Moving forward Multidimensionally: Five Lessons on Child Poverty from Malawi and More

In the past four years that I have been working on several national child poverty studies, either conducting the analysis directly or providing assistance and feedback, a few key findings have emerged as consistent poverty-related or poverty-driven issues that need to be addressed. These five areas should be considered priorities in meeting Sustainable Development Goal 1 – ending child poverty everywhere:

  1. Children who live in rural areas* AND in a family that works in the agricultural sector are more likely to be deprived, and poor.*Although it is important to note that it’s not only that children in rural areas are more likely to be poor and deprived, but children in families whose main activity is agricultural work, especially where heads of household are engaged in casual agricultural work, or the household  is able to produce only the bare necessary for survival, and does not have the means to produce enough to sell their products on the market (what is known as ‘subsistence farming’). In Malawi, if the head of the household is a casual laborer in agriculture, children are 7 percentage points more likely to be deprived than in the case that the head of household is inactive. In Tanzania, children of small farmers are equally more likely to be deprived. A similar pattern is found in Mozambique. In Armenia, children in rural households are 20 percentage points more likely to be deprived, and all else being equal, children of farmers are again at higher risk of deprivation, if to a smaller degree (1.2 pp). Households that live in remote rural areas, not well connected to markets, and/or without enough means to have a productive activity, are not only poorer in terms of money, but their children will be more likely to be malnourished, miss school, not have access to water, and so on. If we needed another reason to invest in agriculture and make it more productive and profitable for small rural households, this is definitely one.
  2. Orphan children are not in surveys. The issue of children who live without their biological parents, either because they are dead or because they live elsewhere, is a complicated and delicate one. In most cases, orphan or abandoned children who are really destitute and living on the streets or in shelters, are not captured by standard surveys. This means we are missing their stories, their data, as relevant to child poverty. On the contrary, some of the children we find in household surveys, and who live without their parents, tend to do better than other children. For example, in Tanzania, children living without parents are slightly less likely to be deprived, and the families they live with are less likely to be poor in monetary terms. What we capture is those children who have been sent to richer relatives for education purposes, or have been taken in after the death of their parents. In some cases, these children can be house servants, working for the family they live with: while they are certainly deprived in other, sometime invisible, ways, they are likely to score better on an array of deprivations (since a family that hires a household servant will likely be richer than others). In either case, the share of these children found in household surveys tend to be small. We need to find better, more adequate instruments to understand the lives of these children and how they can be made better.
  3. Gender differences are complicated. First, we have to accept that sometimes gender differences do not go in the direction we expect, or are simply not there. We need to understand that gender differentials are a) specific to the context. While a general discrimination against girls and women is common everywhere, there are specific ways in which this happens. Gender discrimination in India is not the same as in Eastern Africa b) happening later in life: gender differences in deprivation tend to appear in adolescence, this is when expectations around gender roles really begin to weigh on children. c) hidden: many instances of discrimination against girls are hidden by the data we use. For example, the definition and data we use for child labor hides the many hours of household chores and caring that girls are more likely to do. At the same time, many gender-specific deprivation may be hidden and not collected, like gender based violence, FGM, and so on. At the same time, it’s important to be receptive to the fact that boys can be more vulnerable to deprivation, due to the same gender roles expectations – for example, boys are more likely to substitute for adult labor, and drop or delay school.
  4. Education is key. Across the board, education of parents and guardians is one of the strongest correlates of child poverty. For the same level of household expenditures, and the same type of employment, one year more of education of the mother or the head of household is strongly related to a decrease in the likelihood of deprivation of the child. In Tanzania, if the mother has even some incomplete primary education, the probability of a child to be deprived can decrease up to 11 percentage points for children under five; in Malawi, a head of household who has completed primary education decreases the probability of an adolescent to be deprived by 20 percentage points. Across eleven Arab countries, the education of the head of household was the strongest correlate of deprivation. This means that investing in education is not only a matter of fairness and equity for children now, but also for their future children.

    Jena, 15, uses chalk to write on the wall of the abandoned house she shares with her mother and four siblings, in Zanzibar Island. Education is key to reducing child poverty.
  5. Infrastructural problems result in high child deprivation. Whether it’s lack of adequate sanitation in Tanzania, overcrowding of people per room in Malawi, or heating devices in Armenia, issues pertaining to housing, sanitation, and access to information infrastructure are often the major drivers of child deprivation. Far from being a case to just use structural, household indicators to measure multidimensional child poverty, it’s a reminder that not all deprivations can be tackled by demand-side intervention, and particularly by increasing household monetary power. While policies that support families’ income, such as social protection programs, are critical, children also need access to basic services, and supply side interventions to improve the availability of proper sanitation, water supply, connection to infrastructure, and so on.


These points highlight important facts about child poverty that need to be kept in mind moving forward with measuring and, most importantly, addressing child poverty. Some of these highlight the limitations of the data we routinely use. This is not a call to stop using them, but to work to improve them, and to complement that information with additional evidence.

Read more:

Lucia Ferrone is a research consultant on child poverty at UNICEF Innocenti. You can follow her on Twitter @lucyferr85.

Explore the UNICEF Innocenti research catalogue for new publications. Follow UNICEF Innocenti on Twitter and sign up for e-newsletters on any page of the UNICEF Innocenti website.

Leave a reply

Your email address will not be published. Required fields are marked with “required.”