Module/Week 6 -- Financial Management

Niels Opstrup is assistant professor in

the Department of Political Science and

Public Management at the University of

Southern Denmark. His research focuses

on the management and organization of

public organizations and, specifi cally, on the

governance of universities and management

of academics. He has also investigated

developments in the backgrounds and

career paths of top civil servants.

E-mail: [email protected]

Anders R. Villadsen is associ-

ate professor at Aarhus University,

Denmark. His research interests include

public management and diversity in

public organizations. He has published

in journals such as Journal of Public

Administrative Research and Theory,

Public Administrative Review, and

International Public Management

Journal.

E-mail: [email protected]

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 291

Public Administration Review,

Vol. 75, Iss. 2, pp. 291–301. © 2014 by

The American Society for Public Administration.

DOI: 10.1111/puar.12310.

Niels Opstrup University of Southern Denmark, Denmark

Anders R. Villadsen Aarhus University, Denmark

Abstract: Recent research has illustrated that demographic diversity infl uences the outcomes of public sector organiza- tions. Most studies have focused on workforce diversity; by comparison, little is known about how managerial diversity aff ects organizational outcomes. Th is article focuses on gender diversity in the top management teams of public organi- zations and its relationship to fi nancial performance. Th eory suggests that management diversity can be a positive asset for organizations, allowing for the use of more diverse knowledge and human skill sets. Results of this study, however, suggest that organizations may only be able to leverage these advantages if they have a supporting management structure. In a longitudinal study of top management teams in Danish municipalities, the authors fi nd that gender diversity in top management teams is associated with higher fi nancial performance, but only in municipalities with a management structure that supports cross-functional team work. Th ese results are interpreted in light of existing theory, and implications are suggested.

Practitioner Points • Little is known about how the composition of the top management team in public organizations matters for

organizational outcomes. • Greater gender diversity has the potential to lead to superior organizational outcomes but does not automati-

cally do so. • To reap the benefi ts of diverse top management teams, the organizational structure must facilitate behavioral

integration between members of the top management team. • Gender diversity in the top management team is related to better fi nancial outcomes when organizational

structures promote integration and discretion of the top management team. • Gender diversity in top management positions has no eff ect in traditional hierarchical organizations.

contracting (Hefetz and Warner 2004, 2012), and performance-information use (Moynihan and Ingraham 2004; Moynihan and Pandey 2010). Th is substantial line of research has provided valuable insights about how managers can aff ect the outcomes of public organizations. Much less research has been devoted to the composition of management teams and the dynamics among individuals occupying manage- rial roles in organizations. Th is is a critical gap in our understanding of public management. Top manage- ment is composed of individuals who apply their knowledge, perspectives, and worldviews to contribute to decision making as well as the overall direction of organizations (Hambrick, Cho, and Chen 1996).

In this article, we explore how gender diversity in top management teams (TMTs) aff ects fi nancial performance in public organizations. Research focusing on private fi rms indicates that managerial gender diversity is related to positive performance outcomes (Auh and Menguc 2006; Carter, Simkins,

Th e Right Mix? Gender Diversity in Top Management Teams and Financial Performance

Recent research has explored how characteris-tics such as tenure (Juenke 2005), race (Pitts 2005), experience (Villadsen 2012), and gender (Meier, O’Toole Jr, and Goerdel 2006) infl u- ence the actions of public managers. Th is research, however, generally fails to allow for the fact that management is rarely an individual endeavor; while a single person may top the hierarchy and serve as the public face of an organization, actual top management is often better described as a team eff ort (Finkelstein and Hambrick 1990). In this article, we move the focus from the individual manager to the top manage- ment team and examine how its composition aff ects organizational outcomes.

A considerable amount of research has investigated how various management practices aff ect public organizations. Th ese practices include leadership (Fernandez 2005; Van Wart 2013), managerial net- working (Meier and O’Toole 2003), organizational strategies (Andrews, Boyne, and Walker 2006),

292 Public Administration Review • March | April 2015

steering of public spending (Serritzlew 2005; Wildavsky and Caiden 2004).

More specifi cally, we study the relationship between TMT gender diversity and fi nancial performance in a longitudinal analysis of 91 Danish municipalities. Our results indicate the importance of organizational structure for accruing the benefi ts of TMT diversity. Only in municipalities where the management team is organized according to an executive board model do higher levels of gender diversity correlate with better fi nancial results (measured in terms of operational results and outperforming the budget in the course of the year). No such eff ect is found for municipalities using a tradi- tional specialized organization of management.

With this study, we make three important contributions. First, we expand the very small number of studies that call attention to the diff erences between male and female public managers. While the Weberian public manager is a genderless bureaucrat, in the real world, men and women bring diff erent backgrounds, experiences, and expectations to the executive fl oors of ministries, agencies, and city halls. Th is study points clearly to the importance of increasing our understanding of how gender diversity aff ects top management decision making and outcomes. Second, we address a call for more research exploring the importance of diversity for organizational outcome (Pitts and Wise 2010). Most diversity research in public organizations has studied workforce diversity using the theory of representative bureaucracy (Kingsley 1944; Park 2013; Roch and Pitts 2012). We add to this knowledge by showing that top manage- ment diversity can have a predictable, signifi cant impact on fi nancial performance. Th ird, we shed light on one of the mechanisms that connects gender diversity to positive outcomes. We suggest that

in order to accrue the benefi ts of diversity, structures must be put in place to facilitate behavioral integration of the team. In this way, our study helps clarify one of the reasons why previous research on diversity yielded mixed results.

Next, we lay out our arguments for a positive relationship between TMT gender diversity and fi nancial performance and set forth our hypotheses. Following this, we describe the

research setting and methods used. After reporting the results of our analyses, we discuss our results with regard to existing research and managerial relevance.

Gender Diversity in Top Management Teams A top management team is the “dominant coalition” responsible for setting an organization’s direction (Cyert and March 1963). Th e TMT is charged with identifying opportunities and problems, gathering and interpreting information, and making decisions that propel the organization forward (Mintzberg 1979). In this process, TMT members must draw on individual as well as collective human and social capital. Each individual brings a diff erent background and perspectives to the table. However, decisions are also shaped in cooperation. While there is a relatively limited number of stud- ies on management team composition and functioning in the public sector, a substantial line of research has investigated TMT and board composition in the private sector and its relationship to

and Simpson 2003; Dwyer, Richard, and Chadwick 2003; Erhardt, Werbel, and Shrader 2003). However, little research has yet explored the outcomes of TMT gender composition in public sector organi- zations (Pitts and Wise 2010), and to our knowledge, no studies have focused on how diversity may aff ect fi nancial performance. We respond to an increasing interest in diversity in public organiza- tions (Park 2013; Pitts 2005; Pitts and Wise 2010) and draw on recent theories about team diversity (for reviews, see Jackson, Joshi, and Erhardt 2003; Van Knippenberg and Schippers 2007) in order to explore how the gender composition of TMTs aff ects fi nancial outcomes in public organizations.

In this article, we argue that TMT gender diversity is a positive asset for public sector organizations, for three reasons. First, draw- ing on the theory of representative bureaucracy, gender diversity is likely to generate a better understanding of the organization’s environment because a collective of diverse team members can relate to many diff erent aspects of increasingly heterogeneous organizational surroundings (Keiser et al. 2002; Park 2013). Second, diversity may inspire more creative and innovative output because it enables organizations to draw on a wider range of social and human capital as well as diff erent cognitive templates (Østergaard, Timmermans, and Kristinsson 2011). Finally, diversity may provide more eff ective problem solving because diverse teams evaluate more alternatives and explore more potential consequences (Carter, Simkins, and Simpson 2003). Further, we draw on theories of group decision-making processes to suggest that male- dominated TMTs may be excessively risk seeking compared with those with a more even gender balance. Based on this, we predict that fi nancial performance is likely to be better in organizations with a gender-integrated TMT.

We propose that the benefi ts of diversity are contingent on organizational factors. Gender diversity in the top management team is likely to be related to superior perform- ance specifi cally when management struc- tures facilitate behavioral integration and cross-gender TMT work. We suggest that the mixed results of previous research were caused by disregarding the role of organiza- tional structure as an important mediating factor. Th is is consistent with research pointing to the impor- tance of the structural context for achieving active representation (Keiser et al. 2002).

Previous studies of the outcomes of diversity and representation in public organizations have predominantly taken education as their point of departure, relying on various measures of student performance (Pitts and Wise 2010). In this article, we focus on the fi nancial performance of multipurpose public organizations. Of course, we recognize that “making money” is not the prime objective of public sector organizations. However, sound fi nancial management is a fundamental prerequisite for eff ective service delivery, particularly in the post–New Public Management period, when many organizations, agencies, and units have gained greater fi nancial autonomy. Th e ability to control fi nances eff ectively is an important factor underlying the work public organizations carry out. Defi cits and budget overruns may impede the eff ective

Gender diversity in the top management team is likely to be related to superior performance specifi cally when management structures facilitate behavioral integration and cross-gender

TMT work.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 293

diverse and complex environments, which may make female leaders an asset (Kent and Moss 1994). Second, women in top management are likely to be highly skilled, as they typically have to overcome signifi cant challenges to break the glass ceiling and get promoted to higher ranks. (Th is point is particularly salient in the context of the present study, as only around 20 percent of municipality top man- agers in Denmark are women.) Th is means that not only are women who attain TMT positions likely to be skilled and knowledgeable, but also they are adept at working the politics of the organiza- tion. Th ird, compared with men, women’s cognitive style tends to emphasize harmony (Krishnan and Park 2005). Th is gender diff er- ence may lead to superior team dynamics when men and women are brought together in continuous interaction, where the distinct strengths of each gender can be leveraged. Further, it can be argued that the degree of gender diversity in top management constitutes a signal to lower hierarchical levels. Female middle managers become more motivated if they see women in the TMT and, consequently, see a viable promotion path (Dezsö and Ross 2012), and female employees in general are likely to feel more accepted and to feel that diff erent gender roles are accepted in the organization (Guy and Newman 2004).

Th e foregoing arguments suggest that there may be positive eff ects of gender-diverse TMTs. Th is has generally been supported in studies of private sector fi rms (for a review, see Jackson, Joshi, and Erhardt 2003). Krishnan and Park (2005) fi nd that TMT gender diversity has a positive eff ect on organizational performance in a study of Fortune 1000 companies. In a study of 127 large U.S. companies, Erhardt, Werbel, and Shrader (2003) report a posi- tive association between board diversity and fi nancial performance indicators. As mentioned, relatively few studies have explored outcomes of diversity in public organizations (Pitts and Wise 2010). Still, representative bureaucracy theory would predict that a repre- sentative TMT will be more attuned to the demands and needs of citizens and, consequently, likely to make better fi nancial decisions (Park 2013).

In this article, we focus specifi cally on how TMT gender diversity is related to the fi nancial performance of public sector organizations. Assessing the performance of public sector organizations is notori- ously diffi cult because such organizations often serve multiple objec- tives and must balance the interests of multiple stakeholders. Unlike research on private sector performance, where fi nancial indicators are the sine qua non, research in public management has tended to focus on service-related performance measures such as student test scores (Andersen and Serritzlew 2007; Meier and O’Toole 2001), composite measures (Andrews, Boyne, and Walker 2006), or self- reported assessments. Diversity research is no exception: as already mentioned, relatively little diversity research is outcome focused, and the few studies that do exist have largely focused on the educational sector and student outcomes (e.g., Johansen 2007; Pitts 2005).

To further understand how TMT dynam- ics may aff ect fi nancial outcomes, it is worth considering how group composition may aff ect risky behavior in groups. Research in economic psychology indicates diff erences in men’s and women’s fi nancial decision mak- ing. Using experimental methods, Powell

organizational performance (Finkelstein and Hambrick 1990; Miller and Triana 2009; Zahra and Pearce 1989). Pitts argues that diversity is a strength when “[t]he ‘whole’ of the group’s eff ort will be greater than the sum of the parts” (2005, 618).

Given that management may be stereotypically thought of as a male task, an important challenge for research has been to investigate how the gender composition of top management matters for the outcomes of organizations. In a comprehensive review of private sector research on diversity, Jackson, Joshi, and Erhardt (2003) fi nd gender to be the most widely studied dimension of diversity, ahead of age and functional background. Research has long been interested in gender segregation in public administration as well (Guy and Newman 2004; Stivers 1995). However, in a systematic review of studies of workforce diversity in 12 core public adminis- tration journals, Pitts and Wise (2010) fi nd that only a few studies have explored the outcomes of diversity. Of these, even fewer have focused on gender. Th is is a critical gap in our understanding of public management.

Th e central premise underlying the hypothesis that gender diver- sity matters to top management teams is that male and female managers diff er, not only in general but specifi cally in terms of the way they perform in their (management) jobs. Research has provided numerous studies of how leadership styles diff er between men and women. Generally, studies fi nd that women engage in leadership behavior that is more participatory, collaborative, and democratic than men’s (Dezsö and Ross 2012; Eagly and Carli 2003). While there are, to our knowledge, no studies focusing on diversity in the top management teams of public sector organi- zations, a small amount of research has illuminated the conse- quences of gender diff erences among public managers. In a study of Texas school districts, Meier, O’Toole, and Goerdel (2006) fi nd gender diff erences in the performance eff ects of managerial networking. Johansen’s (2007) study of the same empirical setting suggests that prospector strategies may be especially eff ective when carried out by women managers. In a study of state agency heads in the United States, Jacobsen, Palus, and Bowling (2010) fi nd that women managers devote less time to internal management and engage in less external networking behavior than their male counterparts.

Top Management Gender Diversity and Financial Performance Building on the assumption that male and female managers exhibit distinct characteristics, there are further reasons to believe that gen- der diversity is related to organizational outcomes. Next we develop the relationship between TMT gender diversity and fi nancial outcomes by drawing on the general diversity literature as well as dedicated fi nancial literature.

Krishnan and Park (2005), drawing on social identity theory, highlight three arguments why TMTs may benefi t from gender integra- tion after a history of favoring men. First, women are more likely than men to be per- ceived as leaders in environments that involve a signifi cant degree of social interaction. Public sector organizations exist in highly

Women are more likely than men to be perceived as leaders in environments that involve a signifi cant degree of social

interaction.

294 Public Administration Review • March | April 2015

TMTs must have a structure allowing integration, which, in turn, can facilitate interaction and cultivate mutual exchange of knowl- edge and opinions. We expect that organizations will be better able to reap the benefi ts of gender diversity when they adopt an organizing approach that allows for higher degrees of integration. Conversely, if a TMT’s responsibilities are disintegrated and distrib- uted across sectorial areas, diversity will be much less important, if at all. Th e benefi t of bringing together members from diff erent groups (in this case, diff erent genders) will simply not be realized because managers will maintain a strong focus on their distinct area of responsibility with little benefi t accrued from cooperation.

Besides integration, team discretion may also aff ect the outcomes of diversity. If the management structure is hierarchical, with a dominating top executive, gender integration is likely to matter less than if the management structure emphasizes teamwork and shared

responsibilities. In the latter situation, in which management discretion resides at the TMT level rather than the individual level, group dynamics matter more for decision quality and outcomes. Management discre- tion has been shown to positively aff ect

organizational effi ciency in local governments (Garrone, Grilli, and Rousseau 2013).

Th is reasoning can also been interpreted in light of representative bureaucracy research. Studies have distinguished between passive and active forms of representation (Park 2013; Wilkins and Keiser 2006). Passive representation may increase the legitimacy of public agencies (Riccucci, Van Ryzin, and Lavena 2014), but active repre- sentation may create more substantial outcomes. Such more active representation, however, may depend on structural conditions of the organization. As argued by Keiser and colleagues, “Within the organization, a [...] necessary condition for representative bureauc- racy is that the bureaucrats in question have discretion to infl uence outcomes” (2002, 558). With regard to overall fi nancial outcomes, discretion is greater when management structures promote real team work and shared responsibilities.

We predict that the organization of management may signifi cantly aff ect the organizational outcomes of TMT gender diversity. When management is structured around team integration and team discre- tion, the benefi t is likely to be higher.

Hypothesis 2: Th e positive relationship between TMT gender diversity and fi nancial performance is moderated by management structure, such that the relationship is stronger when TMT integration and discretion are higher.

Research Setting Danish municipalities (local governments) provide an excellent setting for studying the impact of gender diversity in top manage- ment teams. Th e 98 municipalities are highly comparable, semi- autonomous units having the same tasks and subject to the same regulations (Christensen 1998; Greve 2006). Local government is responsible for the larger part of public service, including primary schooling, child care, elder care, local infrastructure, and social security benefi ts. Th e municipalities levy their own taxes and have considerable autonomy in service provision.1 Government rests with

and Ansic (1997) fi nd that women are less risk seeking than men and that they adopt diff erent strategies in their fi nancial decision making. A similar point is made by Francis et al. (2009) in a study of chief fi nancial offi cers (CFOs) showing that female CFOs are more conservative than male CFOs. Th ese individual-level gender diff erences aggregate through group composition to create system- atic diff erences in organizational-level outcomes. Group behaviors conform to the assumptions, values, and norms that are shared by the majority of the group’s members (Cartwright and Zander 1960). Th is means that if the prevailing view held by the majority favors a more risky decision, the minority group members are likely to conform. Stoner (1968), for instance, fi nds that groups give more weight to risk-related attitudes shared by more members. If men and women tend to hold diff erent attitudes toward risk, this research suggests that TMT gender segregation may lead to fi nancial deci- sions that are either too risky or too safe depending on the majority gender of the group. A gender-balanced team is less susceptible to the weaknesses of such groupthink. In public organizations looking to fi nd the right balance between being fi nan- cially progressive and conservative, a balance of male and female TMT members is likely to be positive.

In total, we base our hypothesis on two lines of argument: fi rst, on general research claiming that gender diversity is positively related to organizational outcomes, and second, on specifi c research pointing to gender diff erences in fi nancial decision making.

Hypothesis 1: Higher TMT gender diversity is positively related to fi nancial performance.

The Moderating Role of Management Structure Th e arguments outlined earlier imply that for a group to obtain the positive outcomes of diversity, its members must be able to interact productively and constructively and feed off each other’s strengths. Previous research has pointed out how structures can facilitate posi- tive outcomes from representation (Meier and Bohte 2001). Th e expectation expressed in hypothesis 1 may therefore be contingent on the organization of the top management team. To release posi- tive TMT dynamics, the TMT must work in a way that facilitates such interaction. Specifi cally, we propose that team integration and team discretion positively moderate the relationship between gender diversity and fi nancial performance.

Upper-echelon theory has argued that the outcome of a TMT’s work depends on its organization and internal dynamics. Hambrick points out that TMTs “often consist of semiautonomous ‘barons,’ each engaging in bilateral relations with the CEO but having little to do with each other and hardly constituting a team” (2007, 336). In light of this, he argues that TMTs with more behavioral integra- tion are likely to perform better. Behavioral integration describes the extent of a team’s collaboration, knowledge exchange, and emphasis on joint decision making (Hambrick 1994). TMTs that are more behaviorally integrated have been shown to be able to embrace ambidexterity to a larger degree and also exhibit higher performance (Lubatkin et al. 2006). Another study fi nds higher TMT behavioral integration to be positively related to decision quality (Carmeli and Schaubroeck 2006). For diversity to aff ect organizational outcomes,

A gender-balanced team is less susceptible to the weaknesses of

such groupthink.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 295

However, the full eff ect of an executive board is presumably only apparent in the pure version of the executive board model, in which the managing directors are formally totally independent of sector interests (Bækgaard 2011, 1068). We explore these diff erences in the organization of management to assess whether the eff ect of manage- ment diversity is contingent on the structure of management.

Methods Data and Analysis To investigate the impact of gender diversity on fi nancial performance in the TMTs of Danish municipalities, we analyze yearly data for the period 2008 to 2012. Th e starting point for the analysis is given by a major structural reform (Strukturreformen) implemented in 2007, which reduced the number of municipalities from 271 to 98 and changed their responsibilities and fi nancing (see Vrangbæk 2010).2

Th e data are compiled from register data and offi cial statistics. Th e composition of municipality top management teams and the municipalities’ administrative structure are coded based on the offi cial local government handbook (Mostrup Kommunal Håndbog), which is issued annually and contains, among other information, the names of the local government CEOs and the other manag- ing directors in each of the 98 municipalities. According to this handbook, the size of the municipality TMTs varied between 1 and 12 members in the analyzed period. In order to make a measure of diversity meaningful, we only include observations with more than two registered municipal managers, eliminating seven municipali- ties. Th erefore, only 91 municipalities feature in the analysis. Data on fi nancial performance is obtained from statements made by the Danish Evaluation Institute for Local Government.3 Furthermore, a number of control variables are integrated from “the yearly munici- pal key indicators” (De Kommunale Nøgletal) published by the Ministry of Economy and the Interior.4

In our analysis, we apply generalized least squares (GLS) regres- sion in order to deal with the potential problems of heteroscedas- ticity attributable to the longitudinal data design (Frees 2004). Furthermore, because each municipality appears multiple times in the analysis, observations may not be independent. Th erefore, we cluster the analysis on the observations (municipalities) and calculate robust standard errors. When analyzing panel data, there is always a choice between random- and fi xed-eff ects specifi cations. Various reasons justify using a random-eff ects specifi cation for this study. First, there is relatively little variation in our independ- ent variable across the years, as changes only happen when there is a replacement in the TMT, rendering within-observation vari- ation modest and fi xed eff ects unfeasible. Further, we conducted a Hausman test, which clearly rejected fi xed eff ects as superior to random eff ects (p < .43). Th erefore, we use random-eff ects estima- tions in all models.

To ensure that multicollinearity is not problematic, we calculated variance infl ation factors (VIFs), which did not reach any levels of concern. Th e mean VIF score is 1.67, and the highest value is 2.45.

To assess whether results are sensitive to autocorrelation, we ran additional models with an AR1 autocorrelation structure. We obtained substantially very similar results (available from the authors upon request).

the elected municipal council, and executive power is shared among a compulsory fi nancial committee, various standing committees, and the mayor. Th e mayor chairs the council and the fi nancial com- mittee meetings and is formally the “daily leader of the municipal administration” (Mouritzen and Svara 2002). Most mayors come from one of the two biggest political parties, the Social Democrats or the Liberals (Berg and Kjær 2005). Th e other members of the council are “part-time politicians” who usually also hold a day job.

While municipal councilors are formally in charge of providing direction and overall decision making, actual operations are left to the administrative organization. Top management consists of career bureaucrats who are promoted from the ranks of civil servants. In practice, municipal administrative top management teams consist of a chief executive offi cer (CEO or kommunaldirektøren) and a number of managing directors. Th e formal hierarchical relationship within the TMT is somewhat ambiguous because of the sharing of executive power among the fi nancial committee, the standing committees, and the mayor (Ejersbo, Hansen, and Mouritzen 1998, 109). However, the CEO is usually regarded as the person in charge of the administrative part of the municipal organization (Bækgaard 2011, 1066).

Two basic yet distinct forms of organization of the top management are used in Danish municipalities (Bækgaard 2011). Traditionally, most Danish municipalities have organized their administration according to a sector-based model, in which the municipality’s tasks are divided among a number of administrative departments, each responsible for a defi ned service area (which typically mirrors the responsibilities of a specifi c standing committee). Each depart- ment has a department head. However, since the beginning of the 1990s, many municipalities have reorganized (Villadsen 2012). Emphasizing the need for coordination at the municipal apex and drawing inspiration from private fi rms, some municipalities have organized the top management as executive boards with cross- departmental responsibilities. Th e main diff erences between the two types of models lie in the roles and the focus of the managing directors. In the sector-based model, the department heads’ primary responsibility is the running of their department. Th is type of organizing is based on traditional bureaucratic thinking, in which each top manager is responsible and accountable for a particular line of work. Th e top managers of each department together form a top management group, but responsibilities are relatively clearly divided. Consequently, little TMT integration is needed, and overall mana- gerial discretion is likely to rest with the top executive.

In the executive board model, the managing directors attend to intraorganizational coordination, strategic decision making, and the economic steering of the municipality. While TMTs organized in this way have a CEO who is ultimately in charge, they are a collective in which decision making is a shared task across the various functional areas of the organization. Th is type of organizing is more time-con- suming, but it allows the team to benefi t from members’ collective human and social capital and provides a higher degree of functional integration among diff erent lines of work. TMTs organized this way will be more integrated and have more team-based discretion.

Some municipalities apply a mixed model, with an executive and one or more of the managing directors also being department heads.

296 Public Administration Review • March | April 2015

each municipality’s TMT for each year. Next, we used the Blau index of dissimilarity to create the variable gender diversity. The Blau index is calculated by the formula 1 – ∑pi2, where p equals the proportion of each group of interest (i.e., male and female) and i represents the number of groups (in this case, two). For gender diversity, the index ranges from 0 to 0.5. A value of 0 represents complete homogeneity (thus, a TMT consisting of either all male or all female directors); a value of 0.5 represents maximum gender diversity (an equal number of men and woman in the TMT). Both of these extreme values are present in our data set. In the period under analysis, the municipalities’ average index score was 0.257, with a standard deviation of 0.196. In order to better refl ect the TMT’s impact on the municipality’s fi nancial performance, the independent variable is lagged by one year.

Moderating variable. The municipalities’ administrative structure in the given year is determined based on information from the local government handbook. Inspired by Bækgaard (2011), we distinguish between sector-based models, mixed models, and executive board models. A municipality is coded as having a sector-based model when the local government handbook lists a number of administrative departments with defi ned service areas (e.g., Department of Children and Youth, Department of Elderly and Disabled, etc.). When managing directors are presented as heading a defi ned service area but no administrative departments are listed explicitly, the municipality is registered as having a mixed model. Finally, when managing directors are presented without any associated area of responsibility and no administrative departments are listed, the municipality is coded as using an executive board model. This is a crude measure that does not necessarily capture the actual daily functioning of each TMT, but it does refl ect how the municipalities choose to present themselves in the annual local government handbook.

Because we are particularly interested in the use of executive board models, we create a dummy variable where executive board models are coded as 1 and the other two board types are coded as 0. Next, we create an interaction term between the measure of TMT gender diversity and the dummy variable for executive board models in order to test its moderating eff ect. In the analyzed period, 17.4 percent of the municipalities were registered as using an executive board model. By comparison, Bækgaard (2011, 1069–70) fi nds, based on short interviews with the CEOs, that 16.0 percent used an executive board model in 2008.

Control variables. In order to better isolate the impact of gender diversity in top management teams on the municipalities’ fi nancial performance, we include a number of control variables in the analy- sis. First, it is likely that structural characteristics both infl uence the tendency to employ woman in top management positions and impact the fi nancial performance of the municipality. To account for this, we control for the municipalities’ population size, propor- tion of inhabitants living in an urban area (urbanization), tax base (a measure of municipal wealth), socioeconomic conditions (using an indicator calculated by the Ministry of the Interior), and long- term debts. All variables are time varying and lagged by a year to better refl ect causality. Second, we include a dummy variable indi- cating whether the municipality resulted from the 2007 structural reform, which led to an “oversupply” of managing directors in the newly merged municipalities, potentially aff ecting the composition

Measurement Dependent variables. We operationalize fi nancial performance using two measures to account for the fact that “good fi nancial performance” may mean different things in different organizations. First, we use accounted operating result because it refl ects the organization’s ability to generate suffi cient income and control expenditures. Second, we use budget overrun because it refl ects an organization’s ability to follow a fi nancial plan laid out by elected politicians. Each of these variables is further specifi ed next.

Th e fi rst dependent variable is the accounted operating result in the given year. Th is is a widely used indicator of the fi nancial perform- ance of Danish municipalities (Hansen, Houlberg, and Pedersen 2014). Th e operating result is measured as the municipality’s accounted operating revenues (tax revenue and block grants from the national government) per citizen minus its accounted operating expenses and net interest payments per citizen—all in thousands of Danish kroner. As stated earlier, these fi gures are obtained from yearly statements made by the Danish Evaluation Institute for Local Government. Between 2008 and 2012, the average accounted operating result of the municipalities that we analyzed was a 1,527 Danish kroner surplus per citizen (approximately US$284). However, there were major variations between “municipality years”; the standard deviation is 1,406 Danish kroner per citizen, refl ecting that both defi cits and surpluses are common.

Th e second dependent variable, a municipality’s budget overrun, measures the diff erence between the accounted operating result and the budgeted operating result in a given year, both per citizen in thousands of Danish kroner. It is calculated according to the follow- ing formula for municipality m in year t:

Budget Overrun m,t

= Budgeted Operating Result m,t

– Accounted Operating Result

m,t

Positive values indicate that a municipality has failed to reach its budgeted fi nancial results (i.e., an overrun). Negative values indicate that the realized result is better than budgeted (i.e., a budget “under- run”). Such a measure has previously been used to examine Danish municipalities’ fi nancial performance (Serritslew 2005). Again, these fi gures are obtained from the yearly statements made by the Danish Evaluation Institute for Local Government. In the period under analysis, the municipalities’ average budget overrun was a modest negative 84 Danish kroner per citizen, refl ecting that, on average, municipalities did a little better than budgeted. Th e standard devia- tion is 1,117 Danish kroner per citizen, indicating that the modest average overrun covers substantial variation across municipalities. Some did far better than the budget; others did far worse.

It can be argued that budget overrun is not symmetrical around zero, as doing better than budgeted is not necessarily positive and may imply that inadequate service levels were provided. To address this, we con- ducted an alternative analysis with only budget overruns (i.e., realized results worse than budgeted). We found substantially similar results.

Independent variable. On the basis of the listings in the local government handbook, we registered the names of all CEOs and managing directors in each of the 98 municipalities for each year of the analysis. From this, we then coded the gender composition of

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 297

As earlier, the municipality’s administrative structure is found to moderate the relationship between TMT gender diversity and fi nancial performance, lending further support to hypothesis 2. In municipalities applying an executive board model, more diverse TMTs produce smaller budget overruns (or even budget underruns,

of the TMT. Moreover, being merged has also been found to aff ect the municipalities’ fi nancial management (Hansen, Houlberg, and Pedersen 2014). Th ird, we control for political factors by includ- ing a dummy variable for the party of the mayor. Liberal mayors are used as the reference category in relation to Social Democratic mayors and mayors from other parties. We also control for potential eff ects of municipal council elections by including a dummy vari- able for the year 2009. Furthermore, we include a measure for the gender diversity on the council. We also control for the extent to which the municipality uses private contractors for service delivery. Finally, we control for the size of the TMT.

Results Table 1 reports the descriptive statistics and correlations among variables. Th e impact of top management team gender diversity on the municipalities’ operational results is presented in table 2. Th ree models are reported: in the fi rst model, only the control variables are entered; in model 2, the variable measuring TMT gender diversity is added; and in the third model, the interaction term between TMT gender diversity and the use of an executive board model is included.

Hypothesis 1 proposed that higher TMT gender diversity is related to better fi nancial performance. Th is hypothesis is not supported with regard to the municipalities’ operating results. Model 2 indicates no signifi cant relationship between TMT gender diversity and operating results. However, the degree of diversity of the TMT is found to have a positive impact on operating results when the municipality’s top management is organized as an executive board. In model 3, the interaction term between TMT gender diversity and the use of an executive board model has a signifi cant positive coef- fi cient. Th is lends support to hypothesis 2, which proposed that the relationship between TMT gender diversity and fi nancial perform- ance is moderated by the organization’s administrative structure.

In table 3, the impact of TMT gender diversity on the municipality’s budget overrun is presented. Following the same steps as earlier, three models are reported. Model 2 indicates no signifi cant relationships between TMT gender diversity and the size of municipal budget overrun. Th us, hypothesis 1 is not supported in this regard either.

Table 1 Correlations and Descriptive Statistics

Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Operating resulta 1.48 1.47 1 Budget overruna –0.10 1.14 –0.71 1 TMT gender diversity 0.26 0.19 0.10 –0.05 1 Population 53,565 56,911 0.13 –0.04 0.05 1 Population density 551 1,347 0.17 –0.07 0.12 0.39 1 Tax base 152,595 28,536 0.18 –0.19 0.21 –0.03 0.34 1 Socioeconomic index 0.94 0.25 0.13 0.01 0.07 0.19 0.20 –0.34 1 Debt 15,622 9,200 0.11 –0.06 0.06 –0.17 –0.01 0.06 0.03 1 Contracting 24.00 4.07 0.08 –0.17 0.14 0.01 0.07 0.21 0.04 0.14 1 Reformed 0.66 0.46 0.13 –0.13 0.15 0.09 0.55 0.38 0.30 –0.18 0.19 1 Social Democrat mayor 0.47 0.49 0.06 0.03 –0.02 0.14 0.04 –0.20 0.32 0.01 –0.12 0.09 1 “Other” mayor 0.19 0.39 0.01 –0.01 0.13 0.16 0.19 0.19 –0.13 0.05 0.05 0.19 –0.46 1 Elected board gender

diversity 0.40 0.08 0.18 –0.22 0.21 –0.04 0.29 0.45 –0.04 –0.01 0.17 0.31 0.04 0.13 1 TMT size 4.88 1.82 –0.06 0.15 0.09 0.29 0.09 –0.09 0.10 –0.04 –0.27 –0.16 0.17 –0.05 –0.07 1 Executive board 0.17 0.38 –0.06 –0.10 –0.09 –0.13 –0.11 0.05 –0.11 –0.08 0.10 0.04 –0.07 –0.01 –0.01 0.40 1 Election year 0.20 0.40 0.02 –0.14 0.01 0.00 –0.00 –0.03 –0.00 –0.01 0.01 0.01 –0.01 –0.01 0.09 0.03 0.00 1

aIn thousands of Danish kroner. Note: All correlations > |0.1| are signifi cant with p < .05.

Table 2 GLS Regressions: Dependent Variable, Operating Result

Model 1 Model 2 Model 3

Population 0.00*** 0.00*** 0.00*** (0.00) (0.00) (0.00)

Population density 0.00 0.00 0.00 (0.00) (0.00) (0.00)

Tax base 0.00 0.00 0.00 (0.00) (0.00) (0.00)

Socioeconomic index 0.95* 0.93* 1.04* (0.40) (0.40) (0.41)

Debt 0.00* 0.00 0.00* (0.00) (0.00) (0.00)

Contracting 0.03 0.02 0.02 (0.02) (0.02) (0.02)

Merged –0.01 –0.05 –0.07 (0.30) (0.31) (0.30)

Social Democrat mayor 0.17 0.17 0.17 (0.19) (0.19) (0.19)

“Other” mayor –0.15 –0.17 –0.19 (0.25) (0.25) (0.25)

Elected board gender diversity 2.53 2.42 2.29 (1.43) (1.46) (1.46)

TMT size –0.23** –0.23** –0.24*** (0.07) (0.07) (0.07)

Executive board –0.81*** –0.76** –1.27*** (0.25) (0.24) (0.32)

Election year –0.06 –0.06 –0.05 (0.12) (0.12) (0.12)

TMT gender diversity 0.52 0.01 (0.44) (0.46)

TMT gender diversity * Executive board 2.61* (1.08)

Constant –2.08 –2.00 –1.58 (1.21) (1.22) (1.21)

Number of groups 91 91 91 Number of observations 442 442 442 Wald chi-square 128.15*** 136.45*** 143.76*** R2 0.14 0.14 0.15

Note: Robust standard errors in parentheses. ***p < .001; **p < .01; *p < .05; two-sided tests.

298 Public Administration Review • March | April 2015

by an almost fl at line. Turning to municipalities using an execu- tive board model, we see that the line is decreasing indicating that diversity’s eff ect on budget overrun decreases as diversity increases. Together, the graphs in fi gure 1 underline how the eff ects of diversity are contingent on management structure. In both panels, the eff ect of diversity does not aff ect fi nancial performance when traditional structures are used, as illustrated by the almost fl at lines in panels 1 and 2.

To further scrutinize the importance of gender diversity for diff er- ent management structures and assess the substantial eff ect sizes, in table 4, the interactions are further illustrated. Based on our regressions we have calculated the predicted diff erences in operating results and budget overruns for municipalities with high versus low levels of TMT gender diversity.

Turning fi rst to the operating results, we observe a diff erence of 1,050 Danish kroner per citizen between those municipalities with a high level of executive board diversity and those with a low level (the eff ect illustrated by the rising line in fi gure 1, panel 1). Th is is

where the result is better than budgeted). In model 3, the interac- tion term between TMT gender diversity and the executive board model has a signifi cant negative coeffi cient.

Figure 1 provides a graphical depiction of the impact of gender diversity on fi nancial performance. Th e graphs show marginal eff ects for diff erent values of diversity. One line depicts municipalities with an executive board model and the other municipalities without such a management structure. In panel 1, we see how the eff ect of TMT diversity on the operational result is contingent on manage- ment structure. Overall, municipalities applying an executive board model produce poorer operational results compared with munici- palities with other administrative structures (cf. models 1 and 2). For low levels of diversity, the traditional management model has a higher eff ect on operational result than the board of director model. However, a more gender-diverse TMT makes the latter munici- palities “catch up” with the former ones as the eff ect of diversity increases for municipalities with the executive board model when diversity is higher. For high levels of diversity, the two models have a similar eff ect on operating results.

In fi gure 1, panel 2, we off er a visual presentation of the marginal eff ects of the results investigating budget overrun. Th e eff ect of diversity on overrun is near constant for municipalities using tradi- tional administrative structures for all levels of diversity indicated

Table 3 GLS Regressions: Dependent Variable, Budget Overrun

Model 1 Model 2 Model 3

Population 0.00 0.00 –0.00* (0.00) (0.00) (0.00)

Population density 0.00 0.00 0.00 (0.00) (0.00) (0.00)

Tax base –0.00* –0.00* 0.00 (0.00) (0.00) (0.00)

Socioeconomic index –0.27 –0.27 –0.37 (0.30) (0.30) (0.31)

Debt 0.00 0.00 0.00 (0.00) (0.00) (0.00)

Contracting –0.03 –0.03 –0.03 (0.02) (0.02) (0.02)

Merged 0.00 0.01 0.04 (0.20) (0.20) (0.19)

Social Democrat mayor 0.09 0.09 0.1 (0.16) (0.16) (0.16)

“Other” mayor 0.30 0.30 0.32 (0.23) (0.23) (0.22)

Elected board gender diversity –3.17** –3.16** –2.92* (1.21) (1.22) (1.24)

TMT size 0.12* 0.12* 0.13* (0.05) (0.05) (0.05)

Executive board 0.38 0.38 0.85*** (0.21) (0.20) (0.25)

Election year 0.14 0.14 0.13 (0.11) (0.11) (0.11)

TMT gender diversity –0.08 0.37 (0.39) (0.40)

TMT gender diversity * Executive board –2.45** (0.78)

Constant 2.55*** 2.56*** 2.11** (0.74) (0.77) (0.81)

Number of groups 91 91 91 Number of observations 442 442 442 Wald chi-square 72.08*** 72.54*** 78.64*** R2 0.09 0.09 0.12

Note: Robust standard errors in parentheses. ***p < .001; **p < .01; *p < .05; two-sided tests.

Figure 1 Interaction between TMT Gender Diversity and Organization Structure

0

.5

1

1.5

2

2.5

L in

e a r

P re

d ic

tio n

0 .1 .2 .3 .4 .5

Diversity Index

Other management structure Executive board model

Predictive Margins with 95% Confidence Intervals

-1

-.5

0

.5

1

L in

e a r

P re

d ic

tio n

0 .1 .2 .3 .4 .5

Diversity Index

Other management structure Executive board model

Predictive Margins with 95% Confidence Intervals

Panel 1 Dependent Variable: Operational Result

Panel 2 Dependent Variable: Budget Overrun

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 299

Discussion Top management team gender diversity presents a paradox for organizations. On the one hand, it may spark creativity and the development and use of diverse knowledge and perspectives on new ideas. On the other hand, diversity is likely to hamper eff ec- tive team work because of potential between-group diff erences in values, perceptions, and cognitive styles. Th is study has advanced the understanding of TMT diversity into the domain of public sector organizations. It provides evidence that not only individual managerial attributes but also the constellation of a management team aff ect core organizational outcomes. Focusing on gender, this study has suggested that organizations are likely to accrue posi- tive outcomes from gender diversity in management, but only if the management structure facilitates the eff ective use of individual diff erences. Th e results of our analysis of the TMTs of more than 90 Danish municipalities indicate that municipalities with gender- diverse management groups obtain better fi nancial performance— but only if they use the executive board management model, in which the top management group shares responsibilities broadly. Conversely, our results indicate that the impact of gender diversity is neutral when top managers are specialized and have clearly divided task responsibilities. Th ese results shed light on the mixed results of previous research into team diversity, highlighting how diff erent structures may, to varying degrees, facilitate the productive use of

diverse human resources. Th e present analysis only deals with gender diversity, but whether the management structure facilitates behav- ioral integration is potentially important for how TMT diversity along various dimensions aff ects organizational outcomes.

As pointed out by Pitts and Wise (2010), we urgently need more knowledge about the out-

comes of diversity. Th is study has contributed to this agenda with an explicit focus on management. Gender has previously been estab- lished as one relevant dimension of diversity, with research fi nding notable diff erences between male and female public managers. Th e results of this study further establish the relevance of gender, show- ing that given the right structure, a gender-integrated management team can outperform a gender-homogenous TMT.

As with any other study, there are limitations that should be noted. Focusing on gender diversity is, as argued earlier, an important task for research. However, it may be that other management charac- teristics amplify or dampen the eff ects of gender. Data limitations inhibited the inclusion of other relevant dimensions in the present study. Future studies should look into the eff ects of other characteris- tics and how diff erent dimensions of diversity interact. For instance, research on private sector organizations has suggested that diversity in education, experience, and age are all relevant for understanding the operational dynamics in top management teams (e.g., Knight et al. 1999; Kochan et al. 2003; Van Knippenberg and Schippers 2007).

Th is study has focused on the association between TMT diversity and fi nancial performance. While the analyses have allowed us to speculate about diversity’s impact on group processes, future studies should investigate underlying mechanisms in fi ner detail. Th is would ideally involve laboratory of fi eld experiments that can identify causal dynamics. Such studies could follow the work of

a substantial result. Th e average municipality size is around 57,000 citizens. Th is indicates that with an executive board management structure gender diversity is a distinct asset. When other manage- ment models are used, we observe no signifi cant diff erence in operating results between municipalities with high and low levels of TMT diversity (as indicated by the fl at line in fi gure 1, panel 1).

Focusing next on budget overruns, we see a similar picture. When the executive board model is in use, there is a diff erence of –830 Danish kroner between high- and low-diversity municipalities. Th e negative number indicates that low-diversity municipalities run larger budget overruns. Again, no signifi cant diff erence is found when other management models are in place. Together, the parts of the table clearly indicate the relationship between TMT gender diver- sity and fi nancial results. For both operating results and budget overrun, the relationship is contingent on the management model. When the executive board model is in use, signifi - cant and substantial diff erences exist between high and low levels of gender diversity.

In the next section, we discuss the results and their implications. While the control variables are not of explicit interest here, they do yield two relevant results. First, it should be noted that greater gender diversity among elected politicians is negatively related to budget overruns. It is beyond the scope of the present article to speculate about the relationship between the political and admin- istrative systems, but this provides further support for the general relevance of gender to fi nancial decision making. Th e data do not show a positive impact of elected board diversity on fi nancial per- formance with regard to operating results, however. An explana- tion for this could be that elected politicians are more attentive to the budget than the actual fi nancial accounts. Th e second notable result is that TMT size is negatively related to the municipalities’ fi nancial performance; larger TMTs, on average, produce poorer operational results and bigger budget overruns. Th ese results could suggest that behavioral integration is impeded when the number of TMT members increases. Although we were not explicitly interested in TMT size, to further control for a possible moderat- ing eff ect of TMT size on the hypothesized relationship between TMT gender diversity and fi nancial performance, we included an interaction term for TMT gender diversity and TMT size to the analysis. However, the original eff ects remained unchanged, and no additional eff ects of TMT size for either the municipali- ties’ operational results or budget overruns were found (results not reported).

When the executive board model is in use, signifi cant and

substantial diff erences exist between high and low levels of

gender diversity.

Table 4 Predicted Differences in Operating Result and Budget Overrun between Municipalities with High/Low Values of Gender Diversity

Operating Result in DKK per Citizen: Difference between “High” Diversity and “Low” Diversity

Budget Overrun in DKK per Citizen: Difference between “High” Diversity and “Low” Diversity

Executive board model 1,050** –830**

Other model 10 150

Notes: High diversity is defi ned as mean +1 standard deviation; low diversity is defi ned as mean –1 standard deviation. **p < .01.

300 Public Administration Review • March | April 2015

Berg, Rikke, and Ulrik Kjær. 2005. Den danske borgmester [Th e Danish Mayor]. Odense: Syddansk Universitetsforlag.

Blom-Hans en, Jens, Marius Ibsen, Th orkil Juul, and Poul Erik Mouritzen. 2012. Fra sogn til velfærdsproducent. Kommunestyret gennem fi re årtier [From Parish to Welfare Producer: Municipalities through Four Decades]. Odense: Syddansk Universitetsforlag.

Carmeli, Abraham, and John Schaubroeck. 2006. Top Management Team Behavioral Integration, Decision Quality, and Organizational Decline. Leadership Quarterly 17(5): 441–53.

Carter, Dav id A., Betty J. Simkins, and W. Gary Simpson. 2003. Corporate Governance, Board Diversity, and Firm Value. Financial Review 38(1): 33–53.

Cartwright, Dorwin, and Alvin Zander, eds. 1960. Group Dynamics: Research and Th eory. 2nd ed. Evanston, IL: Row, Peterson.

Christensen , Jørgen Grønnegård. 1998. Det kommunale mikrokosmos [Th e Municipal Micro-Cosmos]. Politica 30(3): 339–48.

Cyert, Richard M., and James G. March. 1963. A Behavioral Th eory of the Firm. Englewood Cliff s, NJ: Prentice Hall.

Dezsö, Cristia n L., and David Gaddis Ross. 2012. Does Female Representation in Top Management Improve Firm Performance? A Panel Data Investigation. Strategic Management Journal 33(9): 1072–89.

Dwyer, Sean, Or lando C. Richard, and Ken Chadwick. 2003. Gender Diversity in Management and Firm Performance: Th e Infl uence of Growth Orientation and Organizational Culture. Journal of Business Research 56(12): 1009–19.

Eagly, Alice H. , and Linda L. Carli. 2003. Th e Female Leadership Advantage: An Evaluation of the Evidence. Leadership Quarterly 14(6): 807–34.

Ejersbo, Niels, Morten Balle Hansen, and Poul Erik Mouritzen. 1998. Th e Danish Local Government CEO: From Town Clerk to City Manager. In Th e Anonymous Leader: Appointed CEOs in Western Local Government, edited by Kurt Klaudi Klausen and Annick Magnier, 97–112. Odense: University Press of Southern Denmark.

Erhardt, Niclas L., James D. Werbel, and Charles B. Shrader. 2003. Board of Director Diversity and Firm Financial Performance. Corporate Governance: An International Review 11(2): 102–11.

Fernandez, Serg io. 2005. Developing and Testing an Integrative Framework of Public Sector Leadership: Evidence from the Public Education Arena. Journal of Public Administration Research and Th eory 15(2): 197–217.

Finkelstein, Sy dney, and Donald C. Hambrick. 1990. Top-Management-Team Tenure and Organizational Outcomes: Th e Moderating Role of Managerial Discretion. Administrative Science Quarterly 35(3): 484–503.

Francis, Bill, Iftekhar Hasan, Jong Chool Park, and Qiang Wu. 2009. Gender Diff erences in Financial Reporting Decision-Making: Evidence from Accounting Conservatism. Paper presented at the Annual Meeting of the Financial Management Association, Reno, NV, October 21–14.

Frees, Edward W. 20 04. Longitudinal and Panel Data: Analysis and Applications in the Social Sciences. New York: Cambridge University Press.

Garrone, Paola, Luc a Grilli, and Xavier Rousseau. 2013. Management Discretion and Political Interference in Municipal Enterprises: Evidence from Italian Utilities. Local Government Studies 39(4): 514–40.

Greve, Carsten. 200 6. Public Management Reform in Denmark. Public Management Review 8(1): 161–69.

Guy, Mary Ellen, an d Meredith A. Newman. 2004. Women’s Jobs, Men’s Jobs: Sex Segregation and Emotional Labor. Public Administration Review 64(3): 289–98.

Hambrick, Donald C. 1994. Top Management Groups: A Conceptual Integration and Reconsideration of the “Team” Label. Research in Organizational Behavior 6: 171–214.

———. 2007. Upper Echelons Th eory: An Update. Academy of Management Review 32(2): 334–43.

Hambrick, Donald C. , Th eresa Seung Cho, and Ming-Jer Chen. 1996. Th e Infl uence of Top Management Team Heterogeneity on Firms’ Competitive Moves. Administrative Science Quarterly 41(4): 659–84.

TMTs in detail and outline why and how human diversity aff ects organizational outcomes.

Further, in this study, we explored team dynamics but with the organization as the eff ective unit of analysis. We urge future research to consider teams as units of analysis. While there is a long tradition of team-based research in the general organizational behavior litera- ture (Van Knippenberg and Schippers 2007), this type of research is rare in public administration. Consequently, we know too little about team dynamics in public sector organizations. Exploring team diversity would be a well-suited starting point for such research.

Finally, an avenue for future research relates not to the outcome of TMT diversity but to antecedents. With the demographic charac- teristics of managers being important for organizational outcomes, we should increase our understanding of the rationales organizations use in hiring managers with diverse characteristics. Factors such as current TMT composition, administrative structure and strategy, and environmental demands could be expected to relate to the degree of TMT diversity pursued by public organizations.

Gender diversity has the potential to lead to superior organizational outcomes, but it does not automatically do so. Th e present analysis implies that to reap the benefi ts of diverse top management teams, the management structure must facilitate behavioral integration between TMT members. However, more knowledge of other con- tingencies and pitfalls is needed.

Notes 1. Besides their tax revenue, municipalities are subsidized by block grants from the

national government (Greve 2006). 2. Th e basis of the current system of local government in Denmark was created

in the 1970 municipal reform (Kommunalreformen), in which 1,388 so-called parish municipals (sognekommuner) and market towns (købstæder) were organized into 275 municipalities. Th is structure existed until 2007, when a new structural reform (Strukturreformen) reduced the number of municipalities to 98. Th irty- three of the existing municipalities continued unchanged, whereas the remaining 65 were new entities based on (more or less) voluntary amalgamations. Th e fi ve municipalities on the island of Bornholm “jumped the gun” on the structural reform and merged into one municipality in 2003 (Blom-Hansen et al. 2012).

3. See http://krevi.dk/noegletal [in Danish]. Since June 1, 2012, the Danish Evaluation Institute for Local Government has been a part of the National Institute for Local and Regional Analysis and Research, an independent institu- tion under the Ministry of Economy and the Interior.

4. See http://www.noegletal.dk/ [in Danish].

References Ande rsen, Simon Calmar, and Søren Serritzlew. 2007. Th e Unintended Eff ects of

Private School Competition. Journal of Public Administration Research and Th eory 17(2): 335–56.

Andre ws, Rhys, George A. Boyne, and Richard M. Walker. 2006. Strategy Content and Organizational Performance: An Empirical Analysis. Public Administration Review 66(1): 52–63.

Auh, Seigyoung, and Bulent Menguc. 2006. Diversity at the Executive Suite: A Resource-Based Approach to the Customer Orientation–Organizational Performance Relationship. Journal of Business Research 59(5): 564–72.

Bækga ard, Martin. 2011. Th e Impact of Formal Organizational Structure on Politico-Administrative Interaction: Evidence from a Natural Experiment. Public Administration 89(3): 1063–80.

The Right Mix? Gender Diversity in Top Management Teams and Financial Performance 301

Mouritzen, Poul Erik , and James H. Svara. 2002. Leadership at the Apex: Politicians and Administrators in Western Local Governments. Pittsburgh, PA: University of Pittsburgh Press.

Moynihan, Donald P., and Patricia Wallace Ingraham. 2004. Integrative Leadership in the Public Sector: A Model of Performance-Information Use. Administration & Society 36(4): 427–53.

Moynihan, Donald P., and Sanjay K. Pandey. 2010. Th e Big Question for Performance Management: Why Do Managers Use Performance Information? Journal of Public Administration Research and Th eory 20(4): 849–66.

Østergaard, Christian R., Bram Timmermans, and Kari Kristinsson. 2011. Does a Diff erent View Create Something New? Th e Eff ect of Employee Diversity on Innovation. Research Policy 40(3): 500–509.

Park, Sanghee. 2013. Does Gender Matter? Th e Eff ect of Gender Representation of Public Bureaucracy on Governmental Performance. American Review of Public Administration 43(2): 221–42.

Pitts, David W. 2005. Diversity, Representation, and Performance: Evidence about Race and Ethnicity in Public Organizations. Journal of Public Administration Research and Th eory 15(4): 615–31.

Pitts, David W., and Lois Recascino Wise. 2010. Workforce Diversity in the New Millennium: Prospects for Research. Review of Public Personnel Administration 30(1): 44–69.

Powell, Melanie, and David Ansic. 1997. Gender Diff erences in Risk Behaviour in Financial Decision-Making: An Experimental Analysis. Journal of Economic Psychology 18(6): 605–28.

Riccucci, Norma M., G regg G. Van Ryzin, and Cecilia F. Lavena. 2014. Representative Bureaucracy in Policing: Does It Increase Perceived Legitimacy? Journal of Public Administration Research and Th eory. Published electronically on March 6. doi:10.1093/jopart/muu006.

Roch, Christine H., a nd David W. Pitts. 2012. Diff ering Eff ects of Representative Bureaucracy in Charter Schools and Traditional Public Schools. American Review of Public Administration 42(3): 282–302.

Serritzlew, Søren. 20 05. Breaking Budgets: An Empirical Examination of Danish Municipalities. Financial Accountability and Management 21(4): 413–35.

Stivers, Camilla. 1995 . Settlement Women and Bureau Men: Constructing a Usable Past for Public Administration. Public Administration Review 55(6): 522–29.

Stoner, James A. F. 19 68. Risky and Cautious Shifts in Group Decisions: Th e Infl uence of Widely Held Values. Journal of Experimental Social Psychology 4(4): 442–59.

Van Knippenberg, Daan, and Michaela C. Schippers. 2007. Work Group Diversity. Annual Review of Psychology 58: 515–41.

Van Wart, Montgomery. 2013. Lessons from Leadership Th eory and the Contemporary Challenges of Leaders. Public Administration Review 73(4): 553–65.

Villadsen, Anders R. 2 012. New Executives from Inside or Outside? Th e Eff ect of Executive Replacement on Organizational Change. Public Administration Review 72(5): 731–40.

Vrangbæk, Karsten. 201 0. Structural Reform in Denmark, 2007–09: Central Reform Processes in a Decentralised Environment. Local Government Studies 36(2): 205–21.

Wildavsky, Aaron B, and Naomi Caiden. 2004. Th e New Politics of the Budgetary Process. New York: Pearson/Longman.

Wilkins, Vicky M., and Lael R Keiser. 2006. Linking Passive and Active Representation by Gender: Th e Case of Child Support Agencies. Journal of Public Administration Research and Th eory 16(1): 87–102.

Zahra, Shaker A., and J ohn A. Pearce. 1989. Boards of Directors and Corporate Financial Performance: A Review and Integrative Model. Journal of Management 15(2): 291–334.

Hansen, Sune Wellin g, Kurt Houlberg, and Lene Holm Pedersen. 2014. Do Municipal Mergers Improve Fiscal Outcomes? Scandinavian Political Studies 37(2): 196–214.

Hefetz, Amir, and M ildred Warner. 2004. Privatization and Its Reverse: Explaining the Dynamics of the Government Contracting Process. Journal of Public Administration Research and Th eory 14(2): 171–90.

———. 2012. Contract ing or Public Delivery? Th e Importance of Service, Market, and Management Characteristics. Journal of Public Administration Research and Th eory 22(2): 289–317.

Jackson, Susan E., Aparna Joshi, and Niclas L. Erhardt. 2003. Recent Research on Team and Organizational Diversity: SWOT Analysis and Implications. Journal of Management 29(6): 801–30.

Jacobson, Willow S. , Christine Kelleher Palus, and Cynthia J. Bowling. 2010. A Woman’s Touch? Gendered Management and Performance in State Administration. Journal of Public Administration Research and Th eory 20(2): 477–504.

Johansen, Morgen S. 2007. Th e Eff ect of Female Strategic Managers on Organizational Performance. Public Organization Review 7(3): 269–79.

Juenke, Eric G. 200 5. Management Tenure and Network Time: How Experience Aff ects Bureaucratic Dynamics. Journal of Public Administration Research and Th eory 15(1): 113 –31.

Keiser, Lael R., Vi cky M. Wilkins, Kenneth J. Meier, and Catherine A. Holland. 2002. Lipstick and Logarithms: Gender, Institutional Context, and Representative Bureaucracy. American Political Science Review 96(3): 553–64.

Kent, Russell L., a nd Sherry E. Moss. 1994. Eff ects of Sex and Gender Role on Leader Emergence. Academy of Management Journal 37(5): 1335–46.

Ki ngsley, John Dona ld. 1944. Representative Bureaucracy: An Interpretation of the British Civil Service. Yellow Springs, OH: Antioch Press.

Knight, Don, Craig L. Pearce, Ken G. Smith, Judy D. Olian, Henry P. Sims, Ken A. Smith, and Patrick Flood. 1999. Top Management Team Diversity, Group Process, and Strategic Consensus. Strategic Management Journal 20(5): 445–65.

Kochan, Th omas, Kat erina Bezrukova, Robin Ely, Susan Jackson, Aparna Joshi, Karen Jehn, Jonathan Leonard, David Levine, and David Th omas. 2003. Th e Eff ects of Diversity on Business Performance: Report of the Diversity Research Network. Human Resource Management 42(1): 3–21.

Krishnan, Hema A., and Daewoo Park. 2005. A Few Good Women—On Top Management Teams. Journal of Business Research 58(12): 1712–20.

Lubatkin, Michael H ., Zeki Simsek, Yan Ling, and John F Veiga. 2006. Ambidexterity and Performance in Small- to Medium-Sized Firms: Th e Pivotal Role of Top Management Team Behavioral Integration. Journal of Management 32(5): 646–72.

Meier, Kenneth J., and John Bohte. 2001. Structure and Discretion: Missing Links in Representative Bureaucracy. Journal of Public Administration Research and Th eory 11(4): 455–70.

Meier, Kenneth J., and Laurence J. O’Toole, Jr. 2001. Managerial Strategies and Behavior in Networks: A Model with Evidence from U.S. Public Education. Journal of Public Administration Research and Th eory 11(3): 271–94.

———. 2003. Public M anagement and Educational Performance: Th e Impact of Managerial Networking. Public Administration Review 63(6): 689–99.

Meier, Kenneth J., Laurence J. O’Toole, Jr., and Holly T. Goerdel. 2006. Management Activity and Program Performance: Gender as Management Capital. Public Administration Review 66(1): 24–36.

Miller, Toyah, and María del Carmen Triana. 2009. Demographic Diversity in the Boardroom: Mediators of the Board Diversity–Firm Performance Relationship. Journal of Management Studies 46(5): 755–86.

Mintzberg, Henry. 19 79. Th e Structuring of Organizations: A Synthesis of the Research. Englewood Cliff s, NJ: Prentice Hall.