Research

Working Papers

Income and the Demand for Food Among the Poor

with Marc F. Bellemare and Eeshani Kandpal (submitted)

[CGD working paper]

How much do the poor spend on food when their income increases? We estimate a key economic parameter - the income elasticity of food expenditures - using data from the randomized evaluations of five conditional cash transfer programs in Mexico, Nicaragua, the Philippines, and Uganda. The transfers provided routine, exogenous increases of 12 to 23 percent of baseline income for at least a year to recipients at or below the global poverty line. Using pooled ordinary least squares and Bayesian hierarchical models, we first show that expenditures on all food categories increase with income. But even among some of the poorest people in the world, all of whom are experiencing high hunger levels, our estimated income elasticity for food is 0.03, i.e., much smaller than many published estimates that either rely on cross-sectional variation or study responses to large income shocks. Next, we run the first credible test of Bennett's Law - the empirical regularity whereby poor households respond to income increases by (i) shifting spending from coarse to fine staples, or (ii) spending more on protein than staples - and find partial support for it. While income increases lead consumers to substitute fine grains for coarse grains and protein for staples, again the estimated shifts are smaller than previous estimates. Quantifying how small and routine income changes affect food demand in low- and middle-income countries can inform the policy discourse on poverty reduction, nutrition, and social protection, as well as the debate on the impact of economic growth on global carbon emission patterns.


How to Close the Skill Gap?  Parental Background and Children's Skill Development in Indonesia

[Working paper] [Slides]

Preexisting inequalities in socioeconomic status can drive differences in children's cognitive skill development and parents' reactions to child development policies influencing policy effectiveness. To analyze the role of parental background and investments (nutrition diversity and schooling expenditure) in this process, I estimate a dynamic structural model using data from Indonesia. Using the model, I simulate three policies: unconditional cash transfers, nutrition, and schooling price subsidies. To compare their long-run effects on adult skills, I account for parents adjusting their investment behavior in response to policies. Given the same cost, a) subsidizing food prices is more effective than subsidizing schooling expenditure, and b) both are more effective than cash transfers. As I find nutrition and schooling to be complements, a price decrease incentivizes parents to increase both inputs. With cash transfers, parents also increase investments but increase consumption relatively more as price incentives do not change. Nutrition subsidies reduce inequality most effectively, as parents with lower education react stronger to food price changes and, consequently, increase child investments more than parents with higher education. They do so as they spend a larger share of investments on nutrition. Further, nutrition subsidies implemented alone are more cost-effective than any combination of the three policies.


Parenting Styles and Children's Skill Development

with Jacek Barszczewski

[Working paper]

This paper studies how parental behaviors, specifically warmth, inconsistency, reasoning, and hostility, influence the development of cognitive and non-cognitive skills during middle childhood and adolescence. Using rich Australian panel data, we present novel evidence that reporting bias in parent-reported measures of children's skills is driven by parenting style. To address this bias and consistently estimate the impact of parenting style on skill development, we employ fixed effects and use past investments as instruments for current investments. To demonstrate that our approach mitigates the bias, we also present results using teacher-reported measures. We find that parental hostility, lack of praise and anger during punishments, negatively impacts non-cognitive skills, decreasing them by 0.12 to 0.23 SD depending on age. Inconsistency in enforcing rules negatively impacts skills in middle childhood but not adolescence, decreasing skills by 0.08-0.10 SD. While parental warmth and reasoning do not influence emotional or behavioral problems, warmth does have a positive impact on prosocial behaviors of children. Cognitive skills are less affected by parenting behavior variations, parental warmth reduces skills by 0.03 SD and inconsistency by 0.07 SD for vocabulary and matrix reasoning tests. In contrast, we find impacts for hostility on school performance, similar in direction as for non-cognitive skills suggesting that non-cognitive skills influence performance. These results highlight the potential effectiveness of interventions focused on reducing parental hostility and enhancing consistency in boosting skill development, thereby contributing to children's human capital formation.

Work in Progress


Pre-school Education and Long-Run Human Capital Formation

with Elisabetta Aurino, Sergi Quintana and Sharon Wolf

Early childhood interventions such as preschools are increasingly used to enhance skill development and reduce learning poverty. However, evidence on their long-term effects remains limited, with some studies indicating a fade-out of initial gains. In this paper, we aim to investigate the mechanisms that may contribute to the fade-out of these interventions by examining the long-term impacts of a preschool program in Ghana, targeted at improving teaching quality and parental awareness of the importance of parent's role in learning. Leveraging the exogenous variation from a randomized controlled trial of the preschool intervention, jointly with a structural model of skill formation and parental investments, we measure the effects on both cognitive and socio-emotional skills over time. Using rich collected panel data and variation in treatment, we aim to explore several potential mechanisms behind the fade-out effect, including low self-productivity of skills, parental beliefs about the production function, changes in parental investments in response to skill levels, and shifts in the production function due to the intervention.  We also propose a novel framework to estimate skill production functions that accounts for heterogeneous information across skill measures, thereby increasing precision in the estimation of long-term effects.


Gender norms and adolescents’ educational and career aspirations and expectations: Evidence from a survey experiment in Ghana

with Elisabetta Aurino and Chukwunonye Emenalo

[Slides]

This study examines the role of gender norms in shaping adolescents' educational and career aspirations and expectations in Ghana, using a survey experiment with 2,400 participants aged around 13. Adolescents were randomly assigned to complete either a gender norms module or an aspirations module first to test whether priming gender norms influences their reported aspirations and expectations. The findings indicate that gender norm priming had no significant overall effect on aspirations or expectations related to university education, STEM careers, or female-dominated jobs. However, there were heterogeneous effects among girls based on the educational background and gender bias of their caregivers. Girls whose caregivers were less educated or held stronger gender biases reported lower educational expectations, while those with more educated or less biased caregivers expressed higher expectations after treatment. These findings suggest that priming gender norms alone may not significantly impact adolescents' aspirations, indicating that other factors contribute to a larger extend to gender inequalities.


Developing and Piloting a Context-driven Intervention for Adolescent Development in Ghana: The Pempamsie Family Program

with Elisabetta Aurino, Noelle M. Suntheimer, Richard Appiah, Sharon Wolf, Esinam Ami Avornyo, Jessica Fishman, Katie Brennan, David Djani, and Eric Howusu-Kumi

Culturally relevant parenting programs hold great potential to support adolescent health, yet evidence from low-income settings remains limited. Additionally, there is a lack of methodological guidance on the cross-cultural development of such programs, restricting their broader application. This paper details the interdisciplinary process used to develop, assess the feasibility of, pilot-test, and refine the Pempamsie Family Program (PFP), a novel, multi-component family-based intervention designed to promote adolescent health and family well-being in Ghana. Following the Medical Research Council’s framework for developing complex interventions, we proceeded step-wise. The first two steps involved identifying key parenting behaviors and implementation modalities, drawing on insights from positive and developmental psychology, cognitive-behavioral interventions, a review of the evidence, and formative research employing qualitative and behavioral science tools. Based on these insights, the PFP targeted stress management and resilience, positive family relationships, and strategies for managing adolescent problem behaviors. The program was delivered through a ‘hybrid’ model combining three manualized in-person group sessions with parents and adolescents, and biweekly SMS messages to parents to reduce costs and increase accessibility. The third and fourth steps involved drafting and pilot-testing a preliminary version of the program with 52 parent-adolescent dyads using a pre-post, mixed-method design. Endline evaluation showed improvements in parental knowledge, reduced stress, and decreased use of violence, although other target areas showed no significant change. Following expert reviews, the program was further refined. This paper highlights the potential of a ‘hybrid’ approach, combining digital and in-person components, to improve adolescent health and family functioning in urban and semi-urban Ghana. More broadly, our step-wise interdisciplinary methodology offers a useful framework for researchers and practitioners aiming to develop contextually relevant, evidence-based, and cost-effective behavioral health interventions.


Food Prices, Program Saturation and Cash Transfers

with Jed Friedman, Eeshani Kandpal and Patrick Premand


Ongoing projects


The Origins of Parenting: the Role of Beliefs, Preferences, and Constraints

with Elisabetta Aurino, John Egyir and Sharon Wolf

[Slides]

In this project, we explore the factors that influence parental choices of parenting styles in a low- and middle-income country context, with a focus on beliefs, preferences, and constraints. Using a sample of 2,400 parents of adolescents in Ghana, we combine a lab-in-the-field experiment with a theoretical model to examine how parents choose between authoritarian (characterized by low warmth and harsh discipline) and authoritative (marked by high warmth and firm but non-harsh discipline) styles. Using experimental variation, we elicit parental beliefs regarding the effectiveness of these styles, the costs associated with implementing them, and the trade-offs between time spent with children and other activities. Our goal is to identify the key determinants of parenting style decisions. Additionally, our sample is part of a randomized controlled trial of a parenting program that promotes positive strategies, enabling us to analyze how such interventions affect beliefs and behaviors. Doing so, we aim to offer insights into the mechanisms underlying parenting programs, contributing to the design of more effective interventions.