Labor Demand for Workers with a Criminal Conviction

How to cite

Dobbie, Will; Cullen, Zoë; Hoffman, Mitchell. 2019. "Labor Demand for Workers with a Criminal Conviction." AEA RCT Registry. AEARCTR-0005200. https://doi.org/10.1257/rct.5200-2.0

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

Will Dobbie* , Zoë Cullen , Mitchell Hoffman

* Primary investigator

Registry ID
AEARCTR-0005200
DOI
https://doi.org/10.1257/rct.5200-2.0
First registered
Last updated
Status
Completed
Trial dates
2019-11-01 – 2022-10-01
Countries
United States of America ()
JEL codes
C93; J23; J71; K14
Registry record
http://www.socialscienceregistry.org/trials/5200

Abstract

We employ a randomized discrete choice experiment on a large on-demand staffing platform to estimate the labor demand for workers with a past criminal conviction (WCCs) under different policies and market conditions. Employers using the platform will have the option to have WCCs assigned to their vacancies at a randomly chosen wage subsidy, both with and without the option of crime and safety insurance and increased performance screening, and under different hypothetical labor market conditions. Employer choices at different wage subsidies will yield willingness to pay (WTP) estimates for this population of jobseekers under different counterfactual policies. These estimates will also allow us to estimate a labor demand function.

Experimental design

We will recruit participants through the online platform’s account managers.

In the survey experiment, we ask firms about their interest in hiring individuals with a criminal record. We cross randomize values of each of the following dimensions:

• Wage discount
• Level of commercial crime insurance
• Number of positive past performance scores on the platform
• Hypothetical local unemployment rate
• Years since the most recent criminal act

The survey is incentive-compatible because subjects are informed that choices will affect the workers assigned to their jobs.

Design details

Randomization method
The randomization will be conducted using a computer random number generator.
Randomization unit
We will randomize at the level of the firm, where all managers from the same firm will be randomized into the same group.
Clusters
The survey will be administered to approximately 9,000 individuals at 7,600 firms. We will cluster at the firm level (where there are several managers from the same firm). Based on previous response rates to the online platform’s surveys, we expect between 760 and 2,000 firms to respond.
Observations
See above.
Arms
We are cross randomizing all parameter values equally allocating firms to each value listed for each parameter.
Primary outcomes
Employers’ choices over whether to have workers with a criminal record assigned to their vacancies at randomly chosen wage subsidies.

Materials

Papers from this trial

  1. Zoë Cullen, Will Dobbie, Mitchell Hoffman, Increasing the Demand for Workers with a Criminal Record, The Quarterly Journal of Economics, 2022;, qjac029, https://doi.org/10.1093/qje/qjac029 https://doi.org/10.1093/qje/qjac029

Cited by

No citation data available for this registration yet.

Keywords