Investigating Hiring Frictions in Small Firms: Evidence from an Internet Platform-based Experiment

Last registered on January 12, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

Investigating Hiring Frictions in Small Firms: Evidence from an Internet Platform-based Experiment

RCT ID

AEARCTR-0003527

Initial registration date

November 27, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

November 30, 2018, 10:25 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated

January 12, 2023, 12:03 PM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

India

Region

Primary Investigator

Name

Niharika Singh

Affiliation

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

A. Nilesh Fernando

PI Affiliation

University of Notre Dame

Contact Investigator

PI Name

Gabriel Tourek

PI Affiliation

Harvard University

Contact Investigator

Additional Trial Information

Status

Completed

Start date

2018-11-28

End date

2020-02-29

Keywords

Firms & Productivity, Labor

Additional Keywords

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

We investigate the role of technology in addressing challenges faced by small firms in recruiting low-skilled workers in Bangalore, India. Firms in our setting report a desire to expand their operations but face difficulties in finding credible candidates outside their personal networks. Online recruitment platforms provide an opportunity for such firms to expand their recruitment pools; yet, the information available on candidates recruited via online platforms may not be of the same quality as those recruited from personal networks. This lack of credible information may limit external hiring, lead to high rates of employee turnover, and limit job opportunities for candidates outside of employer networks. In partnership with QuikrJobs, a leading online job portal for low- and semi-skilled recruitment in India, we investigate whether and how these ‘hiring frictions’ restrict firm growth and performance. First, we expand the pool of applicants available to firms recruiting on the platform by giving them access to premium recruitment services. Second, we introduce applicant background verifications for employers, intended to mimic the screening function of personal networks, and vary employer access to this information. Combining administrative data and detailed firm and jobseeker surveys, we then assess the impact of these interventions over a 12-month period on a range of firm- and worker-level outcomes.

External Link(s)

Registration Citation

Citation

Fernando, A. Nilesh, Niharika Singh and Gabriel Tourek. 2023. "Investigating Hiring Frictions in Small Firms: Evidence from an Internet Platform-based Experiment." AEA RCT Registry. January 12. https://doi.org/10.1257/rct.3527-1.1

Former Citation

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Treatment 1 - We expand the recruitment pool available to an employer by providing access to ‘premium’ recruitment services on the QuikrJobs platform. The treatment increases the salience of a job advertisement via ‘top-of-page’ placement and ‘gold’-tagged advertising for a fixed time period. It also provides priority access to additional human resources support for employers.

Treatment 2 - This treatment reveals verified information about applicant identity to employers. We first verify applicant identity using details they submit from government-issued ID during the application process and then randomly vary whether we reveal the outcome of their verification to the employer. An applicant’s verification status is only privately revealed to the employer via a badge shown on the candidate’s application within 24 hours of submission; that is, this information is not visible to other employers recruiting on the platform. Furthermore, we randomly offer some employers an additional type of background verification– employers receive an offer to verify the residential addresses of up to 3 applicants and must initiate the verification request. This verification is conducted through physical visits or postal checks by a third-party provider and takes an average of 5 days to complete.

Intervention Start Date

2018-11-28

Intervention End Date

2019-02-28

Primary Outcomes

Primary Outcomes (end points)

Applications, callback rate, hires, firm size, recruitment sources/costs, applicant quality, employee tenure, take-up of address verification

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The experimental design cross-randomizes our two treatments--- premium recruitment and background verification services. Our sample is split between those receiving the 'Regular (R)' service available at no-cost on the website and those receiving access to 'Premium (P)' recruitment services. Each employer in these two groups is further assigned to one of three verification intensities for their pool of applicants: (1) 0%, verification status is never revealed; (2) 50%, verification status is revealed for half of their applicants; and (3) 100%, verification status is revealed for all applicants to a job post. The verification status is only displayed to the employer posting the advertisement; it is not publicly available to other employers. Finally, for half of all employers randomized into the 100% verification arms, we verify residential addresses of up to 3 applicants per employer in addition to the ID documents that have already been verified. A summary list of all of the experimental groups is given below.

R1- Regular, 0% ID verification (control group)
R2- Regular, 50% ID verification
R3a- Regular, 100% ID verification
R3b- Regular, 100% ID verification and addresses verified for up to 3 hired applicants
P1- Premium, 0% ID verification
P2- Premium, 50% ID verification
P3a- Premium, 100% ID verification
P3b- Premium, 100% ID verification and addresses verified for up to 3 hired applicants

Employers for the study are recruited from the QuikrJobs platform. We sample Bangalore-based firms with fewer than 50 employees recruiting low and semi-skilled workers for sales, marketing or delivery positions. Employers recruiting in these categories receive a screening questionnaire during the job posting process and conditional on eligibility, their job post is randomized into one of our experimental groups.

Experimental Design Details

Randomization Method

Randomization will be conducted in real time on the platform using pre-assigned probability thresholds to achieve the desired sample sizes for our experimental groups.

Randomization Unit

We follow a two-stage randomization. In the first stage, we assign employer job postings to one of our experimental groups. Then, based on this assignment, we further randomize whether an applicant's verification status is revealed or not. Randomization in the first stage is also stratified by role, previous usage of the platform, and a firm size dummy for whether firm is above or below 10 employees.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

2,100 employers

Sample size: planned number of observations

2,100 employers

Sample size (or number of clusters) by treatment arms

R1- Regular, 0% ID verification (400 employers)
R2- Regular, 50% ID verification (300 employers)
R3a- Regular, 100% ID verification (150 employers)
R3b- Regular, 100% ID verification and addresses verified for up to 3 hired applicants (150 employers)
P1- Premium, 0% ID verification (400 employers)
P2- Premium, 50% ID verification (300 employers)
P3a- Premium, 100% ID verification (200 employers)
P3b- Premium, 100% ID verification and addresses verified for up to 3 hired applicants (200 employers)

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Supporting Documents and Materials

IRB