The effects of employee referrals on firm performance: Evidence from a retail chain

Last registered on August 10, 2018

Pre-Trial

Trial Information

General Information

Title
The effects of employee referrals on firm performance: Evidence from a retail chain
RCT ID
AEARCTR-0000964
Initial registration date
November 23, 2015

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 23, 2015, 12:16 PM EST

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

Last updated
August 10, 2018, 3:11 AM EDT

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

Locations

Region

Primary Investigator

Affiliation
University of Cologne

Other Primary Investigator(s)

PI Affiliation
University of Toronto Rotman School of Management
PI Affiliation
Goethe University, Frankfurt
PI Affiliation
Goethe University, Frankfurt

Additional Trial Information

Status
Completed
Start date
2013-01-01
End date
2016-12-31
Secondary IDs
Abstract
Employee referrals, when existing employees recommend their connections for open vacancies, are a popular hiring practice. There is also a substantial theoretical literature from sociology and economics about referrals. Yet, causal evidence on the effects of referrals on firm performance remains scarce. We run a field experiment in a retail network of 238 shops in an Eastern European EU country, in which we encourage employees in the treatment shops to recommend people they know for vacancies within the network. Employees in the control shops can also recommend but are not encouraged to do so.

The main outcome variable of our interest is employee turnover. We choose to target turnover with our experimental treatment because it has blighted our study firm for years, averaging at an upwards of 80% per year and costing about 400 Euros per quit. In general many economies feature high levels of turnover which involves the risk of sub-optimal investment of companies in the human capital of their workers. Hence, it seems an interesting question to investigate whether single employers can reduce the turnover or whether this is an economy wide phenomenon. Our second main outcome variable is absenteeism. We choose to target absenteeism with our treatment because absenteeism rates are also very high and expensive for our study firm.

The treatment shops have been divided into four groups, each with a different amount of the bonus offered to the employees whose referrals will lead to a successful employment of five months or longer. The first group will receive a bonus of 50 Euros, the second 90 Euros, the third 120 Euros, and the fourth will receive no bonus. The size of the bonus and the minimum length of employment required for a bonus have been determined based on the existing length of employment data and on the survey of the production employees of the same firm who will not be part of our experiment.

Treatment and control stores were selected randomly. We work with store and regional managers and an employee survey to ensure that we can detect and minimize information spillovers between stores in different treatment groups.

The experiment starts at the end of November 2015.
External Link(s)

Registration Citation

Citation
Friebel, Guido et al. 2018. "The effects of employee referrals on firm performance: Evidence from a retail chain." AEA RCT Registry. August 10. https://doi.org/10.1257/rct.964-4.0
Former Citation
Friebel, Guido et al. 2018. "The effects of employee referrals on firm performance: Evidence from a retail chain." AEA RCT Registry. August 10. https://www.socialscienceregistry.org/trials/964/history/32934
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2015-11-30
Intervention End Date
2016-12-31

Primary Outcomes

Primary Outcomes (end points)
Personnel turnover

Absenteeism

Sales

Other shop performance data (e.g. number of customers, mystery shopping scores),
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Employee referrals, when existing employees recommend their connections for open vacancies, are a popular hiring practice. There is also a substantial theoretical literature from sociology and economics about referrals. Yet, causal evidence on the effects of referrals on firm performance remains scarce. We run a field experiment in a retail network of 238 shops in an Eastern European EU country, in which we encourage employees in the treatment shops to recommend people they know for vacancies within the network. Employees in the control shops can also recommend but are not encouraged to do so.

The main outcome variable of our interest is employee turnover. We choose to target turnover with our experimental treatment because it has blighted our study firm for years, averaging at an upwards of 80% per year and costing about 400 Euros per quit. In general many economies feature high levels of turnover which involves the risk of sub-optimal investment of companies in the human capital of their workers. Hence, it seems an interesting question to investigate whether single employers can reduce the turnover or whether this is an economy wide phenomenon. Our second main outcome variable is absenteeism. We choose to target absenteeism with our treatment because absenteeism rates are also very high and expensive for our study firm.

The treatment shops have been divided into four groups, each with a different amount of the bonus offered to the employees whose referrals will lead to a successful employment of five months or longer. The first group will receive a bonus of 50 Euros, the second 90 Euros, the third 120 Euros, and the fourth will receive no bonus. The size of the bonus and the minimum length of employment required for a bonus have been determined based on the existing length of employment data and on the survey of the production employees of the same firm who will not be part of our experiment.

Treatment and control stores were selected randomly. We work with store and regional managers and an employee survey to ensure that we can detect and minimize information spillovers between stores in different treatment groups.

The experiment starts at the end of November 2015.
Experimental Design Details
Randomization Method
Randomization done by computer.
Randomization Unit
Shops
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
238 (=number of shops)
Sample size: planned number of observations
238
Sample size (or number of clusters) by treatment arms
48 in treatment 1, 48 in treatment 2, 48 in treatment 3, 48 in treatment 4, 46 in the control group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
Yes
Intervention Completion Date
December 31, 2016, 12:00 +00:00
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials