Responses to CEO Activism

How to cite

Sandvik, Jason. 2020. "Responses to CEO Activism." AEA RCT Registry. AEARCTR-0006303. https://doi.org/10.1257/rct.6303-1.1

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

Jason Sandvik*

* Primary investigator

Registry ID
AEARCTR-0006303
DOI
https://doi.org/10.1257/rct.6303-1.1
First registered
Last updated
Status
In development
Trial dates
2020-08-23 – 2020-08-27
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/6303

Abstract

We investigate the impact of CEO activism, the increasingly more common practice of CEOs speaking out on controversial social and political issues, on firm value. CEO activism may be beneficial for shareholders, as it can bolster firms' relationships with customers and employees. Alternatively, CEO activism may be detrimental to firm value if it alienates important stakeholders with opposing views. Consistent with the former, we find that CEO activism results in a positive market reaction and higher valuations. Firms with CEO activism have higher sales, exhibit higher employee productivity, and innovate more. Additionally, activist CEOs are rewarded with more directorships and a reduced likelihood of turnover.

Intervention

We will recruit our experiment participants from Amazon Mechanical Turk. To qualify for the study, participants have to provide satisfactory responses to a screening survey, which will ask about their experience on the platform, their experience making investment decisions, and their understanding of corporate decision-making. We will adhere to the current best practices to ensure our sample contains only high-quality participants. From the screening survey, approximately 900 participants will be invited to participate in the randomized controlled trial.

The experimental design of the randomized controlled trial will be as follows. Each participant will be told that they are either (i) a mutual fund manager who is deciding whether to buy the stock of Firm A, (ii) a customer who is deciding whether to purchase a hi-tech television from Firm A, or (iii) a potential employee of Firm A who is on the job market (300 participants will be placed in each bin). Each is given the same information about Firm A's past stock return performance. Within each group of participants, one third of the participants will be given no additional information about the company and will be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm. They will respond using a 100-point slider scale. Another third of the participants will be told that it has become increasingly common for CEOs to take a public stance on social issues (CEO activism), and then they will be told that the CEO of Firm A regularly engages in CEO activism. They will then be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm. The last third of the participants in each group will be told that it has become increasingly common for CEOs to take a public stance on social issues (CEO activism), and then they will be told the CEO of Firm A avoids engaging in CEO activism. They will then be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm.

After each participant makes their decision, we will then ask them several follow-up questions to provide insight into the effect of CEO activism on other stakeholders. Answers will be given on a 7-point Likert scale from “Strongly Disagree” to “Strongly Agree.”

Experimental design

We will recruit our experiment participants from Amazon Mechanical Turk. To qualify for the study, participants have to provide satisfactory responses to a screening survey, which will ask about their experience on the platform, their experience making investment decisions, and their understanding of corporate decision-making. We will adhere to the current best practices to ensure our sample contains only high-quality participants. From the screening survey, approximately 900 participants will be invited to participate in the randomized controlled trial.

The experimental design of the randomized controlled trial will be as follows. Each participant will be told that they are either (i) a mutual fund manager who is deciding whether to buy the stock of Firm A, (ii) a customer who is deciding whether to purchase a hi-tech television from Firm A, or (iii) a potential employee of Firm A who is on the job market (300 participants will be placed in each bin). Each is given the same information about Firm A's past stock return performance. Within each group of participants, one third of the participants will be given no additional information about the company and will be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm. They will respond using a 100-point slider scale. Another third of the participants will be told that it has become increasingly common for CEOs to take a public stance on social issues (CEO activism), and then they will be told that the CEO of Firm A regularly engages in CEO activism. They will then be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm. The last third of the participants in each group will be told that it has become increasingly common for CEOs to take a public stance on social issues (CEO activism), and then they will be told the CEO of Firm A avoids engaging in CEO activism. They will then be asked the extent to which they want to (i) invest in the firm, (ii) purchase a television from the firm, or (iii) pursue employment with the firm.

After each participant makes their decision, we will then ask them several follow-up questions to provide insight into the effect of CEO activism on other stakeholders. Answers will be given on a 7-point Likert scale from “Strongly Disagree” to “Strongly Agree.”

Design details

Randomization method
Randomization into one of the nine treatment cells (3 x 3) at the individual level. Participants work independently from each other and are very unlikely to communicate or be influenced by one another in any way. The randomization will be done automatically through Qualtrics.
Randomization unit
Individual experiment participant
Clusters
450-900 individual participants
Observations
450-900 individual participants
Arms
50-100 individual participants in each of the 9 treatment arms (3 x 3 design)
Primary outcomes
Responses to questions about the extent to which participants want to (i) invest in the firm, (ii) purchase goods from the firm, or (iii) pursue employment with the firm.
Secondary outcomes
Responses to questions about what factors influenced the particpants' decisions.

Cited by

No citation data available for this registration yet.

Keywords