Ask poor people what poverty is like, and they typically talk about fear, humiliation and ill health, at least as much as money. But can the non-income dimensions of poverty be measured in a way that allows policy makers to weigh priorities and allocate resources? If not, the danger (as often happens) is that decision makers and documents initially nod towards the many dimensions of poverty, but by paragraph two, you’re back in $ per day territory. And all too often, in policy terms, if it can’t be measured, it gets ignored.
The Oxford Poverty and Human Development Initiative (OPHI) has been working for years to try and develop such metrics, and they recently launched the ‘Multidimensional Poverty Index’ (MPI), which will feature in this year’s UNDP Human Development Report, celebrating its 20th anniversary. I’ll briefly summarize it here, before unleashing an exchange of guest blogs between the World Bank and OPHI.
The MPI brings together 10 indicators of health (child mortality and nutrition), education (years of schooling and child enrolment) and standard of living (access to electricity, drinking water, sanitation, flooring, cooking fuel and basic assets like a radio or bicycle). It’s thus a logical extension of its predecessor, UNDP’s pioneering Human Development Index, launched in the first Human Development Report back in 1990, which combined life expectancy, education (literacy + enrolment rates) and GDP per capita.
What were the results when they crunched the numbers? Here’s the blurb from the launch press release:
“OPHI researchers analysed data from 104 countries with a combined population of 5.2 billion (78 per cent of the world total). About 1.7 billion people in the countries covered – a third of their entire population – live in multidimensional poverty, according to the MPI. This exceeds the 1.3 billion people, in those same countries, estimated to live on $1.25 a day or less, the more commonly accepted measure of ‘extreme’ poverty.
The MPI also captures distinct and broader aspects of poverty. For example, in Ethiopia 90 per cent of people are ‘MPI poor’ compared to the 39 per cent who are classified as living in ‘extreme poverty’ under income terms alone. Conversely, 89 per cent of Tanzanians are extreme income-poor, compared to 65 per cent who are MPI poor. The MPI captures deprivations directly – in health and educational outcomes and key services, such as water, sanitation and electricity. In some countries these resources are provided free or at low cost; in others they are out of reach even for many working people with an income.
Half of the world’s poor as measured by the MPI live in South Asia (51 per cent or 844 million people) and one quarter in Africa (28 per cent or 458 million). Niger has the greatest intensity and incidence of poverty in any country, with 93 per cent of the population classified as poor in MPI terms.
Even in countries with strong economic growth in recent years, the MPI analysis reveals the persistence of acute poverty. India is a major case in point. There are more MPI poor people in eight Indian states alone (421 million in Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan, Uttar Pradesh, and West Bengal) than in the 26 poorest African countries combined (410 million). The MPI also reveals great variations within countries: Nairobi has the same level of MPI poverty as the Dominican Republic, whereas Kenya’s rural northeast is poorer in MPI terms than Niger.”
My views on all this (largely stolen from my colleague Claire Hutchings)? It’s a step forward on the previous Human Development Index, but only a limited one. There are still many facets of poverty that it doesn’t touch on, such as conflict, personal security, domestic and social violence, issues of power/ empowerment, or intra-household dynamics. This is partly because it still relies on existing data sets, focusing on how to use differently the data we are already collecting, rather than proposing/ starting from a fresh conceptual framework on critical dimensions of poverty. That makes the proposal more practical, but less radical.
The comparison of extreme income poverty scores vs multidimensional poverty scores is interesting (see chart – the bar is the MPI score, the line is the income poverty score) – it would be great to see further research into possible explanations for the divergences, such as the role of social services and social protection- both formal and informal, and the potential implications for policy development.
Another advantage for policy development and assessment is that this index responds more rapidly than income to different policy interventions. A child feeding programme or scrapping user fees will have an immediate impact, whereas it may take years for government policies to filter through into income stats.
Great that it’s all open source – as with all measures there is scope to choose the mix of indicators to back up your particular argument, but at least making this data open source allows other people to challenge your analysis.
Finally, while it does allow for comparisons of groups within countries it is still a very aggregate picture, designed primarily to enable comparisons between countries.