Abstract
Existing evidence indicates that racial discrimination is a common, if not pervasive, feature of Black Americans’ experiences in U.S. consumer markets. However, few studies have quantitatively explored specific social psychological and interactional consequences of consumer racial discrimination. In response, we draw from literatures on experiences, attributions, and reactions to racial discrimination to posit and test for Black-White differences in consumers’ behavioral responses to dissatisfactory dining experiences. Specifically, past research shows that Black Americans’ dissatisfactory experiences in consumer markets are more often perceived to be the result of consumer racial discrimination. Given their increased exposure to racial discrimination in consumer markets and the United States more broadly, we posit that Black customers will react more punitively to dissatisfactory restaurant experiences than White customers. We test this notion using a within-subject experimental design and regression analyses of survey data collected from a consumer panel of White and Black U.S. adults (n = 307). Results indicate that Black respondents in this sample are more likely than White respondents to penalize their server’s tips and lodge a complaint when dissatisfied with restaurant food and/or service. These findings are consistent with the prediction that Black American consumers tend to react more punitively on average to dissatisfactory consumer experiences than Whites and are consistent with prior scholarship linking coercive and unjust experiences to retributive behaviors. We conclude by discussing implications of these results and the need for further research on racial discrimination in U.S. restaurants and related consumer markets.
Keywords
Historically, a central aim of sociological inquiry has been to illuminate the existence, causes, and consequences of anti-Black stereotypes, prejudices, and discrimination in the United States (Brunsma, Embrick, and Nanney 2015). Likewise, a vast body of research documents the pervasiveness of discriminatory experiences in the everyday lives of many Black Americans who, as a matter of necessity, must regularly navigate white spaces (e.g., Anderson 2015; Essed 1991; Feagin 1991; Feagin and Sikes 1994; Roscigno 2007; Rosino Forthcoming). For example, racial discrimination is a common, if not pervasive, feature of Black Americans’ experiences in frequently utilized U.S. consumer markets (Brewster and Rusche 2012; Gabbidon 2003; Gabbidon et al. 2008; Gabbidon and Higgins 2007; Harris, Henderson, and Williams 2005; Pager and Shepherd 2008; Williams, Henderson, and Harris 2001). In fact, according Crockett, Grier, and Williams (2003:6), Black consumers “come to define their marketplace experiences by the presence or absence of discriminatory treatment.” Racial discrimination in consumer markets thus reflects and reinforces the everyday nature of contemporary racism more generally (e.g., shopping while Black, hailing a cab while Black, etc.). According to Essed (1991), everyday racism includes the racial attitudes and actions infused in the fabric of society that become manifest and are reproduced in the familiar, mundane, and recurrent practices of everyday living in a White-dominated society (see also Lewis 2003; Rosino Forthcoming). One familiar social situation in which Black consumers appear particularly vulnerable to such discriminatory treatment is while dining in full-service restaurants, or what is colloquially dubbed “tableside racism” or “dining while black/brown” (e.g., Brewster, Brauer, and Lynn 2015; Brewster and Rusche 2012; Dirks and Rice 2004; Feagin 1991; Feagin and Sikes 1994; Wang 2014).
Black Americans’ enhanced vulnerability to mistreatment in this market milieu is undoubtedly a manifestation of the documented blatant and pervasive racialization of full-service restaurant consumers (cf. Brewster 2012b). According to Lewis (2003:287), racialization unfolds during everyday interactions and involves “the assignment of bodies to racial categories (assigning identities to people and groups) and the association of symbols, attributes, qualities, and other meaning with those categories (which then are understood to belong to those bodies in a primordial or natural way).” In full-service restaurants, racial categories are explicitly linked with expectations about customers’ tipping practices, such that Whiteness is commonly associated with superior tipping practices vis-à-vis Blackness. For instance, a recent survey of waitstaff from across the United States found that nearly 70 percent of servers perceived Blacks to be below average tippers. In stark contrast, a mere 2 percent of servers in this study reported White customers to be below average tippers (Brewster et al. 2015).
The fact that one racial category (Whites) is normatively perceived by servers to tip better than another category (Blacks) would be inconsequential in itself, but as Lewis (2003:294; see also Rosino Forthcoming) elaborates, “Racial identifications are not merely about thought processes but about action; acts of inclusion and exclusion are part of the racialization process.” Thus, in addition to creating racialized boundaries demarcating who is and is not predicted to tip adequately, the racialization of restaurant consumers designates who is and is not considered a desirable customer that is worthy of severs’ respect and optimal efforts (Brewster 2012b; Brewster et al. 2015). As such, to be identified as Black by restaurant severs is in itself often a sufficient condition for mistreatment while dining in full-service restaurants.
In addition to extant research on these topics, Blacks’ tipping practices and the associated risks of mistreatment are also frequent topics of discussion in the mass media, particularly in outlets oriented toward and/or produced by Black communities (e.g., The Root, Madame Noire, Ebony, Clutch, etc.). For example, in partnership with Black Public Media, Angela Tucker (director and producer) broached the topic in the inaugural episode of the satirical but candid web series titled, “Black Folk Don’t . . . [Tip]” (http://www.pbs.org/show/black-folk-dont/episodes/). Similarly, Linda Wallace (2008a, 2008b), an African American cultural coach and freelance writer, published a series of articles in prominent Black newspapers (e.g., The New Tri-State Defender and Pittsburgh Courier) delineating the implications (e.g., discriminatory service) of the interracial tipping gap for Black American communities. The Black-White tipping difference has also been underscored in several motion pictures centering on strained U.S. race relations including Crash (2004) and more recently Dear White People (2014). Given Black Americans’ personal (and vicarious) experiences with discriminatory restaurant service, that the topic is acknowledged and discussed in the media is perhaps not very surprising. In fact, according to Robinson (2014:98), the “restaurant scene”—where Blacks are perceived to be poor tippers and as such are given less than optimal service—is so familiar to Black Americans that it has become a prototypical piece of the metanarrative of interracial interactions in public spaces in the United States—alongside other well-known scenes including the “elevator,” “street,” and “retail shopping” scenes.
Despite the scientific and public attention focused on the racialization of restaurant consumers, there have been few attempts to systematically explore consequent racialized reactions to dissatisfactory marketplace experiences. For instance, if Black consumers experience pervasive racial discrimination in consumer markets, then we might expect Blacks to react more punitively to dissatisfactory experiences in consumer markets than Whites. This argument follows from research that frequently documents racial discrimination in consumer settings (e.g., Brewster and Rusche 2012; Feagin 1991; Gabbidon 2003; Harris et al. 2005; Williams, Henderson, and Harris 2001). Likewise, this expectation follows from research showing that experiencing racial discrimination can provoke anger and trigger a physiological “fight” response, foster a hostile view of relationships, and encourage disengagement from conventional norms (Burt, Simons, and Gibbons 2012; Evett et al. 2013; Feagin 1991; Klinner and Walsh 2013; Krieger 1999). Finally, this expectation flows from social psychological research that suggests Blacks and other disadvantaged groups may be particularly sensitive to discrimination cues and as such are more likely to attribute dissatisfactory events to racial prejudice (Allport [1954] 1979; Anderson 2015; Feagin 1991; Major, Quinton, and McCoy 2002; see also Baker, Meyer, and Johnson 2008).
Thus, based on these findings, Black diners faced with dissatisfactory service may be more likely than Whites to experience anger and hostility and in response reject conventional norms of the setting (e.g., to “stiff” a server by not tipping). Specifically, in this study, we posit that Black consumers will react more punitively to dissatisfactory encounters, such as by verbalizing their dissatisfaction (i.e., complaining) and/or by leaving smaller tips for their servers. We begin by reviewing the theoretical and empirical evidence underpinning our predictions, then test these predictions by analyzing responses to survey questions and an experimental vignette collected from a geographically diverse sample of Black and White U.S. consumers.
Background
Results from recent nationally representative polls identify perceptions of racial discrimination as an enduring characteristic of Blacks’ experiences in consumer markets. For instance, a 2015 Gallup Poll of consumers found that 24 percent of Black Americans reported being treated unfairly in the past month because of their race while shopping in a store, while 20 percent reported experiencing unfair treatment in a restaurant, bar, theater, or other entertainment venue. Further, perceived discriminatory treatment in these markets has been remarkably stable since the question was first asked by Gallup in 1997 and are comparable to Black Americans’ perceptions of unfair treatment in other more widely studied domains including the workplace (18 percent) and interactions with the police (18 percent) (see Newport 2015). Another recent national survey conducted by the Pew Research Center (2016) estimates that nearly half (49 percent) of Black Americans believe that Blacks are treated less fairly than Whites in stores and restaurants. In contrast, only 21 percent of Whites reported to believe that Blacks and Whites are treated unequally in these consumer settings (Pew Research Center 2016).
Substantiating Blacks’ perceptions of mistreatment in full-service restaurants in particular are studies documenting restaurant servers’ self-reported admissions to providing disparate service to their White and Black clientele. For instance, a survey of 200 servers working in 18 restaurants showed that nearly 40 percent of servers admit the quality of their service is at least sometimes predicated on their customer’s race (Brewster and Rusche 2012). Over half (52.8 percent) the servers in that particular study also reported observing their coworkers treating Black clientele poorly. Similarly, a more recent survey of nearly one thousand U.S. waitstaff revealed that 53 percent of servers admit to not always giving their Black clientele their best efforts (Brewster et al. 2015). Furthermore, qualitative accounts from both servers (e.g., Dirks and Rice 2004; Mallinson and Brewster 2005) and customers (e.g., Feagin 1991; Feagin and Sikes 1994; Kwate and Goodman 2015) point toward high rates of service discrimination against Black restaurant consumers.
While overt and egregious incidents of discrimination against Black diners continue to surface, subtle and covert forms are more typical (Bonilla-Silva 2010). For instance, Black customers may be made to wait relatively longer periods of time to be seated and served relative to their White counterparts (Ainscough and Motley 2000; Dirks and Rice 2004; Feagin and Sikes 1994; Harris et al. 2005; Lewis 2003; Walsh 2009). Discrimination against Black diners can also take the form of inattentive, discourteous, or inauthentic service delivery, such as by using avoidance tactics; giving depreciating, dismissive, or patronizing comments and looks; or invoking less interpersonal warmth or service with a “fake” smile (see Brewster, Lynn, and Crocroft 2014; Crockett et al. 2003; Feagin 1991; Klinner and Walsh 2013). Racial discrimination may further manifest as less friendliness, professionalism, appreciation, and enthusiasm embodied in servers’ interactions with their Black clientele (Hebl et al. 2002; King et al. 2006; Walsh 2009).
The subtle forms of discrimination that are levied against Black restaurant patrons in many cases reflect the explicit anti-Black server sentiments that have been well documented (e.g., see Brewster and Rusche 2012; Dirks and Rice 2000). In other cases, the dining experiences of Black consumers may be affected by unconsciously conveyed nonverbal expressions signifying unfriendliness (e.g., not smiling, blinking, lack of eye contact, etc.) that stem from servers’ implicit or unconscious racial biases (Dovidio et al. 2002; Vanman et al. 2004). Additionally, Blacks’ restaurant experiences may be tarnished in subtle ways by racial differences in communicative styles, preferences, and expectations that result in interactions between Black and White Americans characterized by misunderstanding, dissatisfaction, and negative stereotype affirmation (Rawls 2000). For instance, according to Rawls (2000:255; see also Rosino Forthcoming), “When persons from the ‘other’ group violate Interaction Order expectations” (e.g., Black diners fail to tip 15 percent to 20 percent), narratives are generated to explain these deviations, which “quickly develop into negative stereotypes” (e.g., Blacks are poor tippers because they are selfish or cheap). This process of interactional misunderstanding and narrative interpretation may then contribute to the reproduction of White hegemony (e.g., Blacks experience chronic prejudice and discrimination).
Given the elusiveness of servers’ explicit and implicit discriminatory behaviors, Black diners may not always detect racially discriminate service (see Brewster and Brauer 2016). However, when faced with the attributional ambiguity inherent in most service failures (see Baker et al. 2008), 1 Black customers’ awareness of their disadvantaged positioning in the U.S. racial hierarchy, coupled with their accumulated experiences with mistreatment in public accommodations and society at large, may lead many to attribute dissatisfactory dining experiences (e.g., inattentiveness) to racial discrimination. Black Americans, in other words, may be on guard, suspicious, and sensitive to the real possibility that their perceived race is the cause of service shortcomings they experience (see Allport [1954] 1979; Anderson 2015; Baker et al. 2008; Dovidio et al. 2002; Essed 1991; Feagin 1991; Major et al. 2002; Robinson 2014).
Results from one innovative online experiment lend credence to this line of argumentation. Specifically, Baker and colleagues (2008) presented Black and White respondents with a hypothetical dining scenario involving a Black couple that was given inattentive service by a White waitress. Black respondents were significantly more likely than their White counterparts to attribute the service failure to racial discrimination. Black respondents were also more likely than Whites to believe that the customers’ dark skin played a role in the service failure and that the waitress was likely racist. Furthermore, compared to Whites, Black respondents on average perceived the service failure as more severe, experienced higher levels of anger in response to the poor service portrayed in the scenario, and reported that they would require greater service recovery efforts (e.g., apology and discount or full refund) in order to be satisfied, return to the restaurant, or recommend the restaurant to friends.
Further, when service shortcomings are attributed to racial discrimination, Black customers might respond by actively confronting the transgressor and/or lodging a formal complaint with a company representative or external governing body such as the Better Business Bureau or Chamber of Commerce (Crockett et al. 2003; Evett et al. 2013; Feagin 1991; Gabbidon et al. 2008; Walsh 2009). According to Walsh (2009:158), “Challenging the discrimination at the time [it occurs] allows [victims] to settle the complaint in person and to let go of the negative thoughts.” Lodging formal complaints or confronting the perpetrator of the mistreatment further allows victims to experience a sense of retribution for the transgression they have endured (see Crockett et al. 2003).
Nevertheless, confrontational reactions to perceptions of racial discrimination are not without accompanying costs (Crockett et al. 2003; Feagin and Sikes 1994; Kaiser and Miller 2001). Such responses not only require that the victim expend time and energy but also renders them vulnerable to other negative social consequences, including being labeled a troublemaker or a complainer and accused of being argumentative, hypersensitive, paranoid, or unduly emotional (Feagin and Sikes 1994; Kaiser and Miller 2001). Further, victims who actively react to racial discrimination by confronting their transgressor run the risk of being accused of playing the “race card” to advance their own self-interest. For instance, one of Feagin and Sikes’s (1994:279) Black middle-class respondents explains: “Now, you can’t holler discrimination, because they’re going to think that you’re trying to get a lawsuit, or trying to [get] a free meal or something, so you really can’t come out and say, ‘I’m being discriminated against.’”
In an attempt to avoid the costs associated with confrontational reactions to mistreatment, victims of perceived discrimination may respond by simply ignoring the incident, withdrawing from the situation, vowing not to return to the establishment, or seeking support from their friends and family (Crockett et al. 2003; Evett et al. 2013; Feagin 1991; Gabbidon et al. 2008; Walsh 2009). In the restaurant context, customers who perceive to be victims of discriminatory treatment have access to another means of recourse—they can reduce or eliminate a server’s tip. Penalizing a server’s tip may mitigate the negative emotions triggered by experiencing unfair or discriminatory service (Evett et al. 2013; Higgins and Gabbidon 2009; Wang, Leu, and Shoda 2011) while also allowing victims to avoid the social costs associated with directly confronting their transgressor.
Of course, many customers are likely to penalize a server’s tip following a dissatisfactory restaurant experience irrespective of race or the perception of discrimination; this type of economic tit-for-tat logically flows from the instrumental server-client interactions shaped by the institution of tipping. However, we expect Black customers to be more likely than Whites to penalize servers’ tips and to do so more severely in response to dissatisfactory service. This is because Blacks are more frequently and systemically exposed—in restaurants and many other U.S. institutions and social domains—to coercion and discrimination. As noted previously, racial discrimination can trigger a physiological “fight” response and angry or hostile reactions (Baker et al. 2008; Burt et al. 2012; Krieger 1999; Schmitt and Branscombe 2002). Also, frequent or prolonged exposure to coercion and discrimination may foster the internalization of cultural “codes” or “cognitive schemas” that emphasize the competitive and instrumental nature of human interaction and the importance of exacting retribution following disrespect (Anderson 2000, 2015; Jacobs and Wright 2006; Simons and Burt 2011; see also Essed 1991).
Following this logic, we expect that Black customers’ reactions to dissatisfactory service encounters will be more punitive or retributive on average compared to Whites. In other words, we expect that tip penalties following dissatisfactory restaurant service might, at least for some Black customers, reflect the retributive meting out of “tableside justice” in reaction to disrespectful and perceived racially discriminatory treatment. While we are unaware of prior academic research exploring potential racial differences in reactions to dissatisfactory consumer experiences, the results of a 10-item online survey of 646 Black and 121 White readers of The Root, an online Black American–focused daily news site, are consistent with our predictions. According to Sharif (2011), results from The Root poll showed that White respondents were more forgiving of server transgressions than their Black counterparts. Specifically, when asked whether there were ever reasons not to tip restaurant servers, 49 percent of White respondents indicated that they would always tip even if the food and service was poor. In contrast, 37 percent of Black respondents indicated that they would always leave a tip regardless of the food and service quality.
In sum, existing evidence derived from self-reports from both victims and perpetrators point to high rates of racial discrimination in U.S. full-service restaurants. Nevertheless, studies assessing the consequences of restaurant servers’ discriminatory practices are quite rare. For instance, while prior research indicates that Black customers are more likely than Whites to attribute negative service encounters to racial discrimination, it remains unclear how such attributions affect Black customers’ behavioral reactions to dissatisfactory restaurant encounters. In response, the aim of the current study is to test for differences in Black and White consumers’ reactions to (dis)satisfactory dining experiences. We posit that Black respondents will be more likely than Whites to react punitively to dissatisfactory service by actively lodging a complaint and/or leaving their server a smaller tip.
Method
Sample
This study analyzes survey data anonymously collected from an online sample of Survey Sampling International’s U.S. consumer panel. The survey asked about consumers’ attitudes, opinions, and experiences with dining in full-service restaurants. Stratified sampling was used to attain an approximately equal number of responses from Asian, Black, Hispanic, and White consumers. In total, 914 respondents completed the survey. Those reporting their racial/ethnic affiliation to be something other than Black or White (n = 442) were omitted from the current analysis. 2 Of the remaining 472 Black and White respondents, 57 cases were omitted due to missing values on one or more of the dependent or independent variables in our analysis. Participants who provided illogical and/or questionable tip amounts across the within-subject dissatisfaction conditions (n = 82) were also deleted from the sample. 3 Data quality concerns with participants’ responses to questions used to create our propensity toward complaining led us to omit 26 additional cases. 4 These procedures resulted in a geographically diverse (48 states and the District of Columbia) final analytic sample of 129 Black and 178 White participants (see Table 1). 5
Descriptive Statistics for Variables in Analysis (n = 307).
Dependent Variables
Reactions to dissatisfaction
This study includes two measures of respondents’ reactions to dissatisfactory restaurant experiences. The first is based on responses to a hypothetical experimental vignette that assesses within-subject changes in the tips that participants report they would leave their server under various conditions of satisfaction or dissatisfaction with the food and service. Respondents were shown a nondescript picture of a restaurant dining room and were randomly assigned a hypothetical dining scenario involving a server who was a White male, Black male, White female, or Black female. 6 Participants’ perceptions of their servers’ race and sex were manipulated by using African American (e.g., Jamal and Tamika) or White-sounding female and male names (e.g., Brad and Emily; see Bertrand and Mullainathan 2004). In each scenario, the size of the bill was held constant ($45.50). Following a prompt, respondents were asked to indicate how much in dollars and cents they would tip their server if they were (1) satisfied with both the food and quality of service, (2) dissatisfied with both the food and quality of service, (3) satisfied with the food but dissatisfied with the quality of service, and (4) dissatisfied with the food but satisfied with the quality of service. 7 Thus, tip amount was measured across four within-subject levels of dissatisfaction, two between-subject levels of servers’ race (White and Black), and two between-subject levels of servers’ sex. In the current study, respondents’ race (Black or White) is the primary between-subject factor of interest.
We also assess respondents’ reactions to dissatisfaction with a measure of respondents’ propensity to complain when dissatisfied with food and/or service. This measure was created using four items asking respondents to indicate their level of agreement (1 = strongly disagree to 7 = strongly agree) with the following statements: “I always let my server know if a mistake has been made with my order,” “When the service or food in a restaurant is unsatisfactory I have a duty to complain,” “I typically do not complain even if I am dissatisfied with the food and/or service (reverse coded),” and “I tend to eat my meal even if it has not been prepared correctly (reverse coded).” Principle component analysis indicated all four items loaded on a single factor with an eigenvalue greater than 1 (2.39) that accounted for 60 percent of the total item variance. Thus, responses to these items were averaged to create an index (α = .77) on which higher values indicate a greater propensity to complain when dissatisfied with food or service in restaurants.
Independent and Control Variables
Our primary independent variable in this study—participants’ race—is a dichotomous variable indicating the respondent is Black (=1) or White (=0). Our analyses also include controls for exogenous factors that might account for associations between participants’ race, restaurant experiences, and reactions to dissatisfactory conditions, such as sociodemographic characteristics, experiences with working for tips, and restaurant patronage. Respondent’s sex is coded as female (=1) or male (=0). Age is a continuous variable measured in years. Educational attainment is measured as an ordinal variable ranging from high school/GED or less (=1) to professional or doctoral degree (=6).
Annual household income is measured on an 8-point scale ranging from below $20,000 (=1) to $80,000 or more (=8). Potential regional differences in participants’ attitudes, experiences, and behaviors are accounted for with nested dummy variables contrasting South (=1), Midwest (=1), and West (=1) with those residing in the Northeast (=0). Participants were also asked to indicate if they are currently or have in the past been a restaurant server or bartender and whether they are currently or have in the past worked for tips outside of the restaurant industry (e.g., taxi driver, barber, hair stylist, food delivery driver, etc.). Those indicating they have worked for tips by answering affirmatively to either question (=1) are contrasted with those who reported having no direct experience with working for tips (=0).
Dining frequency is measured with an item indicating how frequently (from less than once a year = 1 to daily = 8) the respondent visits full-service restaurants. Participants were also asked to report whether they typically visit inexpensive (=1, a meal tends to be less than $10), moderately expensive (=1, a meal tends to be $10 to $20), or expensive (=0, a meal tends to be more than $20) restaurants. Respondents also identified the most common meal consumed (dinner = 1; breakfast/brunch/lunch = 0) and the main types of food served (North American food = 1, all other types of cuisine = 0) at the restaurants they typically visit. In addition, we control for length of recall period with an ordinal item assessing the time since last visit to a full-service restaurant (from today = 1 to more than 6 months ago = 6). Finally, in the within-subject analysis predicting tipping behaviors, we also include controls for servers’ race (Black = 1) and sex (female = 1). Table 1 presents descriptive statistics for the analytic sample used in this study.
Results
In the first analysis, we estimate a repeated measure general linear model predicting the amount (in dollars and cents) a respondent would reportedly tip across four hypothetical experimental conditions involving satisfactory or dissatisfactory dining experiences after receiving a $45.50 bill in a full-service restaurant. In this model, dissatisfaction is a variably manipulated within-subjects factor, and participant’s race is the primary between-subjects factor of interest. The model also includes between-subjects controls for server’s race and sex (variably manipulated between-subjects factors) and participants’ sex, education, income, age, experience working for tips, and restaurant usage patterns (e.g., dining frequency, typical meal, time since last visit, expensiveness of restaurants typically visited).
A test of between-subjects effects produced by this analysis (not shown here) revealed significant differences in tip amounts across Black and White respondents (F = 18.02, p < .001). Further, a test of within-subjects effects (not shown here) showed significant differences in tip amounts across the dissatisfaction with food and/or service quality conditions (F = 35.05, p < .001). Moreover, consistent with our predictions, these tests indicated the existence of significant racial differences in the effects of dissatisfaction on tipping (F = 3.56, p < .05). The direction of these observed tipping differences is revealed in Models 1 through 4 of Table 2.
Parameter Estimates from Repeated Measures GLM and OLS Analyses Predicting Participants’ Reactions to Dissatisfaction.
Note: Standard errors in parentheses. GLM = generalized linear model; OLS = ordinary least squares.
Repeated measures GLM estimates.
OLS estimates.
p < .05.
Table 2 presents separate regression models (1–4) predicting tip amount in each satisfaction/dissatisfaction condition. These models show that Black respondents tend to tip less on average than Whites across all four conditions. As predicted, racial differences in reported tip amounts are smallest when respondents are satisfied with food and service, whereas Blacks react more punitively than Whites, as indicated by harsher tip penalties, when faced with dissatisfactory food and/or service. For instance, when the quality of food and service are both satisfactory (Model 1), the average tip at mean/baseline values for covariates on a $45.50 bill for Black respondents in our sample is $8.12 (17.8 percent of bill), compared to $8.78 for Whites (19.3 percent of bill). In contrast, when the quality of food and service are both dissatisfactory (Model 4), the average tip for Black respondents is $2.11 (4.6 percent of bill), compared to $3.68 for Whites (8.1 percent of bill).
Figure 1 graphically displays the racial differences in tipping across dissatisfaction conditions observed in our sample and summarized in Table 2. The bar graphs display the predicted average tip amounts (at mean/baseline values for covariates) for White (white bar) and Black (shaded bar) respondents in each condition. These bars show that the relatively small difference in average tips between Blacks and Whites in the most satisfactory condition ($0.66; far left) increases in the most dissatisfactory condition (by 91 cents to $1.57; far right). In addition, the line graphs in Figure 1 display the percentage reduction in tip for Whites (dashed line) and Blacks (solid line) across dissatisfactory conditions relative to a completely satisfactory restaurant experience. As these lines indicate, the average tip for White respondents is reduced by 58 percent when dissatisfied with the food and service quality compared to a 74 percent reduction in the average tip among Black respondents.

White-Black tipping differences in response to dissatisfactory food and service.
Finally, Model 5 in Table 2 presents results from an ordinary least squares regression model predicting respondents’ propensity to complain when dissatisfied with food and/or service in full-service restaurants. Consistent with predictions, this model indicates that Black respondents in our sample are significantly more likely than their White counterparts to report that they would complain if the food/service is dissatisfactory (b = .469, p < .01). Taken together, results from this model and the analysis of racial differences in the effects of dissatisfaction on tipping support our theoretical expectations regarding Black-White differences in consumers’ retributive reactions to dissatisfactory restaurant experiences.
Discussion
This study finds that Black Americans are more likely than Whites to lodge a complaint and/or harshly penalize their servers’ tips when dissatisfied with their dining experience. 8
Drawing from existing theory (e.g., Major et al. 2002), we interpret these observed racial differences in consumers’ behavioral reactions to dissatisfactory market experiences as reflecting a greater tendency among Black consumers to attribute dissatisfactory service to racial prejudice and discrimination (see also Baker et al. 2008). That is, as a result of their cumulative exposure to racial prejudice and mistreatment, Black Americans may be more sensitive to the genuine possibility that their race plays a salient causal role in dissatisfactory market experiences and as such are more likely than Whites to react retributively (e.g., by leaving a small tip or lodging a complaint) in an attempt to seek catharsis or redress for unjust treatment.
The racial differences in consumers’ reactions to dissatisfactory experiences documented in this study likely play a significant role in sustaining industry-wide racial stereotypes that depict Black consumers as relatively undesirable clientele who are undeserving of servers’ optimal service efforts. Specifically, consistent with extant anecdotal evidence (see Sharif 2011), our results suggest that the documented widespread server cynicism toward Blacks’ tipping practices may actually stem from racial differences in consumers’ responses to relatively rare dissatisfactory service encounters more so than from racial differences in tipping practices following the much more typical scenario wherein Black and White consumers are satisfied with their dining experiences (see Brewster and Brauer 2016; Lynn 2006; Lynn and Brewster 2015). To underscore this implication, consider that 94.6 percent of Black participants in our study reported that they would tip at least $4.50 if satisfied with both food and service quality, or 10 percent of the hypothetical $45.50 bill, compared to 98.3 percent of our White participants. However, 40.3 percent of Black participants reportedly would tip $1.00 or less if dissatisfied with both the food and service quality, compared to 16.3 percent of Whites who would tip $1.00 or less in similar circumstances. Furthermore, White participants are twice as likely as Blacks (65.4 percent of Whites vs. 31.1 percent of Blacks) to report that they would still tip equivalent to 10 percent ($4.50) or more of the hypothetical $45.50 bill when dissatisfied with the food and service.
Additionally, our finding that White diners are reportedly less likely to voice their discontent when dissatisfied suggests that servers may more frequently avoid culpability for poor service when serving White customers, including expending the requisite effort needed to mitigate the adverse effects of customers’ dissatisfaction (see Mattila 2006). Complicating matters further, in our sample, the propensity to complain when dissatisfied was only weakly correlated (r = .207, p < .001) with participants’ tipping practices (i.e., tip penalty) under dissatisfactory conditions. As such, servers may sometimes receive what they perceive to be a grossly inadequate tip from a Black customer and yet never be forced to associate the poor tip with the customer’s dissatisfaction; rather, the poor tip may simply be attributed the customers’ dark skin. Ironically, if our interpretation is correct that lower tips and more frequent complaints among Black consumers often reflect retaliatory responses to perceived discriminatory treatment (aka “tableside justice”), then these consumer behaviors, which are often used by servers to justify racially disparate treatment, could have been avoided had servers historically and contemporarily refrained from racially discriminating against their clientele in the first place.
While our findings contribute important insights into the racialized nature of restaurant establishments and the resultant risk of mistreatment faced by Black consumers, we nevertheless encourage readers to refrain from making strong conclusions about these results prior to replication and assessment of specific causal mechanisms underlying the observed patterns. Specifically, we encourage future research that tests whether racial differences in consumers’ behavioral reactions to dissatisfactory experiences indeed reflect racial differences in cognitive attributions of discrimination (see Baker et al. 2008) and hence differential motivations to retaliate or seek redress following perceptions of unjust or disrespectful treatment (Anderson 2000, 2015; Jacobs and Wright 2006; Simons and Burt 2011). Relatedly, future research should attempt to specifically isolate the effects of consumers’ attributions of dissatisfactory experiences to racial discrimination on retributive reactions from the potentially suppressing effects of alternative attributional confounds (e.g., server is having a bad day; restaurant is busy). Research that specifically accounts for an array of causal attributions might find Black-White differences in retributive behavioral reactions to dissatisfactory consumer experiences that are more pronounced than those observed in the current study. In addition, research that includes measures of consumers’ emotional reactions to dissatisfactory market experiences might allow for more direct tests of the posited theoretical mechanisms since prior work suggests that consumers’ behavioral responses to perceived racial discrimination in consumer markets may be mediated by negative emotions such as anger and frustration (cf. Baker et al. 2008; Evett et al. 2013; Klinner and Walsh 2013).
Research assessing the effects of front-line employees’ race on customers’ attributions and reactions to dissatisfactory experiences in consumer markets is also needed. Prior research indicates that Black Americans may be more likely to make attributions to racial discrimination when they experience negative outcomes that stem from their interactions with White Americans (e.g., Dovidio et al. 2002; Major et al. 2002; Mendes et al. 2008; Rawls 2000; Shelton and Sellers 2000; but see Brewster and Lynn 2014). Based on this body of work, Black Americans might be more likely to make attributions to racial prejudice and react more punitively (e.g., with harsher tip penalties or by lodging complaints) when dissatisfactory service encounters involve a White rather than a Black server. We assessed these ideas in supplementary repeated measure general linear models that included servers’ race and participants’ race as between-subject factors in predicting tip amounts. These models did not produce a reliable three-way interaction effect between dissatisfaction, participants’ race, and servers’ race on tip, which suggests Black (and White) consumers penalize the tips of Black and White servers similarly in response to dissatisfactory dining experiences.
It is possible that Black customers’ motivations to penalize White servers in response to attributions of discrimination are frequently offset by a desire to avoid affirming stereotypes depicting their racial group as “bad tippers” (see Brewster and Lynn 2014). Alternatively, studies have shown racial profiling practices to be common among both White and non-White front-line service employees (Brewster 2012a; Brewster et al. 2015; Gabbidon et al. 2008). As such, it is possible that many Black Americans have experienced less than satisfactory service as a result of the discriminatory behaviors of Black and White servers alike (Brewster et al. 2015; Robinson 2014). Thus, Black consumers may have developed heightened sensitivities to anti-Black sentiments and discriminatory behaviors that are generalized to the full-service restaurant context or consumer markets more broadly. In other words, like Black Americans’ suspicious attitudes toward “blue cops” (see Weitzer 2000), their attributions and reactions to dissatisfactory service in restaurants may be predicated less on the perceived racial identity of the server involved in any specific incident and more on their attitudes toward and cumulative experiences with service workers in U.S. market settings. Finally, while it is common in the social sciences to use Black and White sounding names to signify race (e.g., Bertrand and Mullainathan 2004), this approach may have undermined our ability to unequivocally convey the race of the servers portrayed in the experimental vignettes (cf. Gaddis 2016). Thus, we encourage future studies to explore the use of alternative manipulations (e.g., pictures) to signify servers’ race. 9
Research aimed at identifying and testing alternative causal mechanisms potentially underlying racial differences in consumers’ reactions to dissatisfactory restaurant experiences is also needed. For instance, consumers’ rights to complain and/or refrain from tipping their server can be conceptualized as institutionalized mechanisms giving consumers the discretion to adjust the cost of their meal under the condition that they are dissatisfied with their dining experience (see Holland 2009). As such, rather than reflecting racial differences in discrimination, attributions, and desires for redress or “tableside justice” as we suggest, it is possible that our findings instead stem from racial differences in the degree to which consumers’ deliberative reactions to dissatisfactory experiences are governed by economic considerations (e.g., cost consciousness or wealth accumulation; see Oliver and Shapiro 2006). In other words, by seeking redress for service failures (e.g., complaining and/or reducing tips), Black Americans on average will pay less than their White counterparts for food and/or service that they find dissatisfactory. Results from a recent study of individual differences in motives governing tipping behaviors lend some credence to this alternative interpretation of our results. Specifically, in a survey of consumers, Lynn (2009) found that White respondents were more likely than Black respondents to report their tipping behaviors were governed in part by self-presentation concerns including a desire to impress others and avoid appearing poor or cheap (see also Azar 2004).
More generally, we encourage scholarship focused on the antecedents, boundaries, and downstream effects of racial differences in consumers’ reactions to dissatisfactory experiences in consumer markets. Analyses that consider the nexus of race and other subordinate/superordinate social positions (e.g., class, gender, sexual orientation, able-bodiedness) might be especially enlightening (see Schwalbe et al. 2000). If the generic operant mechanism that we implicate in the observed Black-White differences in consumers’ retaliatory responses to dissatisfactory dining experiences is confirmed to stem from Black Americans’ greater exposure to coercive and unfair treatment, then such reactions might be amplified or perhaps tempered by other forms of inequality. Thus, descriptive research might document how varied forms of social inequality intersect to shape Black Americans’ interactions with service providers and their everyday consumption experiences. Additionally, research aimed at modeling causal chains within interactional sequences might help uncover how race subtly yet saliently structures everyday interactions between service providers and their customers and at times perpetuates consumer (and server) dissatisfaction, racial stereotyping, service discrimination, and retributive encounters. Studies have shown that Black-White differences in communicative styles play a salient role in producing interracial misunderstanding, distrust, negative stereotypes, and discrimination (cf. Dovidio et al. 2002; Rawls 2000; Rosino Forthcoming). Efforts to document how specific interactional differences become manifest in Black and White consumers’ (and servers’) perceived dissatisfactory experiences, their causal attributions, and subsequent reactions would add important insights into the everyday nature of racial discrimination in consumer markets. In closing, ample evidence indicates racial discrimination is a pervasive and enduring feature of Blacks’ everyday consumption experiences; we encourage more research devoted to illuminating the causes and consequences of such undesirable and socially damaging human interactions. We hope that this article encourages such efforts.
Footnotes
Acknowledgements
The authors also thank David Merolla and three SRE reviewers for their helpful comments on earlier versions of this manuscript.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The writing of this article was supported by a Humanities Center Faculty Fellowship and University Research Grant at Wayne State University.
