¿Eres Amigo o Enemigo? Contextual Determinants of Latinos’ Perceived Competition with African-Americans

Abstract

Prior research has examined how racial and socioeconomic environments influence the racial attitudes of Whites and Blacks. We extend this line of research to explore how local context influences Latinos’ perceived competition with African-Americans. We use the 2006 Latino National Survey and a newly appended contextual data set. We find that having more Black neighbors heightens Latinos’ perceived competition with Blacks, but only when Latinos are economically vulnerable. The presence of African-Americans within Latinos’ neighborhoods is a necessary but insufficient condition for increasing their perceived competition with Blacks; Latinos’ perceived threat is shaped by the combined effect of the neighborhood racial composition and their group’s economic status.

Keywords

racial–ethnic relations Latino politics political behavior

Latinos and African-Americans often face similar economic, political, and social circumstances (e.g., Aud, Fox, and KewalRamani 2010; de Leeuw et al. 2008; Wilson 2001). They also frequently hold similar policy views regarding the need for increased government services, especially support for job training and educational assistance (Lopez and Pantoja 2004; but see Meier et al. 2004). Given these similarities, many people have contemplated the prospects for political alliances between the groups (Kaufmann 2003; McClain and Karnig 1990; Meier et al. 2004).

Yet, there may be more compelling reasons to anticipate increased competition between Blacks and Browns. Although economists say the Great Recession ended in 2009 and the overall unemployment rate has declined (Washington 2011), the jobless situation has only marginally improved for Latinos and Blacks (U.S. Department of Labor, Bureau of Labor Statistics 2011). Meanwhile, the latest Census and migration data show more Latinos living in urban centers, often inhabiting the same or adjacent neighborhoods as large populations of African-Americans (Karafin 2009; U.S. Census Bureau 2011). The influx of Latinos potentially changes the socioeconomic and political balance in places previously defined by White–Black relationships (Camarillo 2004; McClain et al. 2006; Mohl 2003).¹ In particular, living among large proportions of African-Americans may force Latinos to compete with Blacks for material resources, group status, and political influence.

Given these recent demographic and economic trends, this research examines the relationship between Latinos’ residential contexts and their perceived competition with African-Americans. Prior efforts to understand the effect of racial and economic contexts on Latinos’ attitudes toward African-Americans are limited for a few reasons. First, most studies that explore the relationship between racial and economic environments to intergroup conflict have focused predominantly on the attitudes of White-Americans. However, given the evidence demonstrating the increasing residential propinquity of Latinos to African-Americans in metropolitan areas, there is reason to believe the racial and economic environment may have a similar—or perhaps even greater—influence on relations between Blacks and Latinos (Johnson and Oliver 1989). Second, studies of Latino attitudes toward other racial groups are frequently confined to select regions or metropolitan areas (see Barreto, Sanchez, and Morin 2010; Bobo and Hutchings 1996; McClain et al. 2006; Oliver and Wong 2003). Third, they tend to exclude contextual variables as predictors of Latinos’ attitudes. As a result, prior studies leave us uncertain about how several competing views of the relationship between the racial and economic context and interracial attitudes (e.g., the threat and contact hypotheses) can be applied to explain Latinos’ attitudes on African-Americans. Using the 2006 Latino National Survey (LNS) and a newly appended contextual data set, we are able to overcome these limitations and highlight the importance of contextual influences.

Ultimately, we think this project sheds further light on the potential for Black–Latino alliances to occur. While some researchers argue political alliances between Blacks and Latinos are critical for minorities to have a voice in the political process (Kaufmann 2003), our findings suggest that Latinos and African-Americans who are in conflict over material resources may be less likely to work together to achieve their common interests.

Competing Theories of Racial and Economic Contexts

This section summarizes past research and identifies the hypotheses to be tested in this study. The literature on racial threat offers a useful theoretical framework for exploring how the size of African-Americans might affect Latinos’ perceived competition with them. The racial threat hypothesis is an operationalization of realistic group conflict theory, which proposes that the source of intergroup conflict is rooted in group competition over scarce resources (Levine and Campbell 1972). Studies of racial threat capture the presence of intergroup competition for economic and political resources by measuring the proportion of out-group members living within defined geographic units (Blalock 1967). The proportion of out-group members may also signal a potential threat to another group’s status within an area (Blumer 1958; Bobo and Hutchings 1996). Ultimately, the racial threat hypothesis proposes that living among large proportions of out-group members increases racial animus and interracial conflict. A number of studies confirm the relationship between the concentration of Blacks and Whites’ anti-Black attitudes and political behavior (e.g., Giles and Buckner 1993; Glaser 1994; Taylor 1998).

Nevertheless, other studies offer differing views of the racial threat hypothesis. First, the racial contact hypothesis (Allport 1954) proposes that greater contact with out-group members reduces racial animosity and facilitates intergroup cooperation. Accordingly, living among large Black populations may encourage Latinos to recognize their shared interests with Blacks. Some evidence does reveal that living among high concentrations of African-Americans increases Whites’ casual interactions with Blacks (e.g., chatting on the street, social visiting, etc.) (Welch et al. 2001). To date, however, evidence showing the development of racial tolerance with increased residential integration is underwhelming. This is likely because while living among a large African-American population increases the probability of having casual exchanges with Blacks, those exchanges may not satisfy all of the conditions necessary for contact to reduce racial animus (Allport 1954). With this in mind, we believe that our data do not offer an adequate test of the contact hypothesis and, therefore, we do not test the influence of racial contact in the analysis.²

Other work suggests that living in racially segregated communities intensifies negative attitudes toward out-groups. The level of Black residential segregation is currently higher than that for Latinos; however, after 30 years of steady growth, Latino segregation is decreasing except for Mexicans (Lee, Iceland, and Sharp 2010). Oliver and Wong (2003) found that negative perceptions of out-groups are higher among those who live in areas with more of their own racial group than those with fewer in-group members.³ Unlike the racial threat hypothesis, the authors do not use the proportion of in-group members to capture competition; instead, the proportion of in-group members reflects the potential for meaningful interactions between co-ethnic members as well as participation in organizations that are composed of predominantly in-group members. Oliver (2010) argued that sizable co-ethnic populations increase in-group solidarity and encourage group members to challenge the racial order. Oliver and Wong took their findings to imply that racial integration may work toward alleviating interracial tensions.⁴ Accordingly, we explore whether the relationship between the proportion of co-ethnic members to Latinos’ perceived threat from African-Americans is consistent with the findings of Oliver and Wong.

Accordingly, we test several competing hypotheses from the existing literature to explain how racial contexts shape Latinos’ perceived competition with African-Americans. The racial threat literature suggests that living among sizable Black populations should heighten Latinos’ perceived threat from African-Americans. Thus, the first hypothesis is as follows:

Racial Threat Hypothesis: Latinos living in areas with large Black populations perceive greater group competition with African-Americans than those living in areas with smaller Black populations.

However, we consider the possibility that the proportion of in-group members plays a more important role in the formation of out-group perceptions than the proportion of out-group members. Thus, the second racial context hypothesis is as follows:

Racial Isolation Hypothesis: Latinos living in areas with large Latino populations perceive greater group competition with African-Americans than those living in areas with smaller Latino populations.

The racial isolation hypothesis is consistent with the focus of Oliver and Wong (2003). By placing measures in our model for the proportion of Latinos and Blacks, we can assess their comparative association with Latinos’ perceived competition with Blacks.

In addition to the racial context, prior studies find that economic conditions shape racial attitudes. Negative reactions to out-group members are strongest in low-status environments (Cohen and Dawson 1993; Oliver and Mendelberg 2000) or in worsening economic situations (Quillian 1995). For instance, Oliver and Mendelberg (2000) showed that Whites living among greater numbers of less educated Whites display a higher level of racial hostility toward African-Americans. Most importantly, the effect of neighborhood educational attainment remains even after the racial composition is taken into account. For the most part, the literature is silent on the influence of economic conditions on Latinos’ racial attitudes and behavior. However, we believe the influence of economic conditions on Latinos will be consistent with its impact on other groups, particularly as economic motives appear to drive, to some extent, migration among the Latino population (Borjas 2002; Fraga et al. 2010). We do recognize that, just as individuals vary in their perceptions of personal circumstances, there could also be differences of opinion regarding perceived neighborhood deprivation. Unfortunately, the LNS does not query respondents regarding the perceived conditions of their neighborhood so we must rely on indicators of the real material conditions in neighborhoods. Overall, we develop the hypothesis as follows:

Neighborhood Deprivation Hypothesis: General economic distress within neighborhoods is associated with increased levels of Latinos’ perceived competition with African-Americans.

Furthermore, there is good reason to distinguish the general economic conditions of an environment from the economic circumstances of different groups within those areas. Resources tend to be distributed unequally across neighborhoods, resulting in residential disparities in the quality of education, access to affordable housing, and social services. Yet, at the same time, within neighborhoods, resources may be distributed between groups disproportionately (Gay 2006). The groups with better economic circumstances enjoy greater social, economic, and political power within their communities. More importantly, group members are more likely to be aware of their economic circumstances within their neighborhoods, especially if their group is economically disadvantaged. Given these considerations, the hypothesis is as follows:

Group Deprivation Hypothesis: Economic stress among Latinos within neighborhoods is associated with increased levels of Latinos’ perceived competition with African-Americans.

In addition to measuring group deprivation by an absolute measure of economic conditions (i.e., Latino unemployment rate), we also consider that group relations may be impacted by relative economic differences between groups rather than their absolute conditions. Gay (2006) demonstrated that the influence of the proportion of Latinos on African-Americans’ attitudes toward Latinos is conditioned on their relative economic status to Latinos within their census tracts. When living among large Latino populations, Blacks were more likely to express negative attitudes toward Latinos when they lived in communities in which Latinos enjoyed an economic advantage relative to African-Americans. In a similar fashion, we explore whether Latinos are more likely to express competition with Blacks when they are at an economic disadvantage to Blacks in their area. Thus, the hypothesis is as follows:

Relative Group Disparity Hypothesis: A relative economic disadvantage to Blacks within their neighborhoods is associated with increased levels of Latinos’ perceived competition with African-Americans.

In particular, several studies suggest conflict between Blacks and Latinos in urban areas largely stems from competition over gainful employment (Johnson and Oliver 1989; Waldinger 1997). These studies propose that, on average, Latinos have an advantage over Blacks for private-sector jobs especially in certain areas of the country due to their stronger social networks and employers’ perception that Latinos are better workers,⁵ though other work indicates Blacks hold an advantage in securing municipal jobs (Lim 2001; McClain and Karnig 1990). We include two measures of unemployment in our analysis to gauge whether joblessness heightens Latinos’ perceived competition with Blacks. We also include measures of income to assess economic conditions.

In many cases, economic conditions are difficult to separate from the racial environment. Branton and Jones (2005) illustrated that White-Americans’ racial attitudes and race-based policy preferences are influenced not by the racial or economic context independently but by their conditional relationship. Anglo-Americans living in highly diverse neighborhoods with large numbers of college-educated people expressed more liberalized racial views because they have more favorable exposure to minority group members. In contrast, Anglo-Americans in diverse but lowly educated environments expressed lower support for race-based policies because they have unfavorable exposure with minority group members and feel threatened by both economic and racial forces. Gay (2006) also found that the impact of the racial context on Blacks’ anti-Latino attitudes is conditional upon the relative economic disparity between Blacks and Latinos within their respective neighborhoods.

Building on this research, we propose the combined effect of Latinos’ proximity to large Black populations and dire group neighborhood material conditions will reveal higher perceived group competition with African-Americans for jobs and other material resources. The final hypothesis is as follows:

Conditional Threat Hypothesis: Latinos’ perceived group competition with African-Americans is strongest in neighborhoods with large Black populations and where Latinos are economically deprived.

Data and Method

The data come from the 2006 LNS. The LNS was a 40-minute bilingual telephone survey of 8,634 self-identified Latino/Hispanic adult residents of the United States. Representative samples of Latino households were drawn from Arizona, Arkansas, California, Colorado, Florida, Georgia, Illinois, Iowa, Nevada, New Jersey, New Mexico, New York, North Carolina, Texas, Washington, and the District of Columbia metropolitan area (including counties and municipalities in Virginia and Maryland). Each LNS respondent was geocoded to allow for the appending of economic and demographic contextual data. Our analyses use data that are weighted to be nationally representative.⁶

Our dependent variable is based on four questions regarding whether respondents perceived competition with African-Americans in getting jobs in the private sector, having access to education and quality schools, getting jobs with the city or state government, and having representatives in elected office. In answering the questions, respondents chose one of the following categories: (1) no competition, (2) weak competition, or (3) strong competition. Table 1 reports the distribution of perceived competition in four areas. Interestingly, the results show higher levels of perceived competition in getting government jobs and having a representative in elected offices, in comparison with getting private jobs and having access to education.

Table 1.

Latinos’ Perceived Competition with African-Americans in Four Areas.

	Jobs	Education	Government Jobs	Elected Offices
No competition	51.95	46.38	39.97	33.93
Weak competition	21.30	24.55	26.14	30.04
Strong competition	26.75	29.07	33.90	36.02
Total	100%	100%	100%	100%

Note. The number of observations is 7,939. Values in the table do not necessarily correspond to the total value of 100% becase they are rounded. Data come from the Latino National Survey in 2006.

For our regression analysis, we computed an additive index using the responses to the four areas of competition as a measure of the overall perception of Latinos toward Blacks.⁷ The index has a continuous scale with higher scores denoting a higher level of perceived competition. It ranges from 1 to 3 with the mean equal to 1.88. Figure 1 shows the distribution of Latinos’ perceived competition with African-Americans. We note that the modal response is “no competition” for all of the areas except for the query about elected officials. More importantly for our purposes, however, Latinos’ perceived competition is evenly distributed across the scale. Because our dependent variable is measured by a numerical scale, we use a linear regression model. We replicate the analysis using a hierarchical linear model and find almost identical results as reported in the subsequent section.⁸ Also important to note, we find identical results when Latinos’ perceived competition in the four areas is analyzed separately.⁹

Figure 1.

The distributions of Latinos’ perceived competition with African-Americans.

We regress the scale of perceived competition on the contextual-level variables that measure racial and economic characteristics. Prior research of contextual influences on racial attitudes has used a variety of geographical units, including census tracts and blocks, zip codes, cities, counties, and metropolitan areas. For our analysis, we primarily use contextual characteristics of the census tracts where the respondents resided at the time of the survey. Gay (2006) highlighted the two fundamental reasons for utilizing measures at the census tract level. First, the availability of economic resources within neighborhoods gives a strong cue of the economic vulnerability of the residents. Second, residents are more likely to be better aware of the economic conditions within their neighborhoods than in larger geographic areas. Accordingly, we use the terms census tract and neighborhood interchangeably.

For testing the effects of the racial context at the census tract level, we use the proportion of African-Americans and Latinos.¹⁰ As indicators of the general economic condition for the economic deprivation hypothesis, our analysis uses measures of the unemployment rate and the logged median income for all residents within the census tract.¹¹ As an indicator of group material conditions for the group deprivation hypothesis, we use a measure of Latinos’ unemployment rate at the census tract level. We use the unemployment rates as a measure of economic conditions because they capture the level of economic stress at the time of the survey more accurately than other indicators such as educational attainment, median income, and poverty. We also use unemployment because we think that levels of unemployment are more easily observed from one’s social networks than levels of income, a point consistent with research that we already cite in the article as well as an extensive line of research indicating the important role that social networks play in people’s search for employment (e.g., Granovetter 1995; Ioannides and Loury 2004). These networks should provide individuals with information about the local labor market, making tract-level unemployment a good proxy for individuals’ information about the group’s economic conditions.

In addition, we test the group deprivation hypothesis by measuring the disparity in material conditions between Latinos and African-Americans. Specifically, we compute the difference between the Latino unemployment rate and the Black unemployment rate at the tract level. This measure falls below 0 if Latinos’ unemployment rate is lower and thus better off than Blacks, whereas it exceeds 0 if Latinos’ unemployment rate is higher and thus worse off than Blacks.

To test the conditional threat hypothesis, we generate the interaction terms between the proportion of African-Americans and the overall unemployment rate, between the proportion of African-Americans and the unemployment rate among Latinos, and between the proportion of African-Americans and the difference in the group unemployment rates. We include these interaction terms individually in the model.

A major concern within the literature is whether larger or smaller units are better measures of group size. The geographic area selected as the unit of analysis may lead to distinctly different conclusions to be drawn from the analysis (Baybeck 2006; Cho and Baer 2011; Ha 2010). While examining Blacks’ racial attitudes toward Latinos, Cho and Baer (2011) demonstrated that the geographic unit used to measure contextual variables can change the results that are drawn from the analysis. In their analysis, measuring the racial context at the census tract level produced results that corroborated the intergroup contact hypothesis, while larger units, such as measures of the city, seemed to support the intergroup threat hypothesis. Ultimately, the authors conclude that researchers should consider the theoretical justification for utilizing measures of the social context at different geographic units.¹²

Given long-standing concerns about the appropriate geographic unit of analysis, we include county-level measures to account for social forces taking place in the broader context. Specifically, our regression model includes the proportion of African-Americans and Latinos as measures of the county racial composition.¹³ In addition, it includes the unemployment rate among all residents and the log median income as measures of the overall economic conditions.¹⁴ Unfortunately, the data on the group-specific unemployment rates at the county level are unavailable, and thus, we cannot test the group deprivation hypothesis and the conditional threat hypothesis at the county level. We believe our specification allows us to determine whether the effects at the census tract level stand after controlling for characteristics of the broader racial and economic environment.¹⁵

Our regression model also includes several individual-level demographic variables as controls.¹⁶ Those variables are respondents’ education, household income, reported employment status, personal financial situation, homeownership, age, gender, marital status, self-reported racial identity, generations in the United States, and national origins. Education is measured by an 8-point scale from 0 to 7. Household income is measured by four indicator variables that equal 1 if respondents belong to one of the quartiles on the scale of income and 0 otherwise. For the estimation, we denote the lowest quartile as a baseline group. To avoid dropping respondents who refused to report their income status from our analysis, we create an indicator variable that equals 1 if respondents refused to do so and 0 otherwise. Respondents’ assessment of their personal financial situation is measured by a 3-point scale ranging from (1) gotten worse, (2) stays about the same, and (3) gotten better. Employment status equals 1 if a respondent is unemployed and 0 otherwise. Homeownership is equal to 1 if respondents report owning homes. Age is measured by the respondent’s actual age at the year of the survey, whereas gender and marital status are measured by an indicator variable that equals 1 if the respondent is female or married and 0 otherwise.

Self-reported racial identity equals 1 if a respondent identifies oneself as Black. As a measure of generational status, we created the five indicator variables that separate respondents into categories such as foreign born and unnaturalized, foreign born and naturalized, second generation, third generation, and fourth generation. For estimation, we use “foreign born and unnaturalized” as a baseline category. National origins are measured by dummy variables for 19 countries of immediate origin.

Furthermore, we include state fixed effects to control the impact of differences across states. Our model includes 15 dummy variables representing each state. The state fixed effects allow us to eliminate the impacts of all socioeconomic and political characteristics at the state level on the estimation results reported below. Summary statistics for the contextual and individual-level variables are provided in Table 2. In total, our analysis includes 7,939 respondents in 6,734 tracts, 613 counties, and 15 states and the District of Columbia metropolitan area.¹⁷

Table 2.

Summary Statistics.

	Mean	SD	Minimum	Maximum
Tract-level variables
Proportion Blacks	0.133	0.195	0.000	1.000
Proportion Latinos	0.440	0.312	0.000	1.000
Overall unemployment rate	0.061	0.025	0.015	0.236
Latino unemployment rate	0.051	0.033	0.000	1.000
Difference in unemployment rate	0.000	0.058	−0.965	0.929
Log median income	10.515	0.428	8.792	12.231
County-level variables
Proportion Blacks	0.117	0.119	0.000	0.621
Proportion Latinos	0.256	0.194	0.002	0.981
Overall unemployment rate	0.047	0.015	0.019	0.154
Log median income	10.646	0.243	9.711	11.326
Individual-level variables
Perceived competition with African-Americans	1.884	0.664	1.000	3.000
Education	3.626	1.935	0.000	7.000
Income Quartile 1	0.328	0.469	0.000	1.000
Income Quartile 2	0.141	0.348	0.000	1.000
Income Quartile 3	0.172	0.377	0.000	1.000
Income Quartile 4	0.174	0.379	0.000	1.000
Income no report	0.185	0.389	0.000	1.000
Financial situation	2.039	0.721	1.000	3.000
Unemployed	0.078	0.268	0.000	1.000
Own home	0.521	0.500	0.000	1.000
Age	40.507	15.418	18.000	97.000
Female	0.543	0.498	0.000	1.000
Married	0.550	0.498	0.000	1.000
Self-identified Black	0.008	0.087	0.000	1.000
First generation, unnaturalized	0.482	0.500	0.000	1.000
First generation, naturalized	0.217	0.412	0.000	1.000
Second generation	0.119	0.324	0.000	1.000
Third generation	0.099	0.298	0.000	1.000
Fourth generation	0.080	0.272	0.000	1.000

Note. The number of observations is 7,939. Data come from the Latino National Survey in 2006.

Findings: Black Concentration and the Economic Environment

Table 3 reports the results for four different models. Column (1) is useful for assessing the unconditional relationship between the racial and economic environment. Column (2) includes the same items from Column (1) in addition to an interaction term between the proportion of the Black population and the neighborhood overall unemployment rate. Column (3) includes an interaction term between the proportion of the Black population and the Latino unemployment rate, whereas Column (4) includes an interaction term between the proportion of the Black population and the difference in the unemployment rates between Latinos and African-Americans. In addition to the census tract variables, the models include the variables measured at the county level and individual level for control.¹⁸ For each model, the standard errors are clustered by census tracts.¹⁹ Note that differences across states and national origins are controlled by the state fixed effects and national origin fixed effects, though these results are not shown in the tables.

Table 3.

The Effects of Racial and Economic Contexts at the Tract and County Levels.

	(1)		(2)
	B	(SE)	b	(SE)
Tract-level variables
Proportion Blacks	0.005	(0.080)	0.083	(0.145)
Proportion Latinos	0.013	(0.045)	0.008	(0.045)
Overall unemployment rate	1.018	(0.883)	1.353	(1.114)
Latino unemployment rate	−0.425	(0.348)	−0.452	(0.355)
Proportion Blacks × Overall unemployment rate			−0.962	(1.605)
Log median income	0.001	(0.030)	0.001	(0.030)
County-level variables
Proportion Blacks	0.210	(0.135)	0.356	(0.485)
Proportion Latinos	0.147	(0.104)	0.141	(0.105)
Overall unemployment rate	0.806	(1.109)	0.788	(1.259)
Proportion Blacks × Overall unemployment rate			−2.897	(8.656)
Log median income	0.075	(0.084)	0.078	(0.086)
Individual-level variables
Education	0.005	(0.006)	0.005	(0.006)
Income Quartile 2	0.001	(0.030)	0.000	(0.030)
Income Quartile 3	0.007	(0.029)	0.007	(0.029)
Income Quartile 4	−0.004	(0.033)	−0.004	(0.033)
Income no report	−0.104**	(0.027)	−0.104**	(0.027)
Financial situation	−0.005	(0.013)	−0.005	(0.013)
Unemployed	0.022	(0.035)	0.022	(0.035)
Own home	0.013	(0.022)	0.013	(0.022)
Age	0.001*	(0.001)	0.001*	(0.001)
Female	−0.050**	(0.019)	−0.051**	(0.019)
Married	−0.060**	(0.019)	−0.060**	(0.019)
Self-identified Black	−0.032	(0.104)	−0.034	(0.104)
First generation, naturalized	0.035	(0.026)	0.035	(0.026)
Second generation	−0.062	(0.052)	−0.062	(0.052)
Third generation	−0.045	(0.055)	−0.046	(0.055)
Fourth generation	0.010	(0.057)	0.010	(0.057)
Constant	1.006	(0.878)	0.965	(0.899)
R ²	.031		.031
	(3)		(4)
	B	(SE)	b	(SE)
Tract-level variables
Proportion Blacks	−0.224**	(0.103)	0.027	(0.079)
Proportion Latinos	0.013	(0.045)	0.012	(0.045)
Overall unemployment rate	1.143	(0.878)	1.374	(0.912)
Latino unemployment rate	−1.024**	(0.336)
Proportion Blacks × Latino unemployment rate	4.334**	(1.147)
Difference in unemployment rate			−0.298*	(0.163)
Proportion Blacks × Difference in unemployment rate			3.084**	(0.999)
Log median income	0.003	(0.030)	0.004	(0.030)
County-level variables
Proportion Blacks	0.190	(0.136)	0.174	(0.136)
Proportion Latinos	0.141	(0.104)	0.140	(0.104)
Unemployment rate	0.924	(1.110)	0.571	(1.127)
Log median income	0.076	(0.084)	0.103	(0.084)
Individual-level variables
Education	0.005	(0.006)	0.005	(0.006)
Income Quartile 2	−0.002	(0.029)	−0.003	(0.030)
Income Quartile 3	0.005	(0.029)	0.005	(0.029)
Income Quartile 4	−0.006	(0.033)	−0.006	(0.033)
Income no report	−0.104**	(0.027)	−0.105**	(0.027)
Financial situation	−0.005	(0.013)	−0.005	(0.013)
Unemployed	0.024	(0.035)	0.024	(0.035)
Own home	0.014	(0.022)	0.013	(0.022)
Age	0.001*	(0.001)	0.001*	(0.001)
Female	−0.051**	(0.019)	−0.051**	(0.019)
Married	−0.059**	(0.019)	−0.059**	(0.019)
Self-identified Black	−0.032	(0.104)	−0.034	(0.105)
First generation, naturalized	0.036	(0.026)	0.036	(0.026)
Second generation	−0.059	(0.052)	−0.058	(0.052)
Third generation	−0.043	(0.055)	−0.044	(0.055)
Fourth generation	0.013	(0.057)	0.015	(0.057)
Constant	0.999	(0.878)	0.665	(0.883)
R ²	.033		.032

Note. Table entities are fixed effect regression estimates with standard errors in parentheses. Standard errors are clustered by census tracts. The results for the state and national origin dummies are omitted from the table. In total, our analysis includes 7,939 respondents in 6,734 tracts in 613 counties in 15 states and the District of Columbia metropolitan area. The dependent variable is an additive index of perceived competition in four areas. Data come from the Latino National Survey in 2006.

p < .10. **p < .05.

Column (1) in Table 3 indicates that the proportions of African-Americans and Latinos have no statistically significant impact on the level of perceived competition with African-Americans. The result for the proportion of African-Americans offers no support for the traditional operationalization of the racial threat hypothesis. The coefficient falls far short of reaching conventional levels of statistical significance. Furthermore, there is no evidence that supports the racial isolation hypothesis; the coefficient for the proportion of Latinos within the census tract is statistically insignificant.

Similarly, the economic context seems to play no important role on Latinos’ perceived competition with African-Americans. Neither the general unemployment rate nor Latino-specific unemployment rate have a statistically significant impact on their perceived competition with Blacks.

To test the hypothesis that Latinos’ perceived group competition with African-Americans is strongest in areas with large Black populations and poor economic conditions, we include an interaction term between the racial and economic contexts in Column (2). The interaction term is the product of the proportion of African-Americans and the overall unemployment rate in the census tract. The result shows no interactive relation between these two terms at the tract and county levels, which would seem to offer no evidence for the conditional threat hypothesis.²⁰

Column (3) examines the possibility that group material conditions may influence the relationship between the proportion of the Black population and Latinos’ perceived group competition more than the general neighborhood economic conditions. Column (3) reports that the proportion of African-Americans at the tract level is estimated to be negative and statistically significant at the .05 level, while the interaction term is positive and statistically significant. Considered jointly, the estimates indicate that the relationship between African-American group size and the level of Latino perceived competition is conditional on Latinos’ economic situation, which gives support for the conditional threat hypothesis.²¹

To show variation in the relationship between African-American population and perceived competition at different levels of Latino unemployment, we plot the marginal effect of the proportion of African-Americans depending on the Latino unemployment rate. The values of the Latino unemployment rate are set to range from 0 to 0.3.²² Figure 2 reports the marginal effects with the 95% confidence interval as the broken lines and shows that when Latino unemployment is low, higher levels of African-American population are associated with lower levels of perceived competition that are statistically significant.²³ For example, a neighborhood with 10% greater African-American population is predicted to have a 0.025 reduction in perceived competition on the 3-point scale when Latino unemployment is close to 0. When Latino unemployment exceeds 10%, increases in the African-American population are associated with increased levels of perceived competition that are statistically significant.

Figure 2.

The marginal effect of the proportion Blacks conditional on the Latino unemployment rate (solid line) with the 95% cofidence interval (broken lines).

When the proportion of unemployed Latinos in the census tract is equal to the mean (0.05), the marginal effect of the proportion of African-Americans becomes almost 0. As the confidence interval overlaps the line of 0, the effect of the proportion of African-Americans is not statistically significant.

Notably, as the proportion of unemployed Latinos surpasses the mean, the association between the proportion of African-Americans and perceived competition becomes positive and statistically significant. For example, the estimated relationship is about 0.25 when the proportion of unemployed Latinos is equal to 0.1. The 95% confidence interval does not overlap the line of 0, suggesting that the relationship is statistically significant. A 10% increase in the size of African-Americans leads to a predicted increase in level of competition by 0.025 on the 3-point scale when the unemployment rate among Latinos is equal to 0. This positive effect of the interaction term follows the pattern predicted by the conditional threat hypothesis.

Our results are consistent with the findings of Branton and Jones (2005) regarding White attitudes. They discover that areas where there is a strong out-group presence and the in-group faces economic difficulty constitute a “maximum threat environment.” This argument directly applies to our findings on the intergroup relationship between African-Americans and Latinos. At the local level, the presence of a large Black population is not necessarily associated with Latino perceptions of Blacks as competitors unless Latinos are also economically vulnerable. Thus, the presence of out-group members is a necessary but insufficient condition for intergroup conflict.

We also explore whether economic vulnerability should be operationalized as the relative economic differences between Latinos and Blacks. Column (4) of Table 3 reports the result that includes the interaction between the proportion of the Black population and the difference in the unemployment rates between Latinos and African-Americans. It shows that the coefficient associated with the proportion Blacks at the tract level is not significant, but the interaction is positive and significant. To evaluate the interactive effect visually, we plot the marginal effect of the proportion of African-Americans depending on the difference in the unemployment rates between two groups. The values of the difference in the unemployment rates are set to range from −0.1 to 0.1.

Figure 3 reports the marginal effects with the 95% confidence interval as the broken lines. When the difference in the unemployment rates is negative, that is the Black unemployment rate is higher than the Latino unemployment rate, the proportion of African-Americans has a negative and statistically significant impact on the level of perceived competition. In other words, the presence of African-Americans in the neighborhood decreases Latinos’ perceived competition with African-Americans when Latinos experience an employment advantage to African-Americans. When the difference in the unemployment rates is smaller or Latinos are worse off than African-Americans, the presence of African-Americans increases Latinos’ perceived competition with African-Americans.

Figure 3.

The marginal effect of the proportion Blacks conditional on the difference between the Latino and Black unemployment rate (solid line) with the 95% cofidence interval (broken lines).

Our results are consistent with the findings of Gay (2006) regarding African-American attitudes toward Latinos. Her research finds that African-Americans tend to show more negative attitudes toward Latinos when they reside in an environment where Latinos are economically more advantageous to African-Americans. This article reports the exact relationship of Latinos’ attitudes toward African-Americans, although the measure of economic disparity between Gay’s research and our study is different.²⁴

Table 3 displays that most of the county-level and individual-level variables have no statistically significant relationship with the perceived level of competition. When controlling for the racial and economic environment at the census tract level, the size of the Black and Latino population and the economic variables at the county level are unrelated to the feelings of group competition toward Blacks. At the individual level, Latinos who refused to report their income status are less likely to see competition with Blacks. There are also strong gender and age effects. Older Latinos perceive a higher level of competition than younger Latinos, whereas female and married Latinos perceive a lower level of competition than male and unmarried Latinos.

Residential Self-Selection

Finally, one issue that continues to emerge in studies that use contextual measures to predict racial attitudes is the role of self-selection in residential patterns. Bobo and Zubrinsky (1996) found prejudiced Whites were less likely to live among large Black populations and instead preferred to live in predominantly White neighborhoods. Their study suggests that studies of racial contexts may reveal the influence of racial attitudes on residential patterns rather than the effect of neighborhood composition on racial attitudes. Other studies, however, show self-selection accounts significantly more for the residential patterns of Whites than it does for members of racial and ethnic minority groups (Oliver and Wong 2003; Powers and Ellison 1995). The disparity between the impact of self-selection between Whites and minorities is likely due to group differences in income and wealth (South, Crowder, and Chavez 2005). While prejudiced Whites typically have the financial resources to relocate to predominantly White neighborhoods, Latinos are less likely to be in a financial position to relocate when living in neighborhoods with large Black populations.

Our findings in Table 3 cast doubt on the self-selection hypothesis. If self-selection is a problem, Latinos with a greater sense of group solidarity would move to neighborhoods with greater Latino populations. As such, we should expect there to be a positive relationship between the proportion of Latinos and perceived competition. Whether we look at the tract or county level in Table 3, there is a positive relationship, but it is substantively trivial and not statistically significant. Ultimately our findings provide suggestive evidence that self-selection is unlikely to pose a problem for our analysis, even though we cannot rule out those effects entirely.²⁵

Conclusion

As the Latino population becomes a significantly larger portion of the U.S. population, their views of, and interactions with, other racial and ethnic groups are of increasing importance to the tone and quality of American political and social discourse. To shed further light on the interethnic relationship, we examined what contextual factors influenced Latinos’ views of African-Americans as competitors using the most comprehensive data of Latinos currently available.

Overall, our analysis highlights the importance and the complexity of local contexts for intergroup relations. Having more Black neighbors is associated with increased feelings of competition with Blacks, but only when Latinos are economically vulnerable. Bad economic conditions, specifically high unemployment, play a larger role in Latinos’ feelings of competition than the racial composition of their neighborhoods. Ultimately, our study extends the findings of prior research, which focused on attitudes of Blacks (Gay 2006) and Whites (Branton and Jones 2005). Now, we have greater reason to believe that racial attitudes, irrespective of the race and ethnicity of individuals, are significantly influenced by the interaction of racial and economic conditions in the local environment.

In addition, we were able to explore the racial and economic environment at both the census tract and county levels. The analysis suggests that the influence of out-group size varies across geographic units. Although the relationship between the size of the Black population and Latinos’ perceived group competition was conditional upon Latinos’ economic status at the census tract level, there was no significant relationship at the county level. Previous studies suggest that counties are a better indicator of realistic group competition, while the neighborhood is a better measure of social interaction (Forbes 1997; Ha 2010); these studies imply living within neighborhoods with more out-group members is more likely to facilitate cooperation than conflict. However, our findings corroborate Gay’s (2006) work that shows the racial and economic environment within neighborhoods can fuel interminority tensions. While there is certainly greater potential for neighborhoods to encourage more meaningful interactions between groups, racial integration may exacerbate racial conflict if either of the groups is economically vulnerable. We suspect this phenomenon arises because Latinos may be acutely aware of their economic standing within their own neighborhoods.

Although we cannot be sure that our findings reflect a causal relationship between Black population, Latino unemployment, and Latinos’ perceptions of competition with Blacks, our findings do speak to the prospect of cooperation between Blacks and Latinos. In areas where the Black population is large, we predict that alliances between the groups will be less common if Latino unemployment is high. In contrast, cooperation between Latinos and African-Americans is most favored in areas with a low sense of perceived competition, and our findings lead us to expect that this will occur most frequently in areas where economic conditions are relatively favorable for Latinos.

There are always unanswered questions for future research. Obviously, this study focused on the relationship between racial and economic conditions and Latinos’ perceived competition with African-Americans. Nevertheless, as Latinos in urban areas are often exposed to a multiethnic context, it would be useful to explore how they react to the size of Asian-American communities or even the size of Latino populations of different national origins. Finally, future research needs to examine whether the conditional relationship between racial and economic contexts predicts other attitudinal and behavioral variables such as racial stereotypes, political participation, and vote choice.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Tony E. Carey Jr. is an Assistant Professor in the Department of Political Science at the University of North Texas. His research focuses on intergroup and intragroup politics, African American politics, gender politics, and political psychology. Some of his other work has appeared in the American Journal of Political Science, Political Research Quarterly, and Politics, Groups and Identities.

Valerie Martinez-Ebers is professor of Political Science at the University of North Texas and Co-Editor of the American Political Science Review. She has published widely on education policy, Latino/a politics, women in politics, and methods of survey research. Her publications have appeared in American Journal of Political Science, American Political Science Review, Journal of Politics and Political Research Quarterly.

Tetsuya Matsubayashi is an Associate Professor in Osaka School of International Public Policy at Osaka University, Japan. His research focuses on mass political behavior, political representation, and public health. Some of his other work has appeared in American Journal of Political Science, American Political Science Review, British Journal of Political Science, Political Research Quarterly, and Politics, Groups and Identities.

Philip Paolino is an Associate Professor in the Department of Political Science at the University of North Texas. His research focuses on public opinion, voter behavior, and political methodology. Some of his other work has appeared in the American Journal of Political Science, Political Research Quarterly, and Political Analysis.

References

Allport

Gordon W.

1954. The Nature of Prejudice. Reading: Addison-Wesley.

Aud

Susan

Fox

Mary Ann

KewalRamani

Angelina

. 2010. “Status and Trends in the Education of Racial and Ethnic Groups.” Technical Report, U.S. Department of Education, Washington, DC.

Baybeck

Brady

. 2006. “Sorting Out the Competing Effects of Racial Context.” Journal of Politics 68:386–96.

Blalock

Herbert M.

1967. Toward a Theory of Minority Group Relations. New York: Wiley.

Blumer

Herbert

. 1958. “Racial Prejudice as a Sense of Group Position.” The Pacific Sociological Review 1 (1): 3–7.

Bobo

Lawrence D.

Hutchings

Vincent L.

1996. “Perceptions of Racial Group Competition: Extending Blumer’s Theory of Group Position to a Multiracial Social Context.” American Sociological Review 61:951–72.

Bobo

Lawrence D.

Zubrinsky

Camille L.

1996. “Attitudes on Residential Integration: Perceived Status Differences, Mere In-group Preference, or Racial Prejudice?” Social Forces 74:883–909.

Borjas

George

. 2002. “Welfare Reform and Immigration.” In The New World of Welfare, edited by Blank

Rebecca

Haskins

Ron

, 369–90. Washington, DC: Brookings Institution Press.

Branton

Regina P.

Jones

Bradford S.

2005. “Reexamining Racial Attitudes: The Conditional Relationship Between Diversity and Socioeconomic Environment.” American Journal of Political Science 49 (2): 359–72.

10.

Camarillo

Albert M.

2004. “Cities of Color: The New Racial Frontier in California’s Minority-Majority Cities.” Pacific Historical Review 76 (1): 1–28.

11.

Cho

Wendy K. Tam

Baer

Neil

. 2011. “Environmental Determinants of Racial Attitudes Redux: The Critical Decisions Related to Operationalizing Context.” American Politics Research 39 (2): 414–36.

12.

Cohen

Cathy J.

Dawson

Michael C.

1993. “Neighborhood Poverty and African American Politics.” American Political Science Review 87 (2): 286–302.

13.

de Leeuw

Michael B.

Whyte

Megan K.

Dale

Meza

Catherine

Karteron

Alexis

. 2008. “The Current State of Residential Segregation and Housing Discrimination: The United States’ Obligations Under the International Convention on the Elimination of All Forms of Racial Discrimination.” Michigan Journal of Race & Law 13:337–40.

14.

Forbes

Hughes Donald

. 1997. Ethnic Conflict: Commerce, Culture, and the Contact Hypothesis. New Haven: Yale Univ. Press.

15.

Fraga

Luis R.

John

Garcia

Hero

Rodney

Jones-Correa

Michael

Martinez-Ebers

Valerie

Segura

Gary M.

2010. Latino Lives in America. Philadelphia: Temple Univ. Press.

16.

Gay

Claudine

. 2006. “Seeing Difference: The Effect of Economic Disparity on Black Attitudes Toward Latinos.” American Journal of Political Science 50:982–97.

17.

Giles

Michael W.

Buckner

Melanie A.

1993. “David Duke and Black Threat: An Old Hypothesis Revisited.” Journal of Politics 55 (3): 702–13.

18.

Glaser

James M.

1994. “Back to the Black Belt: Racial Environment and White Racial Attitudes in the South.” Journal of Politics 56:21–41.

19.

Granovetter

Mark

. 1995. Getting a Job: A Study of Contacts and Careers. Chicago: Univ. of Chicago Press.

20.

Shang E.

2010. “The Consequences of Multiracial Contexts on Public Attitudes Toward Immigration.” Political Research Quarterly 63 (1): 29–42.

21.

Iacus

Stepahno

King

Gary

Porro

Giuseppe

. 2009. CEM: Coarsened Exact Matching Software. Available from http://j.mp/oGIFsf.

22.

Ioannides

Yannis M.

Loury

Linda Datcher

. 2004. “Job Information Networks, Neighborhood Effects, and Inequality.” Journal of Economic Literature 42 (4): 1056–93.

23.

Johnson

James H.

Oliver

Melvin L.

1989. “Interethnic Minority Conflict in Urban America: The Effect of Economic and Social Dislocations.” Urban Geography 10:449–63.

24.

Karafin

Diana L.

2009. “Racial and Ethnic Integration in U.S. Metropolitan Neighborhoods: Patterns, Complexities and Consequences.” PhD dissertation, The Ohio State Univ., Columbus.

25.

Kaufmann

Karen M.

2003. “Cracks in the Rainbow: Group Commonality as a Basis for Latino and African-American Political Coalitions.” Political Research Quarterly (2):199–210.

26.

Lee

Barrett

Iceland

John

Sharp

Gregory

. 2010. “Racial and Ethnic Diversity Goes Local: Charting Change in American Communities Over Three Decades.” http://www.s4.brown.edu/us2010/Data/Report/report08292012.pdf (accessed December 15, 2015).

27.

Levine

Robert A.

Campbell

Donald T.

1972. Ethnocentrism: Theories of Conflict, Ethnic Attitudes and Group Behavior. New York: John Wiley.

28.

Lim

Nelson

. 2001. “On the Back of Blacks?” In Strangers at the Gates: New Immigrants in Urban America, edited by Waldinger

Roger

, 186–227. Berkeley, CA: Univ. of California Press.

29.

Lopez

Linda

Pantoja

Adrian D.

2004. “Beyond Black and White: General Support for Race-Conscious Policies Among African American, Latinos, Asian Americans, and Whites.” Political Research Quarterly 57 (4): 633–42.

30.

McClain

Paula D.

Carter

Niambi M.

DeFrancesco Soto

Victoria M.

Lyle

Monique L.

Grynaviski

Jeffrey D.

Nunnally

Shayla C.

Scotto

Thomas J.

Kendrick

J. Allan

Lackey

Gerald F.

Cotton

Kendra Davenport

. 2006. “Racial Distancing in a Southern City.” Journal of Politics 68:572–84.

31.

McClain

Paula D.

Karnig

Albert K.

1990. “Black and Hispanic Socioeconomic and Political Competition.” American Political Science Review 84 (2): 535–45.

32.

Meier

Kenneth J.

McClain

Paula

Polinard

J. L.

Winkle

Robert D.

2004. “Divided or Together? Conflict and Cooperation Between African Americans and Latinos.” Political Research Quarterly 57 (3): 399–409.

33.

Mohl

Raymond A.

2003. “Globalization, Latinization, and the Nuevo New South.” Journal of American Ethnic History 22 (4): 31–66.

34.

Morin

Jason L.

Sanchez

Gabriel

Barreto

Matt A.

2010. “Perceptions of Competition Between Latinos and Blacks: The Development of a Relative Measure of Inter-group Competition.” In Black-Brown Relations, edited by Rivera-Salgado

Gaspar

Telles

Edward

Zamora

Sylvia

, 96–124. New York: Russell Sage Foundation.

35.

Oliver

J. Eric

. 2010. The Paradoxes of Integration: Race, Neighborhood, and Civic Life in Multiethnic America. Chicago: Univ. of Chicago.

36.

Oliver

J. Eric

Mendelberg

Tali

. 2000. “Reconsidering the Environmental Determinants of Racial Attitudes.” American Journal of Political Science 44 (3): 574–89.

37.

Oliver

J. Eric

Wong

Janelle

. 2003. “Intergroup Prejudice in Multiethnic Settings.” American Journal of Political Science 47 (4): 567–82.

38.

Pettigrew

Thomas F.

Tropp

Linda R.

Wagner

Ulrich

Christ

Oliver

. 2011. “Recent Advances in Intergroup Contact Theory.” International Journal of Intercultural Relations 35:271–80.

39.

Powers

Daniel A.

Ellison

Christopher G.

1995. “Interracial Contact and Black Racial Attitudes: The Contact Hypothesis and Selectivity Bias.” Social Forces 74 (1): 205–26.

40.

Quillian

Lincoln

. 1995. “Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-immigrant and Racial Prejudice in Europe.” American Sociological Review 60 (4): 586–611.

41.

South

Scott J.

Crowder

Kyle

Chavez

Eric

. 2005. “Exiting and Entering High-Poverty Neighborhoods: Latinos, Blacks, and Anglos Compared.” Social Forces 84 (2): 873–900.

42.

Taylor

Marylee C.

1998. “How White Attitudes Vary with the Racial Composition of Local Populations: Numbers Count.” American Sociological Review 63 (4): 512–35.

43.

U.S. Census Bureau. 2011. “‘American Fact Finder.” African American and Hispanic Populations in Metropolitan Statistical Areas by States Data. Available from http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=DEC_10_SF1_GCTP3.US21PR&prodType=table (accessed July 11, 2011).

44.

U.S. Department of Labor, Bureau of Labor Statistics. 2011. “Unemployment Databases, Tables and Calculators by Subject.” Historical Unemployment Data for African Americans and Hispanics. http://data.bls.gov/pdq/SurveyOutputServlet (accessed July 11, 2011).

45.

Voss

D. Stephen

. 1996. “Beyond Racial Threat: Failure of an Old Hypothesis in the New South.” Journal of Politics 58 (4): 1156–70.

46.

Waldinger

Roger

. 1997. “Black/Immigrant Competition Re-assessed: New Evidence from Los Angeles.” Sociological Perspectives 40 (3): 365–86.

47.

Waldinger

Roger

. 2001. “Up from Poverty?: ‘Race,’ Immigration, and the Fate of Low-Skilled Workers.” In Strangers at the Gates: New Immigrants in Urban America, edited by Waldinger

Roger

, 80–116. Berkeley: Univ. of California Press.

48.

Washington

Jesse

. 2011. “Despite End of Great Recession, Jobless Rate Still Rising for Blacks.” Fort Worth Star Telegram, July 10, p. 6A.

49.

Welch

Susan

Sigelman

Lee

Bledsoe

Timothy

Combs

Michael

. 2001. Race and Place: Race Relations in an American City. New York: Cambridge Univ. Press.

50.

Wilson

William J.

2001. The Bridge over the Racial Divide: Rising Inequality and Coalition Politics. Berkeley: Univ. of California Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB