May 27, 2016

Why Changing How We Measure Poverty Matters

By David Stein

Over the past two decades, there has been a dramatic transformation in developing countries: the number of people living in extreme poverty has been more than cut in half and, as a result, there are one billion fewer people living in extreme poverty today.

But if we are going to achieve the United Nation’s sustainable development goal and eradicate extreme poverty by 2030, we must first understand how the landscape of poverty has changed.

Today, over 70% of the world’s poorest people live not in low-income, but in middle-income countries, which the World Bank defines as having a per capita gross national income (GNI) of between $1,000 and $12,500. This dichotomy illustrates that measuring poverty by average national income can be misleading in today’s world, overlooking millions of extremely poor people.

Take Nigeria, a lower-middle income country with a per capita GNI of around $3,000, or nearly twice that of Sub-Saharan Africa as a whole. Yet, over 50 percent of Nigerians – more than 80 million people – live in extreme poverty.

Changing the definition of poverty might seem technical, but many global development institutions – like the World Bank and other multilateral development banks – use GNI as an eligibility requirement for assistance. Bill Gates noted, “Once countries cross the threshold from low-income to middle-income status, the grants and below-market loans that have helped them rise often come to an end. Countries with huge pockets of poverty like Nigeria, India, Pakistan, Ghana, and Vietnam could lose as much as 40% of their development assistance in the next few years.”

These definitions also have implications for U.S. foreign assistance. The Millennium Challenge Corporation, for example, is only authorized to enter into compacts with countries that are “lower-middle income” and below (meaning a GNI of around $4,000 or less). While that currently includes a country like Nigeria, it won’t be long before the country crosses the threshold into “upper-middle income,” especially given its current GNI growth rate of over 7 percent.

To better capture global poverty, researchers at the Center for Global Development propose using the median (or middle figure) of household income, which – unlike per capita GNI – is less likely to be skewed by income inequality in a country. Further, because it is gathered from household surveys, it will not be influenced by growth in resource sectors like mining or oil, except as they affect household income. This is significant, because while these sectors can greatly increase GNI, they will not necessarily lower the rate of extreme poverty.

When we examine the poverty rate of Nigeria as measured by the median of household income, we see that a Nigerian in a the middle of the income distribution lives on roughly $650 per year – a far cry from the $3,000 average national income measured by per capita GNI.

This is not to say that measuring per capita GNI is unnecessary. According to data scientists at the World Bank, “While clearly not perfect, GNI correlates well with several other indicators commonly used to assess the progress of countries.” The data is also widely available and has been captured for over 50 years, providing a useful measure for historical trends and changes over time.

Still, as the economies of low-income countries grow, the development community will increasingly have to focus on those hard to reach pockets of poverty within countries. Helping these countries provide for their own poor populations by, for example, providing technical expertise for tax collection and helping them implement their own social programs is a big part of the solution. But we must also improve our measurements for poverty. If we don’t, we may miss many of the world’s poor.