Reviewing The Economic Lives of the Poor

In their paper, “The Economic Lives of the Poor,” Esther Duflo and Abhijit Banerjee take on the exceptionally difficult and delicate task of profiling the financial situations of those in the world with the lowest incomes. Because the methodology used is surprisingly simple, the paper makes for an accessible and illuminating read.

Attempting to describe the “economic lives of the poor” is an undertaking that lends itself to methodological criticism. The authors clearly understand this and discuss the limitations of the data cited often and almost all data used is laid out in appendices. The authors use household survey data from 13 countries. For 9 countries, the data utilized is from the Living Standard Measurement Surveys (LSMS) which out put together by the World Bank. For 3 other countries, the data comes from RAND’s Family Life Surveys. The last, but not least, country is India, for which Duflo and Banerjee choose to use their own data, collected with the Poverty Action Lab (PAL) for previous studies. Comparing that data from different survey instruments is clearly problematic, and even comparing LSMS surveys across countries is fraught with unknowns (were all questions understood in the same manner in every country? Are there cultural differences that make people more or less likely to exaggerate their incomes or consumption?). If anyone has any thoughts on the consistency of LSMS across countries, I would love to hear them.

The professors choose to define the extremely poor as “those living in households where the consumption per capita is less than $1.08 per person per day,” and the poor as “those who live under $2.16 a day using the PPP [purchasing power parity] in 1993 as benchmark.”

Phew, so now that all the caveats are out of the way, lets get back to the purpose of this post, which was to point out the conclusions that I found most surprising and/or worthy of further consideration:

- In the section titled “How the Poor Spend Their Money,” the authors point out that “spending on festivals is an important part of the budget for many poor households.” The numbers vary greatly across countries. The data from a survey conducted by Duflo and Banerjee in Udaipur, shows that for the extremely poor “the median household spent 10 percent of its annual budgets on festivals.” In only 3 of the 13 countries did festivals not make up a substantial portion of expenditures.

- Prior to discussing spending on festivals, the authors exhibit data on food consumption, “Among our thirteen countries, food typically represents from 56 to 78 percent [of consumption] among rural households and 56 to 74 percent in urban areas.” The authors see this number as quite low considering the low body mass index of the extremely poor and the likely need for more calories. Duflo and Banerjee go on to assert that poor households “do see themselves as having a significant amount of choices and choose not to exercise it in the direction of food.” They cite spending on alcohol, tobacco and festivals as money that could be spent on food. I am not sure I agree with the authors here. From personal observation, it seems like their might be a great deal of social pressure on households in rural settings to spend money on festivals, and to reciptocate the spending of their neighbors. Tobacco and alcohol are both addictive and people may be conditioned into spending money on that type of consumption. Certainly people have choices, but they are probably tougher choices than the professors make them out to be.

- The data presented that I personally find most fascinating is that on reported stress by the extremely poor. The authors reveal that while “self-reported happiness or self-reported health levels are not particularly low” it is a different story for stress. In a paper by Case and Deaton (2005), the authors “find that the answers of poor South Africans and poor Indians about stress look very similar, while reported levels of stress are very much lower in the United States.” Duflo and Banerjee go on to mention that “Cutting meals is also strongly correlated with unhappiness.” The argument is that it is the lack of safety nets (i.e. government programs, insurance, savings) that causes stress, not simply the fact of being poor. This is a concept certainly worth considering as more financial products are developed for the poor.

- In the section “The Market for Insurance and the Poor,” the professors use data from Udaipur to show that informal insurance through social networks is insufficient and that when the “the poor come under economic stress, their form of insurance is often eating less or taking their children out of school.” It is also sadly pointed out that in drought years, “the gap in mortality of girls relative to boys is much larger” than usual.”

There is a great deal more I would like to cite, but if you have gotten this far in the post, you will probably take a look at the paper itself.