We survey workers to understand their outside job options and so infer their employers' market power. We then use these answers to estimate a structural model of the labor market.
Experimental design
We plan to use the online survey to collect information on worker rents and rent sharing.
Design details
Randomization method
Randomization is done by the survey software, Qualtrics.
Randomization unit
Randomization is at the individual level.
Clusters
N/A
Observations
Our target sample size is 2000 respondents. However we may sample fewer than 2000 respondents if the Prolific platform provides fewer respondents that meet our criteria.
Arms
For both the question on worker rents and the question on rent sharing, each of the 9 cutoffs will be drawn with equal probability.
Primary outcomes
There are two outcomes. The first is the respondent’s choice to stay or leave a lower-paying version of their current job. The second is the respondent’s indication that their employer would be willing to give them a pay raise of a certain amount. (See “Experimental Design” for more details)
Cited by
No citation data available for this registration yet.