Labor-Market Information, Job Postings, and Employer Beliefs: Experimental Evidence from Austria

How to cite

Steinmayr, Andreas; Butschek, Sebastian; Schwab, Marina; Gilbert, Mattis; Rathelot, Roland. 2026. "Labor-Market Information, Job Postings, and Employer Beliefs: Experimental Evidence from Austria." AEA RCT Registry. AEARCTR-0018379. https://doi.org/10.1257/rct.18379-1.0

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

Andreas Steinmayr* , Sebastian Butschek , Marina Schwab , Mattis Gilbert , Roland Rathelot

* Primary investigator

Registry ID
AEARCTR-0018379
DOI
https://doi.org/10.1257/rct.18379-1.0
First registered
Last updated
Status
In development
Trial dates
2026-04-16 – 2027-06-30
Countries
Private
JEL codes
J31, J63, J64, C93, D83, J68
Registry record
http://www.socialscienceregistry.org/trials/18379

Abstract

Many firms report difficulties in filling vacancies. To study the importance of information frictions in the job posting process, we test the effect of giving employers information on the wages posted by competing firms. Together with the Austrian Public Employment Service, we field an employer survey linked to administrative employer–employee data. We embed a randomized information experiment that provides half of the recruiters with recent occupation-specific posted wages for the position most relevant to them. We measure belief updating and link it to subsequent stated and realised posting choices.

Intervention

We provide a random subset of participating employers with information about posted wages in their regional occupational labour market. The information treatment consists of an interactive graph showing the distribution of recent posted wages in job ads on a major job board for an occupation chosen by the respondent.

Experimental design

We run a between-subject survey experiment, with our information treatment embedded in an online questionnaire. The questionnaire is anchored on the employer's next planned job posting and the corresponding occupation. It elicits beliefs about the typical posted wage in this occupation and then, (for the treatment group), contrasts this with actual posted wages in the market. The design allows us to study the effect of this information on employers' stated market beliefs and intended posted wages on the one hand and their realized posting choices for the next vacancy (in the chosen occupation) on the other.

Design details

Randomization method
Stratified block randomization, automated in oTree, with real-time assignment to strata during survey participation.
Randomization unit
Individual respondent (one per establishment)
Clusters
600 respondents = establishments
Observations
600 respondents = establishments
Arms
300 respondents = establishments treatment; 300 respondents = establishments control.
Primary outcomes
1. Log intended posted wage 2. Belief on the elasticity of vacancy duration with respect to the posted wage 3. Belief on the elasticity of the number of qualified applicants with respect to the posted wage 4. Posting at least one AMS vacancy for the chosen occupation in the 6 months after survey participation 5. Log realized posted wage in the AMS vacancy/vacancies for chosen occupation (if any)

Cited by

No citation data available for this registration yet.

Keywords