The Multidimensional Nature of Poverty in Developing Countries: A Comparative Study of Mexico, South Africa, and Turkey

Poverty is still one of the major challenges of contemporary societies, both at the national and the global level. How to conceptualize and to measure poverty, however, is much debated. Traditionally, poverty has been considered as monetary deprivation and has been measured using metrics focusing on income or consumption. But over the last few decades, alternative perspectives have been developed conceptualizing poverty as a deprivation from multiple aspects of life, such as Amartya Sen’s capability approach (1981, 1987), the United Nations’ “End Poverty Millennium Development Goals and Beyond 2015”, or the recently launched Multidimensional Poverty Index (MPI, 2010).

However, current multidimensional poverty indexes such as the MPI (2010) suffer from serious shortcomings, notably regarding target groups, indicators, thresholds, weights and aggregate functions. In particular, there is no consensus on poverty thresholds and on the dimensions of poverty to be included in these multidimensional approaches. Furthermore, most multidimensional poverty studies are selecting indicators as a function of the availability of data, and due to differences in the dimensions included, it is not possible to compare poverty rates among countries.

My paper demonstrates that, from a sociological perspective, the use of multidimensional poverty indexes is not based on a robust method. The paper, therefore, proposes a pattern based approach focusing on poverty statuses and not on a binary classification of poverty (poor or non-poor). These patterns designate different forms of poverty, conceptualized by a combination of ‘relative needs’ (due to interaction of individual’s behaviors in society), ‘survival needs’ (to distinguish absolute poverty) and ‘well-being related subjective psychological needs’. Finally, poverty is measured and empirically analyzed based on this pattern approach by using micro data from Mexican (Socio-Economic Conditions Survey, 2012), South African (General Household Survey, 2011), and Turkish (Living Conditions and Income Survey, 2011) surveys.