Kinship Pressure and Firm-Worker Matching Distortions

How to cite

Swanson, Nicholas. 2023. "Kinship Pressure and Firm-Worker Matching Distortions." AEA RCT Registry. AEARCTR-0011328. https://doi.org/10.1257/rct.11328-2.1

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

Nicholas Swanson*

* Primary investigator

Registry ID
AEARCTR-0011328
DOI
https://doi.org/10.1257/rct.11328-2.1
First registered
Last updated
Status
In development
Trial dates
2023-04-25 – 2023-11-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/11328

Abstract

Many businesses in Sub-Saharan Africa choose to employ relatives. A common understanding of this phenomenon is that this reflects information assymetries or benefits to hiring from one’s network that alleviate moral hazard. In this project, I propose a different hypothesis: that some of this hiring may be driven by social norms or redistributive pressures, which maybe exacerbated by greater moral hazard among members of one’s network. I design a field experiment to examine the relevance of these redistributive pressures generating hiring distortions among Zambian microenterprises. I offer microenterprise owners a subsidy to hire an additional employee, and ask business owners to decide between hiring a relative or a non-relative candidate for the position. Microenterprise owners then make their choice under two conditions: a control condition, in the owner receives a poster explaining that they received a subsidy to hire a person and a treatment condition, in which employers are told that if they receive a subsidy, they will be given a poster explaining the subsidy AND that many business owners offered this subsidy do not have complete control over who is hired with the subsidy. The treatment poster is designed to provide business owners with plausible deniability for not hiring a particular employee, as compared to natural hiring conditions. I then test how the identity of the employee chosen changes under this treatment. The experiment outlined in this pre-registration is one that explores the hypothesis. An additional experiment involving rural employers may also be conducted as part of the same project (subject to budget and logistical constraints).

Intervention

The treatment offers microenterprises different posters that vary the degree to which their involvement in who they choose to hire for their business is signalled to others.

Experimental design

In the experiment I offer microenterprises a subsidy and elicit preferences of who to hire under different conditions.

Design details

Randomization method
Randomization by computer
Randomization unit
Individual, stratified by gender, market and whether the business has employees
Clusters
N/A
Observations
400-600 businesses depending on eligibility, take up of subidy opportunity. An additional 200 businesses in the follow up survey.
Arms
Equal numbers will be assigned to the treatment and control poster conditions
Primary outcomes
The primary outcomes from the experiment are: 1) Identity of chosen employee when offered equal subsidies for the relative and non-relative candidate 2) Compensating differential in terms of subsidies required to hire a relative (rather than a non-relative) 3) Audit score (proportion of times that the business is audited and the employee is there) for businesses that have the subsidy implemented. In addition, to ensure that in the BDM the hiring decisions are made seriously, businesses will be audited after the the subsidy is implemented, to check whether the employee that the subsidy was given for is actually there.
Secondary outcomes
To the degree I have sufficient variation in the measures, I will look at heterogeneity in primary outcomes (except audit score) above by: 1) the amount of redistributive pressure that the business owner faces (elicited from survey questions) and 2) the degree that they for expect the relative employee to work harder/better or less hard/less well for them as compared to other non-related employers (elicited from survey questions) as well as 3) the characteristics of the job that the employer is hiring for (to the degree there is sufficient variation in this outcome). Additional exploratory heterogeneity might also be conducted (for example with respect to characteristics of the job, employer, employees etc.) Finally, I will elicit the willingness to pay for a treatment poster (rather than a control poster) among all business owners and correlate this also with the primary outcomes above. Outcome subject to feasibility/budget: If possible, I will also ask business owners who do not have the subsidy implemented to bring in their relative/non-relative and do a task with them. In this task I will measure 1) the business owner’s beliefs about task performance and 2) the actual performance under different conditions

Cited by

No citation data available for this registration yet.

Keywords