Xenophobia in Juror Decision-Making

Abstract

With the upsurge of anti-globalizing ideologies and politics, the increasing institutionalization of xenophobia within the legal system has emerged as a pressing concern. Existing law and social science research has underexplored xenophobic bias in the US legal system. This article conceptualizes xenophobic bias as consisting of racism and nationalism. It investigates whether mock jurors reach different verdicts on defendant companies from foreign countries of origin (Japan, France, and China) compared to domestic (US) companies. Using a test simulating a patent lawsuit, the research finds no evidence of general xenophobic bias in juror liability verdict decisions, yet there is a specific bias against the Chinese company when granting damage awards. The similarity-leniency effect that has been established in the previous literature is corroborated in this article. Additionally, political views moderate the effects of the company’s country of origin on juror decisions. This research offers a more nuanced conceptual framework of xenophobic bias in juror decision-making for future law and social science research and informs judicial policies seeking to improve jury instructions and jury selection to reduce xenophobic bias.

Introduction

The recent upsurge of anti-globalizing ideologies and politics worldwide has led to a significant rise in hostility toward foreign individuals and entities. As part of the trend, the US Federal Government has recognized that “it has played a role in furthering these xenophobic sentiments through the actions of political leaders” (Biden 2021). This changing political landscape raises questions about whether foreign parties may encounter bias when interacting with domestic legal institutions, thereby potentially compromising the principle of legal equality.

Despite the significance and urgency of this question, there is a lack of sociolegal research on xenophobic bias in courtrooms. The prevailing body of research on this topic, primarily conducted in the late 1990s and 2000s (Clermont and Eisenberg 1996, 2007; Moore 2002), failed to reflect the current sociopolitical landscape where Americans’ views toward different countries have considerably evolved. In addition, existing research has yielded mixed results on whether foreign parties are disadvantaged relative to domestic parties in the US courts (Moore 2002; Clermont and Eisenberg 1996, 2007).

The exploration of xenophobic bias in juror decision-making research is notably limited. Existing research on juror decision-making has primarily focused on racial bias in the domestic context (Hunt 2015), and only a few studies have explored bias against alien individuals and entities (Espinoza et al. 2015). In addition, there has been insufficient research on how jurors’ identities and views affect their decisions (Devine 2012), especially in the context where one party is foreign. Given the substantial shifts in the sociopolitical landscape and the lack of academic inquiry, it is imperative to study xenophobic bias in juror decision-making.

To address the gap in the existing literature, this article presents the first simulation study to identify juror bias in the form of xenophobia against foreign companies through an online experiment. Using a nationally representative sample of US jury-eligible adults from the data collection platform Prolific, I examined whether the country of origin of a defendant company (US, Japan, France, or China) influences the mock jurors’ verdicts in a civil case of patent dispute. Building upon a conceptualization of xenophobia comprising racism and nationalism, I additionally tested the similarity-leniency effect and the influence of political views on mock juror decision-making. The findings show no evidence of a general xenophobic bias against foreign companies in liability verdict decisions. However, the plaintiff was granted a discernibly higher damage award against a Chinese company. Jurors demonstrated a significant level of leniency toward companies that shared their ethnic origin. Moreover, right-leaning jurors displayed a strong preference for a domestic company, suggesting that juror xenophobic bias is driven by political ideology. These results have substantial implications for juror decision-making research and, more broadly, for judicial decision-making research, as they underscore the need for continued efforts to study and address courtroom biases in various forms to ensure a fair and just legal process for all parties involved.

Xenophobia and legal institutions

Xenophobia is defined as negative attitudinal sentiments such as fear and hatred toward outgroup members perceived as unknown or alien (Yakushko 2009). Existing literature suggests that xenophobia is multifaceted and multicausal (Yakushko 2009). It is inherently interwoven with racism (Wimmer 1997; Rydgren 2003) and is associated with civic nationalism, which indicates the superiority of one’s nation to others (Hjerm 2001; Esses et al. 2005). Consequently, researchers differentiate between ethnic and civic types of xenophobia (Yakushko 2009; Lewin-Epstein and Levanon 2005; Esses et al. 2005). Due to its complexities, this study conceptualizes xenophobia as a general negative sentiment toward foreigners that encompasses various forms of prejudice, including racism and nationalism.

Sociolegal scholars have predominantly situated the academic discourse of xenophobia in the context of immigration (Calavita 2005; Arrocha 2019). Another line of inquiry into xenophobia in the legal system has been pursued in the context of lawsuits involving foreign businesses. Early research was undertaken in response to a widespread perception among practitioners that “American courts are hostile to foreign parties” (Moore 2002, 1497). During the 1990s, jury consulting firms were among the pioneers in developing detailed bias scales and utilizing simulations to test juror bias against foreign companies. For instance, DecisionQuest, a leading jury-consulting firm, reported differences in mock juror decisions when the nationality of the defendant changed from American to Japanese in simulation tests (Moore 2002). This bias was found to be “more common in venues suffering economic difficulties due to foreign competition and becomes much more pronounced when Japanese witnesses use translators on the stand” (Moore 2002, 1499).

However, subsequent sociolegal research has not followed the steps of these firms in conducting simulation studies to test xenophobia; instead, it has relied on the analysis of observational data and yielded inconclusive results on whether judges and jurors in the United States harbor xenophobic bias toward foreign parties. Moore’s (2002) analysis of patent cases adjudicated in federal courts between 1983 and 2000 reveals a “substantial disparity” in win rates between domestic and foreign parties in patent litigation, with domestic parties faring significantly better than foreign ones. She attributes this susceptibility of patent cases to prejudicial decision-making partially to the “liberation hypothesis” from psychology literature, which suggests that the decision outcomes of adjudicators are more likely to be influenced by their prejudice and bias when they face both complex and close cases. Similarly, Bhattacharya, Galpin, and Haslem’s (2007) analysis of 2,361 US corporate defendants and 715 foreign corporate defendants from 1995 to 2000 reveals that US firms are less likely than foreign corporations to lose at trial, although dismissal rates show no marked difference. More recent research has shown adverse “government bias” against Chinese multinational companies (Li 2021) and the disadvantaged position of Chinese defendants in business crime cases (Kim 2018; Fang and Li 2021) amidst growing tension between the United States and China.

By contrast, Clermont and Eisenberg (1996; 2007) find no evidence of xenophobic bias in their two landmark studies of civil cases coded by the Administration Office of the US Courts. In their detailed critique of Moore’s study, they noted that her statistical figures were generated solely from the data of patent trial cases, and she only used single percentages to compare foreign and domestic litigants. This approach was criticized for not employing a standard method of comparing “foreign to domestic plaintiffs when facing a domestic defendant” and “foreign to domestic defendants when facing a domestic plaintiff” to show the foreigner effect (Clermont and Eisenberg 2007, 448). They also questioned Bhattacharya, Galpin, and Haslem’s findings on the ground that the bias was only found “in judge trials, not jury trials” (Clermont and Eisenberg 2007, 451). Clermont and Eisenberg’s (1996, 2007) analyses of two large samples (92,142 federal actions in their first study and 171,710 federal cases in their second) suggest that foreign parties fare better than domestic parties in civil cases. However, they concede that the reluctance of foreign parties to litigate in the United States may result in only stronger cases being brought to court by such parties in the aftermath of 9/11. Consistent with Clermont and Eisenberg’s conclusions, subsequent studies on patent litigation do not confirm bias against foreign parties (Janicke and Ren 2006; Allison, Lemley, and Schwartz 2014).

While these studies represent essential efforts to explore xenophobia in the US legal system, conclusive findings on whether decision-makers exhibit antagonism toward foreign parties are still lacking. Scholars’ analyses of observational data to identify xenophobia are subject to significant limitations as they cannot fully control case inputs, and the results may be compromised by confounding factors such as case merit. Meanwhile, xenophobia is complicated by sociopolitical, legal, and economic factors to the extent that bias may manifest in different ways under different circumstances, contributing to the inconsistencies in previous findings. Vidmar (2008) suggests that “[e]ven with statistical controls we cannot be sure if apples were being compared to oranges” and, therefore, “[a] good laboratory simulation would add insight to Moore’s findings” (p. 62).

Juror bias, race, and political views

Jurors are inevitably susceptible to bias in their decision-making processes (Kalven and Zeisel 1966). Juror simulation research has consistently shown that racial bias influences juror decisions in criminal justice (Sweeney and Haney 1992; Mitchell et al. 2005). An early meta-analysis of fourteen experimental studies identified juror bias against minority defendants, mainly Black defendants (Sweeney and Haney 1992). Meanwhile, a small but growing body of work has recognized anti-Latinx bias (Espinoza and Willis-Esqueda 2008; Willis-Esqueda, Espinoza, and Culhane 2008; Espinoza et al. 2015; Minero and Espinoza 2016) and anti-Asian bias (Phan, Espinoza, and Sy 2022) in criminal cases. Research has found that the defendant’s country of origin and ethnicity influence levels of punitiveness both independently and in interaction with residency status (undocumented immigrant, refugee, and US citizen) (Costelloe, Stenger, and Arazan 2021).

Racial bias has been found in civil jury verdicts as well. Studies have shown racial disparities in juror liability verdicts and damage awards: minorities receive smaller awards than Whites (Chin and Peterson 1985; Greene and Bornstein 2003). African-American plaintiffs win less often and receive smaller awards in damages than their White counterparts in torts and sexual harassment cases (Bothwell et al. 2006; Cardi, Hans, and Parks 2020).

Yet, racism plays out in complex and nuanced manners. For instance, according to aversive racism theory, individuals with egalitarian values tend to mitigate their overt bias toward outgroups (Dovidio, Gaertner, and Pearson 2016). People correct their racial bias and thus make verdicts less discriminatory, or even favorable, to minorities, especially when the subject of race is salient (Sommers and Ellsworth 2000, 2001). Even when concerns for social desirability are reduced, liberals still tend to judge minority defendants more favorably (Salerno et al. 2023).

A well-established pattern in mock-juror simulation studies is the similarity-leniency effect derived from social identity theory: jurors are inclined to render more lenient decisions toward defendants with the same racial or ethnic identity as themselves, while they treat defendants from a different racial or ethnic background more harshly (Kerr et al. 1995; Mitchell et al. 2005; Devine and Caughlin 2014). The effect of similarity leniency is muted when the evidence that disadvantages the defendant is strong or conclusive (Kerr et al. 1995). Outgroup bias has been found in mock juror studies involving immigrant defendants as well (Minero and Espinoza 2016; Navarro, Heath, and Stein 2022). In addition, Schwartz and Hunt (2011) report that Latinx mock jurors exhibit a bias in favor of Latinx plaintiffs compared with European American mock jurors in the civil suit setting.

Beyond racial bias against defendants, juror research has investigated other litigant or juror factors that affect decision-making. Local plaintiffs are more likely than remote plaintiffs to receive large awards, but there is no statistically reliable evidence suggesting that the geographic location of a defendant corporation affects damage awards (Hastie, Schkade, and Payne 1999). When it comes to the effects of political ideology on juror decisions, legal authoritarianism, construed as the perception that government power trumps individual rights, has a modest positive association with conviction rates and penal severity (Narby, Cutler, and Moran 1993; Barnett, Brodsky, and Davis 2004; Butler and Moran 2007; Devine and Caughlin 2014; Jones, Jones, and Penrod 2015). However, little research has investigated the effects of legal authoritarianism in the civil legal setting (Greene and Bornstein 2003; Vinson, Costanzo, and Berger 2008).

Notwithstanding its impressive size, the literature on bias and juror decision-making suffers from a few gaps. First, studies of juror racial bias are primarily anchored in the Black-White binary, and bias stemming from perceptions of other races and ethnicities or other factors has been underexplored (Hunt 2015). Second, the role of racial bias in shaping juror decisions in civil litigation remains underexamined. Third, existing research is often devoid of sociocultural and ideological considerations, so little is known about the moderating effects of mock jurors’ traits other than ethnicity and race.

The present study: juror xenophobic bias in a patent trial

Given the insufficient examination of xenophobic bias in juror decision-making, the present research examines xenophobic bias among mock jurors in a civil trial. The simulation design of the study is expected to overcome the shortcomings of previous observational studies and provide more compelling evidence regarding xenophobic bias. Additionally, this study is grounded in the conceptualization of xenophobic bias as being constituted by racism (bias arising from the interplay between juror ethnicity and the ethnic background of the litigant) and nationalism (a sentiment of superiority highly susceptible to individual political views).

Xenophobia is socially constructed. This study draws on Devine and Caughlin’s (2014) conceptual model and posits that juror xenophobic bias is a product of a confluence of sociopolitical context, juror traits and views, and case factors and settings. The contemporary political milieu in which nationalist ideology and anti-globalization policies are prominent provides the backdrop for this examination. Seminal research on xenophobia has situated its investigation in the context of patent disputes (Moore 2002; Allison Lemley, Schwartz 2014), and the politicization of intellectual property protection amidst anti-globalization in recent years suggests that a patent dispute is ideal for testing the foreigner effect. Thus, a patent dispute was chosen for the simulation study to test whether mock jurors hold a bias against foreign companies and whether mock jurors’ race/ethnicity and political ideology moderate their verdicts.

Specifically, I tested seven hypotheses in this study. First, I tested the general xenophobia hypothesis that mock jurors would demonstrate bias in their decisions against foreign companies in a patent infringement case. The first hypothesis goes as follows:

H1 A foreign company as the defendant will produce higher rates of liability verdicts than a domestic company as the defendant.

The next question of interest goes to whether mock jurors’ prejudice against foreign companies specifically targets companies of a particular ethnic or national origin. Previous research indicates the association between Japanophobic sentiments, which peaked in the 1980s and 1990s in the US, and biased treatment of foreign companies in American courtrooms (Moore 2002). Recent studies report the antagonism toward Chinese individuals and companies since the US-China trade war (Kim 2018; Fang and Li 2021; Li 2021). Given these backgrounds, I selected Japan and China as the two nations of origin for the experiment. The second hypothesis concerns mock juror bias against Asian companies, and the third hypothesis involves bias against a Chinese company in particular.

H2 An Asian company as the defendant will produce higher rates of liability verdicts than the domestic company condition.

H3 A Chinese company as the defendant will produce the highest rates of liability verdicts of all the conditions.

Per the similarity-leniency effect (Mitchell et al. 2005; Devine and Caughlin 2014; Schwartz and Hunt 2011), the fourth and fifth hypotheses postulate that jurors will show leniency toward companies from countries whose majority populations share their ethnic identity and will disfavor companies from countries with citizenry whose majority differs in ethnic background.

H4 Asian jurors will be less likely to find Asian companies infringe, and White jurors will be less likely to find companies from White-majority countries infringe.

H5 Asian jurors will be more likely to find White companies infringe, and White jurors will be more likely to find Asian companies infringe.

I then examine how political views shape jurors’ decisions regarding foreign companies. Political conservatism is characterized by “resistance to change” and “acceptance of inequality” (Jost et al. 2018, 135). Right-wing authoritarianism is associated with economic protectionist attitudes (Jedinger and Burger 2020). A Gallup survey from 2021 documented that only 44 percent of Republicans viewed foreign trade as an economic opportunity, compared to 79 percent of Democrats, and 51 percent of Republicans saw foreign trade as more of a threat than an opportunity (Younis 2021). Given the association between political conservatism and protectionism, I hypothesize that jurors with a conservative nationalist political ideology are prone to bias against foreign companies.

H6 Jurors identifying with conservative political ideology will be more likely to find liability against a foreign company.

In addition, drawing on the associations between faith in Trump and punitiveness toward Chinese defendants charged with white-collar crimes (Reisig, Holtfreter, and Cullen 2024), it is postulated that political conservatism produces an anti-Chinese bias in foreign business lawsuits. I hypothesize that politically conservative jurors will be more likely to find against Chinese companies relative to other jurors:

H7 Jurors identifying with conservative political ideology will be more likely to find liability against a Chinese company.

Data and method

Recruitment and participants

The online experiment employed the Prolific platform to present stimulus materials and collect data. Pilot studies were performed from December 2020 to October 2021 to test materials and measure the potential effect size. Based on the results from the pilots, a power analysis indicated that the total sample should consist of 1500 jury-eligible participants, with 375 participants per cell, to have sufficient power to uncover a small (10 percent) effect. Data for the full study was collected between October 25 and November 2, 2021.

A representative sample was utilized to enhance the study’s validity and generalizability. I used Prolific’s representative-sample service to match the sample’s demographic distribution to US Census data on age, sex, and ethnicity. Employing this service provides a more diverse and representative sample with respect to these demographic characteristics than community member samples and student samples. I recruited 1659 jury-eligible adults by screening for citizenship, age, language spoken, and enfranchisement. One hundred and ninety-seven participants either quit the study before completion or failed to pass at least one quality-check question, with an attrition rate of 11.87 percent, which fell within the normal range (Cullen and Monds 2020). The final sample comprised 1462 juror-eligible adults, and participants were compensated at a rate of $8.19 per hour.

The average age of participants in the sample was forty-four years old, with a range of eighteen to ninety-two. Of these, 50.21 percent were female. In terms of race, 73.94 percent self-identified as White, 13.27 percent as Black or African American, 4.31 percent as Hispanic or Latinx, 6.57 percent as Asian, and the remainder as American Indian or Alaska Native, Pacific Islander, or other. Regarding other demographic characteristics of the sample, 64.09 percent had a college degree, 59.44 percent worked full-time or part-time, and 45.35 percent were married or in a domestic partnership. The median annual personal income fell in the range of $40,000–$49,000 (US).

In terms of political ideology, 45.96 percent of the participants identified themselves as Democrats, 15.53 percent as Republicans, 35.98 percent as Independents, and 2.53 percent as having other political affiliations. As for political attitudes, 7.59 percent identified as strongly conservative, 13.13 percent as moderate-leaning conservative, 18.95 percent as moderate, 32.15 percent as moderate-leaning liberal, 26.33 percent as strongly liberal, and 1.85 percent as holding other political attitudes.

Design and case synopsis

The experiment was a one-factorial stimulus design with four conditions, where only the defendant company’s country of origin varied (US, Japan, France, or China). The trial material consisted of texts and images, including one GIF image. To improve the ecological validity of the design (Bornstein 1999; Breau and Brook 2007), the trial utilized materials from a real patent case in which a foreign party was the defendant. The simulation provided elaborated arguments from both parties rather than just a description of the case’s fact pattern. To enhance the participants’ awareness of the defendant company’s country of origin, pictures of a middle-aged businessman, varying in ethnicity, were presented and described as the company’s president. The pictures in the four conditions were selected based on matched ethnicity, age, gesture, and appearance. All of the “company presidents” were men wearing business suits and smiling.

In the simulation, the defendant company was accused of infringing the No. 1,234,567 patent owned by the plaintiff company (a US company) for making and importing to the United States a chainsaw (the Accused Product) and selling it to customers and Walmart. The synopsis provided participants with (1) the background of the two parties, including their names, residences, and businesses, (2) a concise description of the dispute brought to the court, and (3) a description of the Accused Product. The trial material was then presented to participants, starting with opening statements by the attorneys.

Patent trials always involve complex and intellectually challenging issues concerning law and technology. To simplify the technical aspects of this simulation, the case was reduced to one central issue: whether the defendant’s Accused Product infringed the plaintiff’s patent. After a GIF-based introduction to the patent-design mechanism, the plaintiff’s attorney presented evidence in the form of pictures of X-ray tests and high-resolution photos of the product, arguing that the Accused Product matched the patent’s description. Then, the defendant’s attorney challenged the reliability of the plaintiff’s tests and their interpretation of the results. The trial was then concluded.

Mock jurors were asked to take two tests to confirm they understood who the parties were, the nature of the case, and the defendant’s country of origin in the case. The judicial instructions then appeared on the screen, asking the participants to decide whether they believed the plaintiff had proved by a preponderance of the evidence that the Accused Product infringed the patent. After submitting their liability verdict, the participants indicated their confidence in their verdict on a one to ten scale. The average confidence level was seven out of ten. Mock jurors who voted in favor of the plaintiff were subsequently asked to select a damage award between $0 and $3,000,000 based on the plaintiff’s claim. A bounded numeric scale was used to reduce the drastic variation in dollar awards (Sunstein et al. 2008).

Upon completion of the liability verdict and damage award decisions, participants were asked to answer a series of questions on a seven-point Likert scale about their perceptions of the two parties. These questions included assessments of each party’s credibility and evaluations of the statements such as “The defendant was a patent thief” and “The defendant’s acts were fair competition.” Participants were also asked for demographic background information, including age, gender, education, employment status, personal income, marital status, and religious beliefs. Additionally, geographic location (West, Midwest, South, Northeast), area of residence (urban, suburban, rural), political ideology (party affiliation and attitudes), and exposure to political news outlets were recorded. On average, it took participants a median of 18.24 minutes to complete the simulation procedure. Each participant was compensated with $2.50 on Prolific.

Variables

The key dependent variable measures the juror verdict: whether the plaintiff proved by a preponderance of the evidence that the defendant company infringed the patent. The second dependent variable is the damage award granted to the plaintiff, ranging from $0 to $3,000,000. Only mock jurors who had submitted a yes verdict on the liability issue were asked to determine the damage award. The case scenario was primarily designed to focus on liability, and the variable of the damage award was considered supplemental.

The key independent variable is the country of origin of the defendant company. I created the variable Company Nationality to represent the four conditions (1 = US company, 2 = Japanese company, 3 = French company, and 4 = Chinese company). To test the hypotheses regarding differences between the domestic-company condition and the foreign-company condition, the Asian-company condition versus the non-Asian-company condition, and the non-Chinese-company condition versus the Chinese-company condition, I created three binary variables: Foreign Company (1 = Foreign-company condition, 0 = Domestic-company condition), Asian Company (1 = Asian-company condition, 0 = non-Asian-company condition), and Chinese Company (1 = Chinese-company condition, 0 = non-Chinese-company condition).

The moderating variables are mock jurors’ race/ethnicity and political views. The race measure was recoded as a three-category variable: Asian, White, and other. Mock jurors’ political views were recoded as liberal, conservative, and moderate or other. For regression analyses, I also created the following dummies to examine the interaction effects: Asian (1 = Asian, 0 = non-Asian), White (1 = White, 0 = non-White), and Conservative (1 = Conservative, 0 = non-Conservative).

Results

Xenophobic bias

Overall, 43.43 percent (n = 635/1,462) of the participants voted against the defendant company and found that the company infringed the plaintiff’s patent right (see Table 1). Among the 635 participants who voted against the defendant, the mean damage award was $1,877,022, and the median damage award was $2,001,300 (see Table 2).

Table 1. Juror Liability Verdicts by Condition

Table 2. Mean and Median Juror Damage Awards by Condition

A general xenophobic bias in liability verdict decisions was not detected in cross-group comparisons. A model that tested manipulation differences across the four groups showed no significance for the liability verdict [χ²(3) = 2.951, p = 0.399]. Regarding general xenophobic bias against the foreign company (H1), descriptive statistics show that 40.44 percent of the mock jurors voted against the domestic defendant company, whereas the rates for the Japanese company, the French company, and the Chinese company are higher—42.27 percent, 44.66 percent, and 46.26 percent respectively. The domestic-versus-foreign group difference is not significant [χ²(1) = 1.7448, p = 0.187]. Likewise, regarding the liability verdict, no statistically significant results were observed about bias against Asian companies [χ²(1) = 1.4647, p = 0.226] or the Chinese company [χ²(1) = 2.5273, p = 0.112] relative to the domestic company. A Chinese company being the defendant did not produce the highest rates of liability verdicts among all the conditions at a significant level [χ²(1) = 1.6302, p = 0.202]. No statistically significant between-group effects were identified, refuting the first three hypotheses.

Regarding the damage granted to the plaintiff, the median damage for the domestic company ($1,825,402) is higher than that for the French company ($1,712,614) but lower than the Japanese company ($1,978,552) and the Chinese company ($2,024,707). I conducted a nonparametric median test to assess the between-group differences in damage awards. The test revealed significant differences in damage awards across groups [χ²(3) = 12.398, p = 0.006]. Pairwise comparisons with the Bonferroni correction showed that the Chinese company was specifically penalized, as the results indicated a significant difference between the Chinese company condition and the domestic condition (p = 0.026). Neither the Japanese company nor the French company had significantly different damages compared with the domestic defendant with the Bonferroni correction (p = 1.000, p = 1.000). Therefore, there was a specific bias against the Chinese company when jurors determined the amount of damages to be awarded to the plaintiff.

Similarity-leniency effects

The second primary issue is whether ingroup/outgroup effects exist in this simulation. It was hypothesized that jurors would be less likely to find against companies from countries with majority populations of similar ethnicities (H4) and more likely to find against companies coming from countries that have different ethnic backgrounds (H5). Figure 1 visualizes the cross-group juror liability verdicts by condition and shows stark contrasts in liability verdict patterns between Asian jurors and White jurors. Whereas 62.50 percent of the Asian jurors assigned to the non-Asian-company condition (the United States and France) voted for the plaintiff, only 38.58 percent of the White jurors assigned to the same experiment conditions did so. In the Asian-company conditions (Japan and China), 31.25 percent of the Asian jurors voted against the defendant company, in contrast to 45.70 percent of the White jurors. Conversely, White jurors tended to be lenient with the US and French defendant companies (37.64 percent and 39.48 percent, respectively) but punitive toward the Japanese and Chinese companies (43.27 percent and 48.16 percent, respectively). The patterns of Asian jurors’ verdicts were reversed: only 24.00 percent and 39.13 percent were found against the Japanese company and the Chinese company, respectively, much lower percentages than those found against the US and French companies (58.33 percent and 66.67 percent, respectively). ¹ Asian jurors’ treatment of Asian companies was far more lenient than their treatment of non-Asian companies.

Figure 1. Interaction Effects Between Asian Company Condition and Juror Race on Liability Verdicts

To further examine ingroup/outgroup effects (H4 and H5), I performed logistic regression tests to predict the likelihood of liability verdict preferences. Two interaction terms were created to measure the interplay between Asian jurors and the Asian company condition and the interplay between White jurors and the Asian company condition. As illustrated in the first model of Table 3, the logistic regression test corroborated the similarity-leniency effect, as the interaction terms (juror race × company conditions) were statistically significant at a high confidence level. According to Model 1, the interaction between Asian jurors and Asian companies was statistically significant (OR = 0.23, 95% CI [0.10, 0.55], p = .001), suggesting that the likelihood of Asian participants voting against Chinese and Japanese defendants was much lower than that for all other combinations. The interaction between White jurors and the Asian-company condition is significant (OR = 2.33, 95% CI [1.45, 3.75], p < 0.001), suggesting that decision outcomes for White jurors facing Asian companies were less favorable than for other combinations. Similar patterns were observed for the interaction of juror race with the non-Asian-company condition: White jurors demonstrated great leniency toward the US and French companies (OR = 0.43, 95% CI [0.27, 0.69], p < 0.001), while Asian jurors displayed punitiveness against the US and French companies (OR = 4.33, 95% CI [1.81, 10.35], p = 0.001). ² Hence, H4 and H5 testing for the similarity-leniency effects and outgroup bias were supported. Notably, both ingroup and outgroup effects were much stronger for Asian participants than for White participants, as Asian participants were more lenient with companies sharing an ethnic background and more punitive toward companies with a different background than White jurors were.

Table 3. Logistic Regression: Interaction Effects between Asian-Company Condition and Juror Race on Liability Verdicts (N = 1,462)

For mock jurors who found liability, the tests on the damage award yielded no significant effects of similarity leniency. A factorial ANOVA test was conducted with the damage award as the dependent variable and juror ethnicity, the company’s country of origin, and the interaction term between Asian jurors and the Asian-company condition as the independent variables. The interaction term between Asian jurors and the Asian-company condition did not reach statistical significance [F(1, 634) = 0.86, p = 0.3532]. Descriptive statistics reveal relatively comparable median damage awards granted by White participants for different company conditions: $1,942,359 (US), $2,002,681 (Japan), $1,997,399 (France), and $2,032,370 (China), respectively. By contrast, Asian mock jurors assigned to the Japanese condition awarded notably lower median damages ($966,039), in contrast to a substantially higher median damages award given to the Chinese company ($2,001,300). The penalization of the Chinese company by Asian jurors could account for the absence of the expected similarity-leniency effect in damage awards.

Effects of Juror Political Views

My third goal is to assess how political views impact mock jurors’ decisions regarding companies of different countries of origin. H6 and H7 postulated that juror conservative political views predict xenophobic bias against foreign companies and the Chinese company. Figure 2 breaks down voting rates based on the interaction of experimental conditions and political views. Liberal jurors, moderate jurors, and jurors with other political views did not display different voting patterns regarding foreign companies versus the domestic company. Of the mock jurors, 42.65 percent who self-identified as liberal voted for the plaintiff in the US-company condition, compared with 42.39 percent in the foreign-company condition. Likewise, participants identifying themselves as moderate or other treated the domestic company not much differently from foreign companies (46.15 percent versus 42.92 percent). Yet, regarding conservative mock jurors, verdict rates differed significantly between the domestic-company condition and the foreign-company condition: only 27.78 percent of the conservative jurors found for the plaintiff when the domestic company was the defendant, but the rate nearly doubled (51.52 percent) when a foreign company was the defendant. The exceedingly low verdict rates in the domestic-company condition suggest conservative jurors have strong protectionism toward the domestic company.

Figure 2. Interaction Effects between Foreign-Company Condition and Juror Political View on Liability Verdicts

To further predict the effects of conservative political ideology on juror verdicts toward foreign companies, logistic regression analyses were performed, with the liability verdict being the dependent variable and the defendant’s country of origin (domestic chosen as the referent), the juror’s political views (liberal chosen as the referent), and the interaction between the foreign-company condition and juror conservativeness being the predictors. Logistic regression analyses corroborated the postulated protectionism. The model reported in Table 4 exhibits the interaction effects between the company’s nationality and the jurors’ conservatism. Generally, conservative participants were less likely than their liberal counterparts to find against the defendant company (OR = 0.50, 95% CI [0.29, 0.89], p = 0.02), probably due to their tolerance of infringement informed by the conservative ideology of “economic freedom.” The interaction term was statistically significant (OR = 2.88, 95% CI [1.53, 5.45], p = 0.001). Hence, liability verdict decisions rendered by self-identified conservative jurors in the foreign-company condition were less favorable to the defendant company compared to all other combinations. These results, consistent with Figure 2, support H6: conservative jurors penalize foreign companies and favor the domestic company defendant.

Table 4. Logistic Regression: Interaction Effects between Foreign-Company Condition and Juror Political View on Liability Verdicts (N = 1,462)

Less clear-cut were the findings regarding the interaction between juror conservatism and the Chinese-company condition (H7). As illustrated in Figure 3, liberals, moderates, and jurors with other political views did not markedly differ in their liability verdicts across the Chinese-company condition and the non-Chinese-company condition. The verdict rates by the liberal mock jurors are 41.47 percent (the non-Chinese company) versus 45.18 percent (the Chinese company). Among participants with moderate or other political views, the rates are 45.02 percent (non-Chinese company) and 39.73 percent (Chinese company). Conservative jurors appeared to exhibit more punitiveness toward the Chinese company than other companies (56.16 percent versus 42.61 percent).

Figure 3. Interaction Effects Between Chinese-Company Condition and Juror Political View on Liability Verdicts

However, a logistic regression test (Table 5) indicates no evidence for the hypothesis that conservative jurors are distinguishably more likely to find against the Chinese company. The interaction between the jurors’ conservative ideology and the Chinese company was not significant (p = 0.10). All of the aforementioned effects remained consistent in the omnibus model, as presented in Supplemental Table 6 in the supplemental document.

Table 5. Logistic Regression: Interaction Effects between Chinese-Company Condition and Juror Political View on Liability Verdicts (N = 1,462)

There is no evidence that political views moderate the effects of a company’s country of origin on mock jurors’ determination of damage awards. Results of ANOVA tests show that the interaction between conservative ideology and the company’s foreign origin was not significant [F (1, 634) < 0.01, p = 0.9834]. In addition, there was no evidence to suggest that conservative jurors were likely to award a higher amount in the Chinese company condition [F (1, 634) = 2.61, p = 0.1069]. Pairwise comparison tests revealed that participants with non-conservative ideologies held anti-Chinese bias in awarding damages (p = 0.0348).

Juror Perceptions of Credibility and Acts

To delve into the reasons why no main effects of bias were found, I investigated how mock jurors’ perceptions of the company’s credibility and acts might inform their liability verdict patterns. Kruskal-Wallis tests show no between-group differences in credibility ratings of both the plaintiff company and the defendant company [χ²(3) = 4.184, p = 0.2423; χ²(3) = 6.646, p = 0.0841] (Table 6). This shows no evidence of the association between the company’s country of origin and the participant’s perception of the credibility of both parties in this simulation.

Table 6. Perceptions of Plaintiff and Defendant Credibility and Acts by Condition

Notes: Plaintiff credibility ratings range from 1 (“not at all credible”) to 7 (“very credible”). Defendant credibility ratings range from 1 (“not at all credible”) to 7 (“very credible”). Ratings of whether the defendant was a thief range from 1 (“strongly disagree”) to 7 (“strongly agree”). Ratings of whether the defendant engaged in fair competition range from 1 (“strongly disagree”) to 7 (“strongly agree”).

In terms of the statements “The defendant is a thief” and “The defendant’s act is fair competition,” the between-group differences are significant [χ²(3) = 13.228, p = 0.0042; χ²(3) = 8.444, p = 0.0377] (Table 6). In a series of follow-up Bonferroni tests, I found that participants assigned to the Chinese company condition (compared with those assigned to the domestic company condition) were more likely to endorse the statement that the defendant company was a thief (p=0.0012) and less likely to support the statement that the defendant’s act counted as fair competition (p = 0.0162). Mock jurors’ evaluations of the defendant company’s acts did speak to their antipathy toward the Chinese company. The anti-Chinese sentiments remained strong compared to the domestic condition in damage awards (p = 0.0020) among the mock jurors when asked their opinions about the defendant being a thief, which may account for the anti-Chinese effect in granting damage awards.

Discussion

This study investigated whether mock jurors exhibited xenophobic bias against foreign companies through a patent simulation. The experimental design of the study addresses the selection bias issue present in previous studies and offers a more valid and reliable method for testing xenophobic bias in mock juror decision-making. I found no evidence of a general bias toward foreign companies in liability verdicts rendered by mock jurors. However, a specific bias against the Chinese company in awarding damages was identified. Additionally, the results corroborated the similarity-leniency effects and the conservatist protection effects. These findings suggest that while there is no direct crass xenophobia, xenophobic bias is manifested in a more nuanced way.

The nuances in the findings lend support to the conceptualization of xenophobia as consisting of ethnic racism and civic nationalism. As proposed earlier, this article theorizes two components of xenophobia in the context of juror bias: racism and nationalism. The former manifests in the form of the similarity-leniency effect, while nationalism, representing a type of prejudice associated with modern state-building, is highly sensitive to personal political views and sociopolitical milieu.

The conceptual framework refutes a monolithic view of xenophobic bias among juror decision-making that previous empirical research has adopted (Moore 2002; Clermont and Eisenberg 1996; 2007). Instead, this study supports the idea that xenophobia has multidimensional facets (Wimmer 1997; Hjerm 2001; Esses et al. 2005; Yakushko 2009) and is context-dependent. Xenophobia manifests as a form of tribalism by which subgroups tend to be more lenient toward those who share their identities and values and harsher toward those who do not (Clark et al. 2019). The ingroup-outgroup effects in the encounter of foreign parties take the forms of both racism and nationalism in inducing xenophobic bias, as racial or ethnic homogeneity produces similarity leniency and conservative ideology leads to protectionism. This theorization of xenophobic bias in juror decision-making offers a conceptual framework for future law and social science research to further probe into the interrelationship between racism and nationalism in juror decision-making.

Despite not finding anti-Chinese bias in mock juror liability verdicts, the study reveals that mock jurors penalized the Chinese company by awarding higher damages to the plaintiff. This finding, to some extent, supports the presence of anti-Chinese bias previously identified in criminal justice (Kim 2018; Fang and Li 2021; Reisig et al. 2024) and provides further evidence of racial bias in civil justice (Chin and Peterson 1985; Greene and Bornstein 2003; Bothwell et al. 2006; Cardi, Hans, and Parks 2020). However, contrary to the expectations, this study finds no evidence of an association between jurors’ conservatism and anti-Chinese bias, which may conflict with previous findings (Reisig et al. 2024). The null result could be because participants with non-conservative political views also hold a bias against the Chinese company when determining the damages award.

The bias against the Chinese company in damage awards likely reflects the prevailing narratives surrounding China’s inadequate protection of intellectual property and the increasing number of incidents related to economic espionage and security threats (Bateman 2022). Animosity toward the Chinese company in this study can be further complicated by the COVID-19 pandemic (Reny and Barreto 2022). The results underscore the idea that juror decision-making is influenced by the contemporary sociopolitical context, highlighting how xenophobia is socially constructed at different times.

The results of the study support the similarity-leniency effect identified in juror decision-making research (Mitchell et al. 2005; Schwartz and Hunt 2011; Devine and Caughlin 2014). This article affirms the applicability of the similarity-leniency effects to Asian-versus-White mock juror decision-making. The findings also demonstrate the existence of the similarity-leniency effects in the realm of civil suits.

A notable pattern observed among Asian participants aligns with the prior observation that the similarity-leniency effect is more pronounced for minority participants (Hunt 2015). Comparing the odds ratio of White jurors assessing Asian companies (OR = 2.33) to Asian jurors evaluating non-Asian companies (OR = 4.33), the magnitude of outgroup bias among Asian jurors was nearly twice as strong as among White jurors. Similarly, ingroup favoritism was more pronounced among Asian jurors (OR = 0.23) than White jurors (OR = 0.43). These findings are consistent with the stronger similarity-leniency effect among minority decision-makers observed in previous studies (Mitchell et al. 2005; Devine and Caughlin 2014) and provide evidence of its presence among Asian participants.

The understudied effect of personal political views on juror decision-making in civil trials (Greene and Bornstein 2003; Vinson, Costanzo, and Berger 2008) was assessed in this study. The proposition that political conservatism elicits juror hostility toward foreign company defendants was evidenced in this simulation. Consistent with earlier studies (Jost et al. 2018; Jedinger and Burger 2020), my findings suggest that conservatives favor domestic companies and penalize foreign companies through their verdict preferences to safeguard the interests of domestic businesses. Hence, xenophobia manifests itself as a form of economic protectionism driven by conservative ideologies. The findings add evidence to the existing literature regarding how jurors’ political views moderate their verdicts in civil justice. While previous research highlights the direct impact of jurors’ political attitudes on their decisions, the results of this simulation suggest that individual political views significantly influence juror verdicts through interactions with litigant characteristics. This reveals a more complex and nuanced role of political attitudes in courtroom decision-making.

Aversive racism and task specificity are considered in this study to account for the complexities of the overall effects of xenophobic bias. The Aversive Racism Theory is deemed plausible for this study, given the overrepresentation of liberals in the sample. Additionally, the task specificity theory posits that information richness and instruction specificity in the simulation may reduce the degree to which negative sentiments translate into juror decisions (Diamond 1997; Pfeifer and Bernstein 2003; Shaked-Schroer, Costanzo, and Marcus-Newhall 2008). The null effects of general xenophobic bias (H1–H3) in the liability verdict may be attributed to both participants’ bias correction and the task specificity of the design regarding the liability verdict.

Nonetheless, the Aversive Racism Theory does not refute the idea that individuals with egalitarian values still harbor bias. Unlike dominative racism, liberals exert bias in “subtle, indirect, and often rationalizable ways” (Dovidio, Gaertner, and Pearson 2016, 270). The presence of similarity leniency or conservatist protectionism in this study can be rationalized by the participants’ inclination to favor their ethnic and national ingroups. Similarly, the contemporary sociopolitical milieu may have contributed to the rationalization of bias against the Chinese company in the damages award, which may have been made more salient by the lack of specificity in the design for the damages award, as the vignette did not provide arguments about the dispute over the damages award.

Such bias against the Chinese company might have nullified the effects of similarity leniency and political views in the damage award. The penalization of the Chinese company by Asian participants might have counteracted ethnic ingroup favoritism, thereby nullifying the effect of similarly leniency in damages. Meanwhile, the manifestation of anti-Chinese bias among non-conservatists may have negated discernible impacts of political views on hostility toward foreign companies.

This article has several limitations. First, like many other mock-juror experiments, using an online sample and procedure can raise concerns about ecological validity. Although I used the materials of the real case to improve ecological validity, I had to tailor the original case materials and technical explanations to ensure the lay participants’ comprehension and their decision-making within a reasonable time frame. Consequently, this simulation may differ from real juror experience. Additionally, there is a substantial gap between the online setting and the real-world courtroom settings (Nuñez, McCrea, and Culhane 2011). The scope of this investigation was restricted to individual-level mock juror decision-making, while, in reality, juror verdicts are products of group deliberations. Future research is needed to understand the extent to which the observations from this simulation can be extended to real-world juror decision-making.

As for the damages awards, although the bounded damage scale was implemented to reduce outliers, it compromised ecological validity by deviating from actual court procedures. Another issue is that requiring the liability verdict as a prerequisite for the damages award decision narrowed and reshaped the participant pool in each condition. Only a subset of participants—those who found for the plaintiff in the liability verdict—proceeded to determine the damages. The subset reduced the statistical power of this analysis. Future research could mitigate this limitation by employing a vignette in which liability is already established, allowing participants to focus exclusively on determining damages.

This research faces generalizability issues due to the nature of the simulation. The patent dispute design may have intensified the effects examined in this article (Moore 2002). The peculiarities of this vignette may raise questions about the extent to which the simulation results of a patent case can be generalized to other civil case settings, such as contract disputes. Further research is needed to generalize the findings to other contexts. Moreover, this study has implications for the sociolegal inquiry into general judicial decision bias, as judges, like mock jurors, are susceptible to the influences of a range of social, political, and economic factors in their decision-making. Yet, the extent to which this study’s findings apply to judges’ decisions needs further investigation.

Finally, despite efforts to gather a participant group from Prolific that represented the national jury-eligible population in terms of age, sex, and race, the sample ended up being more liberal and more educated than the general population, with individuals fifty-eight years old and older slightly underrepresented.

Conclusion

This article investigates whether and how xenophobic bias plays out in juror decision-making. While the findings do not offer sufficient evidence of a general xenophobic bias in liability verdict decisions, they do reveal a specific bias against the Chinese company in damages awards. The study introduces a conceptual framework for understanding xenophobia as consisting of racism and nationalism, with results corroborating ingroup-outgroup bias that manifests through the interactions between the defendant’s nationality and juror characteristics and attitudes.

As the first mock juror simulation study examining xenophobic bias in civil trials, this article advances both xenophobia literature and juror-bias literature in law and social science research. It contributes to the existing mock juror research by offering compelling evidence of racial bias against a non-Black minority group, nationality-based bias, and the moderating effects of similarity leniency and political views in civil trials. With a simulation design, the study overcomes the selection bias present in earlier analyses of observational data.

The study raises important questions about ensuring access to fair trials for foreign parties, particularly when jury panels lack ethnic diversity or when conservative members are overrepresented. These findings have significant implications for judicial policies aimed at curtailing xenophobic bias in the jury system, including improving diversity and representation in jury selection, as well as increasing task specificity in jury instructions.

Amidst escalating anti-globalization sentiments and policies, xenophobia poses a significant barrier to international exchange and global peace. There is a pressing need for more law and social science research to thoroughly examine xenophobia in the legal system to better understand whether and how various forms of xenophobia are institutionalized through laws. Future research should delve deeper into the intricacies of xenophobia by further exploring racism, nationalism, and other discriminatory decision-making in the legal setting. Methodologically, research on juror xenophobic bias must enhance its ecological and external validity by incorporating in-person simulation designs, videotaped case materials, refined measures of xenophobia, and group deliberations.