Fundamentals

Multidimensional Poverty: What It Is, How It Is Measured, and Its Limits

By Daniel Sardá · Published on October 22, 2026

6 min read1,265 words

Multidimensional poverty identifies simultaneous deprivations in essential areas of life. Understanding its rules makes it possible to use the indicator without mistaking it for a complete measure of well-being or an explanation of its causes.

Two households may have similar incomes yet live in very different circumstances. One may have safe water, adequate housing, and uninterrupted access to education; the other may face shortages in several of these areas at once. A measure based only on money may not directly capture those differences. Multidimensional poverty seeks to make them visible.

This does not mean income has ceased to matter. Monetary resources and actual living conditions answer related, but not identical, questions. The multidimensional approach looks at a defined set of deprivations, establishes when a person or household has accumulated enough of them to be identified as poor, and summarizes the results in one or more indicators.

Key idea: Multidimensional poverty is neither a universal list of needs nor a complete measure of well-being. It is an approach that makes explicit which deprivations will be observed and the rules used to combine them.

Monetary poverty and multidimensional poverty

Monetary poverty compares a person’s or household’s income or consumption with a specified line. Its strength is clear: money broadens the ability to choose among different goods and services, and it can summarize diverse resources in a common unit. But it does not always directly show whether services are available, whether they function, or whether a particular deprivation persists.

Multidimensional measurement begins from another angle. Rather than asking only how many resources a household has, it examines whether the household faces selected deprivations in areas such as health, education, or housing conditions. It also considers simultaneity: experiencing one isolated deprivation is not the same as accumulating several.

The two measures can complement each other. A household may sit above a monetary poverty line while facing significant deprivations; another may have temporarily low income while retaining certain basic conditions. Neither result alone exhausts the reality of people’s lives. Mexico’s CONEVAL methodology, for example, retains economic well-being and relates it to social deprivations within its own design.

Nor should poverty be confused with inequality. Poverty refers to shortfalls or deprivations relative to certain thresholds; inequality describes how resources or outcomes are distributed among people. A society may reduce one measure of poverty without reducing income inequality to the same extent, and vice versa.

How a multidimensional index is built

There is no single multidimensional poverty index for every place and purpose. The general approach becomes a specific measurement through several connected choices:

• Unit of analysis: this may be the individual or the household, depending on the data and the objective.
• Dimensions: the broad areas the measurement considers relevant.
• Indicators: observable features that operationalize each dimension.
• Deprivation thresholds: the points at which an indicator counts as a deprivation.
• Weights: the importance assigned to each indicator in the combined result.
• Identification rule: the level of accumulated deprivation that allows someone to be classified as multidimensionally poor.

The Alkire–Foster method, used in numerous indices, is often described as a dual-cutoff procedure. First, it determines who is deprived in each indicator. It then adds weighted deprivations and applies a second threshold to identify the population in multidimensional poverty. OPHI’s handbook for national indices explains these choices and shows how a national index can be adapted to local priorities and data.

For that reason, having one deprivation does not automatically make someone multidimensionally poor. Everything depends on the identification rule adopted. Nor should the dimensions of a particular index be presented as the universal definition of the concept.

Useful distinction: The multidimensional approach is the general idea; a multidimensional poverty index is a specific application, with defined indicators, cutoffs, and weights.

Incidence, intensity, and composition

Once the population in multidimensional poverty has been identified, it matters both how many people are in that situation and how many deprivations they face.

Incidence is the proportion of the population identified as poor. Intensity is the average share of weighted deprivations faced by that group. Two territories can have the same incidence while showing different intensities: in one, the identified population may accumulate more deprivations.

In the standard Alkire–Foster formulation, the adjusted headcount ratio—known as M0 or the MPI—combines incidence and intensity through the product H × A. This makes it possible to track both dimensions in a single figure. Yet the aggregate number should be read alongside its components and the contribution of individual indicators. Otherwise, improvement concentrated in one area may obscure stagnation or deterioration in another.

The UNDP and OPHI global Multidimensional Poverty Index is one specific application: it uses ten indicators grouped into three dimensions and a weighted cutoff of one third. Those rules serve that index; they do not define every possible MPI. National indices may select other indicators and thresholds to respond to their institutional frameworks and statistical needs.

Why two indices can tell different stories

Imagine two countries seeking to measure multiple deprivations. Country A gives greater weight to housing conditions and basic services; country B includes school attendance, employment, and social protection in greater detail. Even if both use a similar aggregation structure, their results do not answer exactly the same question.

Surveys, units of analysis, thresholds, and weights may also differ. Colombia’s methodological experience, with five dimensions and fifteen indicators in the design documented by DANE, illustrates how a country can operationalize the approach through its own architecture.

For this reason, comparisons are stronger when definitions, data sources, and calculation rules remain constant. Comparing the same index over time may be reasonable if the methodology and data quality remain compatible. Directly comparing different national indices requires much greater caution.

What the indicator reveals—and what it cannot prove

A well-designed multidimensional index can identify concurrent deprivations, show their intensity, and reveal which indicators contribute most to the result. It can guide questions, help prioritize information, and assess descriptive changes. Its transparency also makes it possible to discuss whether the dimensions and thresholds adequately reflect what a society seeks to observe.

But an index cannot by itself prove why deprivations exist. A decline in the indicator after a reform is not enough to establish causation: the economy, the composition of the population, the data, or several policies may have changed at the same time. Establishing causes requires evaluation designs and additional evidence.

The selection of dimensions, cutoffs, and weights combines technical knowledge with normative judgments. That does not make measurement useless or necessarily arbitrary; it makes documentation and scrutiny essential. Monetary poverty also depends on choices about poverty lines, prices, equivalences across households, and information sources.

From a perspective centered on individual liberty, material conditions must also be understood alongside people’s real capacity to choose. Institutions, incentives, the rule of law, and open access to opportunity shape the context in which people can overcome deprivations. Even so, an MPI does not by itself measure freedom or allow one to infer automatically which policy would best expand those capabilities.

Reading criterion: Before comparing figures, ask which dimensions were measured, where thresholds were set, how deprivations were weighted, and whether the data and rules are compatible.

The value of multidimensional poverty lies precisely in making visible a part of reality that a single variable can conceal. Its best use is not to replace every earlier measure or turn a number into a political verdict, but to describe more precisely who faces simultaneous deprivations and with what intensity. Read alongside income, disaggregated indicators, and institutional context, the index becomes an informative tool; read without its methodology, it can create a certainty the data do not justify.