Introduction

At the heart of the US economy is entrepreneurship. In these uncertain economic times, scholars are reminded of the importance of entrepreneurship specifically to urban economies. Businesses are vital to the development of urban neighborhoods as they are a source of job creation and of income and wages among residents. In recent years, a growing body of research has found that entrepreneurship is a means to solving economic problems for a wide range of groups including immigrants, women, and racial minorities who are subject to considerable levels of poverty, discrimination, and spatial concentration of disadvantages (Bates, 1993; Kloosterman & Rath, 2003; Portes, 2010; Wilson, 1996). However, Black business ownership continues to lag behind other groups due to these same structural barriers. This reality is what W.E.B. Du Bois in The Souls of Black Folk described as “the problem of the color-line” (1903/2012, p. 5). For example, economists Fairley and Robb (2008) demonstrated how a series of race-based forms of discrimination limited the experiences and success of Black business owners much like how Massey and Denton (1993) illustrated that racial residential segregation impacts Black Americans more than other racial/ethnic groups. Nonetheless, Black Americans seek entrepreneurship as a source of self-employment largely out of economic necessity, causing the numbers of Black-owned firms to grow considerably in recent decades. And important to their survival, Black firms are also becoming more diversified across industries and professional sectors, making this group an important economic thread in urban communities (Bates, 1993; Boyd, 1991; Harper-Anderson, 2019). Thus, Black entrepreneurship is relevant and important to any efforts to understand the link between race and the urban economy, as well as changing outcomes within these communities.

We explore the relationship between Black entrepreneurship and the drop in juvenile arrests for violence, while taking into account larger economic trends toward deindustrialization and the concentration of disadvantage in these areas. After conceptualizing the importance of, and growth in Black entrepreneurship in the urban economy, we offer distinct measures of Black business

Activity drawing from SBO (Survey of Business Owners) data over three decades (1990 to 2010). These measures are used to estimate the impact of Black entrepreneurship in youth violence during specific decades (1990, 2000, and 2010), and on changes in urban violence over time, before offering a comparative look at the impact of Black entrepreneurs to their White counterparts in 2010. Our overall goal is to not only diversify the discussion of the urban economy within the literature on crime by incorporating literature about race and self-employment occurring in economics, sociology, and urban policy, but also acknowledge the growing importance of Black entrepreneurship to urban communities, access to wealth, and as Du Bois hoped, racial progress. A review of the Black entrepreneurship literature, which acknowledges the specific barriers they face structurally and historically, is offered before proposing the conceptual connections to urban communities and violence.

Literature Review

Black Entrepreneurship

The role of business owners to urban economy is undeniable and the entrepreneurial spirit remains the benchmark of the US economy. For the community, entrepreneurship represents economic growth and job creation; for the business owner, it means monetary rewards and wealth for them as individuals and their families; and for workers, it is a means to a livelihood and wages. While this has been true for business ownership in general, it is certainly true for the Black entrepreneur who has been uniquely impacted by structural disadvantage, deindustrialization, and discrimination historically (Bates, 1993, 1994; Bonilla-Silva, 2003; Boyd, 2000; Bulter, 1991; Wilson, 1996). Gold (2016) utilized a critical race approach to outline how the arrival of Black persons into the US under conditions of slavery to present day systemic racism has sustained large-scale conditions of oppression that impact the economic life chances of Black Americans, including limits on economic advancement. For example, in Du Bois’ seminal study, The Philadelphia Negro (1899/1996), he observed how prejudice and social exclusion prevented Black business owners (caterers in his example; p. 120) from participating or expanding their businesses in times of increasing capital and economic growth. Additionally, Woodard (1997) showed how Jim Crow laws made doing business outside of predominantly Black areas very difficult, limiting Black Americans from competing against other (i.e., White) entrepreneurs. Feagin and Imani (1994) found through interviews with Black construction contractors that racial discrimination existed in key factors associated with business success (e.g., unions, project conditions, contracting and bidding processes) and, more recently, Harper-Anderson (2019) reported that Black entrepreneurs in Chicago claimed structural racism placed barriers and limitations on business success. Though just a few examples, these studies instead of works provide historical insights to structural barriers facing Black Americans, contributing to weakened economic opportunities today. Recently, when drawing upon research identified as CRT, systematic racism, and colorblind racism, Gold (2016) systematically outlined how these factors help explain Black Americans’ low rates of entrepreneurship specifically.

Both in the past and present, the importance of self-employment among racial and ethnic minority groups is not surprising. In fact, the necessity of self-employment for economic gain becomes more lucid when this historical and structural context is acknowledged. Research shows that foreign-born persons have higher rates of self-employment than natives to the US, and further that the immigrant generation of an ethnic group has significantly higher self- employment rates than US-born children (Gold et al., 2006; Min, 2008). For Black Americans, however, fewer than 8 percent were foreign born by 2007, suggesting that access to inter-generational entrepreneurship and economic resources remains relatively quite low (Roberts, 2010). For many minority groups and recent immigrants, starting a business is one of the few sources of employment, particularly when facing obstacles like discrimination and shrinking labor markets. Self-employment (or business ownership) increasingly served as a viable alternative for employment particularly during times of “mismatch” associated with deindustrialization (Boston, 2006; Boyd, 2000). That is, a central theme in the minority-owned business literature is that Black Americans have increasingly sought business ownership in response to exclusion within the mainstream economy (Boyd, 2000; Bulter, 1991; Butler & Wilson, 1988; Harper-Anderson, 2019; Lee, 1998). However, despite a greater propensity on average to attempt starting a new business relative to White Americans, Black American business ownership and success has still lag behind other racial/ethnic groups (Köllinger & Minniti, 2006).

Additionally, Wilson (1996) and others revealed that governmental policies subsidized the movement of White Americans from urban areas, while confining racial minorities in declining areas economically. Racial segregation and the growing suburbanization of America further concentrated Black Americans in areas with weakened institutions (Massey & Denton, 1988; Rothstein, 2017), including housing and abandoned or disinvested properties, which made resources for and opportunities to grow businesses scare. Walker (1998) concluded that these structural disadvantages largely prevented Black entrepreneurs from achieving high levels of success. Thus, although self-employment is often an avenue for economic advancement when access to labor markets and job opportunities are restricted due to deindustrialization and discrimination, structural barriers play a role in limiting the success of Black entrepreneurs via reducing access to external source of capital, increasing the presence of credit card debt, and contributing to shorter survivability of the business relative to nonminority-owned businesses (Bates, 1993; Boyd, 2000; Casey, 2014).

That said, when Black-owned businesses are successful, they can play an important role in revitalizing Black communities: they are often job generators that largely employ low-income and Black workers from inner-city neighborhoods, they often offer higher quality jobs to Black workers than White firms offer, and they promote additional growth in Black-owned firms (Boston, 2006). Specifically, scholars have illustrated that Black-owned businesses employ predominantly a minority workforce and they are more likely than White-owned firms to operate in central areas of cities and recruit from minority neighborhoods (Bates, 1993; Boston & Ross, 1997; Simms & Allen, 1997). Bates (1993), for example, found that 86.7% of Black firms operating in urban zip codes where residents were largely White employed predominantly minority workforce. For this current study, the significant growth in Black entrepreneurs over time provides an interesting opportunity to test for the impact of entrepreneurship on declines in urban youth violence. Urban violence has long been linked to declining economic conditions within urban areas, as well as a breakdown in social institutions and processes that promote social capital and social control. Business ownership, however, has not received attention as a source of economic advancement or social capital to lower Black urban youth violence.

The Urban Economy, Violence, & Linkages

The literature on urban violence establishes the critical role of the economy, including in pivotal accounts of the 1990s American crime drop. Given the vast nature of the economy-crime research, we focus on core themes in this literature relevant to our current work rather than providing an exhaustive review. The first major theme is the large compilation of papers on the American crime drop where economic conditions have received considerable attention (Blumstein & Wallman, 2006; Farrell et al., 2014; Papachristos et al., 2018; Rosenfeld, 2004). Zimring (2007), for example, attributed the drop in New York City homicide rates to growth in both the economy and imprisonment rates, while others linked economic conditions to crime declines in other major US cities as well (see also Baumer & Wolff, 2014; Parker et al., 2017; Rosenfeld et al., 2007). Rosenfeld and Messner (2009) argued that the crime drop in the US and across Europe was largely due to upturns in the economy. While much of this literature focuses on total homicide or violent crime rates, scholars have examined race-specific crime trends as well, finding disparities in the magnitude of the drop across groups and places (Hawkins, 2011; Parker, 2008; Parker et al., 2016), including the importance of Black businesses to the drop in Black arrests (Parker, 2015), as well as the relevance of both immigration and racial segregation patterns (Feldmeyer, 2010; Feldmeyer et al., 2015; Ousey & Kubrin, 2018; Parker & Stansfield, 2015). Overall, throughout the crime drop literature, including studies that examine variation across racial groups and places, studies clearly illustrate the vital role of economic conditions.

A second major theme gaining widespread attention in the social sciences is the role of concentrated disadvantages on urban crime rates. While there are many notable examples, the most seminal is the work of Wilson (1987) where he described the significance of the spatial concentration of poverty in many urban communities, as well as the crippling effect of economic restructuring (i.e., the industrial shift away from manufacturing sectors toward administrative, service lead industries) in many US cities (Kasarda, 1992; Tienda & Lii, 1987; Wilson, 1987, 1996). Empirical support for the link between concentrated disadvantage and urban violence is found throughout much of the criminological literature (Krivo & Peterson, 2000; Lee & Ousey, 2001; Parker, 2004; Parker & McCall, 1999; Sampson et al., 2018; Semenza et al., 2021; Shihadeh & Ousey, 1998) and others claim economic deprivation/disadvantage is one of the strongest, most robust indicators (Pratt & Cullen, 2005) and/or a “classic covariate” (McCall et al., 2010) of violent crime.

Recently, attention has shifted toward examining other labor market characteristics (Parker, 2008), perceptions of economic improvements via “consumer sentiment” (Rosenfeld, 2009; Rosenfeld & Fornango, 2007), and changes in business activity, particularly in restaurant and retail industries in more active crime areas (Rosenthal & Ross, 2010). These studies are particularly interested in how the US economy expanded during the 1990s, in terms of both reduced unemployment (Gould et al., 2002) and collective perceptions of economic improvement (Rosenfeld & Fornango, 2007) contributing to the 1990 crime drop (see also Rosenfeld & Messner, 2009). Further, Sharkey et al. (2017) linked local nonprofit organizations to the national crime decline from 1990 to 2010. Because community organizations are created partially in response to community problems, in an effort to build communities (e.g., the arts and music), and to address issues of crime and disorder, they proposed that local efforts to promote nonprofit organizations should reduce crime rates. They found that growth in community nonprofits led to significant declines in the murder rate, as well as overall rates of property and violent crime in a panel of 264 cities over the 20-year period. These novel areas of research further enhance the rather extensive and prominent literature on the link between the economy and crime trends. Our current study highlights the link between race and the urban economy by directly measuring entrepreneur activity within US cities among Black Americans. Specifically, we document the importance of the growing presence of Black entrepreneurship as a source of economic power, if not a necessary form of self-employment for Black Americans in cities facing economic restructuring and job loss. Further, Black entrepreneurship is also argued to be a contributor to reduced violence arrests among Black youth within these areas. This linkage is conceptualized in two ways in the next section.

Conceptualizing Entrepreneurship Violence

While there are a number of facets to the economy-crime link to consider, we focus on the presence of Black entrepreneurship. Building off the entrepreneurship literature reviewed above, two possible conceptual linkages are explored: (1) the shift in values associated with rising diversity among business owners in the urban context and (2) the economic opportunities created for jobs and social networks that come with Black entrepreneurial growth. While each linkage draws upon different aspects of the role Black entrepreneurship can play in urban communities, both perspectives exemplify the presence of self-employment via business activity for Black Americans in their communities. As stated over a century ago in Du Bois’ discussion of poverty and crime in Philadelphia, “the question of employment for Negroes is the most pressing of the day…… the object of social reform should be so to diversify Negro employments as to afford proper escape from menial employment for the talented few, and so as to allow the mass some choice in their lifework” (1899/1996, p. 141).

The first linkage proposes that Black-owned businesses, through the lives of their owners, employees, and families, can serve an important function- as role models to urban youth in the community or “social buffers” in the words of Wilson (1987, p. 56). Additionally, Anderson (1999) outlined how structural barriers associated with racial inequality and limited economic opportunities foster a cultural orientation conducive to violence among adolescents. In particular, structural conditions lead to cynicism and hopelessness that become an “institutional feature” of urban neighborhoods that shape values and the nature of public interactions, such as violence (Anderson, 1999; Oliver, 2006; Sampson & Wilson, 1995). Anderson’s work has been supported in numerous studies on youth violence at the individual level (Brezina et al., 2004; McNulty & Bellair, 2003; Stewart & Simons, 2006). At the macro-level, Parker and Reckdenwald (2008) found empirical evidence of an inverse relationship between the presence of African American role models and Black youth violence in a large sample of 157 US cities. Additionally, the relationship between Black concentrated disadvantage and youth violence was largely mediated by the availability of role models within the community. Other studies have documented the role “old heads” and role models play in the urban community, particularly among youth (Harding, 2010; Oliver, 2006; Stewart & Simons, 2006; Stewart et al., 2006), and speculate how Black-owned businesses may be important in this regard (Boyd, 1991; Villemez & Beggs, 1984; Wilson, 1987).

Black entrepreneurs within the community, then, provide role models in the presence of discrimination, ineffective social institutions, and otherwise disadvantaged environments. They reinforce values and expectations about social institutions, which can guide youth away from violence. Furthermore, Eisner (2008) suggested that the crime drop of the 1990s could be linked to cultural changes, such as an increased focus on self-control and discipline among youth specifically during this time period. Boyd (1991) also found that Black entrepreneurship was positively associated with social and political well-being among Black Americans (see also Villemez & Beggs, 1984) and Wilson (1987) argued that the Black middle- and upper-class families served a ‘social buffer’ role in disadvantaged communities, whereby the economically stable Black families and business owners were critical to maintaining viable social institutions (e.g., schools, churches, recreational facilities) that deflect pronounced joblessness and poverty (see further discussion in Wilson, 1987, p. 56). In an attempt to account for the presence of Black entrepreneurs as promoters of mainstream values or social buffers, we hypothesize that the presence of Black entrepreneurs will be inversely related to Black youth violence (H1a).

Businesses are also a means of economic resources into the community, reducing the level of economic disadvantage which has been linked to urban violence. As argued, business activity brings self-employment, which has been well established as a necessity to Black Americans.

Businesses also mean job creation by expanding the job opportunities for residents (Bates, 1994; Boston, 2006). In a study of 28 large metropolitan areas, Bates (1994) found that 96.2 percent of Black employers operating within urban minority communities relied largely on minority employees. That is, the availability of, and growth in, Black-owned businesses within the urban community brings economic power and much more. While Black family income levels typically fall below those of White families, individuals who own their own businesses have higher household income than the general population. Minority-owned businesses increase the revenue base of an area and serve as a source of jobs for workers displaced by the removal of manufacturers associated with deindustrialization. Black Americans have sought out self-employment as an avenue to curtail the detrimental impact of these trends. In the midst of the Great Recession (2007–2009), places with higher unemployment exhibited more business startups (Fairlie, 2013). While Bates (1993) found that these businesses can be situated outside minority neighborhoods, Black entrepreneurs are more likely to hire Black workers than are White-owned businesses (see also Bates, 1994; Boston, 2006; Boyd, 2009). Jennifer Lee’s work (1998) also revealed that small business owners in minority neighborhoods hire minority (Black) workers, and this has little to do with Title VII prohibiting employment discrimination or government pressure via affirmative action. To capture arguments concerning the influx of job opportunities associated with Black entrepreneurship and growth in Black firms, we hypothesize that an increase in the number of Black firms with paid employees will be inversely related to Black youth violence (H1b).

As the above discussion articulates, to urban youth, Black entrepreneurship brings job networks and economic mobility, as well as the presence of role models and cultural expectations for youth within these areas, reducing the social isolation from mainstream institutions. Importantly, this discussion draws upon literature in urban planning and economics concerning the importance of entrepreneurship that has been largely unexplored in the sociological and criminological literatures (for more detailed arguments, see Boyd, 2009; Parker, 2015). To date, Parker (2015) is the only research that directly examined the role of Black entrepreneurship on urban violence. However, this work focused only on the period of the crime drop (1990 to 2000) and did not provide a look comparatively for the potential impact of White business activity in urban areas. In our work, we move beyond previous work both conceptually and empirically. First, we examine Black entrepreneurship as conceptually and empirically distinct from employment, deindustrialization, and concentrated disadvantage. Thus, we estimate the impact of Black entrepreneurship on youth violence cross-sectionally in recent decades while accounting for concentrated disadvantage, deindustrialization, and racial income inequality.

In addition, to acknowledge that the changes in Black entrepreneurs can be directly linked to reduction in Black youth violence over time, we utilize a fixed-effect model to estimate the potential within city changes in these constructs over time. In the change model, we hypothesize that the growth in Black entrepreneurship is directly related to the changes (reduction) in violent arrest of Black youth over time (H2). Finally, because comparable SBO data were available for 2010, we report any similarities and differences in the entrepreneurship–urban violence relationship between Black and White groups in the supplemental analysis. We turn to the data and methods employed to test these claims.

Data and Methods

The unit of analysis is US cities with a population of 100,000 residents or more in 1990 that reported complete data (or 12 months of arrest statistics) on our dependent variables. There are three major sources of data for this research. For the dependent variables, the data are Uniform Crime Reporting program data of arrest counts by age sex and race (United States Department of Justice, 1989). The Uniform Crime Report also serves as the source of information on the number of sworn police officers in each city circa 1990, 2000, and 2010. For many of the economic and social predictors included in the study, the Census of Population: Social and Economic Characteristics serves as the major data source.The third source of data is the business ownership information which was obtained by the Census: Survey of Business Ownership (SBO), which are published every 5 years, with years ending in 2 and 7. The SBO is the only comprehensive and nationwide data collection effort on characteristics of businesses owners by gender, ethnicity, race, and veteran status. It is the best, most up to date and comprehensive data on minority businesses (see Fairlie & Robb, 2008). The SBO data used in this study are gathered from the 1992, 2002, and the 2012 reports and linked to the corresponding Census and crime data.

Dependent Variables

This study utilizes Black juvenile arrests based on three-year averages circa 1990, 2000, and 2010. Consistent with other work (Messner et al., 2005; Villarreal, 2004), arrest counts for each city are based on three-year sums (e.g., 1990, 1991, 1992) to control for any fluctuations in reporting. The dependent variable represents the number of Black (non-Hispanic) juvenile violent arrests, where murder, rape, robbery, and aggravated assaults are combined. Arrest data count each instance in which a person is arrested, which is not the same as the total number of individuals arrested. One difficulty with arrest statistics is that they capture crimes known to the police and involve an arrest; thus, they do not represent the number of crimes committed in a given area because not all crimes are reported to the criminal justice agencies. However, arrest data provide information on the number of arrests that law enforcement agencies made in a given year and, given the systematic nature of the UCR data, it is often a source of data used in studies involving large samples and multiple time periods (see Osgood & Chambers, 2000). According to Weisburd et al. (2009), “arrest reports provide the most accurate listing of the ages of suspects” (p. 450), where juvenile offenders are more easily apprehended that adult offenders (Puzzanchera, 2009) and violent offenses are more reliably measured than other types (Gove et al., 1985; Hindelang, 1981; Osgood & Chambers, 2000). For example, Piquero et al. (2014) not only reported agreement between self-report and official arrests for serious youth offenders in a longitudinal study, but that agreement is also stable over time and similar for race/ethnicity.

Independent Variables

We offer a series of measures to capture the local urban economy- concentrated disadvantage, deindustrialization, and the two indicators of Black American entrepreneurship. First, similar to the extant literature, our measure of concentrated disadvantage is a composite measure generated using principal components analysis and includes the following five indicators: (1) poverty (percentage of Black persons living below the poverty level); (2) income inequality (via the Gini Index of Black Income Concentration; Index can range from 0, representing perfect equality to 1, representing perfect inequality); (3) single parent homes (percentage of Black children aged 18 year or less not living with both parents); (4) racial residential segregation (measured with the Index of Dissimilarity; Index measures 0 to 1, the closer to 1 the more segregated the groups are); and (5) the percentage of female headed households receiving public assistance payments.Footnote 1 Deindustrialization is measured by Black Americans that are employed 16 years of age or older within the service industry out of all Black Americans 16 years of age and older in the population. We include a service industry-specific indicator to tap both the availability of jobs within these industries at a given time point (1990, 2000, 2010) but also the shift in Black employment levels over time.

Central to our study, we offer two distinct measures of Black entrepreneurship—the presence of Black entrepreneurs and the rate of Black firms with paid employees. The first measure is operationalized as the percentage of all businesses or firms that are Black -owned. The second measure is the number of paid employees in Black firms out of the total number of Black firms with employees, converted to a rate per 100,000. Evidence has been produced showing that measures of Black entrepreneurship are distinct from other economic measures like concentrated disadvantage and overall employment (Parker, 2015). A measure of racial inequality is also included in this study to account for the level of economic disparities between Black and White Americans. Racial inequality is measured by the ratio of White-to-Black family median income (the higher the ratio the less racial inequality). Finally, a measure of the presence of community nonprofits, which is the cumulative number of active nonprofits, is also included in this research. This measure is provided by Sharkey who offers detailed nonprofit data for 234 cities for all years from 1990 to 2013 (Sharkey, 2018).

Control Variables

Control measures include the percentage of the population with Hispanic origin (% Hispanic), a dummy code for region (cities located in the southern region are coded as one and all other cities are coded as zero) residential mobility (operationalized as the percentage of residents that reported they were not living in the same residence for the previous five years), and a measure of overall business activity (total number of businesses out of the population per 100,000 residents). We control also for police force size (the number of sworn police officers [excludes civilians] in each city per 100,000 population) given our focus on arrests and variation in total population size. Table 1 displays the descriptive statistics for all measures included in the study, and across time periods, as well as indicators of change in predictors over time.

Table 1 Means, standard deviations (in parentheses) for all variables—1990, 2000, & 2010, and percent change 1990–2000 & 2000–2010
Full size table

Analytic Techniques

For the multiple regression models, different estimation procedures are called for when determining the influence of Black entrepreneurship on youth violence cross-sectionally and in change models. Because the relationship between Black entrepreneurship and urban violence has received little attention in the empirical research, we provide a systematic examination of this relationship by examining key periods of time cross-sectionally before estimating the potential relationship between changes in Black-owned business and Black youth violence over time. For our cross-sectional models, preliminary analysis revealed a non-normal distribution in our dependent variables, suggesting that ordinary least squares (OLS) regression estimations were not appropriate. Work by Osgood (2000) and Osgood and Chambers (2000) who also examined youthful offenders, and numerous other studies since the publication of their work, illustrate the appropriateness of count models. That is, Poisson estimators use counts as the outcome variable, which are a more useful and potentially reliable method when examining crime events among specific subpopulations. Due to evidence of over-dispersion in these data, we use the negative binomial variant of Poisson regression (Greene, 1996). Finally, we converted the Black juvenile arrests counts into the equivalent of a rate by including the logged young Black population size (all Black persons 17 years of age or younger) variable as an offset variable in the model and constraining the coefficient to equal 1 (Osgood, 2000). We obtain robust standard errors for the parameter estimates, as recommended by Cameron and Trivedi (2009), in an attempt to adjust for heterogeneity in all of our models. While R-square is not typically part of the Poisson- based estimation, a pseudo r-square is reported for each model. Additionally, the BIC (Bayesian information criterion) statistics are reported since they offer a useful way to compare maximum likelihood models on the same data, where smaller values on the information criterion are considered to be better (Raftery, 1995). Thus, the BIC value will be compared in models of same data to illustrate the contribution of Black entrepreneurship measures specifically. The Black models that span three decades are presented in Table 2. Later, in Table 3, we offer a comparable model for White Americans in 2010.

Table 2 Negative binomial regression equations using (robust standard errors), 1990, 2000, & 2010. Z scores reported in brackets
Full size table
Table 3 Fixed-effect negative binomial pooled equations, 1990 – 2010 & race-specific comparison model, 2010 with coefficient comparison test. Z scores reported in brackets
Full size table

Our estimation continues with a change model also offered in Table 3, which employs a fixed-effects estimator for pooled time series design. We use fixed-effects estimation for a variety of reasons: (1) it expresses variables only in terms of change within a unit, which is appropriate since we seek to determine if key theoretical constructs (e.g., Black entrepreneurship) influence the change in the dependent variable for a given city; (2) it controls for unobserved characteristics that are time invariant and not included in the model; and (3) the Hausman specification test indicated that the fixed-effects method is more appropriate than random-effect modeling. That is, this approach allows for each city to act as its own control by removing any bias that arises from time-invariant differences between cities. The approach was performed using STATA (version 18.0) and robustness tests were performed and period effects as controls were included.

Results

Descriptive Statistics

Before turning to the multiple regression analysis of this study, we first present some of the key descriptive findings from Table 1. First, Black-owned businesses have out-paced general business growth patterns nationally (see also Boyd, 2009; Fairlie & Robb, 2008; Smith & Tang, 2013). From 1990 to 2000, Black entrepreneurs’ presence grew 27.9% (7.53%, 9.62%, respectively) and the number of Black-owned firms with paid employees increased 15.8%. During more bleak economic times from 2000 to 2010, the number of Black-owned businesses increased by 55.4%, which is more than triple the national rate of 17% for all businesses in our sample of cities (note- the national rate reported by SBO was 18% across all geographic areas). Fairlie (2013) reveals that places with higher unemployment during the Great Recession (2007–2009 specifically) had higher rates of business starts. Looking at our indicators of disadvantage, our large sample of cities experienced significant growths in poverty, households seeking public assistance, and racial segregation during this time period, which supports Fairlie’s findings.

Over the same period, the number of paid employees in Black firms increased 80.7% (from 517,563.7 to 935,109.3, respectively). This growth has also been documented by others (Bates, 1994; Boyd, 2009) and is comparable to national reports and estimates provided by a Commerce report (Minority Business Development Agency, 2011). It is partially due to the ability of Black Americans to move into industries where minority presence has traditionally been lacking (Bates, 1993; Harper-Anderson, 2019). However, an imbalance between growth in numbers and performance in terms of sale receipts must be noted. For example, Harper-Anderson (2019) found that between 2007 and 2012, the number of Black firms in Chicago increased by 32%, but there was a 10% decline in sale receipts reported by Chicago firms during this time period.

In terms of other key measures, we highlight a few additional findings from this table. First, in line with extant literature on the crime drop in US cities, violent offenses involving Black American youth dropped approximately 29.75% from 1990 to 2000 and 31.5% from 2000 to 2010. Our other measures of economic climate indicate a slight drop in concentrated disadvantage from 1990 to 2000, largely attributed to drops in poverty (14%), racial residential segregation (17%), and the removal of public assistance to female headed households (47%). However, from 2000 to 2010, levels of Black concentrated disadvantage grew 13% due to rising poverty, segregation, and households on public assistance. Evidence of deindustrialization, particularly the displacement of Black workers, is also present. The availability of service jobs to Black Americans declined 5% during the ten years spanning from 1990 to 2000 yet increased 14% by 2010.

The Cross-Sectional Models: 1990, 2000, & 2010

Table 2 displays the cross-sectional results of the negative binomial regression estimation in 1990, 2000, and then 2010. For each decennial time period, the first model excludes the measures of Black entrepreneurship while the 2nd model includes these measures for a more detailed look at the contribution of Black business activity on Black youth violent arrests. As shown in Table 2, the BIC statistic is considerably smaller in model 2, when compared to model 1, across all time periods. The model with the lowest BIC is considered the best model and used as an argument against the other model. The Bayesian information criterion (BIC) is also one of the most widely known tools in statistical modeling, which allows for comparing models since the BIC is calculated for each model. Here, the difference in the statistic from model 1 to 2 is large, providing strong evidence favoring model 2 over the alternate. Given this, our interpretation of the results will be organized around key theoretical constructs, focusing largely on the full model (model 2) in this table.

We estimate the impact of the Black entrepreneurship on Black youth arrests within the context of US cities, hypothesizing that Black entrepreneurship will reduce youth violence. As shown in Table 2, we find that the presence of Black-owned businesses has an inverse relationship with violent arrests in both 1990 and 2000. That is, an increase in the percentage of Black-owned businesses is associated with a decrease in juvenile arrests for violence in 1990 and then again in 2000. In fact, Z scores for this construct reveal the strength of this predictor on the decline in Black violence controlling for structural disadvantage and other characteristics of cities. On the other hand, the rate of Black firms with paid employees does not reach statistical significance in 1990 or 2000 but yields a statistically significant inverse effect on Black violent arrests in 2010. These findings provide support our hypothesized expectations and the importance of Black entrepreneurship to the urban economy.

The concentrated disadvantage index documents the combined influence of racial residential segregation with forms of Black economic disadvantage to signify the degree to which Black Americans occupy different spatial niches in American cities. Symptomatic of the high levels of concentrated disadvantage and deindustrialization, Black Americans were vulnerable to economic dislocation (e.g., joblessness) and disadvantage more than others, leading us to expect a positive relationship between these constructs and Black juvenile arrests for violent behaviors. As shown in Table 2, we find that concentrated disadvantage has a positive impact of arrests for youth violence in 1990 but not the other decades. Additionally, looking at model 2 in 1990, the increase in Black employment in the service economy is related to a decrease in arrest of youth for violence in the expected direction. Deindustrialization and disadvantage appear less relevant to Black youth violence in 2000 and 2010, wherein insignificant coefficients (in all models) are reported for our indicators using conventional standards. Racial inequality and the presence of community nonprofit organizations increase arrests for Black violence in 2010 relative to other time periods as well.

The relevance of our control measures show varying effects on Black youth violence across recent decades. The control for regional variation (South) significantly influences violent arrest of urban youth regardless of time period, with arrests for Black violence being lower in the South relative to other regions. This finding is inconsistent with previous research, which typically finds that African American adult arrest rates are higher in the South as a result of heightened racial tensions and racial threat in this region (seeLiska & Chamlin, 1984; Stolzenberg et al., 2004; Parker et al., 2005). Cities with larger police force sizes engage in more arrests in 1990 and 2000 but a null effect was found in 2010. We control for the rate of business activity as well and this indicator is positive and statistically significant in three of six models. No other structural indicators produce significant effects on youth violence consistently across the models, yet we find a statistically significant inverse relationship between residential mobility and youth violence in 1990.

Overall, our results suggest that Black entrepreneurs served as role models and social buffers to curtail Black youth violent arrests during each time period directly associated with the American crime drop (H1a), while their ability to provide job opportunities and economic resources to urban areas (H1b) surfaced as more influential to the decline in Black violence in more recent times. Moreover, when comparing Model 1 and Model 2 across time periods, the inclusion of Black entrepreneurship indicators improved the model fit as indicative of the smaller BIC score and growth in the pseudo-r2 for each of the three decades. For example, the Bayesian information criterion statistics indicate that the inclusion of both Black entrepreneurship measures greatly improved the model, relative to the baseline model (Model 1) in 1990, 2000, and 2010. These findings provide partial support for our claims of the importance of Black-owned business to the local economy of urban areas, and that entrepreneurship is directly related to the decline in Black youth violence within these areas.

Changes in Black Entrepreneurship and Violence Over Time

Having established the significance of Black entrepreneurship to Black arrest for violence in the cross-sectional models, we now turn to estimating whether or not the growth in Black entrepreneurship is relevant to the considerable within city change (drop) in Black youth violence during the period of the American crime drop (1990) and since (2010). Table 3 displays the results from our pooled, cross-sectional time series estimation procedure using a fixed-effects estimator. In reference to our central question, we find that Black entrepreneurism has a statistically significant, inverse relationship with the within city change in Black arrest for youth violence from 1990 to 2010. As hypothesized (H2), the growing presence of Black-owned businesses in urban cities has contributed to the significant decline in arrests for Black youth violence in recent decades, net effects of deindustrialization, and the concentration of disadvantage among the Black population. On the other hand, the change (rise) in paid employees within Black firms does not reach statistical significance in the change model. Overall, we find partial support for our hypothesis that the changes (reduction) in Black youth violence were related to growth in Black entrepreneurship within these urban areas.

The within city change in levels of concentrated disadvantage produces an inverse, statistically significant effect on the change in youth violence arrest over time. The other indicator of the local economy, specifically the rise in Black service industry employment via deindustrialization, does not produce statistically significant change in youth violence over time. Racial inequality also does not significantly influence within city change in youth violence, which is consistent with other studies investigating Black arrest rates (Eitle et al., 2002; Parker & McCall, 1999; Parker et al., 2005; Stolzenberg et al., 2004).

Turning to other control measures, police presence is one of the most consistently cited features in the US crime drop literature (Baumer, 2008; Baumer & Wolff, 2014; Levitt, 2004; Zimring, 2011). During the 1990s, for example, the size of the police force grew in many cities as they implemented strategies to target activities linked to crime, such as carrying a weapon (Kubrin et al., 2010; Rosenfeld et al., 2007). For these reasons, we incorporated an indicator of police force size in our change model. As shown, we find that an increase in police force size produces a reduction in arrest for Black youth violence. As research on the “crime decline” continues to expand, more definitive conclusions may emerge about the role police presence plays in the American crime drop as it relates to variation across racial groups and youth specifically. Finally, as shown in Table 3, changes in the size of the Hispanic population has a positive, statistically significant effect on Black youth violence arrests while controls for population size and region produce an inverse, statistically significant influence in our model. That is, similar to the cross-sectional models, the change model shows strong regional patterns. On the other hand, changes in presence of community nonprofits, levels of racial inequality, and residential mobility are unrelated to drops in Black youth arrests from 1990 to 2010.

Supplemental Analysis: Racial Comparison

Thus far, in our efforts to expand upon the study of race and the urban economy by estimating entrepreneur activity within US cities, we document the importance of the growing presence of Black entrepreneurship as a source of economic power and a significant source of reduced violence within these areas, net other social and economic conditions. Our cross-sectional and fixed-effect models, along with their results, confirm this effort. Because of SBO data availability in 2010, we offer a racial comparison of Black and White models in Table 3. That is, we continue our goal of better understanding the relationship between entrepreneurship and urban violence by offering an investigation of the relationship between business activity and violent arrests of youth when comparing Black and White Americans. Similarly to the above expectations, there is every reason to believe that the growing presence of White businesses and the number of paid employees in White firms will reduce violent arrest of White youth as well.

As shown, the two indicators of White entrepreneurship have an inverse statistically significant relationship with White youth violence. In the Black model, the rate of employment in Black firms significantly reduces Black youth violence arrest in 2010, while the overall presence of Black-owned business is unrelated. During this period of economic recession, entrepreneurship plays a significant role for both groups; however, only partial support was found in the Black model relative to the White model. Another key difference is the impact of racial inequality in the models. Racial income inequality leads to higher Black arrests for youth while racial inequality yields no statistically significant influence on White youth violence. Because large scores on this ratio indicator reflect higher economic competition between White and Black Americans, arrest rates of Black youth (while not White youth) increase, which is consistent with racial threat and competition arguments (Blalock, 1967). This finding is also consistent with previous research that racial disparities economically lead to an increase in arrest rates among Black Americans (Ousey & Lee, 2008; Parker et al., 2005). Other differences also appear in the two models. For example, the size of the Hispanic population contributes to a significant increase in White arrests for violence, while total population size is related to Black arrests for violence. Similarities are found as well, such as the violent arrests for White and Black youth are both lower in the southern region of the United States. Overall, entrepreneurship remains the hallmark of the US economy and an important aspect of the economy to incorporate in future work, while other structural conditions in urban areas remain important as well, specifically across racial groups.

Discussion

The advancements in entrepreneurship over the last decade or two have largely been made by immigrants and Black Americans (Bates, 1994; Butler, 1991; Shim et al., 2000), even though government support for these groups has remained weak. In fact, an estimated 25% of entrepreneur growth in high-tech industries between 1995 and 2005 was founded by immigrants (Shim et al., 2000; Zavodny, 2011). Black entrepreneurship has grown considerably over time, as many scholars suggested out of economic necessity, exceeding that of nonminority-owned businesses in many parts of the United States. The Black entrepreneur has higher levels of wealth and upward wealth mobility than do Black workers (Bradford, 2014), leading to an overall reduction in the wealth disparities between Black and White families. Green and Pryde (1996) offered an account of how the entrepreneur is not only an agent of economic success but also a prime motivator of social reconstruction, where self-employment is sought to ameliorate conditions of poverty and growing disadvantage associated with economic restructuring in urban areas. The aim of this paper was to highlight the growing significance of Black entrepreneurship within US cities and to illustrate the role of Black business activity to the decline in youth violence in recent decades as well as the changes in crime during the well-documented crime drop in American cities and period of economic recession. Our effort to test this claim started with documenting the growing presence of Black-owned businesses in urban areas. In doing so, we generate two indicators of Black entrepreneurship to capture the presence of Black businesses in urban areas as well as the rate of Black firms with paid employees—both showing significant growth from 1990 to 2010. Importantly, Black entrepreneurship is an empirically distinct aspect of the local economy, not captured by current investigations of concentrated disadvantage and deindustrialization (specifically Black employment in service industries) in the study of violence. We then offered conceptual arguments for the role Black-owned businesses may play in understanding Black youth violence, hypothesizing that Black entrepreneurship will be inversely related to violent arrests of Black youth within these areas. The first measure was argued to reflect the changing values within urban communities as a result of role models or social buffers that occur with the presence of minority-owned businesses. The second measure, the rate of Black firms with paid employees, captured social networks and job opportunities through available Black firms. Our conceptual arguments were informed by previous research on Black businesses and the structural barriers they face as well as historical and critical race approaches about the experiences of Black entrepreneurs in urban areas.

We first offered a set of analyses to estimate the influence of Black entrepreneurship on youth violence during three recent decades (1990 to 2010), controlling for larger economic trends toward the concentration of disadvantage and deindustrialization. The presence of Black-owned businesses significantly reduced youth violence in 1990 and 2000 (supporting our first hypothesis), whereas the rate of Black firms with paid employees decreased Black youth violence in 2010.

Although we offer this conclusion with caution, given the inability to establish causal mechanisms with these data, our results suggest greater empirical support for the presence of Black-owned businesses on youth violence through role models and value changes during the 1990s crime drop while social networks and economic resources associated with Black firms who employ persons became more relevant during the economic recession.

Utilizing a fixed-effect design, we then estimated the influence of shifts (growth) in Black business activity on the changes in youth violence over time. These estimations also produced a statistically, significant inverse relationship between the presence of Black entrepreneurs and youth violence, net the effects of other social and economic features of US cities. That is, offering further support for our claims about the role of Black entrepreneurship in urban communities, we find the presence of Black business was a significant predictor of the change in Black youth violence between 1990 and 2010, but the growth in Black firms with paid employee firms did not significantly contribute to the change in youth violence over time. This model provides additional, temporal support that Black businesses serve an important function as role models and source of mainstream culture to urban youth in the community. While this claim has been well documented in urban planning and economic-based studies (Boyd, 2009), the present study reveals the importance of the presence of Black entrepreneurship to the study of urban violence, which expands upon the way economic conditions are examined in the criminological literature and incorporates the uniqueness of Black entrepreneurship to the urban economy.

Limitations and Future Research

The current research is limited in several aspects. Our analysis is restricted to the study of youth Black violence, largely because we were interested in how the presence of Black entrepreneurs in urban areas reduced levels of youth violence through serving as role models and providing economic opportunities. Future research should continue to explore the potential for direct and indirect relationships between entrepreneurship and urban violence, particularly when creating measures to capture how business owners serve as social buffers or role models in disadvantaged communities. Albeit important, whether a similar relationship exists between entrepreneurship and urban violence among adults and for other racial and ethnicity-specific groups in more recent time remains to be determined as well.

Additionally, recent research has illuminated the intersecting identities of business owners—including class, race, and gender—and how these factors shaped their decisions to start a business (Wingfield & Taylor, 2018) as well as the goals of the individuals starting their own businesses—such as overcoming discrimination, economic stability, or revitalizing communities (Boston, 2006; Harper-Anderson, 2019). Unfortunately, we are unable to account for all such factors or the motives of the business owners in this study. The SBO data also do not allow for considerations of whether the business owners are locals or other specifics of those they employ—for instance, we do not know the race of the employees or if the employees are from the neighborhoods where the businesses are located, if they are from neighboring communities, or commute from further away. More definitive conclusions also require analysis based on a longer period of time and more recent time periods to determine whether these impactful results continue.

We further caution the use of arrest rates. Arrest data reflect the number of incidents known to the police across a large selections of US cities, thus an important source of data for our national multiple city study, but arrest statistics may not reflect the true level of crime (i.e., those offenses not known to the police). That is, the limitations of official arrest data have been well documented, thus seeking other reliable data on criminal behavior across locales is warranted. Scholars could lead the way in developing research strategies to better understanding the perceptions and attitudes toward Black American and immigrant business owners and investors. While attitudes and perceptions toward immigrant- and Black-owned business are beyond the scope of this current investigation, future research must acknowledge the greater need for government support toward small business owners and business ownership by immigrants and native minority groups. Governmental support has had positive effects on entrepreneurs overall (Casey, 2014), and community-based organizations have also significantly increased access to credit for lower wealth entrepreneurs which include some minority business owners. Given research that documents discrimination in the lending, an important implication from this research is that additional oversight in the lending community could go a long way to promoting the sustainability and growth in minority-owned businesses (Robb & Fairlie, 2007).

Conclusions

Harper-Anderson (2019) argued that entrepreneurship could help even the playing field between racial groups and reduce poverty and disadvantage through job opportunities. Our research further illustrates the benefits of Black businesses to Black youth residing in urban communities. As we explore the positive and rather distinct role of Black entrepreneurship in US cities, this investigation allows us to move beyond empirical investigations that focus solely on concentrated disadvantage and joblessness of Black Americans. As a result, we are able to champion the importance of Black entrepreneurs by illustrating the positive influence this group has on reducing youth violence in urban areas over time, as well as offer perspectives that present Black resistance to economic decline and discrimination in job opportunities within urban communities.

While we examine Black entrepreneurship as a significant and growing feature of US cities, we acknowledge that this group is neither homogenous nor monolithic. In fact, while this research provides an important starting point, the specific role that Black-owned businesses play is in need of further investigation in criminological literature. Only future research will be able to fully establish the unique role of Black entrepreneurship, whether it is the presence of businesses, or the jobs/employment created through black firms within these communities that influence Black youth or the changes in values and social networks within the community that thrive as minority-owned businesses grow.

While we show the role that business ownership and employment can play in understanding the decline in youth violent crime trends over recent decades, the importance of entrepreneurship has long been discussed. Du Bois (1899) wrote at the turn of the twentieth century that employment was on the most pressing issues Black Americans were facing and that social reforms should focus on diversifying the opportunities available to Black workers that would allow for advancement beyond the most menial positions. And, given discriminatory barriers often limited employment opportunities to Black workers, Du Bois (1899) advocated for the creation of better industrial opportunities for the next generation of Black workers which included the importance of Black Americans going into business for themselves and others in their communities. As such, Du Bois (1912) believed that progress for Black Americans would include Black-owned businesses and employment opportunities available in these businesses among other factors.

Unfortunately, what we have seen in the US is continued structural and racialized barriers to employment and the startup of Black-owned business. That is, despite evidence that Black Americans may be more likely to attempt a business startup (Köllinger & Minniti, 2006), other research shows that they remain less likely to own businesses than other racial/ethnic groups and tend to experience less success on a number of fronts including lower sales numbers, being able to afford fewer employees, and having higher closure rates relative to White business owners (Fairlie & Robb, 2008). Thus, the structural barriers of the past appear to continue to limit opportunities for Black entrepreneurship today despite research showing the benefits of Black business ownership and employment can have on revitalizing communities (Boston, 2006) as well as on lower rates of youth violent crime as illustrated in the current study (see also Parker, 2015). As such, like Du Bois argued during his lifetime, social reform today should work to improve the opportunities and support for business startups among Black Americans given the benefits this can bring to low- income Black communities and their residents as well as city revitalization more generally (Boston, 2006).

For Black American business ownership specifically, it is important to realize that business ownership (or self-employment) is partially driven by a lack of opportunity and discrimination within the larger labor market structure (Holzer et al., 2005; Wilson, 1987). These business owners are largely dependent on their own personal or family capital for startup funds, where their access to external capital relative to nonminority-owned businesses are significantly limited (Bates, 1993; Boyd, 1991; Bulter, 1991). This often leads to higher rates of failure among minority-owned businesses, which will only continue to fail at higher rates without greater institutional support. Understanding this reality and the racial discrimination that has shaped it— past and present—are critical to overcoming this structural barrier (Gold, 2016). Furthermore, in more practical terms, providing assistance to minority-owned businesses, such as technical training, information on how to seek out loans, programs on how to run a business, developing accounts and building assets, could reduce barriers faced by minorities when seeking financial capital or startup funds (see also Robb & Fairlie, 2007). Through greater oversight in lending companies and the development of programs to not only benefit minority-owned businesses but also help low-income families build wealth and assets, sustainable growth, diversity, and expansion in Black American business owners will follow, along with the potential for lower urban violence.