The Economist Blog
By: S.D. London
August 2, 2010
THIS week, we carried a piece about a new cross-country poverty index devised by a group of researchers at the Oxford Poverty and Human Development Initiative, which is designed to capture several dimensions of poverty at once. The idea, as the piece explains, is that some aspects of poverty (say, nutrition) may not always move perfectly with income, so that looking directly at how many people are deprived along several (admittedly subjectively chosen) dimensions at once may give researchers and policymakers a better handle on just what poor people lack, and what could be done to deal with these problems. So, for instance:
Since the MPI is calculated by adding lots of different things up, it is possible to work backwards and see what contributes the most to poverty in specific places. In sub-Saharan Africa, the material measures contribute much more to poverty than in South Asia, where the biggest contributor is malnutrition. The authors argue that having this information readily accessible makes it easier for development agencies and governments to decide what to focus on. The MPI also does a better job of uncovering long-term trends. Successful reforms in health or education increase earnings only many years into the future but will show up quickly in the MPI poverty rate.
Sabina Alkire, the director of OPHI does a nice job in an audio interview she did with us of explaining just what she and her team set out to do. Quite apart from the practical aspects, I found the index interesting because it is inspired by Amartya Sen's ideas about capabilities and development as something that enhances people's capabilities, which I've always found intellectually appealing but hard to turn into something concrete. Obviously, the MPI doesn't manage to fully capture the idea of development as freedom, but it takes a fairly cogent stab at running with Mr Sen's ideas.
Ms Alkire was also at pains to point out that the MPI isn't meant to replace, but rather to supplement, income-based poverty measures like the World Bank's dollar (or $1.25) per day-type measures. I think this is right--at an intuitive level, income is very appealing because it is in fact a measure of people's ability to buy the things that make their lives less onerous. Focusing on income, for all its flaws, has a certain intellectual consistency; one can think of conditions under which income maps directly into some Sen-style notion of choice. At the same time, I just think it's useful to have a clear, transparent way to see what this translating into at the level of the household, and this is what the MPI offers.
But Mr Alkire's protestations notwithstanding, it is perhaps inevitable that the MPI be seen in opposition to income-based measures. What it does is throw a spotlight on things that governments, for example, might argue will be an inevitable outcome of more income--except the MPI shows, in some cases, that this isn't happening, or isn't happening very fast.
Not surprisingly, then, Ms Alkire is engaged in a fascinating debate with the World Bank's Martin Ravallion, who developed and refined the dollar-a-day measures, on Oxfam's blog. Mr Ravallion's main beef is with the MPI's choice of equal weights for its component indices. He argues that:
On closer scrutiny, the embedded trade-offs (stemming from the weights chosen by the analyst) can be questioned, and may be unacceptable to many people. In the context of the HDI, I pointed out 15 years ago that by aggregating GDP per capita with life expectancy the HDI implicitly put a value on an extra year of life, and I showed that this value rises from a very low level in poor countries to a remarkably high level in rich ones (4-5 times GDP per capita). If it was made clearer to users, I expect that they would question this trade-off embedded in the HDI.
The MPI index faces the same problem. How can one contend (as the MPI does implicitly) that the death of a child is equivalent to having a dirt floor, cooking with wood, and not having a radio, TV, telephone, bike or car? Or that attaining these material conditions is equivalent to an extra year of schooling (such that someone has at least 5 years) or to not having any malnourished family member? These are highly questionable value judgments. Sometimes such judgments are needed in policy making at country level, but we would not want to have them buried in some aggregate index. Rather, they should be brought out explicitly in the specific country and policy context, which will determine what trade off is considered appropriate; any given dimension of poverty will have higher priority in some countries and for some policy problems than others.
Ms Alkire has responded, arguing that:
We measure multidimensional poverty as the incidence (or the percentage of the population that is poor) times the intensity (or the average percentage of deprivations poor people experience). Unlike the HDI, this construction does not add up achievement levels, which requires strong assumptions concerning the variables in question as Martin noted. Instead, we add up deprivations, which does not.
OK, now to the issue of weights. Income poverty aggregates within a country using actual or imputed prices (these are critical for fixing the income poverty standard across countries and time). Setting prices is not unproblematic in practice, particularly in Colombia where I am writing from. Indeed the Presidential address to the 2010 American Economic Association raised concerns such as the prices attributed to housing (Deaton 2010). Chen and Ravallion 2008 carefully review the robustness of their results to different pricing approaches.
As Martin observed, instead of using prices, the MPI sets weights as value judgements. Amartya Sen among others sees this feature as a strength not an embarrassment: "There is indeed great merit... in having public discussions on the kind of weights that may be used" (1997a).
I don't know if they'll respond to each other again, but even if not, the full posts are worth reading. What I personally found very interesting about the MPI exercise was not so much the value the index took in different countries, but the ability to quickly decompose the MPI headcount into the contributions of various sub-indices. It would certainly be worth seeing how much fiddling with the weights affects rankings, and lots more discussion about how such a measure could be refined. Is there some way to derive weights that reflect, for instance, the cost of making a marginal improvement in each dimension the MPI measures, just to throw a thought out? At the same time, as Ms Alkire points out, there is a weighting implicit in an income-based measure. In any case, I don't think that this debate needs to be an either-or; neither, I think, do either of the participants.