What Do Workers Want From (Private) Government? Experimental Evidence from the United States

How to cite

Mazumder, Soumyajit. 2020. "What Do Workers Want From (Private) Government? Experimental Evidence from the United States." AEA RCT Registry. AEARCTR-0005777. https://doi.org/10.1257/rct.5777-1.1

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

Soumyajit Mazumder*

* Primary investigator

Registry ID
AEARCTR-0005777
DOI
https://doi.org/10.1257/rct.5777-1.1
First registered
Last updated
Status
In development
Trial dates
2020-05-27 – 2020-06-30
Countries
Private
JEL codes
J54, J30
Registry record
http://www.socialscienceregistry.org/trials/5777

Abstract

Much of the American labor force spends time in ``private governments" over which they have little say during and beyond the work day. Do Americans prefer to work for businesses that look more like democracies or autocracies? We study this question using conjoint experimental techniques on a nationally representative sample of Americans. This design allows us to vary a large number of features of the workplace--especially their governance structures and the degree to which these structures allow for meaningful democratic decision-making. We hypothesize that workers should have a preference for democratic corporate governance structures such as employee ownership, co-determination, and the direct election of management especially so after experimentally manipulating class identity. The results of this project will contribute to literatures across management, economic sociology, and political science by empirically investigating the problem of the ``corporate regime type" from the bottom-up.

Experimental design

We use a conjoint survey to attain a comprehensive sense for Americans' preferences toward the workplace. Each respondent evaluates four comparisons between pairs of firms, each displayed on a new page. We randomly vary the two firms' profiles of 19 attributes that one would might consider when choosing between job offers. From this, we aim to estimate the Average Marginal Component Effect (AMCE), which is the marginal contribution of each attribute on the outcome of interest.

Design details

Randomization method
We randomize conjoint attributes with all combinations having equal probability.
Randomization unit
Individual
Clusters
1000
Observations
4000
Arms
Treatment arms are factorial design so many treatment arm cells are empty. The quantity of interest is the AMCE which can be unbiasedly estimated despite the presence of empty cells via ordinary least squares.
Primary outcomes
After reading the firm profiles, respondents answer five outcomes measures. The first question forces respondents to indicate at which firm they would rather work. We code the responses to the question as a binary variable, \textit{Firm Preferred}, which is 1 if the firm is preferred and 1 otherwise. This is our key dependent variable. This variable approximates the real labor market in that when someone accepts a job offer, they must reject the other offer. We then use two outcome variables to measure whether respondents believe that more democratic workplaces improve their day-to-day work experiences. We ask respondents which firm is ``best at resolving any problems or complaints that arise at work'' and at which firm they think ``workers would have more power.'' Both measures are forced choice. Additionally, we ask respondents at which firm they think they would have more responsibilities to see whether workers realize that more democratic workplaces would require them to take on some managerial responsibilities. Finally, we include an open-ended question for respondents to express why they prefer one firm over the other. This question helps us capture any other reasons that we might miss.

Cited by

No citation data available for this registration yet.

Keywords