Periodic Reporting for period 2 - UNEQUALWITHIN (From Household Allocations to Global Inequality: New Methods, Facts and Policy Implications)
Reporting period: 2025-02-01 to 2027-07-31
To understand and mitigate economic inequalities, both locally and globally, we need to acknowledge inequalities within households. Yet, in most empirical studies, such intra-household inequalities are disregarded mainly because we lack appropriate measurement tools and data. Not only is this problematic for inequality measurement, this lack of understanding hampers the design of cost-effective poverty reduction and child development policies. The ERC-funded UNEQUALWITHIN project examines the allocation of resources and decision making within households in the form of a global outlook combined with an-in depth focus on households in Tanzania. The work of the project will lead to an integrated framework, new tools and data that will shape and improve our understanding of the mechanisms behind consumption inequalities among adults and child development.
The project UNEQUALWITHIN has five general objectives. First, the project’s research results will help to update known facts about inequalities through direct measurements of intra-household consumption allocations, and relate these to spousal income contributions. Second, the project team will develop and validate novel measures of parental resource-allocation preferences and use these to study whether children are likely to benefit more if mothers, rather than fathers, receive cash transfers. Third, the team will develop and validate novel measures of household decision-making and use these to investigate how targeted transfers shape women’s empowerment. Fourth, the project has the objective to study whether cash transfers or an educational parenting program is most cost-efficient for child development. Fifth, results are consolidated in an integrated framework using the new tools and data. This will refine our understanding of the mechanisms behind inequalities among adults and child development.
Concretely, the project will contribute to the knowledge base on intra-household decision making and child development in the following ways. We will carry out an extensive data collection on intra-household allocations, parental-allocation preferences, and women’s empowerment in a set of ten diverse countries, one from each decile of the world income distribution. We will engage in an early child development RCT in Tanzania that tests the effect of regular cash transfers and a parenting intervention. The RCT also involves extensive data collection on household consumption, time use, preferences and decision-making. Finally, we will conduct lab experiments to validate parental-allocation preferences and decision-making measures.
The project UNEQUALWITHIN has five general objectives. First, the project’s research results will help to update known facts about inequalities through direct measurements of intra-household consumption allocations, and relate these to spousal income contributions. Second, the project team will develop and validate novel measures of parental resource-allocation preferences and use these to study whether children are likely to benefit more if mothers, rather than fathers, receive cash transfers. Third, the team will develop and validate novel measures of household decision-making and use these to investigate how targeted transfers shape women’s empowerment. Fourth, the project has the objective to study whether cash transfers or an educational parenting program is most cost-efficient for child development. Fifth, results are consolidated in an integrated framework using the new tools and data. This will refine our understanding of the mechanisms behind inequalities among adults and child development.
Concretely, the project will contribute to the knowledge base on intra-household decision making and child development in the following ways. We will carry out an extensive data collection on intra-household allocations, parental-allocation preferences, and women’s empowerment in a set of ten diverse countries, one from each decile of the world income distribution. We will engage in an early child development RCT in Tanzania that tests the effect of regular cash transfers and a parenting intervention. The RCT also involves extensive data collection on household consumption, time use, preferences and decision-making. Finally, we will conduct lab experiments to validate parental-allocation preferences and decision-making measures.
The work in UNEQUALWITHIN is centered around three main components: (i) the development of novel measurement tools and survey instruments, (ii) the collection of data in a randomized controlled trial and a lab-in-the-field experiment, and (iii) a global study on establishing facts about local and global inequalities.
Several strands of progress have been achieved before the project was terminated. One notable example is the development of new measurements for intra-household decision making and parental resource allocation preferences. This has culminated in a publication in a high-ranking international journal Econometrica, titled “Economics and Measurement: New Measures to Model Decision Making” (Almås, Attanasio and Jervis, 2024), which draws on and discusses work from UNEQUALWITHIN.
Furthermore, the project has developed new measurement tools as the basis for the data collection that will ultimately inform the formulation of an integrated framework to improve the understanding of the mechanisms behind intra-household consumption inequalities and child development. A number of data collection efforts were successfully finalized during the project period; however, the lab-in-the-field experiment in India and the global study remains unimplemented at the time of project completion. First, the project team has contributed to an RCT that implements two interventions to improve early child development outcomes but may also have effects on intra-household inequalities with respect to parental preferences for resource allocation and decision-making: (i) a regular cash transfers, and (ii) a parenting program. The successful implementation of the RCT baseline and midline data collection with more than 3,000 households is another achievement of the project. Secondly, the project team has conceptualized, prepared, and implemented a lab-in-the-field experiment in Tanzania with the related RCT sample, in which the survey instruments were validated for the Tanzanian context. For the lab-in-the-field experiment in Tanzania, participants were randomized to answering the task hypotactically or incentivized. The task consisted of three stages which mean the participants first allocated a budget among themselves, their partner, and the target child, then divided each allocation into spending categories, and those who were incentivized could make immediate purchases from a pop-up shop. The incentivized and hypothetical conditions produced statistically indistinguishable allocations, providing support for the use of cost-effective hypothetical elicitation in future large-scale surveys. Design features where also documented which helped safeguard internal validity, such as recording any reallocations at the shop and conducting follow-up calls to detect resale or crowding-out effects. The lab-in-the-field experiment in Tanzania is a stepping stone towards testing these measures also in other contexts and ultimately developing new research tools that can significantly advance the research discipline beyond the state of the art. Congruently, the interplay of RCT, lab experiment allows us to study behavioral preferences not in the isolation of the laboratory environment but in front of a backdrop of rich contextual data. This marks the first step towards an integrated and universal framework.
Several strands of progress have been achieved before the project was terminated. One notable example is the development of new measurements for intra-household decision making and parental resource allocation preferences. This has culminated in a publication in a high-ranking international journal Econometrica, titled “Economics and Measurement: New Measures to Model Decision Making” (Almås, Attanasio and Jervis, 2024), which draws on and discusses work from UNEQUALWITHIN.
Furthermore, the project has developed new measurement tools as the basis for the data collection that will ultimately inform the formulation of an integrated framework to improve the understanding of the mechanisms behind intra-household consumption inequalities and child development. A number of data collection efforts were successfully finalized during the project period; however, the lab-in-the-field experiment in India and the global study remains unimplemented at the time of project completion. First, the project team has contributed to an RCT that implements two interventions to improve early child development outcomes but may also have effects on intra-household inequalities with respect to parental preferences for resource allocation and decision-making: (i) a regular cash transfers, and (ii) a parenting program. The successful implementation of the RCT baseline and midline data collection with more than 3,000 households is another achievement of the project. Secondly, the project team has conceptualized, prepared, and implemented a lab-in-the-field experiment in Tanzania with the related RCT sample, in which the survey instruments were validated for the Tanzanian context. For the lab-in-the-field experiment in Tanzania, participants were randomized to answering the task hypotactically or incentivized. The task consisted of three stages which mean the participants first allocated a budget among themselves, their partner, and the target child, then divided each allocation into spending categories, and those who were incentivized could make immediate purchases from a pop-up shop. The incentivized and hypothetical conditions produced statistically indistinguishable allocations, providing support for the use of cost-effective hypothetical elicitation in future large-scale surveys. Design features where also documented which helped safeguard internal validity, such as recording any reallocations at the shop and conducting follow-up calls to detect resale or crowding-out effects. The lab-in-the-field experiment in Tanzania is a stepping stone towards testing these measures also in other contexts and ultimately developing new research tools that can significantly advance the research discipline beyond the state of the art. Congruently, the interplay of RCT, lab experiment allows us to study behavioral preferences not in the isolation of the laboratory environment but in front of a backdrop of rich contextual data. This marks the first step towards an integrated and universal framework.
The work in the UNEQUALWITHIN project has focused on the conceptualization and development of novel survey instrument for the measurement of (i) intra-household decision making power and (ii) parental preferences for household resource allocation between oneself, the spouse, and a child.
The development of the survey instruments is expected to generate breakthroughs that advance the research field beyond the current state of the art, particularly once integrated into a comprehensive model of household decision making. To achieve this, we need to collect all necessary data from lab experiments, surveys, and the RCT in Tanzania to calibrate and estimate the model. The framework is then expanded using experimental and survey data from India, ultimately producing a fully integrated model for nine additional countries. Both the Tanzania model and the multi-country extensions are anticipated to represent significant advancements, providing novel, validated tools to study the impact of policies and shocks on households with young children and the consequent effects on child development.
The development of the survey instruments is expected to generate breakthroughs that advance the research field beyond the current state of the art, particularly once integrated into a comprehensive model of household decision making. To achieve this, we need to collect all necessary data from lab experiments, surveys, and the RCT in Tanzania to calibrate and estimate the model. The framework is then expanded using experimental and survey data from India, ultimately producing a fully integrated model for nine additional countries. Both the Tanzania model and the multi-country extensions are anticipated to represent significant advancements, providing novel, validated tools to study the impact of policies and shocks on households with young children and the consequent effects on child development.