The Bank's Poverty and Inequality Practice Group is experimenting with different approaches to measuring poverty
The first, more traditional approach, is to measure poverty based on "snapshots"—how much money does a person live on per day?
The second, the "movie" approach, takes a closer, more nuanced look at various aspects and lengths of poverty
July 10, 2012―When we talk about poverty, it might be useful to borrow from Tolstoy: Rich families are all alike; each poor family is poor in its own way.
Indeed, some folks are born into poverty and may stay there for a lifetime. Some are thrown into poverty by an unexpected death or illness in the family. Then, there are others who choose temporary poverty by going back to school, hoping for a larger pay check in the future.
Circumstances aside, everybody is treated the same in standard poverty measurements. The World Bank’s global poverty update by the Development Research Group counts people as poor if they live under $1.25 a day at the time household surveys are conducted. Its simplicity, along with its rigorous methodology and sheer scale (the latest update drew on 850 household surveys in 125 countries), helped make it one of the most widely-used measures in international development. The United Nation’s first Millennium Development Goal of poverty reduction, for example, was set based on the benchmark, and many developing countries use a similar approach to calculate national poverty rates.
That snapshot approach, however, has its shortcomings. Without taking into account a person’s history of “poverty spells”―the timing, length and severity of poverty episodes―we won’t really know how poor that person is. That’s because people with long poverty spells are less likely to escape from poverty, and are more vulnerable even if they do. Childhood poverty is especially detrimental, because it can hurt children’s cognitive development and lifetime earnings.
So should poverty be measured, instead, more like a movie, which captures the history of poverty spells? But how can one take that into account when measuring poverty? There isn’t an easy fix. Indeed, it’s hard to gauge accurately how previous poverty spells should affect current poverty measurements. For example, is a person who was briefly poor early in life poorer than a person who has been trapped in poverty for a long time as an adult? It’s also rare to find comparable panel data, which track the same people and their poverty status over an extended period of time. Even collecting a series of snapshots, which requires less data, is difficult to carry out in practice.
Some economists are exploring ways to overcome those challenges, and several research papers have been published in a new special issue of the Journal of Economic Inequality. Last month, the issue was officially launched in a workshop organized by the Bank’s Poverty and Inequality Practice Group, which is led by a World Bank team, along with the World Institute for Development Economics Research of the United Nations University in Helsinki, Finland. More than 240 people attended in person or online from countries such as Bolivia, Mauritius, Burundi, and Kazakhstan.
Peter Lanjouw and Luc Christiaensen, both economists in the Development Research Group proposed in the special issue, along with their colleagues, a way to overcome the lack of complete, comparable consumption data. Rather than using a full set of consumption data, they experimented with using components of consumption or even household’s assets (such as cars, TVs and refrigerators) and human capital (such as the level of education) to track poverty. Those data are more comparable and regularly updated than a full set of consumption data. When compared with the actual numbers in Vietnam and China, the methodology worked well.
The special journal issue also focuses on how to develop poverty measures that account for a person’s history of poverty during their lifetime. One study, for example, proposes to measure a person’s poverty by the number of consecutive periods the person is poor. By that standard, nonwhites in the U.S. would be considered much poorer than whites from 1967 to 1992, because they stayed in poverty much longer. Another study, meanwhile, looks at how to count poverty spells that occur at different times of a person’s life.
“The lack of comparable and long panel data shouldn’t stop us from thinking much more seriously about the time element in poverty measurement,” says Christiaensen, a guest editor of the journal issue. Already, the Bank is trying to collect more panel data, he added.
Jaime Saavedra, Poverty Reduction and Equity Group Director of the World Bank's Poverty Reduction and Economic Management Network, says such studies on poverty spells could be useful in the Bank’s operational work. Countries can use them, for example, in their analysis of the middle class, which some define as people who aren’t at risk of falling into poverty for a specific period of time. And if the length of poverty spells affects a person’s ability to escape from poverty, a personal poverty history needs to be considered for the eligibility of safety net benefits.
“But as we do better movies, we should continue to expand our efforts in supporting regular household surveys, which is like taking pictures of welfare in a country,” Saavedra says. The Bank is helping improve household surveys in the regions, especially low-income countries and Africa, where more than half of the countries are still struggling to collect post-2008 poverty data.