Nepal and Bangladesh are star performers in cutting multidimensional poverty
Nepal and Bangladesh are star performers in a new study by Oxford University of how countries have reduced multidimensional poverty over time, achieving some of the strongest decreases in poverty in both absolute and relative terms.
Using a poverty measure that assesses a range of deprivations in health, education and living standards, researchers found that the percentage of poor people in Nepal dropped from 64.7% to 44.2% between 2006 and 2011, 4.1 percentage points per year, while in Bangladesh poverty rates decreased by 3.2 percentage points per year between 2004 and 2007.
In addition to reducing the percentage of poor people, both Nepal and Bangladesh reduced the intensity of poverty. This means that even poor people were on average less poor – deprived in fewer things at the same time – than they had been before, an important element of multidimensional poverty analysis that provides a more balanced picture of poor people’s lives.
India also made significant progress between 1999 and 2006, but at a rate that was less than one-third of the speed of its poorer neighbours, with a reduction in poverty rates of 1.2 percentage points per year [instead of 4.1% (Nepal) or 3.2% (Bangladesh)], the study found. Nonetheless, India reduced multidimensional poverty more than 50% faster than it reduced income poverty in absolute as well as relative terms.
‘The success of Nepal and Bangladesh in reducing poverty despite their relatively low income highlights the effectiveness of social policy investments combined with active civil society engagement,’ said Dr Sabina Alkire, Director of the Oxford Poverty and Human Development Initiative (OPHI), the research centre at Oxford University that conducted the study.
The poverty measure used by OPHI, the global Multidimensional Poverty Index (MPI), is unique in capturing the simultaneous disadvantages experienced by poor people, such as malnutrition, education and sanitation, providing a high-resolution lens on their lives. If people are deprived in a third or more of ten (weighted) indicators they are identified as ‘MPI poor’.
Of 22 countries analysed for changes in MPI poverty over time, Nepal, Rwanda and Bangladesh were found to have made the largest absolute reductions, followed by Ghana, Tanzania, Cambodia and Bolivia. Most ‘top performing’ countries reduced MPI poverty as fast or faster than income poverty, with Nepal in particular making stellar progress in cutting both. The strongest reductions in deprivations in Nepal were made in indicators such as assets, electricity and school attendance, but all ten indicators saw significant reductions. While the rise in income due to increased rural wages and remittances clearly affected the reduction in asset deprivation, the dramatic increase in access to electricity and schooling was largely the result of NGO and government interventions. Nepal’s baseline data were collected in 2006, the year in which the Comprehensive Peace Accord was signed, ending the country’s decade-long civil war.
Bangladesh achieved a strong reduction not only in the number of people experiencing poverty, but also in the intensity of their poverty. The largest improvements were made in sanitation and school attendance, and statistically significant reductions were made in all six of its regions. The Rajshahi region – which has a population larger than Nepal – had annual reductions of MPI poverty that were larger than Nepal’s rate, as did the Chittagong region.
India, like Nepal, made significant reductions in all ten poverty indicators. However, multidimensional poverty was reduced least in the poorest states – such as Bihar, Madhya Pradesh, Rajasthan, Uttar Pradesh and West Bengal - and among the poorest social groups, such as Scheduled Tribes, Muslims, female-headed households and larger households. Moreover, even India’s best-performing states – Kerala and Andhra Pradesh – progressed little more than half as fast as Nepal or Bangladesh in reducing multitidimensional poverty.
“From 1999-2006 India did very well in certain aspects of poverty reduction; for example, MPI among the scheduled caste groups showed a strong reduction, and poverty among the most destitute went down faster than the average,” said Dr Suman Seth, OPHI Researcher. “However, it’s still the case that the benefits of national poverty reduction have been enjoyed least by some of the poorer groups and regions.”
Unfortunately, India has not collected official data on MPI deprivations including malnutrition since 2005/6, making India’s MPI the least up-to-date in South Asia. OPHI anticipates being able to update the MPIs for Bangladesh, Pakistan and Afghanistan again in the coming year, all using 2011 data. The MPI indicators are built from fewer than 40 survey questions, which make up less than 7% of the 600-plus questions present in most Demographic and Health Surveys.
Where is all this leading? OPHI hopes that these measures will spur the eradication of multidimensional poverty. “If Nepal and Bangladesh continue reducing poverty at the current rate, they will halve MPI in less than 10 years and eradicate it in 20,” said OPHI’s Dr José Manuel Roche. “Based on the same assumptions, it will take India 41 years to eradicate acute poverty as measured by the MPI.”
The release of OPHI’s findings on changes in poverty over time follows the publication of the updated MPI in the UN Development Programme’s flagship Human Development Report for 2013 on 14 March 2013. The global MPI, which was developed by OPHI and UNDP in 2010 and has been published in the HDR ever since, assesses multidimensional poverty in 104 countries for which data since 2002 are available.
For interviews and more information:
Sabina Alkire, Director, OPHI: +91 9910 460205, sabina.alkire@qeh.ox.ac.uk
Emma Feeny, Research Communications Officer, OPHI: +44 (0)1865 271528, emma.feeny@qeh.ox.ac.uk
NOTES FOR EDITORS
*STORY IS NOT TO BE PUBLISHED UNTIL EMBARGO LIFTS AT 0:01am (GMT) Monday 18 March 2013
Data sources and constraints
The MPI relies on the most recent data available, mainly from three datasets that are publicly available and comparable for most developing countries: USAID’s Demographic and Health Survey (DHS), UNICEF’s Multiple Indicators Cluster Survey (MICS), and the WHO’s World Health Survey (WHS). OPHI’s analysis of multidimensional poverty reduction in India was done using India’s DHS or National Family Health Survey (NFHS) datasets for 1998/9 and 2005/6. Unfortunately it is not currently possible to update this analysis because the dataset has not been repeated; nor do the National Sample Survey datasets include the required questions.
Background to the Multidimensional Poverty Index (MPI)
The MPI was created in collaboration with the UNDP HDRO by OPHI Director Sabina Alkire and OPHI Research Associate Maria Emma Santos, now at Universidad Nacional del Sur and CONICET, Argentina. In 2013, the MPI has been updated and analyses have expanded, including substantial new analyses of changes to poverty over time and where the world’s poorest billion people really live. The MPI is constructed using a methodology developed by Alkire and Professor James Foster, an OPHI Research Associate and Professor of Economics and International Affairs at George Washington University. For more information, including all MPI 2013 materials, please see www.ophi.org.uk/multidimensional-poverty-index.
Calculation of poverty using the Multidimensional Poverty Index (MPI)
The MPI is based on a deprivation score which reflects each person’s overlapping deprivations in nutrition, child mortality, years of schooling, child school attendance, water, sanitation, electricity, cooking fuel, flooring, and assets. A person is identified as ‘multidimensionally poor’ if he or she is deprived in one-third or more of ten (weighted) indicators. The MPI of a country or region is the product of the proportion of poor people (H) and the average share of deprivations that poor people face at the same time, i.e. the average intensity of their poverty (A). In other words, MPI=HxA. By directly measuring the different types of poverty in each household, the MPI captures how people experience different deprivations simultaneously. See Alkire, Conconi and Roche (2013) ‘Multidimensional Poverty Index 2013: Brief Methodological Note and Results’, at www.ophi.org.uk/multidimensional-poverty-index.
Oxford Poverty and Human Development Initiative (OPHI)
OPHI is a research centre within the Oxford Department of International Development at the University of Oxford. OPHI is led by Sabina Alkire and works to develop and apply new ways of measuring and analysing poverty, human development and welfare, drawing on the work of Nobel Laureate economist Amartya Sen. Alkire and James Foster (see above) developed the Alkire Foster counting approach to multidimensional measurement, which is used to calculate the MPI as well as national measures of poverty - for example in Mexico and Colombia – and of wellbeing in Bhutan, and women’s empowerment in agriculture. For more information about OPHI, please visit www.ophi.org.uk.
