Innocenti Working Papers

This report provides the first comprehensive national estimates of multidimensional child poverty in Armenia, measured using UNICEF’s Multiple Overlapping Deprivation Analysis (MODA) methodology. Dimensions and indicators for three age groups (0-5, 6-14 and 15-17) were selected as the result of a broad consultative process with key stakeholders convened by UNICEF Armenia. Based on nationally representative data from the Armenian Integrated Living Conditions Survey 2013/14, the study finds that 64 per cent of children under 18 are deprived in 2 or more dimensions, with a substantially higher rate in rural than in urban areas. The highest rates of deprivation are in access to utilities, quality housing and leisure activities. More than one in four children are both multidimensionally deprived and live in consumption-poor households, while more than one in three are deprived but do not live in poor households. The findings suggest that to target the most vulnerable children, policies should concentrate on closing the rural/urban divide in infrastructure and on strengthening social safety nets, especially in rural areas.

Worldwide close to 800 million people are reached by state-operated cash transfer programmes. In sub-Saharan Africa, the poorest region in the world, the number of cash transfer programmes has doubled in the last five years and reaches close to 50 million people. What is the impact of these programmes, and do they offer a sustained pathway out of ultra-poverty? In this paper we examine these questions using experimental data from two unconditional cash transfer programmes implemented by the Government of Zambia. We find far-reaching effects of these two programmes, not just on their primary objective, food security and consumption, but also on a range of productive and economic outcomes. After three years, we observe that household spending is 59 per cent larger than the value of the transfer received, implying a sizeable multiplier effect. These multipliers work through increased non-farm business activity and agricultural production.

Seasonality in agricultural production continues to shape intra-annual food availability and prices in low-income countries. Using high-frequency panel data from northern Ethiopia, this study attempts to quantify seasonal fluctuations in children's weights. In line with earlier studies, we document considerable seasonality in children’s age and height adjusted weights. While children located closer to local food markets are better nourished compared to their counterparts residing in more remote areas, their weights are also subject to considerable seasonality. Further analysis provides evidence that children located closer to food markets consume more diverse diets than those located farther away. However, the content of these diets varies across seasons: children are less likely to consume animal source foods during the lean season. This leads us to conclude that households located near these food markets are not able to insulate their children from seasonal weight fluctuations. We discuss some policy options with potential to address this threat to child well-being.

While a long-standing literature analyses cross-country variation in the incidence of child poverty in rich countries in a single year, less is known about children’s individual movements into and out of low household income over a period of time. Using longitudinal data from the European Union Statistics on Income and Living Conditions (EU-SILC), the present study addresses this gap by analysing both income mobility and child poverty dynamics in the EU during the recent economic crisis. It finds that income growth among children has been generally pro-poor but not sufficiently so to put a brake on the increasing income inequality. There is substantial heterogeneity among the EU-SILC countries in the rates of child poverty entry and exit. Scandinavian countries tend to combine lower exit and entry rates, while Southern and Eastern European countries tend to have higher rates of both poverty exit and entry. Household-level income events, i.e. relative growth in employee earnings, are found to be the most important predictors of transitions in and out of poverty, followed by employment events (i.e. changes in the number of adult workers), while the relatively rare demographic events have little bearing on child poverty transitions in the EU-SILC.

Existing evidence is inconclusive on whether a socio-economic gradient in children’s cognitive ability widens, narrows or remains stable over time and there is little research on the extent of ‘cognitive mobility’ of children who had a poor start in life compared to their peers. Using data from five sweeps of the United Kingdom (UK) Millennium Cohort Study (MCS) at the ages of 9 months, 3 years, 5 years, 7 years and 11 years, this paper explores the cognitive ability trajectory of children in the bottom decile of the distribution at a given age, and the factors that drive or hinder their progress relative to their peers. Using discrete-time event history analysis methods, the paper analyses children’s risks of moving in and out of the bottom decile of the cognitive ability distribution. The findings indicate a relatively high level of cognitive mobility between ages 3 and 11, especially in the pre-school period (between ages 3 and 5), with children from income-poor households more likely to get ‘trapped’ in the bottom of the age-specific cognitive ability distribution. Parental education plays a dual buffer role: it protects children from falling behind their peers as well as increasing the chances of moving up the ability distribution. Parental involvement, such as reading to a child and regular bedtimes at the age of 3, was found to protect children from falling into the bottom group of cognitive distribution but not necessarily helping lower scoring children to move up.

In the context of increasing child poverty, deprivation rates and the relative child income gap, and with the most economically vulnerable children hit extensively by the crisis (Chzhen 2014), this paper sets out to understand who are the most disadvantaged children. Analysis of the composition of the children at the bottom end of the income distribution illustrates that households with a lone parent, at least one migrant member, low work intensity, low education, or in large families are overrepresented in the first decile to different degrees in European countries. The analyses also reveal immense differences in living standards for children across Europe. In European countries included in the analyses, at least 1 in 5 children in the poorest decile lives in a deprived household. A closer look at the different dimensions of deprivation at the child-specific level, reveals what living in the poorest decile means for children’s everyday life. Children in the bottom end of the income distribution are prone to a lack of a suitable place to study or do homework. The shares of children in the poorest decile living in a household that cannot afford fruit and vegetables daily or one meal with meat or protein at least once a day are worryingly high. They can also be considered to be deprived in relation to social aspects such as insufficient resources at home to provide regular leisure activity, or to invite friends to play or eat from time to time, or to participate in schools trips and school events that cost money.

A large body of literature has established socio-economic gradients in adolescent health, but few studies have investigated the extent to which these gradients are associated with very poor health outcomes. The current analysis examined the extent to which the socio-economic background of adolescents relates to very poor self-reported health and well-being (the so-called ’bottom end’). For this analysis, we use data from the last four cycles of the Health Behaviour in School-aged Children study (2001/02, 2005/06, 2009/10, and 2013/14) for 32 European and North American countries in the EU and/or OECD. We examined the following as indicators of adolescent health: psychosomatic health complaints; physical activity; healthy eating; unhealthy eating; and life satisfaction. Adolescents who scored below the mean of the lower half of the distribution of a given indicator fall in the “bottom group” on this indicator. Family affluence is used as a measure of adolescents’ socioeconomic background (transformed to ridit score). In the vast majority of the countries studied, adolescents from families with a relatively low socio-economic status had a greater likelihood of reporting poor health. The largest, most persistent and widespread socio-economic gradients are in life satisfaction, physical activity and healthy eating, while the findings are mixed for unhealthy eating and psychosomatic health. Socio-economic inequalities were largely stable, but in a sizeable minority of the countries, socio-economic inequalities in physical activity and healthy eating have widened between 2001/02 and 2013/14, while inequalities in unhealthy eating and life satisfaction have narrowed in several countries.

Measuring inequality of learning outcomes in a way that provides meaningful benchmarks for national policy while retaining a focus on those students who are ‘hard to reach’ and ‘hard to teach’ is a challenging but vital task in the light of the global post-2015 education agenda. Drawing on PISA 2012 data and its earlier rounds, this paper explores alternative approaches to measuring educational inequality at the ‘bottom-end’ of educational distribution within the cross-national context. Its main aim is to understand how far behind children are allowed to fall in their academic achievement compared to what is considered a standard performance in their country. Under the framework of relative (measured as achievement gap between the median and 10th percentile) and absolute (measured by the percentage of students achieving at a given benchmark) educational disadvantage it examines cross-country rankings as well as national trajectories with reference to overall academic progress. We find that on average across OECD countries around 11% of 15- year-olds lacked skills in solving basic reading, mathematical, as well as science, tasks in 2012, but variation across countries was large. The average achievement gap in mathematics across OECD countries between low-achieving and ‘average’ students stood at around 122 score points; in reading, at 131 score points; and in science, at 124 score points. This paper argues that understanding how the reduction in bottom-end inequality is achieved matters no less than the outcome itself, as it often reflects the level of support provided to low-achieving students. As our analysis shows, narrowing the achievement gap might be due to falling academic standards and have no direct benefit to the ‘bottom group’.

The health of children and adolescents in high income countries negatively relates to income inequality. Theoretical interpretations of this association suggest that inequality intensifies social hierarchies, erodes social or material resources that support health, or impacts socioemotional development in childhood and subsequently harms health. The evidence in support of this causal interpretation is limited by a reliance on cross-sectional, ecological studies. Using multilevel panel data from the Health Behaviour in School-aged Children (HBSC) study, this paper examines lagged and contemporaneous associations between national income inequality and health and well-being during adolescence. Health symptoms and life satisfaction were measured in successive surveys of 11- to 15-year-olds in Europe and North America between 1994 and 2014. These data were linked to country-level income inequality for each survey year (contemporaneous effects) and for earlier developmental periods, at 0-4 years and 5-9 years (lagged effects), dating back to 1979 – the birth year of 15-year-olds in the 1994 survey cycle. Societal growth curve modelling was used to pool data from successive survey cycles and to isolate age, period, and cohort effects. The results show evidence of lagged effects of income inequality during childhood (5-9 years) on health symptoms and life satisfaction in adolescents (11-15 years), after differences in concurrent income inequality and income per capita, cohort, time period, and individual gender, age, and affluence were held constant. This period of development for income inequality exposure coincides with the early school years when social relationships extend from the family to school and community settings. Inequality may shape child developmental trajectories in ways that later manifest in reduced health and well-being. Though not causal evidence in the strictest sense, these findings establish antecedent-consequence conditions in the association between income inequality and health. The practical and theoretical implications of these results are discussed.