Racial diversity and its asymmetry within and across hierarchical levels: The effects on financial performance

Abstract

The benefits and drawbacks of diversity inside organizations have been the focus of attention for researchers and practitioners for several decades. In our article, we investigate the business case for racial diversity across different hierarchical levels. More precisely, we ask: How does racial diversity within organizations and its asymmetry across hierarchical levels affect their financial performance? From a sample of 143 US law firms from 2008 to 2012, we provide strong support for the business case and show that greater racial diversity for the entire organization is positively associated with firm financial performance. However, contrary to our initial expectations, the benefits of diversity are not more pronounced at the top of the organization, where its effects should arguably be more clearly observable. Diversity seems to have a similar effect across the three levels in law firms: associates, mid-level and partners. Furthermore, we find that the most profitable firms actually have their racial diversity heavily concentrated at the associate level. We discuss alternative explanations for this surprising finding and why the top-performing law firms have both overall higher degree of racial diversity and more concentration of its diversity at the lower level.

Keywords

financial performance law firms organizational hierarchical levels racial diversity

Introduction

On 11 August 2011, President Obama, the first African American US President, signed an Executive Order establishing a coordinated, government-wide initiative to promote diversity and inclusion in the federal workforce. The Executive Order succinctly encapsulated the so-called ‘business case’ for diversity, namely that promoting racial diversity is both the right thing to do and good for business. In the words of the Executive Order,

Our Nation derives strength from the diversity of its population and from its commitment to equal opportunity for all. We are at our best when we draw on the talents of all parts of our society, and our greatest accomplishments are achieved when diverse perspectives are brought to bear to overcome our greatest challenges.¹

The US government is not alone in promoting the business case for diversity. Large multinational firms have also strongly encouraged, and sometimes mandated, greater demographic diversity among their organizations, often including top management (Miller and del Carmen Triana, 2009).

But while the business case for diversity is widely popular with politicians and companies, it is also controversial. It has been criticized for providing a weak basis of support for, or even undermining, equal opportunity: ‘Once the debate is conducted in the language of what is in the interests of business, then a business case can be articulated against [equal opportunity] action’ (Dickens, 1999: 10). It has also been criticized for making equal opportunity dependent on economic efficacy, thereby undermining moral arguments for promoting diversity (e.g. Noon, 2007).

This last point is particularly important owing to the mixed empirical evidence for the economic efficacy of racial diversity, thereby potentially undermining support for greater racial inclusion. Though research on the effects of demographic diversity, especially racial diversity, has grown in recent years (Jonsen et al., 2013; Mayo et al., 2017), this research has been mostly at the individual (Elfenbein and O’Reilly, 2007), team (van Knippenberg and Schippers, 2007), or the top management team (TMT) level (Certo et al., 2006). There is also research from economists and economic geographers, who have developed a rich body of research examining the effect of increasing diversity owing to migration on the economic performance of cities (Nathan, 2015) and countries (Kerr, 2013; Kerr and Kerr, 2011; Nathan, 2014). In contrast, ‘relatively little attention has been devoted to the firm-level consequences of firm-level diversity in the past literature’ (Joshi et al., 2011: 531), and the empirical findings at the firm level that do exist are inconsistent (Joshi et al., 2011).

These inconsistent empirical findings are not surprising. Theoretically, there are strong reasons to believe that diversity could have either a positive or a negative effect on firm financial performance. On the one hand, based on the information-processing and decision-making perspective (van Knippenberg and Schippers, 2007), diverse groups should outperform homogeneous groups because they are likely to elaborate on task-related information and generate more creative ideas and solutions (Pelled, 1996). On the other hand, based on social categorization theory (Brewer, 1979; Tajfel and Turner, 2004) and the similarity/attraction paradigm (Byrne, 1971), homogenous groups could outperform heterogeneous groups, because in-group/out-group behavior limits the development of a team identity in racially diverse teams (Brodbeck et al., 2011).

Given this debate about the performance implications of racial diversity, it is important to examine the organizational contingencies that may influence the diversity–performance relationship (Joshi and Roh, 2009; van Knippenberg and Schippers, 2007). One contingency that remains underexplored is the effect of diversity across hierarchical organizational levels (Nathan, 2016). In this article, we ask: How does racial diversity within organizations and its asymmetry across hierarchical levels affect their financial performance? In doing so, we extend existing research to study not only the effect of diversity at the firm level, but also the hierarchical levels within the organization at which racial diversity is most likely to have a positive effect.

We examine these effects in a setting uniquely well suited to observe the potential effects on firm financial performance, namely large US law firms. Because of their societal importance, professional services organizations, like law firms, accounting firms and consulting firms, are a particularly important context for studying the performance effects of diversity (McGinn and Milkman, 2013). Moreover, as suggested by earlier research on human capital (Hitt et al., 2001) and racial diversity (Richard, 2000), law firms are an especially appropriate context to study diversity because of the critical role that human resources and workgroup dynamics play in their operations and organizational performance, as we discuss more fully later in this article. Using a sample drawn from the largest law firms in the United States, we examine the effect of racial diversity on firm financial performance in the workforce as a whole and, most critically, within and across organizational levels, that is, from the lower-level associate and mid-level attorneys to the partners at the top of law firms.

Buttressing previous research that racial diversity can indeed be a strategic firm resource and in line with the business case for diversity (Andrevski et al., 2014; Richard et al., 2003, 2007), we find that racial diversity at the firm level is indeed positively associated with financial performance in law firms, including when we account for potential endogeneity of both racial diversity and firm performance. However, we do not find empirical support for our expectations regarding the hypothesized greater effect of diversity at the partner level on firm performance. Instead, our results indicate that racial diversity is valuable across all three of the organizational levels in law firms. However, organizations with the highest performance have a greater percentage of their racial diversity concentrated at the associate level rather than at the partner level, which is contrary to our initial predictions. In the Discussion section, we elaborate on our main findings, especially the positive effect of diversity across all organizational levels and the greater concentration of racial diversity at the lowest level for law firms with the highest financial performance.

Theory and hypotheses

Racial diversity and financial performance

The existing literature on diversity identifies both positive and negative performance effects for racial diversity, usually at the team level (Harrison and Klein, 2007; Polzer et al., 2002; van Dick et al., 2008). On the one hand, diverse groups may be expected to outperform homogeneous groups because they are more likely to elaborate on task-related information (e.g. van Dick et al., 2008; van Knippenberg and Schippers, 2007). This means that diverse groups will be less likely to engage in groupthink, and more thoroughly analyze all available information before coming to a decision. This positive view of diversity as variety (Harrison and Klein, 2007) suggests that diverse groups will have more cognitive resources (i.e. distinct knowledge, skills and abilities), and will thus be able to bring differing opinions and perspectives to problem solving. This is particularly important for critical thinking and developing innovative solutions in challenging tasks, such as designing strategies in a complex legal case.

A key benefit of diverse groups is the necessity to elaborate on conflicting viewpoints, which facilitates friction that enhances deliberation and reduces conformity, groupthink and errors (Levine et al., 2014; Mayo, 1999). In this sense, a diverse configuration of individuals provides the team with different levels of skills, information and perspectives that can make it more productive (Horwitz and Horwitz, 2007). Indeed, several studies confirm that exposure to diverging and potentially surprising perspectives may lead to more creative and innovative ideas and solutions in management teams (Ancona and Caldwell, 1992; Bantel and Jackson, 1989; De Dreu and West, 2001; Miller and del Carmen Triana, 2009). Moreover, a more diverse workforce may also raise firm performance through allowing better matching of people to tasks (Peri and Sparber, 2011). Thus, racial diversity may be associated with better organizational performance, especially when effective team dynamics are highly relevant (Srikanth et al., 2016), such as in law firms.

On the other hand, there is also a more pessimistic view of the effect of racial diversity, which is that it makes diverse teams more difficult to coordinate, and potentially less effective. Individuals in diverse groups may have negative reactions to racially diverse members, thereby negatively affecting team performance. As noted by Harrison and Klein (2007), diversity as separation can be indicative of differences in deep beliefs and values among team members and, as a result, it may invite social categorization dynamics within the team. Social categorization theory holds that similarities and differences are used as a basis for categorizing like and unlike into groups, with the resulting categorizations distinguishing between in-groups and out-groups (Brewer, 1979). People tend to like and trust in-group members more than out-group members, and thus generally tend to favor in-groups and discriminate against out-groups (Tajfel and Turner, 2004).² From this perspective, we could expect lower performance in organizations that make greater use of racially diverse teams, especially when team dynamics are most critical for performance.

This debate has not been settled yet by empirical research (Harrison and Klein, 2007), though we believe that the positive effects of racial diversity are likely to predominate in the context that we are investigating. Workforce-related resources, for example, their racial diversity, should be particularly important in professional service firms, such as law firms, because a firm’s competitiveness depends almost entirely on the characteristics and the skills of the firm’s workforce. Indeed, Barney (1991: 101) stated that one means of establishing competitive advantage is through ‘the training, experience, judgment, intelligence, relationships and insight’ of individuals within the organization. In this sense, the racial composition of a workforce can be a strategic resource to the extent that it is indeed rare, difficult to substitute or imitate, and enables firms to pursue valuable opportunities (Barney, 1991). The presence of minority members can enhance the organization’s ability to reach different market segments, especially those more concerned with racial equality. Furthermore, racial diversity would also be valuable if it can serve to foster an organization’s creativity and improve decision making, thus providing a source of competitive advantage that may lead to superior performance (Cox and Blake, 1991).

Based on these arguments, our first hypothesis suggests that we should observe a positive relationship between racial diversity and financial performance for law firms as a whole. Legal work is intellectually challenging work where creativity and better problem solving are crucial. Law firms create value by protecting the interests of their clients by, for example, crafting persuasive and novel legal arguments in court. Assuming that law firms capture some of this value, the superior creativity and better problem solving found in previous literature should be reflected in law firm profits (Andrevski et al., 2014; Richard et al., 2007). Thus, despite the existence of conflicting results (see Kochan et al., 2003), we formulate our baseline hypothesis based on the benefits of having a racially diverse workforce as a strategic resource in law firms. Accordingly:

Hypothesis 1: Racial diversity for the entire organization will be positively related to financial performance for law firms.

Effects of diversity across hierarchical levels

Though we argue above that greater racial diversity for the entire organization is likely to increase financial performance, the distribution of racial diversity throughout the hierarchy in the organization could also affect firm financial performance, though no prior research has been conducted on this issue. We analyze now how racial diversity may not necessarily have the same positive effect on financial performance across different organizational levels. Building on the ideas presented above about racial diversity as a strategic resource in law firms, we hypothesize that racial diversity at the partner level could be expected to have the greatest effect on performance.

There are several reasons behind our expectation that racial diversity at the top of the organization should have greater consequences on firm performance. First, though all the tasks that lawyers perform should benefit from the increased debate, decision comprehensiveness and creativity that diversity could facilitate, these effects could be more beneficial for some tasks than for others. For example, decisions relating to strategy, such as whether to expand into new practice areas or regions, are made by partners. Partners are the ultimate owners of the firm, and decide issues such as expansion, hiring and mergers. Accordingly, any potential benefits and drawbacks deriving from greater diversity in making these types of strategic decisions should be most critical at the partner level, and they should be more clearly observable when partner diversity increases.

Second, greater diversity at the partner level could benefit the firm financially by providing greater motivation to lawyers lower in the hierarchy. Lower-level employees in law firms may make social comparisons with their superiors (McGinn and Milkman, 2013), and these comparisons with demographically similar superiors who serve as role models are likely to create a positive self-image (Brewer and Weber, 1994). Such a demographic match between a subordinate and superior could signal to lower-level ethnic minority employees that success is possible (Barker et al., 1999), leading to greater effort as they work their way towards promotion (McGinn and Milkman, 2013). Because effort, in the terms of hours billed, should directly translate to financial performance, this should mean greater profitability for firms.

An example of this effect is the experience of the US military during the Vietnam War. Following the Second World War, the number of African American enlisted men had increased greatly, but the number of African American officers was extremely low. Owing to this discrepancy, the military experienced, ‘increased racial polarization, pervasive disciplinary problems, and racially motivated incidents in Vietnam and on posts around the world’ (Wilkins, 2007). In response, the US military launched a strong Affirmative Action program to increase the number of African American officers. The success of this program is one example of the effectiveness of increasing the racial diversity of higher-level managers.

Third, important stakeholders could reward firms that conform to desired social norms and promote racial diversity. Institutional theory holds that firms can gain sociopolitical legitimacy through the value that cultural norms and political authorities place on an activity (Aldrich and Fiol, 1994). Important institutional actors, such as regulators and multinational corporations, encourage the development of such norms. These actions by stakeholders provide institutional pressure on firms to hire and promote diverse attorneys, and also provide strong signals to law firms that they will benefit from doing so. Indeed, much legal work involves representing clients before these various agencies that are promoting diversity. For example, the Office of Diversity and Inclusion (ODI) was specifically established to ‘ensure that Federal departments and agencies recruit and retain talented individuals from all communities.’³ More specifically, agencies of the United States Federal Government, such as the Federal Deposit Insurance Corporation (FDIC), provide preferential contracting and legal referrals to so-called ‘Minority and Women Law Firms,’ that is, those majority-owned by racial minorities and women.⁴ These agencies also regularly report on the number of referrals to minority-owned firms. Following the edicts of these agencies by promoting diversity could give diverse law firms a competitive advantage when dealing with these important institutional actors.

Other important stakeholders, such as large corporations, also provide incentives for firms with greater diversity at the partner level. Wal-Mart provides an example as to the importance of diversity at the partner level to important stakeholders. Wal-Mart required that at least one person of color and one woman must be among the top five relationship attorneys handling its business (Wilkins, 2007).

Overall, the positive performance consequences of racial diversity should be most critical and noticeable for the highest level of the hierarchy, which are in charge of the key strategic decisions, provide motivation to lower-level employees, and bring institutional support and legitimacy to the organization. Hence, we elaborate our second hypothesis about the asymmetric effects of racial diversity across hierarchical levels on law firm performance:

Hypothesis 2: Firms with a substantial concentration of racial minorities at the top of the organization will outperform those in which racial minorities cluster lower down the organizational hierarchy.

Methods

Organizational context and sample

Law firms are generally organized as partnerships based on a hierarchical pyramid structure (Hitt et al., 2001). The bottom is made up of a large number of relatively inexperienced associates, who are salaried employees of the firm. The middle is made up of counsel and non-equity partners, who are more experienced attorneys, but nonetheless still salaried employees of the firm. The top is constituted of equity partners, who have an ownership interest in the firm.

The hierarchical structure is reflected in the different roles of the attorneys in the firm (Hitt et al., 2001). Lawyers are usually members of so-called practice groups, based on their area of expertise. These practice groups vary in size, with the smallest specialties being a few attorneys, and the largest being several dozen lawyers, sometimes well over a hundred (such as litigation, or mergers and acquisitions). Most work in law firms is accomplished by using groups of lawyers from these practice groups for specific projects (deals, cases, etc.). Normally, several associates work on projects in groups, with fewer non-equity partners or ‘of counsel’ (roughly equivalent to mid-level managers) acting as the group leaders (Hitt et al., 2001). Also, equity partners often play multiple roles, with their primary responsibility being to act as the ‘rainmakers,’that is, someone who excels in generating business and attracting clients. They also act as the key contact person with the client, building and maintaining relationships with clients (Hitt et al., 2006). To a lesser extent, they manage attorney teams and provide legal services.

Though work is generally accomplished with teams, these workgroups are temporary, can vary in size over time, and an attorney usually works on multiple projects simultaneously. For instance, a litigation associate might initially work on a case with only another mid-level attorney. However, if that case goes to trial, numerous other attorneys, including multiple associates, more mid-levels and possibly an equity partner, will be brought in to handle the additional work. In sum, at each hierarchical level within a law firm, lawyers have specific roles that they perform, and how well they perform those roles can directly affect the firm’s financial performance.

For our study, we draw our sample from the list of the 200 largest US law firms by total revenue published annually by The American Lawyer. We build a panel dataset from 2008 to 2012 from different data sources, as discussed below. The final sample comprises only 143 distinct US law firms and 444 firm-year observations over this period, owing to missing data for some years because of acquisitions and consolidations, changes in rankings, and lack of data owing to combining data from multiple sources (Hitt et al., 2001).

Measures

Firm performance. We measure law firm financial performance by taking the log of the profits per equity partner (PPP), the prevailing measure of performance in the industry. We also use the log of gross revenue for robustness purposes. Our data on firm financial performance are derived from a profitability index reported annually by The American Lawyer, which has been used previously by other researchers (Hitt et al., 2001, 2006).

Racial Diversity. Race is a controversial category. According to Omi and Winant (2015: 21), ‘[t]he meaning of race is defined and contested throughout society, in both collective action and personal practice. In the process, racial categories themselves are formed, transformed, destroyed, and reformed.’ In this study, we rely on self-reported measures of race and ethnicity. As we argue above, racial diversity will proxy for cognitive diversity, thereby providing more cognitive resources that should improve problem solving and decision making. This is so because people identifying with different racial groups should ‘pull from different wells of experience,’ shaping their identity and potentially guiding their behavior (Page, 2007: 308). We obtained these data from the Law Firm Diversity Database (the ‘Diversity Database’) to compute several racial diversity measures. The Diversity Database was created by Vault.com Inc. and the Minority Corporate Counsel Association (MCCA), and provides statistics and other information on diversity at more than 300 US law firms nationwide, covering most of the same firms for which we have financial data.⁵ This database is compiled from responses to the annual Vault/MCCA Law Firm Diversity Survey.

We use Blau’s (1977) index of heterogeneity to measure racial diversity for the total firm and for each of the hierarchical levels. This measure of diversity is the de facto research standard to measure racial diversity (e.g. Berrone et al., 2016; Julian and Ofori-Dankwa, 2017; Richard et al., 2007). Blau’s index is also known as the inverse Herfindahl-Hirschman/Fractionalization Index (HHI). We calculate Blau’s index as:

R a c i a l D i v e r s i t y = (1 - \sum {P_{k}}^{2})

where P is the proportion of group members in category k. Values of Blau’s index range from zero to (k – 1)/k. Its maximum occurs when members of a team, V_k, are spread equally – called ‘evenness’ or relative abundance of species in ecological literature – over all possible k categories – called ‘richness’ of species (Harrison and Klein, 2007). An even spread of members over the richest number of information sources corresponds to Harrison and Klein’s (2007) definition of maximum variety.

There are seven categories of race that are included in the Diversity Database: White, African American/Black, Hispanic/Latino, Alaska Native/American Indian, Asian, Native Hawaiian/Pacific Islander and Multiracial. As the number of categories increases, the maximum value of Blau’s index score also increases; for seven categories, the index takes on a range from 0 to .86. An index of zero suggests only one category of employees, whereas a value of .86 implies that all seven categories are equally represented in the organization (i.e. more diversity). Following Richard et al. (2007), we created a racial diversity variable for attorneys, which is Blau’s index for the total firm (diverse attorneys), and we also computed Blau’s index for each hierarchical level – diverse associates, diverse mid-levels and diverse partners – to create a variable of racial diversity for each of these organizational levels. We calculated all measures of diversity for each firm for each year.

Asymmetry of diversity. We also use Blau’s index to calculate asymmetry of diversity among the various hierarchical levels, using exactly the same formula, but relying on the percentage of diverse attorneys for each of the three levels as P_k in Blau’s index formula above. We take the inverse of this score, such that a higher score would mean greater concentration, that is, greater asymmetry of diversity across levels, such that the level of diversity is unequally distributed among the three organizational levels. Thus, this measure is now fully equivalent to the Herfindahl index of industry concentration in diversification research, but applied to the percentage of diverse attorneys for each of the three hierarchical levels in law firms. As a measure of Asymmetry of Diversity, a high value for this variable means that the firm has a greater concentration of racial diversity at one specific level, as opposed to having a symmetric level of diversity across the three hierarchical levels.

It should be noted, however, that Blau’s index traditionally used by researchers on diversity as variety (Harrison and Klein, 2007) does not capture which specific group is more prevalent; that is, it does not tell us whether the asymmetry that may exist indeed results from higher diversity at the higher level (partners) or the lower level (associates) within the organization. To fully understand the performance implications of asymmetry of diversity, we need to rely on additional data, particularly the type of asymmetry that we observe in the best-performing firms versus the other law firms. For this analysis, we further compute the Euclidean distance measure, as discussed below.

Euclidean distance. To explore whether there is an optimal amount of racial diversity at each hierarchical level specifically, we also computed diversity as separation through the entire organization (Harrison and Klein, 2007). Thus, we calculated the mean Euclidean distance for each firm (k) in our sample from an optimal point inductively derived from the best performers (j) using the following equation (Harrison and Klein, 2007):

E u c l i d e a n D i s t a n c e_{j k} = \sum \sum {(D_{i j} - D_{i k})}^{2} / n] / n

where the mean distance for each k firm from the diversity profile of the j best-performing firms is computed based on D_ij = the optimal percentage of diverse members at hierarchical level i (associate, mid-level and partner) for the highest performers j, and D_ik = the percentage of diverse members at hierarchical level i for each firm k. The mean distances for each hierarchical level are squared and then averaged among the firms in the sample; finally, their mean for the three levels is computed to obtain the overall distance in diversity profile for the entire firm. We estimated the optimal level of diversity as the average level of diversity at each hierarchical level for the highest-performing (10%) firms in our sample, and computed the distances to this level for the remaining 90% of firms in the sample (e.g. Roth and O’Donnell, 1996). The Euclidean Distance variable captures the extent to which this subsample, excluding the top performers, given the total number of diverse attorneys, has a distribution of attorneys across different organizational levels more or less similar to the optimal profile for the best-performing law firms. A higher score for this measure indicates greater deviance from that ideal measure.

To further interpret the direction in which the 10% top performers differ from the remaining 90% law firms, we have to compare their descriptive statistics, which denote the actually observed deviance from the ideal measure for the firms in our study. As we discuss in the Results section in greater detail and given these descriptive statistics for high performers versus the remaining firms, the sign of the coefficient for the Euclidean Distance variable indicates what happens to firm performance as law firms deviate further from their optimal level (empirically obtained) and reduce their concentration of diverse attorneys at the associate level, decreasing their asymmetry of diversity across different levels. Thus, in contrast to Blau’s index that measures diversity as variety (Harrison and Klein, 2007), this variable provides additional insight about the performance implications of asymmetry of diversity in the specific direction that we observe among law firms, as they move further away from the ideal level of diversity concentration of best performers, which actually have a high concentration of diversity at the associate level.

Control variables

We include numerous variables that may affect the overall profitability of law firms for various reasons, and we removed their effect using these as control variables. We compute these variables for each year. We include USA firm size, the total number of lawyers in the United States in each firm, Leverage, the total number of non-partner lawyers in a firm divided by the total number of partners, and NYC, a binary measure of the existence of an office in New York City. Previous studies have found that these variables are correlated with law firm financial performance (Hitt et al., 2006). We obtained these data either from The American Lawyer, or by reviewing the firm’s website. We also controlled for Domestic and international concentration using the Herfindahl index (Hitt et al., 2001).⁶ We calculate international concentration using the same formula for non-USA office. We also control for time effects with year dummies, keeping 2009 as the base year.

We also created several controls for differences across firms. We controlled for Service diversification (Hitt et al., 2001) by calculating the number of practice areas listed by each firm on its Martindale-Hubbell profile. We reviewed the profile of each firm individually and calculated the number of practices listed for the firm in 2013. We also controlled for education by including the number of associates that each firm in our sample hired from the Top 14 law schools. These are the 14 highest ranked law schools in the US News & World Reports ranking, which is the most important ranking of law schools in the industry. Since the beginning of the rankings, these 14 firms have been at the top of the rankings, in one position or another, and there is generally considered to be a significant gap between the 14th and 15th law schools on the list. We obtained these data from The American Lawyer for the year 2013.

Empirical analysis

To test our hypotheses, we use a pooled ordinary least square (OLS) regression model, which takes advantage of both cross-sectional variations across firms in racial diversity and firm performance as well as time series data. We obtained robust standard errors to compute the significance tests, clustering the observations that belong to the same law firm. We use a one-year lag between our performance variables and the diversity independent variables (Hitt et al., 2001, 2006) to reduce the possible effect of the changes in performance that could affect the firms’ hiring decisions, which would change the composition of the workforce and its racial diversity. Thus, values for the dependent variables refer to 2009–2012, and values for the independent variables, 2008–2011.

We do not control for fixed effects because this estimation requires significant within-panel (i.e. law firm) variation of the variable values to produce consistent and efficient estimates; however, racial diversity within each law firm does not change much during the 4 years under investigation. When the variable of interest does not vary much over time, the fixed-effects estimates would be imprecise (Wooldridge, 2002). Attorneys are generally hired with the expectation of staying with the firm for a significant period of time, such that racial diversity changes very little through 2008 to 2011, which makes fixed-effects regression inappropriate for our short panel.

As a robustness check, we use instrumental variables to explicitly test for endogeneity of diversity across organizations and carry out 2SLS regression estimation.⁷ For this analysis we use the percentage of minority attorneys in three committees (partner review, hiring and diversity committees) that should not be expected to be systematically associated with financial performance, beyond the effect that may occur through the overall relationship between racial diversity for the overall firm and financial performance. Because these committees do not affect performance, and similarly performance does not affect these committees, they should be reasonably exogenous for our analysis of the overall firm diversity–performance relationship. Similar results were obtained when we use the three instruments separately, although we use all of them at once to test for their validity through the Sargan test.

Results

Tables 1 through 3 contain descriptive statistics and zero-order correlations for the variables in our study, respectively, including the measures of racial diversity at the different hierarchical levels. Although approximately 36% of the population belonged to a racial or ethnic minority group according to the 2010 US Census data, we could see in Table 1 that law firms are overwhelmingly and disproportionately majority white. Also, no firms in our sample were ‘minority-majority’ or minority-owned. Furthermore, the proportion of diverse attorneys to all attorneys at each level gradually decreases moving up the hierarchy, going from 21% of minority attorneys at the associate level to only 6% diversity among law firm partners. Thus, these descriptive statistics show that racial diversity is particularly pronounced at the lowest organizational level.

Table 1.

Diversity across levels.

Characteristics	All firms	Top 10%	Bottom 90%	Minimum	Maximum
Diverse partners	9.88	9.75	10.88	0	88
Total partners	163.42 (6.05%)	130.35 (7.48%)	173.23 (6.28%)	23	878
Diverse mid-levels	11.61	5.84	12.91	0	49
Total mid-levels	125.62 (9.24%)	59.26 (9.85%)	135.92 (9.50%)	1	427
Diverse associates	53.93	109.85	48.38	4	274
Total associates	254.93 (21.15%)	422.95 (25.97%)	237.31 (20.39%)	58	1,167
Diverse attorneys in law firm	81.08	125.44	72.17	13	362
Total attorneys in law firm	543.97 (14.90%)	612.56 (20.48%)	546.46 (13.21%)	163	1,860

Mean number of attorneys at each organizational level. The percentage of diverse attorneys at that level is in parentheses. N = 444.

Table 2.

Descriptive statistics.

	Variables	Mean	SD	Min	Max
1	Profits per partner (log)	13.77	.53	12.85	15.42
2	Firm size (log)	6.27	.51	5.12	7.68
3	Partner leverage	3.00	1.43	.76	12.44
4	NYC	.71	.46	.00	1.00
5	Domestic concentration	.34	.22	.04	1.00
6	International concentration	.73	.34	.03	1.00
7	Practice areas (log)	91.30	76.79	1.00	868
8	Education	13.44	20.88	.00	106
9	Asymmetry of diversity	.49	.15	.21	.96
10	Euclidean distance	.05	.05	.00	.24
11	Diverse attorneys	.24	.08	.07	.43
12	Diverse partners	.11	.06	.00	.33
13	Diverse mid-levels	.17	.08	.00	.58
14	Diverse associates	.34	.09	.08	.68

N = 444.

Table 3 shows positive correlations between firm performance and racial diversity across the different levels of the organization. Firm profits are positively correlated with associate racial diversity (r = .49, p < .05), partner racial diversity (r = .40, p < .05), and mid-level racial diversity (r = .20, p < .05). All control variables were also associated with firm performance as expected. More specifically, profits per partner were greater for those law firms that were larger, had greater leverage, had an office in New York, were more focused in the domestic market and more diversified in the international market, had greater specialization in a few practice areas, and employed more attorneys from top law schools.

Table 3.

Correlations.

	Variables	1	2	3	4	5	6	7	8	9	10	11	12	13
1	Profits per partner (log)
2	Firm size (log)	.25
3	Partner leverage	.43	.26
4	NYC	.47	.38	.39
5	Domestic concentration	.55	−.31	.03	.13
6	International concentration	−.52	−.55	−.25	−.35	−.06
7	Practice areas (log)	−.17	.19	−.02	−.01	−.20	.01
8	Education	.66	.45	.26	.35	.35	−.61	−.15
9	Asymmetry of diversity	.55	−.03	.18	.30	.46	−.30	−.21	.56
10	Euclidean distance	−.45	−.06	−.08	−.31	−.31	.26	.20	−.44	−.79
11	Diverse attorneys	.66	.17	.36	.41	.30	−.42	−.26	.55	.47	−.41
12	Diverse associates	.49	.11	.30	.32	.16	−.27	−.21	.38	.42	−.45	.89
13	Diverse partners	.40	.17	.24	.23	.15	−.30	−.15	.31	.10	−.07	.70	.52
14	Diverse mid-levels	.20	.13	.15	.19	−.03	−.17	−.10	.15	−.24	.10	.52	.40	.35

All correlations of .10 or greater are significant at the 95% confidence level. All independent variables are lagged by 1 year. N = 444.

Table 4 presents the results of our regression analyses, where we present the tests of our two hypotheses. We show our results in a hierarchical fashion to better depict the variance explained by the different sets of predictor variables. Hypothesis 1 predicts that racial diversity for the entire organization will be positively associated with firm performance. In Model 2, the coefficient of diverse attorneys for the entire organization is positive and highly significant (β = 1.626, p < .01), thus supporting Hypothesis 1. Law firms with greater overall racial diversity have greater financial performance.

Table 4.

Effect of racial diversity on firm performance (log of PPP).

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Firm size (log)	.102*	.146**	.172**	.148**	.169**	.159**	.169**	.067
Leverage	.080**	.061**	.061**	.069**	.073**	.067**	.064**	.062**
NYC	.160**	.099**	.084*	.119**	.100**	.106**	.093**	.076*
Domestic concentration	1.076**	1.053**	1.004**	.982**	1.020**	1.053**	1.024**	.668**
International concentration	−.312**	−.257**	−.242**	−.260**	−.264**	−.284**	−.259**	−.211**
Practice areas (log)	−.077**	−.046**	−.038*	−.052**	−.051**	−.053**	−.045**	−.003
Education	.006**	.003**	.002^†	.003**	.003*	.003*	.002^†	.007**
Diverse attorneys		1.626**	1.514**					1.647**
Asymmetry of diversity			.444**	.658**	.840**	.406**	.650**
Diverse partners				1.223**			.687*
Diverse mid-levels					.823**		.384^†
Diverse associates						.976**	.577**
Euclidean distance								−.871**
Adjusted R²	.709	.742	.749	.734	.733	.740	.744	.700
F	128.62	143.15	149.51	144.66	126.59	141.99	129.96	91.03
N	444	444	444	444	444	444	444	403

Coefficients for year dummies not shown. All independent variables are lagged by one year.

†

p < .10, * p < .05, ** p < .01.

Our second hypothesis suggests that greater concentration of racial diversity at the higher organizational level (i.e. partners) will have a more pronounced effect on firm financial performance. We test our hypothesis in three steps. First, we investigate in Model 3 if firms that have the same level of racial diversity across organizational levels have greater (or lower) financial performance than those that show differences across levels, that is, asymmetry of diversity. In Model 3, we can see that asymmetry of diversity was positively related to firm performance (β = .444, p < .01). Thus, firms with more asymmetric levels of racial diversity have greater financial performance. However, this variable does not reflect the specific hierarchical level at which racial diversity has greater effect on firm performance.

Second, we investigate the effect of diversity at each hierarchical level separately. To do so, we replace the diverse attorney variable with measures of diversity at the three individual hierarchical levels, first separately in Models 4 through 6, and then all together in Model 7. In Models 4–6, the coefficients for diversity at each specific level are always positive and significant; hence, diversity matters throughout the entire organization and not only at specific levels. To test at which level racial diversity may have a greater performance effect, we include all three measures of diversity in Model 7. We found that there was no significant difference among them, though diversity at mid-level becomes only marginally significant. This marginal effect of diversity at mid-level seems to be owing to its collinearity with the other two measures, given that it is significant when introduced alone in Model 5. In sum, there is no support for our Hypothesis 2 that diversity at the partner level should have a greater effect on performance than at other levels within the organization. Yet, the actual degree of diversity observed across the three hierarchical levels is not similar across the law firms in our sample.

To further explore the type of asymmetry in racial diversity that may be associated with greater performance, we also examine the specific distribution of diversity of the best performers in the industry. For this analysis, we computed the Euclidean distance variable as explained above, which gauges the extent to which firms deviate from the distribution of racial diversity that we observe in the 10% of firms with higher performance in our sample. As shown in the descriptive statistics in Table 1 above, the top performers not only have greater racial diversity across all hierarchical levels (20.48% v 13.21%), but their diversity is most pronounced at the associate level (25.97% v 20.39%). The results from Model 8 indicate that as firms move away from the high concentration of racial diversity at the lowest level, their performance tends to decrease systematically, as shown by the negative and highly significant Euclidean distance coefficient (β = −.871, p < .01). Our results actually go against our Hypothesis 2 in that, even though greater asymmetry is associated with greater performance, the firms with a greater percentage of their racially diverse attorneys at the associate level are the ones that have greater performance.

Additional analyses

The results so far provide general support for the business case for racial diversity and the positive effect of racial diversity at any organizational level on the financial performance of law firms, though we did not find evidence for the greater effect of diversity at the partner level that we initially expected. We further investigated the robustness of our conclusions with additional analyses. First, we replicated our regressions using firm gross revenue (in log form) as an alternative dependent variable, because in our theoretical development we argued that firms with greater racial diversity should be able to attract more business. This effect should be particularly observable at the partner level. The results, which we show in Table 5, are very similar in terms of sign and significance of the coefficients to our results reported earlier. Greater racial diversity for the entire organization and at any hierarchical level is positively associated with firm performance. Diversity at the partner level also has somewhat greater effect on generating revenue than at the associate level (.388 for associates versus .635 for partners), but the difference between them is not statistically significant. Finally, as firms move away from the high concentration of diversity at the associate level that we observe in the best-performing firms, their gross revenue actually decreases (β = −1.183, p < .01; Model 16).

Table 5.

Effect of racial diversity on firm performance (log of gross revenue).

Variables	Model 9	Model 10	Model 11	Model 12	Model 13	Model 14	Model 15	Model 16
Firm size (log)	.910**	.948**	.971**	.950**	.975**	.959**	.975**	.909**
Leverage	.007	−.009	−.010	−.002	.000	−.005	−.007	.001
NYC	.159**	.107**	.093**	.123**	.099**	.112**	.093**	.073**
Domestic concentration	.480**	.460**	.417**	.397**	.433**	.461**	.431**	.208**
International concentration	−.522**	−.474**	−.461**	−.476**	−.475**	−.498**	−.468**	−.469**
Practice areas (log)	−.048**	−.022	−.015	−.026^†	−.023	−.027^†	−.018	.007
Education	.004**	.002**	.001	.002**	.001	.002*	.001	.004**
Diverse attorneys		1.407**	1.310**					1.289**
Asymmetry of diversity			.383**	.571**	.790**	.347**	.665**
Diverse partners				1.107**			.635*
Diverse mid-levels					.904**		.579**
Diverse associates						.868**	.388*
Euclidean distance								−1.183**
Adjusted R²	.882	.897	.900	.894	.896	.896	.900	.908
F	543.12	568.51	539.86	601.02	412.8	529.95	443.5	434.22
N	444	444	444	444	444	444	444	403

Coefficients for year dummies not shown. All independent variables are lagged one year.

†

p < .10, * p < .05, ** p < .01.

A possible concern is that racial diversity and firm performance may be endogenously determined. For instance, reverse causality could still be biasing our results, despite lagging the independent variables one year. If this is the case, racial diversity may not affect performance, but financial performance may predict the law firms’ decisions regarding hiring more or fewer diverse attorneys. To explore this potential concern, we use instrumental variables in 2SLS regression analysis and explicitly test for endogeneity (see Table 6).

Table 6.

Effect of racial diversity on firm performance – 2SLS regression analysis.

Variables	Model 17	Model 18	Model 19
	First stage	Second stage	Second stage
	Diverse Attorneys	Log PPP	Log gross revenue
Firm size (log)	−.046**	.122*	.953**
Partner leverage	.022**	.105**	.017
NYC	.019*	.095**	.053
Domestic concentration	−.026	.811**	.183^†
International concentration	−.028^†	−.118*	−.446**
Practice number (log)	−.012**	.000	.005
Education	.001**	.006**	.001
Asymmetry of diversity	.086**	.281*	.255*
Diverse attorneys	–	1.275*	1.944**
Instruments
Diverse diversity committee	.065**	–	–
Diverse hiring committee	.142**	–	–
Diverse partner review committee	.142**	–	–
R ²	.567	.797	.917
Test of excluded instrument	19.06**	–	–
Over-identification test (Sargan)	–	4.58	.91
Endogeneity test (Hausman)	–	.79	1.63

Coefficients for year dummies not shown. All independent variables are lagged one year.

†

p < .10, * p < .05, ** p < .01. N = 324.

To run the 2SLS analysis, we need instruments that should be associated with the firms’ racial diversity through the organization and at each hierarchical level, but not with their actual performance. Law firms often have several management committees, including a hiring committee, a partner review committee and a diversity committee. We use as exogenous variables the percentage of diverse members in each of these committees, which was available for most of the firms in the sample. Evidently, the percentage of diverse members of these committees should increase as the firms become more diverse. However, there is no theoretical reason why the percentage of diverse members in these specific committees should be directly correlated with firm financial performance, because these committees are created to manage certain aspects of the firm completely unrelated to firm financial performance. If there is a correlation between financial performance and racial diversity in these committees, it would have to happen through the diversity–performance relationship at the firm level, but we should not expect any direct correlation between firm performance and the racial composition of these committees. In other words, an increase in firm performance should not directly increase diversity of those committees, apart from the overall performance–diversity correlation for the entire organization that we are trying to assess. Because we use more instruments than exogenous variables, we can test the validity of our instruments (Semadeni et al., 2014).

Table 6 shows the results for the first and second stages of our 2SLS analysis and the key tests. The first column presents the first-stage regression results for diverse attorneys in the entire organization as an endogenous variable for both models reported later. In Model 17, the significant test of excluded instruments (19.06, p < .01) shows that the three instruments are positively correlated with diverse attorneys and thus may be used as predictors. We also use Sargan’s over-identification test to determine whether the three instruments are indeed exogenous and, thus, not correlated with the error term (Semadeni et al., 2014). The non-significant result for this test (4.58 and .91, p > .10) from the second-stage regressions confirms that the three instruments are valid. Furthermore, the non-significant result for the Hausman endogeneity test (.79, p > .10) suggests that the diverse attorneys variable is not endogenous. Consequently, the OLS results that we previously reported do not seem to be biased by endogeneity. In fact, OLS estimation is preferable to 2SLS when there is no presence of endogeneity, owing to its greater efficiency (Wooldridge, 2002),

Despite rejecting the threat of endogeneity using good instruments, we report for the sake of completeness the second-stage results with the predicted values obtained from the first stage in Models 18 and 19 for our two dependent variables based on profits and revenues, respectively. The two models provide very similar results for the coefficients for diverse attorneys, which are significant and positive in both 2SLS models (β = 1.275, p < .05, Model 18; β = 1.944, p < .01, Model 19). Whereas our short panel does not allow the use of fixed effects and we cannot definitively rule out that unobservable variables could be driving the results, our 2SLS analysis gives greater confidence in the validity of our conclusions about the positive effect of racial diversity on financial performance.

Discussion

While extremely popular among politicians and managers, the business case for increasing racial diversity has been criticized for providing an insufficient basis, or even undermining, calls for greater racial diversity based on equal opportunity (Dickens, 1999; Noon, 2007). Here, our results provide strong empirical support for the idea that racial diversity can indeed be a strategic resource of a firm that improves profitability, aside from social and ethical considerations. However, in contrast to our initial predictions, we also find that racial diversity is not most valuable to law firms when it is concentrated at the partner level. In fact, our results point in the opposite direction to the extent that we observe the best-performing firms having greater concentration of their diversity at the associate level. Overall, the results for the positive effects of racial diversity are robust, using different dependent variables (log of PPP and log of gross revenue), even after checking for potential endogeneity, though not always as expected.

We contribute to the management literature through our novel analysis of racial diversity at different organizational levels within law firms. Our article responds to the call of researchers to better explore the contingencies of diversity effects (Joshi and Roh, 2009; van Knippenberg and Schippers, 2007), specifically across organizational levels. The results confirm that racial diversity influences firm financial performance, not only for the entire organization, but also across the differing hierarchical levels of the organization. The positive effect of racial diversity was clearly observable for all three hierarchical levels in law firms, that is, associates, mid-level and partners, and in any of them racial diversity can indeed be a strategic resource for a firm (Andrevski et al., 2014; Richard et al., 2003, 2007).

Contrary to our second hypothesis, however, the benefits of racial diversity are not greater at higher organizational levels. In fact, we find that the firms with greater percentages of their racial diversity concentrated at the associate level actually have higher financial performance. Though we do find a positive effect of racial diversity at the three hierarchical levels inside law firms, the top-decile-performing firms show substantially greater concentration of their racial diversity at the bottom of the organization (86% of diverse attorneys are associates). In contrast, the remaining firms have much more balanced distribution of diversity across all levels (only 65% their diverse attorneys are associates), and the more the asymmetry is further reduced, the lower their performance. We would like to elaborate on this important and unexpected finding.

One possible explanation is that firms may be able to increase profits by decreasing personnel costs of ethnic minorities, essentially paying minority associates less than their white counterparts or having harder promotion guidelines for their ethnic minorities. We believe that this is very unlikely to be the case, particularly for law firms. This would be considered illegal racial discrimination, and firms would subject themselves to civil lawsuits and possible federal criminal prosecutions if they attempted to pay less to their racial minority employees. Similarly, law firms can hardly sustain a record of being tougher on the promotion of racial minorities, especially among lawyers, who have the training and means to legally fight this potential sort of bias. In fact, law firms generally follow a ‘lockstep’ compensation structure for associates; that is, associates are paid based on their years of seniority, not individual factors. In general, all associates at the firm graduating from law school the same year will be paid the same amount, irrespective of race, gender and other possible individual differences that may be regarded as discriminatory.

A second explanation is that racial diversity may bring more value to the organization at lower levels – exactly the opposite to our initial expectations. For instance, despite the clear benefits of diversity, racial identity may somewhat dilute through time as minority partners could become more similar to other majority partners owing to identifying more strongly over time with their firms and profession. In fact, tenure has been shown to be associated with higher strength of organizational identification (Bartel, 2001; Mael and Ashforth, 1992; Schneider et al., 1971), and partners (and mid-levels) will have much longer tenure with their firms and the legal profession. Whereas associates are relatively young, inexperienced attorneys, mid-level attorneys and partners have almost uniformly spent many years in the profession and at a particular firm. They will generally not be promoted until 8 or 10 years after graduation from law school, and are almost never promoted before spending several years at the target firm.

Instead, we suggest a third interpretation for our unexpected results. Keeping in mind that racial diversity has positive performance effects across the three hierarchical levels in law firms, high performers may be the first organizations to realize and act upon the benefits of increasing racial diversity, which in law firms starts from the bottom. In other words, top-performing law firms may be actively hiring more diverse associates in recent years, as most new attorneys enter the organization at that level. This explanation is consistent with the greater relative number of associates for the top 10 percent performers in the industry, as shown in Table 1 (20% for high performers versus 13% for the remaining firms). If the best performers are actively engaged in hiring more diverse associates, this practice may expand to the rest of the industry and eventually also permeate to higher organizational levels, as these associates gain tenure. Thus, racial diversity is not more valuable at the top as we expected, but it may not be more valuable at the bottom either, considering that the OLS regression coefficients for associates were not greater than for partners. In this sense, the asymmetry of diversity variable may essentially reveal that growing racial diversity starts from the bottom in law firms, such that the greater number of associates of the top performers reflects that they are indeed early adopters in the value-generating decision to increase their diversity.

Implications for research and practice

Our results have clear implications for research on diversity as well as managerial practice, particularly regarding the positive performance implications of racial diversity and the business case for diversity. One of the main criticisms of the business case is that its arguments are contingent and variable (Dickens, 2000). That is, the business case ‘will have relevance for some organizations and not others, it is likely to change over time, and it might benefit some disadvantaged groups but not all’ (Noon, 2007: 778). Accordingly, organizational self-interest may sometimes run counter to, rather than supporting, increased racial inclusion. However, in our study we found strong support for the positive performance implications of racial diversity across all levels in law firms, notwithstanding any considerations about social justice that may also be valid.

Some law firms, particularly underperformers, could be making a suboptimal decision when it comes to recruiting and managing diverse human resources – ‘wasting resources,’ in the words of Dickens (2000). This effect has been documented in other settings (e.g. UK Davies Report, 2011; see Vinnicombe et al., 2018). To the extent that changes in human resources are slow and the potential benefits from increasing racial diversity are realizable years into the future, managers might neglect these benefits in favor of short-term financial targets (Noon, 2007). Our clear results in favor of racial diversity may be useful for those who want to rebuild the workforce to be more in line with the large and growing degree of racial diversity in our society. Law firms in the US seem to be behaving sub-optimally, as the penetration of racial minorities in their organizations (approximately 14.9% in our sample) is well below the level of diversity that exists in their society (approximately 37.9% of the US population in 2014, according to the US Census⁸).

Our results also have implications for promotion decisions. Previous research has shown that law firms often promote attorneys with a certain type of ‘cultural capital,’ irrespective of aptitude, because it signals quality (Ashley and Empson, 2013). As a result, ethnic and racial minorities will only be promoted if they are culturally similar to the majority group. If firms truly recognize the benefits of diversity, they should promote larger numbers of minority attorneys to higher levels, and those firms would secure (at least) similar benefits from this diversity at the partner level, for which we found clear evidence in our analyses. However, it is not entirely clear why the most profitable firms indeed have greater concentrations of their minorities at the associate level, and future research may be necessary to explore this issue further.

Limitations and future research

As with all studies, ours faces several limitations. First, our sample is confined to one industry, the legal services industry in the United States, which limits its generalizability to other settings. This limitation is partly a result of our research design. Because theories predicting the effect of diversity on performance generally look at group dynamics, it is more appropriate to study industries in which group dynamics have greater potential to influence financial performance, either positively or negatively. Thus, though our results should apply to other large professional service firms in other countries, such as accounting and consulting firms, they may overstate the effects of racial diversity on financial performance in other industries where human capital and group dynamics may be somewhat less important. Future research should attempt to clarify the consequences of racial diversity in other contexts.

Second, though we argue that asymmetry of diversity will be important for law firm performance, it could be argued that the effect of having ethnic minority attorneys at senior levels and all (or mostly) white associates would be different than in a firm where the partners are white and all (or most) of the associates are ethnic minority. Our theory does not distinguish between these two possibilities, though we cannot empirically test this alternative supposition because all of the firms in our sample are majority-white at the partner and associate levels. A possible extension of our research would be to examine this possible effect in another cultural context that would allow differentiation between the two.

Finally, we only studied the performance implications of racial diversity. However, our theoretical arguments could be applicable to other types of demographic diversity, such as gender, age and experiential or cognitive differences. We fully agree that future research should explore how these other types of diversity, both throughout the organization and across organizational levels, may affect firm financial performance. Moreover, racial issues are controversial and highly sensitive, and they can be studied from a variety of perspectives, ranging from equality and social justice to performance and market valuation.

Despite these limitations, our article provides strong and novel evidence about the positive effect of racial diversity on the financial performance in law firms. Most critically, we show that this effect can be observed across different organizational levels. Though we expected that the benefits of racial diversity should be particularly important for partners in law firms, we found that racial diversity is similarly important across all three levels, even though firms with higher performance actually have greater concentrations of racial diversity at the lower organizational level.

Footnotes

Funding

This research was partially supported by the Spanish Ministry of Economy and Competitiveness (grant number: ECO2012-33081).

Notes

ORCID iD

Stephen Smulowitz

Stephen Smulowitz is a research fellow at the IMD Global Board Center in Lausanne, Switzerland. He is a graduate of the University of Pennsylvania Law School, and received his PhD from IESE Business School in Barcelona, Spain. His research interests include Corporate Governance, risk, executive compensation incentives and community embeddedness. [ Email: ssmulowi@gmail.com]

Manuel Becerra is Professor of Strategic Management at the University of Queensland Business School, Brisbane, Australia. His research investigates issues of trust, risk and corporate strategy, and it has been published in journals such as Academy of Management Journal, Academy of Management Review, Journal of Finance, Journal of Management, Journal of Management Studies, Organization Science and Strategic Management Journal. He is also the author of the book Theory of the Firm for Strategic Management. [ Email: m.becerra@business.uq.edu.au]

Margarita Mayo is Professor of Organisational Behaviour at IE Business School, Spain, and a Fulbright Alumni of Harvard University. Her research interests revolve around authentic leadership, diversity, change management, and identity processes in organizations. Her work has been published in Human Relations, Journal of Applied Psychology, Academy of Management Journal, Academy of Management Learning & Education, Organizational Research Methods, Leadership Quarterly, Journal of Vocational Behaviour, Human Resource Management and Harvard Business Review. She is also the author of the book Yours Truly: Staying Authentic in Leadership and Life (Bloomsbury, 2018). Margarita has been recognized as finalist for the 2017 Thinker50 Leadership Award. [ Email: margarita.mayo@ie.edu]

References

Ahern

Dittmar

(2012) The changing of the boards: The impact on firm valuation of mandated female board representation. The Quarterly Journal of Economics 127(1): 137–197.

Aldrich

Fiol

(1994) Fools rush in? The institutional context of industry creation. Academy of Management Review 19(4): 645–670.

Ancona

Caldwell

(1992) Bridging the boundary: External activity and performance in organizational teams. Administrative Science Quarterly 37(4): 634–665.

Andrevski

Richard

Shaw

Ferrier

(2014) Racial diversity and firm performance: The mediating role of competitive intensity. Journal of Management 40(3): 820–844.

Ashley

Empson

(2013) Differentiation and discrimination: Understanding social class and social exclusion in leading law firms. Human Relations 66(2): 219–244.

Bantel

Jackson

(1989) Top management and innovations in banking: Does the composition of the top team make a difference? Strategic Management Journal 10(S1): 107–124.

Barker

Monks

Buckley

(1999) The role of mentoring in the career progression of chartered accountants. The British Accounting Review 31(3): 297–312.

Barney

(1991) Firm resources and sustained competitive advantage. Journal of Management 17(1): 99–120.

Bartel

(2001) Social comparisons in boundary-spanning work: Effects of community outreach on members’ organizational identity and identification. Administrative Science Quarterly 46(3): 379–413.

10.

Berrone

Gelabert

Massa-Saluzzo

Rousseau

(2016) Understanding community dynamics in the study of grand challenges: How nonprofits, institutional actors, and the community fabric interact to influence income inequality. Academy of Management Journal 59(6): 1940–1964.

11.

Blau

(1977) Inequality and Heterogeneity: A Primitive Theory of Social Structure. New York: Free Press.

12.

Brewer

(1979) In-group bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin 86(2): 307–324.

13.

Brewer

Weber

(1994) Self-evaluation effects of interpersonal versus intergroup social comparison. Journal of Personality and Social Psychology 66(2): 268–275.

14.

Brodbeck

Guillaume

Lee

(2011) Ethnic diversity as a multilevel construct: The combined effects of dissimilarity, group diversity, and societal status on learning performance in work groups. Journal of Cross-Cultural Psychology 42(7): 1198–1218.

15.

Byrne

(1971) The Attraction Paradigm. New York: Academic Press.

16.

Certo

Lester

Dalton

(2006) Top management teams, strategy and financial performance: A meta-analytic examination. Journal of Management Studies 43(4): 813–839.

17.

Cox

Blake

(1991) Managing cultural diversity: Implications for organizational competitiveness. The Executive 5(3): 45–56.

18.

De Dreu

West

(2001) Minority dissent and team innovation: The importance of participation in decision making. Journal of Applied Psychology 86(6): 1191–1201.

19.

Dickens

(1999) Beyond the business case: A three-pronged approach to equality action. Human Resource Management Journal 9(1): 9–19.

20.

Dickens

(2000) Still wasting resources? Equality in employment. Personnel Management 3: 137–169.

21.

Elfenbein

O’Reilly

(2007) Fitting in: The effects of relational demography and person-culture fit on group process and performance. Group & Organization Management 32(1): 109–142.

22.

Harrison

Klein

(2007) What’s the difference? Diversity constructs as separation, variety, or disparity in organizations. Academy of Management Review 32(4): 1199–1228.

23.

Hitt

Bierman

Shimizu

Kochhar

(2001) Direct and moderating effects of human capital on strategy and performance in professional service firms: A resource-based perspective. Academy of Management Journal 44(1): 13–28.

24.

Hitt

Bierman

Uhlenbruck

Shimizu

(2006) The importance of resources in the internationalization of professional service firms: The good, the bad, and the ugly. Academy of Management Journal 49(6): 1137–1157.

25.

Horwitz

(2007) The effects of team diversity on team outcomes: A meta-analytic review of team demography. Journal of Management 33(6): 987–1015.

26.

Jonsen

Tatli

Özbilgin

Bell

(2013) The tragedy of the uncommons: Reframing workforce diversity. Human Relations 66(2): 271–294.

27.

Joshi

Roh

(2009) The role of context in work team diversity research: A meta-analytic review. Academy of Management Journal 52(3): 599–627.

28.

Joshi

Liao

Roh

(2011) Bridging domains in workplace demography research: A review and reconceptualization. Journal of Management 37(2): 521–552.

29.

Julian

Ofori-Dankwa

(2017) Context matters: Diversity’s short- and long-term effects in fortune’s ‘best companies to work for’. Strategic Management Journal 38(7): 1557–1565.

30.

Kerr

(2011) Economic impacts of immigration: A survey. NBER Working Paper No. 16736. Cambridge, MA: National Bureau of Economic Research.

31.

Kerr

(2013) US high-skilled immigration, innovation, and entrepreneurship: Empirical approaches and evidence. NBER Working Paper No. 19377. Cambridge, MA: National Bureau of Economic Research.

32.

Kochan

Bezrukova

Ely

, et al. (2003) The effects of diversity on business performance: Report of the diversity research network. Human Resource Management 42(1): 3–21.

33.

Levine

Apfelbaum

Bernard

, et al. (2014) Ethnic diversity deflates price bubbles. Proceedings of the National Academy of Sciences 111(52): 18524–18529.

34.

McGinn

Milkman

(2013) Looking up and looking out: Career mobility effects of demographic similarity among professionals. Organization Science 24(4): 1041–1060.

35.

Mael

Ashforth

(1992) Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior 13(2): 103–123.

36.

Mayo

(1999) Capitalizing on a diverse workforce. Ivey Business Journal 64(1): 20–26.

37.

Mayo

Kakarika

Mainemelis

Deuschel

(2017) A metatheoretical framework of diversity in teams. Human Relations 70(8): 911–939.

38.

Miller

del Carmen Triana

(2009) Demographic diversity in the boardroom: Mediators of the board diversity–firm performance relationship. Journal of Management Studies 46(5): 755–786.

39.

Nathan

(2014) The wider economic impacts of high-skilled migrants: A survey of the literature for receiving countries. IZA Journal of Migration 3(1): 4. Available at: https://doi.org/10.1186/2193-9039-3-4.

40.

Nathan

(2015) After Florida: Towards an economics of diversity. European Urban and Regional Studies 22(1): 3–19.

41.

Nathan

(2016) Ethnic diversity, business performance and location: Which firms? Which cities? Environment and Planning A: Economy and Space 48(2): 2462–2483.

42.

Noon

(2007) The fatal flaws of diversity and the business case for ethnic minorities. Work, Employment and Society 21(4): 773–784.

43.

Omi

Winant

(2015) Racial Formation in the United States. New York and Abingdon: Routledge.

44.

Page

(2007) The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies. Princeton, NJ and Woodstock: Princeton University Press.

45.

Pelled

(1996) Demographic diversity, conflict, and work group outcomes: An intervening process theory. Organization Science 7(6): 615–631.

46.

Peri

Sparber

(2011) Highly educated immigrants and native occupational choice. Industrial Relations: A Journal of Economy and Society 50(3): 385–411.

47.

Polzer

Milton

Swann

(2002) Capitalizing on diversity: Interpersonal congruence in small work groups. Administrative Science Quarterly 47(2): 296–324.

48.

Richard

McMillan

Chadwick

Dwyer

(2003) Employing an innovation strategy in racially diverse workforces: Effects on firm performance. Group & Organization Management 28(1): 107–126.

49.

Richard

(2000) Racial diversity, business strategy, and firm performance: A resource-based view. Academy of Management Journal 43(2): 164–177.

50.

Richard

Murthi

Ismail

(2007) The impact of racial diversity on intermediate and long-term performance: The moderating role of environmental context. Strategic Management Journal 28(12): 1213–1233.

51.

Roth

O’Donnell

(1996) Foreign subsidiary compensation strategy: An agency theory perspective. Academy of Management Journal 39(3): 678–703.

52.

Schneider

Hall

Nygren

(1971) Self image and job characteristics as correlates of changing organizational identification. Human Relations 24(5): 397–416.

53.

Semadeni

Withers

Certo

(2014) The perils of endogeneity and instrumental variables in strategy research: Understanding through simulations. Strategic Management Journal 35(7): 1070–1079.

54.

Srikanth

Harvey

Peterson

(2016) A dynamic perspective on diverse teams: Moving from the dual-process model to a dynamic coordination-based model of diverse team performance. The Academy of Management Annals 10(1): 453–493.

55.

Tajfel

Turner

(2004) The social identity theory of intergroup behavior. In: Jost

Sidanius

(eds) Key Readings in Social Psychology: Political Psychology. New York and Hove: Psychology Press, 276–293.

56.

van Dick

van Knippenberg

Hägele

, et al. (2008) Group diversity and group identification: The moderating role of diversity beliefs. Human Relations 61(10): 1463–1492.

57.

van Knippenberg

Schippers

(2007) Work group diversity. Annual Review of Psychology 58: 515–541.

58.

Vinnicombe

Doldor

Sealy

(2018) The female FTSE board report: Busy going nowhere with the female executive pipeline. University of Cranfield, UK.

59.

Wilkins

(2007) Your people. In: Empson

(ed.) Managing the Modern Law Firm: New Challenges, New Perspectives. Oxford: Oxford University Press: 37–63.

60.

Wooldridge

(2002) Introductory Econometrics: A Modern Approach. Mason, OH: South Western Educational Publishing.