Rethinking the Way We Count the World's Poor | Asian Development Bank

Rethinking the Way We Count the World's Poor

Op-Ed / Opinion | 4 September 2008

"The matter with the poor is poverty", said George Bernard Shaw, the Irish playwright born 11 years after the 1845 potato famine that killed 1 million of his countrymen. Shaw knew that to be poor is not necessarily a prescription for a miserable existence. Poverty, on the other hand, can kill you.

This year, poverty is no better for your health. UNICEF estimates that more than 9.7 million children under the age of five die every year from preventable causes, most of them poverty related. Eradicating poverty, then, is among the most important development challenges facing the world today. But how do we identify the poor in our societies? Where, in short, do we draw the poverty line?

Last week saw the release of two reports from opposite sides of the world that shed new light on this issue. The first was the long anticipated revision of the World Bank's 1990 US$1-a-day poverty line. In the report, the bank revised the poverty line up, to US$1.25 in 2005 prices.

On Wednesday, in Hong Kong, the Asian Development Bank, where I work, also released a report on poverty lines. That report, "Comparing Poverty Across Countries: The Role of Purchasing Power Parities", assembled national poverty lines from 16 Asian countries and describes the methods involved in generating internationally comparable estimates of poverty. As part of this study, a new poverty line was drawn at US$1.35. We called it the Asian Poverty Line.

How can that be? How can two multilateral institutions, both committed to fighting poverty, arrive at poverty lines that are priced so far apart? Ten cents might not sound like much but, as every politician knows, apply it to the bottom line and the higher figure will translate into millions more people living in poverty in your country.

The answer is quite straightforward. Our poverty study focused exclusively on Asia, where standards of living are much higher than in the countries surveyed by the World Bank. They used the 15 poorest countries in the world, only two of which came from Asia, to fix their poverty line. They call this an extreme poverty line. Our US$1.35-a-day poverty line represents the median of the 16 countries in our study.

Of most interest, however, was the different method used by each study to count the number of poor. It should be a straightforward task to determine the percentage of people who live in a certain country on less than, say, our US$1.35 a day. But carrying out currency conversions at market exchange rates is impractical. Purchasing power parities (PPPs) are the way out. These are conversion factors that ensure a common purchasing power over a given set of goods and services. But which PPPs should be used?

The World Bank's poverty estimates are based on PPPs developed for comparing household consumption across countries, known as consumption PPPs. However, these represent goods purchased by the general population, not necessarily by the poor.

To provide the most accurate estimate of the number of people living in poverty, we devised a poverty PPP, and this was the first time anyone has used one. With the assistance of our 16 participating countries, we examined what the poor buy, in what quantity, as well as the quality of the products they purchase.

Using the poverty PPPs, we found that 843 million people live on less than US$1.35 a day in the 16 countries in our study. If we had used consumption PPPs to perform the same maths, the number living in poverty would have risen to 1.042 billion - a difference of 199 million people, close to the population of Indonesia.

That's not a blip, that's a balloon, and it suggests that, with the greatest respect to Shaw, the matter with the poor is not only poverty. It's also the way that we count them.