Measuring Poverty in the Indigenous Population of Mexico

12 May, 2020

Eleonora Nun, international consultant.


The measurement of multidimensional poverty in the indigenous population of Mexico stems from two concerns. The first is a general concern about poverty. It has been established that the prevalence of poverty in the indigenous population – extreme poverty in particular – is consistently greater than in the rest of the population, and that it involves much broader factors than income. The end result is an accumulation of disadvantages that work against equal rights.

The second is a concern for what defines this population as ‘indigenous’. In this context, the hypotheses are a) that the indigenous population is more than just a group that speaks an indigenous language, which was how it had been defined up to 2010, b) that poverty indices could vary according to the definition of this group, and c) that different dimensions of the Indigenous could be related differently to the experienced deprivations.

How is it that a social attribute becomes an obstacle to the full realisation of rights? The perspective of multidimensional poverty is well-suited to shed light on which aspects of the indigenous condition best ex- plain the phenomenon of poverty. Inversely, this analysis enables the identification of the attributes of ‘the indigenous’ that allow the segments of the population that share certain deprivations to be grouped together.

Until 2010, there were a variety of sources of statistical information and ways of defining the indigenous population in Mexico. There was no cohesion among the different state institutions with regard to who should be considered indigenous, and the institutions, in turn, had varied their means of identification on multiple occasions. Therefore, a first step was to standardise the questions about ethnicity on the Population and Housing Census and the Socioeconomic Conditions Module of the National Survey of Household Income and Expenditure (MCS-ENIGH). Thanks to a collaboration between the National Institute of Statistics and Geography (INEGI) and the Nation- al Council for the Evaluation of Social Development Policy (CONEVAL), the decision was made for both tools to include a common question about ethnic identification in addition to the questions about spoken language that had, in accordance with the recommendations of international entities, been used for several years to identify the indigenous population.

Based on the information gathered in the 2010 Population and Housing Census and the 2008, 2010, and 2012 MCS-ENIGH, five analytical categories were built for the indigenous population: indigenous language speakers (ILS), the population in indigenous households, the population that identifies as indigenous, the indigenous language-speaking population living in non-indigenous households, and the population that speaks an indigenous language but does not consider itself indigenous. Poverty in the general population and the subset of the indigenous population was characterised based on this data, identifying both the specifics of the latter with respect to the rest of the population and heterogeneity among the analytical groups that make it up.

The measurement of multidimensional poverty in Mexico combines income-based poverty and social deprivations. Income is measured on a well-being line that indicates if income is sufficient to cover basic nutritional and non-nutritional needs, and a minimal well-being line that corresponds to people whose incomes allow them to cover nutritional, but not non-nutritional, needs. Well-being in terms of social rights includes the dimensions of education, health, social security, quality of housing spaces, and nutrition, and the number of deprivations experienced therein.

The more structural the ethnic relationship – language to identity, for example – the greater the probability of presenting deprivations in one or more dimensions.

Based on these measurements, a population is classified as poor (if they have one or more social deprivations and an income beneath the well-being line), extremely poor (three or more social deprivations and an income beneath the minimum well-being line, moderately poor (being poor but not extremely poor), vulnerable due to social deprivations (having an income above or equal to the well-being line, but experiencing one or more social deprivations), vulnerable due to income (no social deprivations, but income is beneath the well-being line), and not poor or vulnerable (no social deprivations and income above or equal to the well-being line) (CONEVAL, 2012).

One initial finding about indigenous poverty is that being indigenous is associated with more precarious living conditions. Being indigenous is associated with a higher likelihood of deprivations in all of the dimensions that comprise multidimensional poverty measurement as well as income-based deprivations (Figures 1 and 2).


With regard to the differences in the experience of poverty between the indigenous population and the general population, it can be observed that, despite presenting greater deprivations across all dimensions, those deprivations have decreased among the indigenous population at a pace similar to that experienced by the rest of the population. This group consistently experiences lower income levels than others in jobs in similar sectors. This would suggest that, while public investment has been effective in its focus on the poorest, certain mechanisms of discrimination work against the indigenous population in the job market that result in restricted access to higher income. This information is of great value to public policy discussions.

With regard to the definition of what constitutes being indigenous, the five analytical categories, based on the break-down of this concept, provide empirical evidence that bring to light important differences among the deprivations experienced within each one. The more structural the ethnic relationship – language to identity, for example – the greater the probability of presenting deprivations in one or more dimensions. In effect, among indigenous language speakers, the prevalence and intensity of poverty is greater, while those who identify as indigenous have characteristics closer to those of rest of the population. This methodology, by including networks and social cohesion indicators, makes it possible also to demonstrate how the environment impacts deprivations experienced in the different dimensions, and reveals the differences between the indigenous populations in rural areas and those who have migrated to the cities.

To summarise, having an official measurement of poverty by ethnic affiliation permits specific up- to-date measurements of their situations of poverty, disaggregated by the five categories – all of which contributes to the ongoing debate about what constitutes the indigenous individual. This work informs the design of more effective and focused public policies for the reduction of multidimensional poverty among indigenous groups.


This article was published in Dimensions 9


Indigenous population México Multidimensional Poverty