The many dimensions of poverty

“Sometimes I’m afraid my son will be killed for something as insignificant as a snack.” That’s what one woman told researchers when they asked her what it means to be poor. Sabina Alkire, director of the Oxford Poverty and Human Development Initiative (OPHI) shared stories like this when she spoke at Oxford Town Hall last night.

The research done by OPHI has made one thing crystal clear: poverty cannot be defined by income alone. For a start, it’s difficult to establish what a certain income level actually means in terms of quality of life. One of the many surprising statistics we heard was that in India, 53% of malnourished children are not defined as income-poor. If you looked at income alone, you would conclude that those malnourished children are just fine.

That’s why OPHI, working with the United Nations Development Programme, developed the Multidimensional Poverty Index (MPI). This is intended to measure acute multidimensional poverty and looks at ten indicators across three dimensions: health, education and standard of living. The indicators are then given different weights in the calculation depending on how important they are. A person is considered poor if they are deprived in at least a third of the weighted indicators.

For example, if a child has died in your family, your house has a dirt floor, you cook with dung and it takes you more than 30 minutes to get clean drinking water, you would be defined as MPI poor, or multidimensionally poor.

The question-and-answer session after Sabina’s talk raised some interesting issues. Sabina told us that OPHI are always painfully aware of how “crude” their measure is and how it isn’t necessarily nuanced enough at local or even country level. One way they tried to remedy this was to ask people in different countries to help them out. “We’ve reached out to countries, saying: ‘This is a crude measure. You can do better. Why don’t you do better?’” One response to this request for help came from Kenya. A woman told OPHI that they had to add cooking with charcoal as an indicator. They already had “cooking with wood or dung” as an indicator, but she insisted that charcoal should be included too, because Kenyan people so often suffer respiratory problems as a result of cooking with it.

This point caused some controversy, because – as chance would have it – someone in the audience had actually helped to design a charcoal stove that was widely used in Africa in the 1970s. He made the point that it’s not charcoal itself, but poor stove design that is the problem.

Sabina agreed, pointing out that what you cook with is weighted relatively lightly on the MPI. (If it wasn’t, you might get Aga owners defined as poor!) The point about the MPI is that it looks at clusters of different factors, not just one factor alone, which is why it’s superior to income-based measurements of poverty.

It was clear from the questions asked that many people in the room were already deeply engaged with the issues, which made for a lively and respectful discussion that could have gone on for much longer. As we ran out of time and drew the meeting to a close, it was at least clear that more sophisticated tools for measuring poverty take us one step closer to eradicating it. And for that, we must thank OPHI for its pioneering work.

Cross-posted to the Oxford WDM blog.