Know your Place - Employment Decisions among Couples

This is a series of incentivized choice experiments (see details in the Experimental Design section).

2026-03-04

2026-05-31

Our primary outcomes are the shares of men and women who choose a norm-conforming bundle over a norm-violating bundle in job offers across treatment arms.

Our secondary outcomes are men's and women's willingness-to-pay to choose a norm-conforming bundle over a norm-violating bundle in job offers across treatment arms.

We recruit married couples (husbands and wives) aged 18 to 60 from rural villages in Bangladesh. Eligible participants must have engaged in income-generating work for at least a month in the recent past and must be permanent residents of the district.

Participating couples are invited to complete a one-day paid job and a survey session at a field office. The job consists of evaluating audio descriptions of images on a tablet. In addition, participants complete a proxy household chore task. Participants receive a salary for completing the job and study. Both husband and wife are required to complete the job and study to receive the salary.

Each spouse makes a series of incentivized binary choices between two bundled job offers. Each bundle specifies a job offer for the husband and a job offer for the wife. Participants cannot choose a job for only one spouse: each decision assigns positions to both.

Each job offer varies along three dimensions:
• Income: Randomized wage levels
• Prestige: Job titles with predefined hierarchy
• Chore allocation: Minutes spent separating stones from rice

Each participant makes a series of binary choices between two bundled job offers: a Benchmark bundle and an Alternate bundle. Each bundle specifies one job offer for the husband and one for the wife.

In the Benchmark bundle the husband is always "better" off than the wife across all dimensions (higher wage level, higher prestige, lower chore allocation), complying with both MBN and the FHN. We exogenously vary whether the wife has a higher salary, a higher prestige, and a lower chore allocation in the Alternate bundle across treatment arms, violating either the MBN, the FHN, or both.

Treatment Dimensions

• T1: Income + Prestige + Chores (I + P + C)
The wife has a higher income, a higher prestige, and a lower chore allocation than the husband in the Alternate bundle.

• T2: Income + Prestige (I + P)
The wife has a higher income, a higher prestige, and a higher chore allocation than the husband in the Alternate bundle.

• T3: Income Only (I)
The wife has a higher income, a lower prestige, and a higher chore allocation than the husband in the Alternate bundle.

• T4: Prestige Only (P)
The wife has a lower income, a higher prestige, and a higher chore allocation than the husband in the Alternate bundle.

• T5: Chores Only (C)
The wife has a lower income, a lower prestige, and a lower chore allocation than the husband in the Alternate bundle.

All choices are incentive compatible: any choice has a positive likelihood of being the final bundled job offer.

In addition to binary choices, participants complete willingness-to-pay (WTP) elicitation tasks using a Becker-DeGroot-Marschak (Becker et al., 1964) mechanism. Participants receive 400 tokens and state the maximum number of tokens they are willing to pay to switch from one bundle to another.

We also test whether participants compensate for violations of either the MBN or the FHN in terms of chore allocation and/or consumption decisions.

We measure baseline attitudes and characteristics to test for heterogeneity.

Not available

Each participant is shown all treatment variations (within-subject design). The order of the treatment comparisons and the order of the bundles within each comparison are randomized at the individual level by the survey program.

Within-subject design

No

Up to 800 couples (800 men and 800 women)

Up to 800 couples (800 men and 800 women)

N/A (choice experiment with many individual-level variations)