Intra-Household (Mis-)Beliefs and Women’s Work Decisions in Pakistan

How to cite

Serra, Danila; Afridi, Farzana; Ahmad, Hamna; Robertson, Raymond. 2026. "Intra-Household (Mis-)Beliefs and Women’s Work Decisions in Pakistan." AEA RCT Registry. AEARCTR-0018327. https://doi.org/10.1257/rct.18327-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Danila Serra* , Farzana Afridi , Hamna Ahmad , Raymond Robertson

* Primary investigator

Registry ID
AEARCTR-0018327
DOI
https://doi.org/10.1257/rct.18327-1.0
First registered
Last updated
Status
In development
Trial dates
2025-04-01 – 2027-06-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/18327

Abstract

Female labor force participation in Pakistan is among the lowest in the world. Many working women are engaged in home-based, informal employment in the garment sector, where they earn low wages and lack protections against exploitation by contractors. We study the role of intra-household misaligned beliefs in shaping women’s labor supply decisions, focusing on whether incorrect perceptions about husbands’ support for outside employment act as a barrier to formal work. In April 2025 (our baseline), we surveyed 301 home-based working women in Lahore along with their spouses, to elicit wives’ and husbands’ preferences for home-based versus outside-home work, as well as their beliefs of their spouse’s preferences. The survey provided evidence of misalignment in intra-household preferences and beliefs. We revisit the household and randomize them to receive information about surveyed husbands’ support for women working outside the home. We evaluate the impact of this information on women’s participation in a community workshop that provides information about local job opportunities, and on subsequent employment outcomes.

Intervention

Approximately two-thirds of the couples surveyed at baseline are randomly assigned to receive an information intervention during household revisits. The intervention provides information on the aggregated responses of all surveyed husbands to a baseline choice experiment, in which husbands chose between a home-based job (Job A) and employment in a garment factory (Job B) for their wives at varying wage levels for Job B (moderate, high, or very high). In addition, conditional on the husband’s prior consent, women receive individualized information about their own husband’s choice between Job A and Job B in the baseline choice experiment. The spouses assigned to control are presented placebo information on the risk of dengue. We assess the impact of the information intervention on women’s participation in a community workshop providing information about job opportunities outside the home, and on subsequent labor market outcomes.

Experimental design

In April 2025, we surveyed 301 married women working for home for the garment sector in Lahore, Pakistan. We employed a randomized choice experiment that allowed participants to choose between hypothetical home-based and factory-based jobs (Job A and Job B, respectively) for the wife, keeping the characteristics of the home-based work constant (similar to the women’s current job), while varying, across subjects, wages of the factory-based job. We elicited own preferences and incentivized beliefs of the spouse’s preferences. We found evidence of misalignment in intra-household preferences and beliefs. In this study, we return to these households and randomize two thirds of them to receive information about the choice experiment’s aggregated answers of all surveyed husbands. In addition, conditional on husband’s consent, we present treated women with the their own’s husband answer in the baseline choice experiment. Control couples receive placebo information on the risk of dengue.

Design details

Randomization method
Randomization automated through Survey CTO. All households who consent to participate in the survey are automatically randomized into either the control (1/3) or the treatment group (2/3).
Randomization unit
Household level. This guarantees that husbands and wives are assigned to the same treatment group.
Clusters
301, assuming no attrition from baseline.
Observations
301 women and 301 men, married to each other.
Arms
Approximately 2/3 households assigned to treatment and 1/3 to control. - Treatment: Assuming no attrition, 200 women and 200 men (their husbands). - Control: Assuming no attrition, 101 women and 101 men (their husbands).
Primary outcomes
The primary outcomes of interest are: (i) stated interest in participating in a community workshop providing information on job opportunities outside the home; (ii) actual participation in the workshop; (iii) registration on an online job-matching platform; iv) job outcomes.
Secondary outcomes
The secondary outcomes are generated by our second household visits (4 to 6 months post intervention). They are measures of: 1) Women’s aspirations to work outside the home; 2) intra-household communication about wife’s work outside the home; 3) men’s and women’s attitudes and perceived norms about “women's freedom to work outside the home”; 4) Intra-household bargaining; 5) women’s mobility.

Cited by

No citation data available for this registration yet.

Keywords