Unpacking the Gender Profit Gap: Evidence from Micro-Businesses in India

How to cite

Delecourt, Solene; Ng, Odyssia. 2018. "Unpacking the Gender Profit Gap: Evidence from Micro-Businesses in India." AEA RCT Registry. AEARCTR-0003180. https://doi.org/10.1257/rct.3180-2.0

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

Solene Delecourt* , Odyssia Ng

* Primary investigator

Registry ID
AEARCTR-0003180
DOI
https://doi.org/10.1257/rct.3180-2.0
First registered
Last updated
Status
In development
Trial dates
2018-08-06 – 2019-01-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/3180

Abstract

Large gender differences in earnings prevail worldwide in almost all occupations, and entrepreneurship is no exception. Women’s businesses are typically less productive, but also smaller and less capital-intensive than men’s. Despite the vast literature reporting the gender gap in profitability, its causes are not well understood. Our study examines the role of business characteristics and discrimination on gender disparities in profitability. We study small-scale produce businesses in India. We first conduct a survey to establish that women’s businesses are both less profitable and smaller than men’s in this context. Using a large-scale field experiment, we test whether giving men and women the same business closes the gap in profitability. We set up our own market stalls, to which we randomly assign male and female vendors. We thus exogenously vary gender, holding the business constant. The evidence from this experiment shows how much of the gender gap in this context is due to business characteristics alone. To examine the role of client discrimination more specifically, we conduct a second experiment with packaged goods instead of vegetables.

Experimental design

Our main experiment consists in setting up vegetable stalls in a retail market, recruit one male and one female vendor every day to sell vegetables that we provide, and randomly assign the vendors to the stalls. Every day, we purchase vegetables from the wholesale market and equally divide them among the two stalls. Our experimental design enables us to exogenously vary gender, holding the business characteristics - quantity, quality of produce, location - constant. It is a form of extreme capital drop intervention: we provide vendors with the exact same business. To incentivize vendors to sell, they keep the day’s profits and are only charged for the vegetables they are able to sell.

We run the same experiment with packaged goods that are sold by men and women. To measure discrimination, we decrease the price of the packaged goods sold by women for a randomly chosen part of the day.

Design details

Randomization method
For recruitment: randomization done on the tablet through the software SurveyCTO. Other randomizations are done in office by a computer (Stata). We select a random pair of one man and one woman to be assigned to our daily stalls.
Randomization unit
We randomize at the individual level as we assign individuals to stalls.
Clusters
250 individuals (125 men and 125 women) for the vegetable experiment. At least 60 individuals (30 men and 30 women) for the packaged goods experiment, depending on funding, we hope to get at least 100.
Observations
250 individuals (125 men and 125 women) for the vegetable experiment. At least 60 individuals (30 men and 30 women) for the packaged goods experiment, depending on funding, we hope to get at least 100.
Arms
125 men and 125 women for the vegetable experiment. At least 60 individuals (30 men and 30 women) for the packaged goods experiment, depending on funding, we hope to get at least 100.
Primary outcomes
Variables at the vendor level: Daily profit, daily revenue,number of "missed" clients, number of purchasing clients Variables at the vegetable level: quoted price, price paid.
Secondary outcomes
Variables at the vendor level: number of male/female clients who approached and buy from the stall, inventory, markup per Rs. Variables at the vegetable level: discount (bargaining).

Cited by

No citation data available for this registration yet.

Keywords