A new Stigler Center working paper finds that the likelihood of someone signing an online petition or contacting their US senators to support corporate bailouts is directly linked to their perception of whether the big corporations are holding up their end of the social contract with their stakeholders.


There is a fundamental debate in the United States about the role of large corporations in society, a debate accelerated by the social unrest that accompanied the Covid-19 pandemic. Now more than ever, the media, regulators, and the public argue about whether large corporations should foster diversity in the workplace, limit wage inequality, protect the environment, and care for local communities. Corporate America appears to be reacting, as most clearly seen in the 2019 statement by the Business Roundtable—the association of chief executive officers of major US companies—which redefined the purpose of a corporation to promote “an economy that serves all Americans,” marking a stark change from the famous statement by Milton Friedman that “the social responsibility of business is to increase its profits.”

The recent redefinition of the social contract between large corporations and their stakeholders raises a number of important questions: Do corporations live up to the moral expectations of the American people? That is, are corporations good citizens? If they are not viewed as good citizens, what are the consequences? Answering these questions seems particularly relevant in recent times, during which the power of public opinion has become increasingly apparent. From boycotts to protests to a number of other social movements and forms of anti-corporate activism, it is undeniable that public views of big business may have important economic consequences. For example, many argue the recent regulatory scrutiny towards the four Big Tech platforms—Apple, Amazon, Google, Facebook— is the result of a deterioration in their public reputation, which suddenly makes targeting Big Tech good politics.

In our new paper, which is part of the Stigler Center’s working papers series, we measure citizens’ perceptions of large corporations’ environmental, social, and governance (ESG) practices and how these affect public support for economics policies. We conduct a series of new large-scale experimental surveys. First, we measure public perceptions of large corporations’ policies that influence society at large. Second, we study how perceptions about corporations correspond to public support for corporate bailouts and related policies during the 2020 coronavirus crisis. Focusing on bailouts at a time of crisis provides an apt setting for investigating the relationship between corporations and society, as the stakes are high and the public is engaged in the policy debate. By varying individual perceptions using multiple experiments, we show that a higher public discontent with big business leads to lower support for bailouts.

To determine perceptions of large corporations, we designed and conducted an online survey of 6,727 US citizens during May 2020, in the midst of the US Covid-19 crisis. We asked about some of the most important dimensions of ESG that respondents can easily relate to, namely executive pay, employee benefits, tax strategy, gender diversity, CO2 emissions, and political donations. We measure perceptions by asking respondents both what they think specific corporate policies are as well as what they think the policies should be. By comparing what respondents think policies are and what they think they should be, we can measure whether corporate actions meet public expectations.

“All respondents perceive corporations as not doing enough for society, relative to what they think the benchmark should be.”

We uncover a strong “big business discontent” spanning the full socioeconomic range. That is, all respondents perceive corporations as not doing enough for society, relative to what they think the benchmark should be. For example, respondents think 70 percent of large corporations donate money to politicians, but they think fewer than 30 percent of corporations should make political contributions. Similarly, respondents think that 40 percent of corporations disclose CO2 gas emissions, but they believe 70 percent of companies ought to. We also found that respondents believe that top executives and managers should be paid less, corporations should pay a larger fraction of employees’ health care costs and more in federal income taxes, and that there should be an equal gender distribution among top managers and executives. The big business discontent is significantly stronger for liberals, but it is also prevalent among conservatives. Older individuals, women, white respondents, and the unemployed also report a large degree of dissatisfaction with current corporate behavior.

We also asked respondents about their views on broad policies the government was considering to support businesses at the time of the survey. We uncover a strong negative association between big business discontent and support for bailouts. Analogously, we find that the higher the discontent, the stricter individuals think the conditions attached to corporate bailouts should be and the higher is their support for policies addressing small businesses’ needs.

We establish a causal link between public discontent with large corporations and the support for government policies using multiple experiments we embed in our survey. Our main experiment consists of showing different animated videos to different subsets of respondents immediately after they complete the socioeconomic background section. Subsets of respondents are shown different videos, so as to illustrate specific corporate policies in either a positive or negative light, while still providing accurate information.

Our primary treatment video aims at altering the perceptions of key dimensions of corporate policies in a negative direction. For example, we emphasize that there are fewer women relative to men in executive and board positions, or that companies are reluctant to cut CO2 emissions. Analogously, a second treatment video aims at altering perceptions in a positive direction. For instance, we emphasize that in recent years there has been a rise in the number of women in executive and board positions, and that several companies are now voluntarily reducing and disclosing CO2 emissions.

We find that both negative and positive videos have a negative effect on the support for corporate bailouts, which indicates that perceptions of large corporations influence individual support for economic policies. The effects are especially strong for liberal respondents, but they are also large and significant for conservatives. The treatment effects on our secondary outcomes of interest, such as support for small businesses and strictness of bailout conditions, are weaker but consistent with the descriptive patterns mentioned above.

To benchmark our results, we add a third treatment arm to our video experiment. A subset of our respondents is shown an animated video where leading economists of all political views agree that bailouts will likely help the economy during this crisis, which was a largely uncontroversial statement at the time of the survey. This is a direct treatment aimed at making corporate bailouts more desirable. Indeed, we do find that this treatment positively affects the support for bailouts, but the effect is less than half that of the negative treatment video effect.

Our main survey focused on self-reported individual preferences and beliefs on government policies and corporate practices. However, these responses do not necessarily reflect respondents’ actual behavior. We address this concern by conducting an additional large-scale experimental survey, in October 2020, where we reached 1,683 new respondents and collected behavioral outcome measures in addition to self-reported answers. This later survey had only one treatment arm, our primary negative video treatment.

The main value added by this final survey is that we measure respondents’ support for bailouts of large corporations in several additional ways. First, we asked respondents whether they would like to sign a petition urging Congress to bail out large corporations in general. Second, we asked permission to include their names in messages to the US senators of their choice expressing either their support or opposition to bailouts. Finally, we also enroll respondents into a lottery, and we ask them whether they would be willing to donate part of their winnings to the Business Roundtable. Consistent with our earlier evidence, a higher big business discontent leads to lower support for bailouts, as respondents are less likely to sign the petition and to email senators to support bailouts, but they are more likely to email senators to oppose them. We also find that treated respondents’ donations are significantly lower relative to the control group.

In conclusion, our thesis is simple: We argue that a social contract exists between US citizens and corporate America. Large corporations undoubtedly have power. Citizens vest this power in them in the hope that doing so is economically efficient but also under the expectation that corporations will not misuse their power and will work to support society as a whole. If corporations fail to meet public expectations, they will be in breach of the social contract and will face public opposition. Our results show that whether and how well corporations uphold the social contract between themselves and their stakeholders may impact the support, design, and subsequent implementation of important economic policies.