Multidimensional approach: Poverty beyond income
Published:
Amartya Sen points out that poverty should also be conceived as the deprivation of basic capabilities and not only as the lack of income (Sen, 2000). Although it is true that income is a means to achieve certain capabilities, it is not the only one; therefore, he suggests that poverty cannot be reduced to being only a monetary phenomenon.
What is multidimensional poverty?
Currently, poverty is one of the most pressing issues on both national and international agendas. To illustrate this, the first Sustainable Development Goal (SDG) set by the United Nations for 2030 is to eradicate poverty in all its forms. This leads us to question whether there are different forms of poverty. The lack of income is the usual criterion for measuring poverty. If an individual’s income falls below a certain threshold, known as the poverty line, then they are considered poor. However, Amartya Sen conceives poverty as a broader phenomenon, basing his analysis on a capabilities approach, which considers the fundamental freedoms a person must have in order to lead a dignified life. In this sense, Sen argues that poverty should also be understood as a deprivation of basic capabilities, not solely as a lack of income (Sen, 2000). While income is indeed a means to achieve certain capabilities, it is not the only one. Therefore, he suggests that poverty cannot be reduced to being merely a monetary phenomenon. Faced with the need to measure poverty comprehensively, the multidimensional approach emerges, encompassing deprivations that an individual may experience in health, education, quality of life, employment, social security, among other non-monetary aspects that reflect the goals and priorities of national development plans.
Multidimensional Poverty Index
The monetary approach is widely used to measure poverty due to its instrumental simplicity. However, not all individuals considered non-monetary poor have acceptable living conditions. According to the National Household Survey (Encuesta Nacional de Hogares or Enaho) in 2019, 25.86% of non-monetary poor households in Loreto lacked hygienic facilities in their homes. Pasco and Ucayali had similar situations, with 19.11% and 15.57%, respectively. When analyzing the survey by geographic region, it becomes evident that 16.98% of non-monetary poor rural households lack this service. Moreover, this percentage is significantly higher than that of non-monetary poor urban households, suggesting evidence that poverty can be experienced differently and unequally depending on the geographical area (see Figure 1).
Figure 1: Non-poor monetary households without toilets by geographical area (%)
In light of the presented evidence, it becomes necessary to introduce a multidimensional poverty approach that complements the income-based approach and can take into account the various deprivations an individual may experience. This new measurement of poverty would reflect social well-being more comprehensively.
The Oxford Poverty and Human Development Initiative (OPHI) has been developing a Multidimensional Poverty Index (MPI) since 2010, which considers three dimensions: education, health, and living standards. Each dimension is a conceptual category composed of various indicators. For instance, the health dimension includes two indicators: nutrition and mortality. These indicators allow for the determination of deprivations an individual may suffer. These deprivations can vary in nature, with some indicators being continuous variables, such as years of education, while others are binary, like the indicator determining whether a household has electricity or not. The criteria under which a household would be considered deprived in an indicator are explained in more detail in Table 1. Additionally, each indicator has a relative weight that reflects its value in the calculation of the MPI compared to other indicator deprivations, as shown in Figure 2. Despite the normative nature of assigning weights to each indicator, meaning it depends on the purpose of the MPI, certain statistical properties must be satisfied [1].
The MPI identifies individuals who are multidimensionally poor (referred to as ‘incidence’) and also measures the average number of deprivations experienced by poor households (also known as ‘intensity’). In other words, the MPI identifies who is poor and how they experience poverty. Additionally, the indicator can be disaggregated by different groups of interest, such as geographical regions, urban or rural areas, gender, age, ethnic groups, and more.
In 2019, the United Nations Development Programme (UNDP), in collaboration with OPHI, produced an MPI report for 101 countries, including 31 low-income countries, 68 middle-income countries, and 2 high-income countries. The study found that 23.1% of the entire considered population (equivalent to 1.3 billion people) were multidimensionally poor, with two-thirds of them residing in middle-income countries. Additionally, it was calculated that half of the multidimensionally poor population were individuals under the age of 18, and one-third were under the age of 10.
According to UNDP and OPHI (2019), Peru is one of the countries that has made significant progress in reducing the MPI (approximately 7.1% reduction each year). Furthermore, when analyzing the MPI by indicator, Peru has made substantial reductions in all ten indicators, albeit in varying proportions (see Figure 3).
Figure 3: Annual Variation in the Peruvian MPI Dimensions (%)
It is also worth noting that while Peru has made significant strides in reducing multidimensional poverty, an analysis by geographical area reveals that this reduction is occurring unevenly. The percentage of multidimensionally poor individuals and those deprived in the indicators in urban areas is much lower than those in rural areas, as illustrated in Figure 4. For instance, only 1% of multidimensionally poor individuals in urban areas are deprived of the indicator related to assets; meaning their households lack items such as a radio, TV, telephone, among others (refer to Table 1). In contrast, in rural areas, 17% experience this deprivation. Additionally, 3.2% of the urban population lacks adequate housing, while in rural areas, 33% of households are deprived in this indicator.
The Importance of the MPI for Public Policies
As observed, the MPI provides more detailed information about poverty and places particular emphasis on how the poor live, that is, what their deprivations are. While it is true that the OPHI and UNDP report is encouraging regarding poverty reduction in Peru, it is recommended that countries develop a national MPI to obtain more precise information about their situation, taking into account their specific context and needs. A national MPI can incorporate more dimensions and indicators than the global MPI, and the relative weights of each indicator could vary depending on the country’s priorities. The Multidimensional Poverty Peer Network (MPPN), an organization comprising 59 countries and 20 institutions, seeks to measure multidimensional poverty. The network aims to provide technical and institutional support to policy makers who are implementing a national MPI (MPPN, 2020). The first country in the region to adopt a national MPI was Mexico, and subsequently, other countries followed suit, including Colombia, Chile, El Salvador, Ecuador, Panama, the Dominican Republic, Costa Rica, and Honduras. While the multidimensional approach complements the monetary one, it also allows for a more precise direction of public policies. Additionally, an MPI enables the assessment of whether policies implemented in a region have been effective and have translated into an improved quality of life for the targeted individuals. A disaggregated analysis of the MPI also facilitates the implementation of policies tailored to specific groups, enabling them to improve their well-being. Furthermore, since the MPI encompasses various dimensions, it presents an opportunity for coordinated efforts among different ministries to eradicate poverty in all its forms.
The process of developing an MPI for Peru is closer than it may seem. The Ministry of Development and Social Inclusion (Midis), in collaboration with the National Institute of Statistics and Informatics (INEI), has initiated discussions with OPHI to learn from the experiences of other Latin American countries and thus embark on the creation of a national MPI (Gestión, 2020). The measurement of multidimensional poverty represents a significant opportunity for all Peruvians, as it will provide policymakers with a broader and more comprehensive understanding of such a complex phenomenon as poverty.
Table 1: Criteria for Determining if a Household is Deprived in an Indicator
[1] According to Alkire & Santos (2014), a necessary statistical property that MPIs must satisfy is robustness. This means that the reliability and stability of the indicator should not be affected by changes in its parameters (such as the relative weight of each indicator, for example).
References:
Alkire, S., & Santos, M. E. (2014). Measuring acute poverty in the developing world: Robustness and scope of the multidimensional poverty index. World Development, 59, 251-274. doi:https://doi.org/10.1016/j.worlddev.2014.01.026
Alkire, S., Suppa, N., & Kanagaratnam, U. (2019). Changes over time in the global Multidimensional Poverty Index: a ten-country study. OPHI MPI Methodological Notes 48, Oxford Poverty and Human Development Initiative, University of Oxford.
Gestión. (February 11, 2020). Peru to Begin Measuring Multidimensional Poverty from 2021 Onward. Gestión. Retrieved from https://gestion.pe/peru/el-peru-empezaria-a-medir-la-pobreza-multidimensional-a-partir-del-2021-noticia/
MPPN, Multidimensional Poverty Peer Network. (2020). Retrieved from https://mppn.org/
OPHI, Oxford Poverty and Human Initiative. (2020). Retrieved from https://ophi.org.uk/
Sen, A. (2000). Development and Freedom, 8th ed. Bogotá: Planeta.
UNDP, United Nations Development Programme, & OPHI, Oxford Poverty and Human Development Initiative. (2019). Global Multidimensional Poverty Index 2019: Illuminating Inequalities.