This article uses conventional Engel curve demand analysis, as well as a double hurdle model, to explore whether there are intrahousehold differentials in the allocation of education expenditure between boys and girls in rural Sri Lanka. Contrary to the case in most developing countries, in Sri Lanka, there was a significant bias favoring girls for 1990-91 and 1995-96 for age group categories 8-9, 14-16, and 17-19, and in 2000 for age group categories 14-16 and 17-19. Significant differences in enrollment favoring girls aged 17-19 explain part of the girl bias observed in 1990-91 and 2000-2001, but most of the bias is driven by positive expenditure given enrollment. The biases favoring girls are observed at critical stages of the schooling career in the run-up to key national exams. The 8-9 age group captures the run-up to the year 5 scholarship exams that are used to gain entry to the better-performing secondary schools. The 14-16 and 17-19 age groups capture those who read for important national-level qualifications that are vital for the job market. This article also looks at various possible explanations for the bias. (c) 2010 by The University of Chicago. All rights reserved..

This paper looks at patterns of growth faltering and catch up of around 1000 children as they moved from 8 to 19 years of age, from middle childhood through adolescence to young adulthood, using Height for Age Difference (HAD) and the more conventional Height for age z-scores (HAZ). It also looks at what individual and household characteristics may have moved these children into or out of situations of nutritional deprivation and how their stunting profile in later childhood correlates with psychosocial outcomes at age 19 and how it may have intergenerational consequences. The paper uses 4 rounds of longitudinal data collected in 2002, 2006, 2009 and 2013 from Andhra Pradesh and Telengana, India when the children were aged 8, 12, 15 and 19. The paper finds that there are significant gender based biases in growth faltering later in childhood disfavouring girls and that becoming newly stunted as an adolescent is strongly correlated with a child reporting to have poorer relationships with peers compared to the group that were never stunted. We also find that a girl experiencing stunting in middle childhood or adolescence (even if they were not stunted at age 8 or eventually moved out of being stunted by age 19) correlates significantly with offspring being shorter and thinner than the offspring of girls never stunted. This is one of few, if any, studies that look at growth patterns in middle childhood and adolescence and outcomes as a young adult and the results are important for their implications for further research and policy.

This paper asks whether an exogenous increase in income in the context of a poverty alleviation program has an impact on child anthropometric outcomes. The study looks at the Samurdhi Program in Sri Lanka and uses household data for 1999/2000. Using propensity score matching to account for selectivity bias, the paper finds that Samurdhi improves the height for age z-score of a child from a grant receiving family by roughly 0.40 standard deviations with the impact driven mainly by children between six to 36 months of age, compared to if they did not receive the grant. It also improves weight for height z-scores by around 0.45 standard deviations of children aged 36-60 months. These results are important for Sri Lanka where child nutrition is a cause for concern.

This study employs a pseudo-panel approach to estimate the returns to education among income earners in Sri Lanka. Pseudo-panel data are constructed from nine repeated crosssections of Sri Lanka's Labor Force Survey data from 1997-2008, for workers born during 1953-1974. The results show that for males, one extra year of education increases monthly earnings by about 5 per cent using the pseudo panel estimation rather than 9 per cent as in the OLS estimation. This indicates that not controlling for unobservables such as ability and motivation, bias the OLS estimation of returns upwards by about 4 per cent on average, driven mainly by what happens in urban areas. It also suggests that males with higher ability seem to be acquiring more years of education. This is contrary to what has been observed recently in countries such as Thailand (Warunsiri and McNown 2010) where the opportunity cost of education seems to be high, such that high ability individuals leave education for the labour market.

This paper investigates whether the death of a parent during middle childhood (ages 7-8 to 11-12) has different effects on a child's schooling and psychosocial outcomes when compared with death during adolescence (ages 11-12 to 14-15) in Ethiopia. The data come from three rounds of the Young Lives longitudinal survey, conducted in 2002, 2006 and 2009, of a sample of around 850 children across 20 sentinel sites in Ethiopia. The results show that when a child's mother dies in middle childhood, it has a significantly negative impact on school enrolment. A parent's death also has a significant negative impact on a child's sense of optimism about the future. These effects are short term in nature and do not persist into adolescence. However, the children orphaned in middle childhood engage in significantly more paid employment and self-employment at age 14-15. In contrast to maternal death in middle childhood, maternal death in adolescence has no impacts on any of the outcomes considered in our sample. However, the death of a father in adolescence has a significant negative impact on school enrolment, maths scores and a child's sense of agency. It is unclear as to why this is the case, as these orphans do not seem to engage in more employment than others and there have been no significant disruption to caregiver arrangements. It is likely that the † I would like to thank Stefan Dercon, John Strauss, Adrian Wood and two anonymous referees for helpful comments on a previous version of this paper. All errors are mine.