Do workers intend to work more if their wage compares well to that of their peers in the economy? A large-scale survey experiment

How to cite

Zheng, Jindi; Fabo, Brian; Tijdens, Kea; Dur, Robert. 2014. "Do workers intend to work more if their wage compares well to that of their peers in the economy? A large-scale survey experiment." AEA RCT Registry. AEARCTR-0000289. https://doi.org/10.1257/rct.289-1.0

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

Jindi Zheng* , Brian Fabo , Kea Tijdens , Robert Dur

* Primary investigator

Registry ID
AEARCTR-0000289
DOI
https://doi.org/10.1257/rct.289-1.0
First registered
Last updated
Status
On going
Trial dates
2014-03-05 – 2015-03-05
Countries
Private
JEL codes
D83, J28, J31
Registry record
http://www.socialscienceregistry.org/trials/289

Abstract

This project explores how workers respond to information about their position in the wage distribution of peers in the economy. We work together with a nationwide survey website in the Netherlands to implement a survey experiment. We ask all respondents about where they think they are in the wage distribution of peers in the economy. Next, we provide a random half of the respondents information about their 'true' position in the relevant wage distribution (using data recently gathered by the survey website). We compare job satisfaction, intention to quit current job, and willingness to exert effort of employees who are informed by us with those who are not. Moreover, we investigate how an employee's reciprocal inclinations and the presence of efficiency wage policies in the employee's workplace influence the treatment effects. Lastly, we examine one possible framing effect.

Intervention

While completing an online questionnaire, respondents will be asked in what decile they think they are in the wage distribution of peers in the economy (defined as those with similar education, tenure, industry, and occupation). After answering that question, a randomized group of respondents will receive information about their 'true' position in this wage distribution. This position is computed by the website using recent historical survey data. The survey receives more than 50,000 responses each year, thus employees from the largest occupations have a big reference group to compare their wages with. We focus on the largest thirteen occupations according to the website's historical data, so that we have a sufficient number of observations to create peer wage distributions. People with other occupations are not included in this experiment.

Experimental design

We work together with a widely used online labor market survey (50.000 annual visitors) in the Netherlands. Survey participants have a chance to win a prize in a monthly lottery worth more than 1,400 euro (monthly minimum wage in the Netherlands). During the survey, participants will be asked about their income, working environment, reciprocity, the way their employer deals with under-performing workers, and so on. We also ask all respondents in what decile they think they are in the wage distribution of peers in the economy (defined as those with similar education, tenure, industry, and occupation). Then we offer the treatment group with information about their 'true' position in the peers' wage distribution using data recently collected by the survey website. The control group does not receive this information. After this stage, each participant answers questions about job and wage satisfaction, intended work effort, and intention to quit current job.

Design details

Randomization method
Randomization done by a computer program on the website: every eight minutes the participants are selected into either treatment group or control group.
Randomization unit
Individual survey respondents who have reported one of the thirteen big occupations.
Clusters
Not applicable.
Observations
We aim for approximately 1,000 observations.
Arms
50/50 randomization into treatment and control.
Primary outcomes
1) Job-related satisfaction (job satisfaction and wage satisfaction); 2) Intention to quit the current job; 3) Intended work effort (the willingness to work unpaid overtime and the willingness to work hard). We test the following hypotheses: 1) A better than expected position in the peers' wage distribution increases job and wage satisfaction and decreases intention to quit. A worse than expected position decreases job and wage satisfaction and increases intention to quit. 2) A better than expected position in the peers' wage distribution increases intended work effort of workers who are reciprocal. A worse than expected position decreases intended work effort of workers who are reciprocal. The size of the response increases with reciprocal inclinations as well as with the difference between expected and true position. 3) A better than expected position in the peers' wage distribution increases intended work effort of workers of whom the employer dismisses under-performing employees. A worse than expected position decreases intended work effort of workers of whom the employer dismisses under-performing employees. The size of the response increases with the difference between expected and true position.

Cited by

No citation data available for this registration yet.

Keywords