bond - for international development Not yet a member? - Join Us

The new 'bottom billion'

Flooding in Pakistan, a middle income country that is fragile/conflict-affected © United NationsWhat if 72 per cent of the world's poor don't live in poor countries?  Andy Sumner of the Institute of Development Studies outlines the findings of his latest research.

The global poverty problem has changed. In the past, poor people lived in poor countries but now there are around 960 million poor people, or a ‘new bottom billion’, living on under $1.25/day in middle-income countries, and most of these are in stable, non-fragile, middle-income countries.

Redefining the ‘bottom billion’

This new bottom billion accounts for 72 per cent - almost three-quarters - of the world’s poor. Only about a quarter of the world’s poor – 370 million people or so − live in the remaining 39 low-income countries, which are largely in sub-Saharan Africa.

These findings are similar across nutrition indicators and the UN Development Programme’s new multidimensional poverty index. The exception is in education where 60 per cent of children out of primary school are in low-income countries compared to just 40 per cent in middle-income countries.

This is a dramatic change from just two decades ago when 93 per cent of poor people lived in low-income countries. This change has major implications for both the achievement of the Millennium Development Goals and global strategies for poverty reduction beyond 2015.

In fact, the change raises questions for donors, governments in middle-income countries, NGOs in North and South and researchers. There is a real prospect of ending poverty by 2025 because the resources exist to do it. How does this new geography of global poverty change things?

The poor live in three types of places:

  • Low-income countries - one quarter of the world's poor
  • Middle-income countries - three quarters of the world’s poor
  • Fragile/conflict-affected states - these are a mix of low and middle income countries and represent around a quarter which overlaps the above

Bottom billion graphic

Which countries do we mean? Democratic Republic of Congo and Burundi are low-income countries that are also fragile/conflict-affected states, while stable low-income countries include Zambia and Tanzania.  Pakistan and Nigeria are both middle-income countries that are fragile/conflict-affected, whereas stable middle-income countries are India and Indonesia.

Why has it taken so long to realise we were getting it wrong? Good data runs two to three years late at least and the numbers reflect a long run pattern that became particularly visible over last year or so as a number of countries reached middle-income status.

Don’t the poor or ‘bottom billion’ live in fragile states? About one in four or five do. Paul Collier’s Bottom Billion were always the billion people not the billion poor people living in 58 countries that were ‘falling apart or falling behind’. Collier’s assumption is that poverty is 100 per cent in these countries.  In the IDS estimates, total fragile/conflict-affected low- and middle-income countries account for about 23 per cent, or about 300 million or so.

Taking Collier’s approach

When we looked more closely at Collier’s 58 countries, we found that we have reasonably good poverty data coverage for those countries. Collier's 58 bottom billion countries represent:

  • close to a billion people in total population but only 27 per cent of the world's poor
  • around 30 per cent of undernourished children
  • around 30 per cent of the poor according to the new UNDP multidimensional poverty measure

Collier’s 58 countries also have a high proportion of the global out of primary school children (57 per cent).

Things have changed too. A third, or 300 million, of the total population of Collier's 58 bottom billion countries now live in World Bank-classified middle-income countries and one in five of the total population of Collier's 58 bottom billion countries no longer live in fragile/conflict-affected states if they ever did (we used the broadest possible definition - 43 countries - by combining the three fragile/conflict-affected states lists).

Paul Colliers 'bottom billion' represents around 30 per cent of undernourished children © iStockphoto/MShep2What does this all mean?

Collier's approach is out of date and overlooks around 70 per cent of the poor people in the world. If development is primarily about fighting poverty, that is a lot.

This doesn’t invalidate Collier’s key point: that the poor in his bottom billion 58 countries (even if only are 27 per cent of the total world’s poor) are trapped and without hope so development needs to focus there; it simply means we would raise a question mark over his assumption that there is no need to worry about the other 70 per cent or the ‘new bottom billion’ because growth will look after them.

The second assumption he makes is also questionable - that the donor community is able to contribute positively to springing these traps in his 58 countries. It is not clear what the evidence is for this.

Poor people haven’t moved of course. What has happened is the countries many poor people live in have got richer – attained middle-income country status - and the data suggests that they have been left behind.

So, what if the poor live in countries that have got richer, why does it matter? It matters because it raises a lot of questions about aid and development policy ‘beyond aid’. If development is about poverty reduction, where poor people live is a crucial question. How should donors, policy makers, NGOs and practitioners respond to the ‘new bottom billion’?

Andy Sumner is a cross-disciplinary Economist and a member of the Institute of Development Studies' Vulnerability and Poverty Reduction Team. Read more about his research on the bottom billion on the IDS website or find more of his work on Scribd.