Our Programmes



Sign up to our newsletter.

Follow ODI

The shifting dynamics of poverty – how do we research poverty and wellbeing in a changing world?

Written by Andrew Norton

Last year the research and evidence division of the Department for International Development (DFID) asked a group of UK-based researchers involved in poverty analysis to come together to consider a shared view of lessons, challenges and frontiers for poverty research. Predictably, it took a while to get agreement on a shared view. But despite the challenges inherent in trying to get a bunch of researchers to agree on the key elements of poverty research, it was great to have a chance to benefit from each others’ experience, and the resulting document is richer for the time and commitment that everyone in the group gave to the process. I think it’s a great summary of lessons and challenges, and we hope it sparks both debate and new thinking.

The key problem the paper addresses is the mismatch between the complex nature of poverty and the reductionist nature of measures used to track it. On top of this familiar paradox comes another: that poverty and inequality are rooted in asymmetric power relations, while the policy machineries we work through to address poverty are very much a part of those power structures. The paper proposes simple, pragmatic approaches to bridging the gap between policy communities and researchers – and points out the dangers of failing to address the ‘inconvenient complexity’ of poverty.  

Three of those involved in producing the paper are offering thoughts today – Andy Sumner of the Institute of Development Studies focuses on methodology, and David Hulme of the Institute for Development Policy and Management on the implications for policy, while I want to focus on the significance of some big changes in the global context, and what they mean for poverty analysis and research.

The first big change in context concerns the shifting geographies of both growth and poverty.  An increasing number of developing countries are comfortably outstripping the growth rates of the developed countries – and sustaining that performance over a substantial period of time (the Organisation for Economic Cooperation and Development’s  (OECD) Development Centre calculates that 83 low-income countries (LIC) or low- and middle-income countries (LMIC) have achieved per capita growth rates of more than double the OECD average over the decade 2000 to 2010).

 In short, it’s no longer just about the dynamism of ‘BRICs’, ‘CIVETs’, Asian tigers or African ‘cheetahs’– there is a clear economic convergence happening on a global scale.  If you take out countries with significant systemic problems, poor countries are growing faster than rich ones, and inter-country inequality is therefore declining. But many of these fast-growing developing economies have high levels of domestic structural inequality, and the growth is not producing jobs fast enough to pull enough people out of poverty.  As Andy Sumner has shown, as a result of these changes, the bulk of the world’s poor are now in middle-income, rather than low-income countries. This means that poverty reduction increasingly requires progress on policies which reduce inequality within countries, and which turn growth into jobs.  This does not mean that inequality between countries does not matter – conflict, insecurity, violence and fragility at the country level are still hugely important as drivers of poverty and deprivation.

Another change in context which became a big focus of discussion in the preparation of the paper concerned the impact of increasing levels of urbanisation on a global scale.  Simple ‘money-metric’ measures of poverty struggle to cope with the difference between urban and rural contexts. It is common practice in consumption-based poverty measurement to make allowances for the fact that the urban poor pay rent, while the rural poor often don’t, or for the fact that the rural poor have some of their own produce that they consume directly (either grown or gathered).

But often these efforts to ‘offset’ one context against another don’t take on board the many other ways that a more or less urbanised context makes a difference for a poor family. Does water have to be paid for?  Can rural families gather fuel? Do household members have to pay for transport?  How does the differing environment affect health and the quality of life? Does living in a deprived community right next to a wealthy suburb bring stigma and discrimination?  And, however carefully technocrats work to take account of these differences, how do you account for peri-urban environments, which are neither clearly one thing nor the other? If shifting geographies of poverty and growth put the spotlight on issues of inequality, then urbanisation points us towards another big theme, namely the multi-dimensional character of poverty, and the need for poverty research to capture issues of dignity and voice, as well as the many different ways in which the material dimensions of deprivation may differ in different contexts.

A final shift in optics which has occurred in the past decade concerns our increasing consciousness of risk, shocks and uncertaintyand the importance of understanding the factors which give poor people and communities resilience in the face of growing risks from an increasingly volatile climate, from rapid changes in the prices of key commodities (particularly food), and from socio-political change which is often ‘hyper-linked’ to change in other domains. This points to the need for poverty research to get better at understanding poverty dynamics – and the factors which can either support or threaten improvements in livelihoods.

The paper concludes with some proposed areas of focus for poverty research and analysis. We need to look back, building on the lessons of the past decade on what works best for understanding the causes of and solutions to poverty. We also need to look forward, and respond to the changing world. This means thinking differently about the ‘what’ of poverty research (understanding the politics of poverty action, for example) and about the ‘how’ (such as putting more resource and effort into longitudinal analysis, which builds an understanding of how the experience of poverty changes over time in different and rapidly changing contexts).  

ODI is a partner in The Chronic Poverty Advisory Network, which was established to produce detailed policy guidance and disseminate research findings on these issues to policy-makers. We hope the paper will provoke further thinking and discussion. Please tell us what you like and don’t like about this approach in the comments below.