty reduction
I’ve been skimming the Department for International Development’s latest annual report and came across the headline claim that over the past two years DFID has been “enabling 11.9 million people to work their way out of poverty by providing access to financial services”. In at least a dozen countries DFID has been supporting micro-credit schemes for the poor and helping them to access banking and insurance.
But is putting a poor person into debt really a good proxy for lifting them out of poverty? Sure, a loan might help someone to run a profitable business, but not in every case, and according to a growing body of research, not in many cases.
In another life I reviewed a micro-credit programme run by the World Bank in Timor-Leste and was struck by what a lemon it was. In this very poor and war ravaged country, widows accessing loans under the scheme were setting up small kiosks in 54 percent of cases. They had little other choice. This was leading to the following typical situation:
In one quiet, dusty clearing in the village of Meligo, in Bobonaro district, five groups of widows had set up five of these small shops next to each other. Here, the customers most likely to purchase some of their imported, packaged goods were the scabby, salt-resistant bushes littering the clearing.
A World Bank review of the project concluded that in 70 per cent of cases, widows who received microcredit wouldn’t make enough money to pay back the original loan.
Growing evidence on the bigger picture is even more concerning. Take this excoriating stuff from Milford Bateman:
The past 30 years has actually shown microfinance to be part of the problem holding back sustainable poverty reduction in developing countries, and not the solution. Not only is there no solid evidence that microfinance has had a positive impact on the well-being of the poor, since 1990 the micro-finance sector has been increasingly marked out by spectacular levels of Wall-Street-style greed, profiteering, client abuse and market chaos.
And even DFID-supported research from last year concludes that: “no clear evidence yet exists that microfinance programmes have positive impacts”.
Bateman’s diagnosis makes sense: microfinance usually just promotes unproductive processes of local job churn, with little or no improvements in employment, income or productivity. To put it another way, I don’t think many of East Asia’s economic miracles were built on what is little more than IOUs for the poor. Countries like Timor-Leste instead need dynamic, job creating enterprises, decent infrastructure, and workers with the right skills supported by social protection and universal health and education systems.
Of course finance has a role to play in all of that, especially in unlocking productive capacity, but it shouldn’t be the keystone of a development strategy. Unfortunately for DFID it is. Under the leadership of Development Secretary Andrew Mitchell DFID has a flagship commitment of providing “50 million people with the means to help work their way out of poverty” measured by the “number of people with access to financial services as a result of DFID support”.
Wouldn’t it make more sense for DFID to be measuring poverty reduction through the number of good jobs created or improvements in income and productivity? Sure such data can be harder to measure and the results might take longer to come to fruition but as Mitchell himself said in June last year:
Don’t be misled into thinking our focus on results means we’ll avoid doing the harder things just because they’re difficult to measure. It doesn’t and we won’t.
To be fair, DFID is already measuring these things in some of its country programmes, such as in Somalia, so why can’t it do it for all of them? Probably because it would forced DFID to challenge its misconception that financial services are some sort of developmental silver bullet. And it is a misconception that I think few hold. At home it would take a brave fool to think that payday loans represent our economic panacea. So why think the same model will work for the world’s poorest?