Investment Decisions with Endogenous Budget Share Allocations Inside the Household

How to cite

Dupas, Pascaline; Buchmann, Nina; Ziparo, Roberta. 2019. "Investment Decisions with Endogenous Budget Share Allocations Inside the Household." AEA RCT Registry. AEARCTR-0004069. https://doi.org/10.1257/rct.4069-1.0

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

Pascaline Dupas* , Nina Buchmann , Roberta Ziparo

* Primary investigator

Registry ID
AEARCTR-0004069
DOI
https://doi.org/10.1257/rct.4069-1.0
First registered
Last updated
Status
In development
Trial dates
2019-05-01 – 2020-07-31
Countries
Private
JEL codes
C91, D13, O12
Registry record
http://www.socialscienceregistry.org/trials/4069

Abstract

This study aims to understand the dynamic relationship between intra-household bargaining and women's consumption decisions. We want to test whether the share of the household budget over which a woman has control is a function of her prior investment decisions and whether these reputation concerns lead women to avoid risky and hide bad decisions. We use a series of lab-in-the-field experiments to test (1) whether husband transfers vary with the salience of the wife’s ability to manage the budget; (2) whether wives are willing to play a game that will potentially send a signal about their performance to their husbands, and whether and how much they are willing to pay a cost to hide signals of poor performance; and (3) whether wives’ willingness to invest varies with the husband’s perception of the cost and the risk of the investment.

Intervention

see experimental design

Experimental design

We play three lab-in-the-field experiments (hereafter, “experiment 1, 2 and 3”). These experiments were designed based on the assumptions (experiment 1) and predictions (experiments 2 and 3) of a theoretical model we developed.

We randomly sample approximately 1,000 couples for experiments 1 (with men) and 2 (with women), and approximately 1,000 married women whose spouse is not present from markets for experiment 3.

In experiment 1, we test whether reputation inside the household matters for the share of the budget over which women have control. For this, we randomly vary the salience of the wife’s ability to manage the household budget to test whether transfers from the husband to the wife in a public goods game vary with the husband’s perception of her ability.

In experiment 2, we test whether women care about their reputation inside the household and the extent to which this influences their investment decisions and behavior. For this, we test whether wives forgo the opportunity to play (for pay) a game that could potentially send a bad signal about their ability to their husbands; and if they play the game for money, whether and how much they are willing to pay to hide bad signals about their ability from their husbands. Wives will first play a game that tests their ability to correctly assess the quality of different products, with either help (hints) or no help. We then test whether wives’ willingness to play a game in which their score might be revealed to their husbands (as well as their willingness to pay to improve their score) vary with their score, which we can instrument with having received help (hints) in the game.

In experiment 3, we will measure how the willingness to invest varies with the husband’s perception of the cost and the risk of the investment. We will elicit women’s willingness to pay for assets that have either low or high effort costs, and test whether their willingness to pay varies with their ability to signal to their husbands that the product was acquired for free (free sticker treatment) or has a safe return (benefits sticker treatment).

Design details

Randomization method
Respondents will be assigned a number based on the order in which they are scheduled to be surveyed. We will randomly pre-assign numbers to treatments using statistical software.
Randomization unit
Individual
Clusters
Around 1,000 couples in experiments 1 and 2 and 1,000 married women in experiment 3
Observations
Around 1,000 couples (husbands in experiment 1 and wives in experiment 2) and 1,000 married women in experiment 3
Arms
- Experiment 1: Salience: 500 men, Control: 500 men - Experiment 2: Easy game: 500 women, Hard game: 500 women - Experiment 3: i) Low-effort-cost product: high perceived cost, high perceived risk: 125, low perceived cost, high perceived risk: 125, high perceived cost, low perceived risk: 125, low perceived cost, low perceived risk: 125 ii) High-effort-cost product: high perceived cost, high perceived risk: 125, low perceived cost, high perceived risk: 125, high perceived cost, low perceived risk: 125, low perceived cost, low perceived risk: 125
Primary outcomes
Experiment 1: decision in public goods game; Experiment 2: willingness to play game, willingness to pay to hide poor performance; Experiment 3: willingness to pay/invest

Cited by

No citation data available for this registration yet.

Keywords