The City Manager’s Role in Policy-Making

Abstract

The city manager’s role in policy-making has long been recognized in the public administrationliterature. However, it remains unclear why city managers participate in policy-making at different levels. Two alternative models have been proposed in the literature: the substitution model and the collaboration model. The present study considers multiple dimensions of political leadership and advances a more comprehensive theoretical framework. It then undertakes an empirical examination, using survey data collected from city managers in Georgia. The results uncover more complicated configurations of political dynamics in local government than described by either the substitution or the collaboration model.

Keywords

policy-making collaboration substitution leadership city manager mayor council

Introduction

The city manager’s role in policy-making is one of the most controversial issues in the public administration literature. Historically, the politics–administration dichotomy has treated city managers as merely policy implementers rather than policy-makers in the local government system. In the beginning of the last century, the major responsibilities of city managers were confined to housekeeping or delivery of public services such as sewage and waste disposal, water supply, public safety, and road construction (Cox III, 2004). However, the role of city managers in the 21st century has become much more complex than it was 100 years ago. Contemporary city managers still handle issues related to public services delivery, but they also participate in policy-making and community politics, and they frequently interact with elected officials (Cox III, 2004; Nalbandian, 1989).

The controversy also stems from the gap between institutional design and real practices. A municipal charter usually does not grant the city manager formal power to make policy. Only elected officials possess formal policy-making authority because they are legally elected and are accountable to the electorate. By institution, a city manager is subordinate to the legislature and needs continuous legislative approval to engage in policy-making because the manager is employed by the legislature or the mayor and does not have the political authority to make policy decisions (Protasel, 1988). However, the city manager’s role in policy-making has been widely recognized in the literature, challenging the perception of a politics–administration dichotomy as well as the institutional arrangement. As early as the 1950s, Bosworth (1958) identified three types of city managers—the administrator manager, the policy researcher and manager, and the community leader and manager. He emphasized that all three roles involved the city manager’s responsibility for policy-making. Kammerer (1964) pointed out that the most expected role of city managers expressed in the public administration literature was policy leader or innovator. Scholars including Booth (1968) and Fannin (1983) have described city managers’ activities and behavior patterns in policy-making.

Later studies further confirm that city managers may regularly participate in policy-making and even play a leadership role (e.g., Ammons & Newell, 1989; Browne, 1985; French & Folz, 2004; Loveridge, 1968; Morgan & Watson, 1992; Mouritzen & Svara, 2002; Nalbandian, 1989, 1991; Newell & Ammons, 1987, 1995; Svara, 1998, 1999a, 1999b, 2001). Even the International City/County Management Association (ICMA) encourages local government managers to participate in policy facilitation, democratic advocacy, and citizen involvement as expressed in the “Practices for Effective Local Government Management.” Wheeland (2004) indicates that two out of the 18 core content areas in ICMA’s “Practices for Effective Local Government Management,” along with the general practices, form the foundation that enables local government managers to participate in policy-making and help elected officials improve their policy-making capacities. Here, a city manager’s leadership, influence, or power in policy-making refers to the situation in which a city manager is deeply involved in policy-making activities and is able to influence policy decisions that are eventually presented to the city council for a vote. Such involvement may include initiating policy problems, writing policy proposals, recommending policy solutions, and participating in policy deliberations.

Nevertheless, city managers from a range of municipalities vary widely in their policy-making roles; they can be strong or weak policy-makers. What determines the extent to which a city manager participates in policy-making? A widespread perception, which has not been systematically tested, is that city managers gain policy leadership because of a combination of elected officials’ inability to make policies and managers’ expertise in policy issues. This untested perception implies a substitution pattern in the politician–administrator relationship with regard to policy-making, known as the “substitution model.” An alternative explanation hinges upon the idea that a stronger local governing body would be more willing to collaborate with the manager and better utilize the manager’s expertise in policy issues; thus, the city manager can influence policy-making to a greater extent. This is the “collaboration model.” Can either of these models capture the general pattern of politics–administration relationships in local policy-making? The present research will summarize the literature that supports each of the models and then develop a more comprehensive framework to specifically examine how political leadership shapes the city manager’s policy-making role.

Two Models in the Literature

Box (1992) describes two general perceptions regarding power relationships in local government operations: conflictual and nonconflictual perceptions. The conflictual perception holds that the relationship between the elected body and the city manager is characterized by conflict and shifts within a relatively fixed sum of power and influence. The nonconflictual perception assumes that the relationship is characterized by a variable sum of power and low levels of conflict. The major difference between the two perceptions is on the question of whether the amount of power held by elected officials is negatively associated with the amount of power held by appointed administrators.

In the sphere of local policy-making, the conflictual relationship perception suggests that appointed administrators would not earn policy-making power unless the elected officials lacked the ability to make policy and thus had to share or even transfer power to appointed administrators. In other words, elected officials and city managers play a zero-sum power game, and managers will substitute for elected officials when the latter cannot effectively function in the policy-making process. We call this type of power relationship the substitution model. In comparison, the nonconflictual perception presumes that the presence and increase of city managers’ policy-making power is assisted by elected officials and does not occur at the expense of elected officials’ policy leadership. In other words, elected officials share the policy-making role with city managers in a cooperative way, and the power of elected officials and that of city managers reinforce each other. We call this the collaboration model.

The substitution model is widely accepted in the literature of local policy-making. Protasel (1988) contends that city managers are deeply involved in policy-making and greatly influence policy outcomes because local politicians are often amateurs while professional managers have considerable policy expertise. He observes that “the role of community leader came to be thrust upon city managers by the circumstance of having to deal with amateur city councils which look to the professional city manager for expert advice” (p. 810). Adrian and Press (1977) propose that city managers may be compelled to play an active policy role largely by necessity when mayors lack substantial formal authority or have difficulty exercising policy leadership. This position is in line with other scholars such as Nalbandian (1991), Hassett and Watson (2002), and Feiock, Jeong, and Kim (2003). Particularly, Nalbandian (1991) states, “The increasing participation of managers in the policy-making process . . . can be attributed to policy voids left by the governing body, the manager’s own predisposition to ‘be in charge,’ and in many instances the governing body’s own expectation” (p. 58). He further points out, “City managers develop agendas for council meetings, bring issues to the council that its members would be unaware of otherwise, prepare an annual budget and make policy recommendations, and authorize the reports from staff to council. The governing body is actually placed in a vulnerable position” (Nalbandian, 2006, pp. 1049-1050).

The collaboration model also appears in the policy-making literature. Svara’s (1998, 1999b, 2001) complementarity model emphasizes the interdependent relationship between elected officials and appointed administrators in the policy-making process. He explains, “There is interdependency and reciprocal influence between elected officials and administrators who fill distinct but overlapping roles in policy and administration” (Svara, 2006, p. 1081). Wikstrom (1979) discovered that city managers usually prefer to work with a mayor who provides policy leadership and bears a strong sense of political responsibility. He suggests that city managers need to understand the political mood of the council and the larger community through effective collaboration with their mayors. Zhang and Yang (2008) use a case study to illustrate that both the mayor and the manager can enhance their individual leadership roles by working together synergistically, consequently making significant changes in their community. This mayor–manager team has the potential to be particularly effective because it utilizes the unique attributes of both parties. The mayor, who understands the specific needs of the community and possesses a promising vision for the way forward, actualizes this vision with the assistance of the manager, who possesses invaluable expertise on policy issues. Zhang and Yang (2008) emphasize, “It is not the case that the mayor contributes policy ideas and directs the manager in implementation, nor is it the case that the mayor simply agrees with the manager’s proposals. Rather, the mayor and the city manager share policy leadership through frequent communication and coordination” (p. 179). In the collaboration model, the assertive manager is not necessarily a replacement or a competitor for mayoral leadership; he or she could be a strong leadership partner.

The substitution and collaboration models provide different theoretical perspectives for thinking about the policy-making power distribution between the governing body and the city manager in local government. The two models conflict on whether political leadership reduces or enhances administrative policy-making power. Building upon existing research, the present article argues that either model might be too simplistic to explain the city manager’s role in policy-making. A more comprehensive framework, along with empirical examination, should be developed to investigate the politics–administration relationship in local policy-making. This study will contribute to the literature that systematically explores factors shaping the city manager’s power in policy-making.

The question of what factors systematically explain the city manager’s power in policy-making has been rarely studied in the literature. Mouritzen and Svara (2002) and Zhang and Feiock (2010) are the only researchers who have published significant studies on the topic. In their pioneering work to empirically investigate the factors leading city managers to exert policy leadership in budgeting and economic development, Mouritzen and Svara (2002) propose four clusters of predictors: (1) mayoral leadership, (2) managers’ perception about ideal politicians, (3) managers’ values, activities, and networking, and (4) governmental and community context. It is judicious to consider a wide variety of factors that may potentially influence managers’ power in policy-making. However, because theirs was an exploratory work, Mouritzen and Svara (2002) did not provide theoretical justification for each relationship in their models. Zhang and Feiock (2010) adjusted Mouritzen and Svara’s (2002) framework to predict city managers’ power in policy-making. They provide theoretical reasoning for the factors of mayoral leadership, manager’s professionalism, manager’s relationship with the council, and administrative authority. One of the limitations of their study is that the dependent variable, city manager’s power in policy-making, is measured by a single survey question. In addition, due to data availability, both studies above do not include council leadership in their models. In the present article, we will measure political leadership, especially the council’s leadership, in various dimensions and examine the impact of political leadership on the city manager’s role in policy-making. We will also test whether a manager’s professionalism and community-level characteristics make a difference in the manager’s policy-making power.

A Proposed Model

In a local government with representative democracy and with a manager position, there are three major policy-makers: the mayor, the city council, and the manager. Policy-making power is distributed among these three parties. The council is the legal legislature that represents all citizens in the community to vote on policy issues. How much policy-making power the manager can possess is largely determined by the extent to which the council shares its policy-making power with the manager and how much power the mayor has appropriated. The manager’s professional identity also plays a role in how much policy-making power he or she chooses to wield. In other words, city managers’ professionalism may shape their power in local policy-making. These relationships are demonstrated as Path 1 in Figure 1. Nevertheless, the decision as to how much power the council will share depends on the council leadership, its relationship with the manager, and the city manager’s professionalism (Path 2).

Figure 1.

The proposed model of city manager’s power in local policy-making.

In the proposed model, we differentiate between council leadership and mayoral leadership in local policy-making and argue that these two types of political leadership exert different influences on the manager’s policy-making power. In other words, the manager–mayor interaction pattern may differ from the manager–council interaction pattern in local policy-making. At the same time, we see the political leadership in two dimensions: formal authority determined by institutional arrangement and informal power rooted in individual ability and desire. We propose that although the share of formal authority shifts within a fixed sum in a municipal government, the informal power of each official may add up to a variable sum of overall leadership in a city.

The formal authority for the manager, mayor, and council members is determined by the form of government. Mixed forms of government are becoming increasingly popular in the United States (Frederickson, Johnson, & Wood, 2004), with many mayor–council governments adding a professional manager position to their city governments and many council–manager cities adopting popular election for the position of mayor. These developments have increasingly blurred the distinction between mayor–council and council–manager forms of government (also see Cox III, 2004). In particular, the weak mayor–council form and the council–manager form have become more alike than distinct. Instead of using the traditional approach of mayor-council form versus council–manager form to indicate the allocation of formal authority within city government, we perceive that the comparison between a strong mayor–council system and other forms of government more strongly indicates how formal power is shared among the mayor, manager, and council members. In a strong mayor–council city, the mayor has greater formal authority in policy-making and administration than mayors in other forms of government. Accordingly, both the council and city manager possess less formal authority than their counterparts in other forms of governments. Therefore, we hypothesize that city managers in strong mayor–council cities have less policy-making power than those in other cities.

“Formal powers are advantages in bargaining relationships, but do not obviate the need to tap other sources to be persuasive” (Wheeland, 2008, p. 328). The effectiveness of political leadership for both mayors and council members also depends on their informal powers—how well they can use their abilities, personality, and political skills to advance their policy goals. Even though the formal authorities described in city charters are identical for mayors in different cities, mayors may differ greatly in their overall influence in their communities due to differences in utilizing informal sources of power. Sparrow (1984) argues that some mayors wield greater power because they have the ability and desire to lead, have a sharp awareness of issues, possess the skills and experience necessary for coalition building, and are more inclined to aggregate formal and informal resources. In council–manager cities, mayors usually have little more power than other council members, but they may be motivated by their ability and desire to lead and thus become relatively strong local political leaders. Svara calls this “facilitative mayoral leadership.” In The Facilitative Leader in City Hall, edited by Svara (2008), scholars use case studies to demonstrate a variety of situations in which the council–manager mayor utilizes facilitative leadership to steer the city to make significant changes toward a better government and a better community (e.g., Klase, 2008; Nalbandian & Negron, 2008; Watson & Schortgen, 2008; Wood, Gabris, & Olson, 2008). As Svara (2008) states, the importance of mayoral leadership in council–manager cities has come to be more widely recognized in the past 15 years.

One aspect of mayoral facilitative leadership is the mayor’s actual influence in policy-making, which may or may not be associated with his or her formal power. How does such mayoral leadership shape the manager’s power in policy-making? Existing studies show that mayors in any form of local government, like other political leaders at all levels of government worldwide, tend to work more closely with the administrative team than they do with council members when setting the policy agenda (Klase, 2008; Svara, 2008; Wheeland, 2008; Zhang & Yang, 2008). Cooperation between the mayor and the manager becomes an easy way for both parties to strengthen their leadership (Klase, 2008; Zhang & Yang, 2008). Therefore, we hypothesize that the city manager’s policy-making power will not diminish as the mayor’s influence in policy-making increases.

Similarly, city councils may have various levels of influence on policy-making depending on how they use their informal power in policy-making—their ability and desire to make policy. Since the existing literature has not examined the council’s leadership in policy-making, this article particularly focuses on measuring and testing the influence of council policy-making leadership with respect to the city manager’s policy-making power. We conceptualize council policy-making leadership in two dimensions: policy-making ability and desire to make policy. The existing literature has been widely concerned with council policy-making ability and observes that city managers are compelled to take on policy-making responsibility when dealing with amateur councils (Adrian & Press, 1977; Feiock, Jeong, & Kim, 2003; Hassett & Watson, 2002; Nalbandian, 1991; Protasel, 1988). From the substitution perspective, council policy-making ability is negatively associated with the city manager’s policy-making power.

The council’s desire to make policy is another important aspect of council leadership. As Svara (2002) indicates, councils differ widely in terms of their desire to lead. In some cities, the governing bodies may perform simply as “delegates”; they merely reflect constituent views in their decisions and transfer citizen preferences into legislative proposals. In other cities, the governing body has a much stronger commitment to act; council members perform proactively as “governors.” One may argue that the council’s desire to make policy should be highly correlated with the council’s ability to make policy. We hypothesize that the city manager’s policy-making power will decrease when the council has a strong desire to lead, assuming the substitution model is true.

Regarding the relationship between the council and the manager, Svara, in an article with Keene, Nalbandian, O’Neil, Portillo, & Svara (2007), compares local government professionals in the United States as a group to those in the United Kingdom and Denmark, and he argues that local professional organizations in the United States give less explicit attention to the politician–administrator relationship than their European equivalents. Keene et al. (2007) generally suggest that the legitimacy of the city manager’s profession is not rooted in institutional structures but rather in external judgments, including council members’ judgment. Browne (1985) shows that managers who report having a good relationship with the council feel that their participation in policy-making is necessary. Mouritzen and Svara (2002) also find that mangers’ communication with council members has a positive impact on their role in policy-making. In sum, we propose that the city manager’s policy-making power will be greater if there is a good council–manager relationship in the local government.

City managers’ professionalism can be indicated by their professional experience, professional education, and participation in professional networks (Zhang & Feiock, 2010; Zhang & Yang, 2009). Experience is important because it translates into immediate outcomes such as work-related knowledge, skills, attitudes, and emotions (Alkadry, 2003; Tesluk & Jacobs, 1998). Professional education is supposed to enrich students with professional skills and professional ethics (DiMaggio & Powell, 1989) as well as leadership skills (Wheeland, 2000). Newell and Ammons (1987) found that managers with MPA degrees were more likely than their counterparts to devote a greater proportion of their time to policy and political roles. Professional networks such as ICMA encourage city managers to participate in policy formulation and help elected officials improve their policy-making capacities (Wheeland, 2004). They can also provide managers with opportunities to learn about best practices and new policy ideas from their colleagues (Ammons & Newell, 1989; Oakerson & Parks, 1988). Selden, Brewer, & Brudney (1999) found that managers with ICMA membership communicated significantly more often with their city councils than managers who were not members of ICMA.

Professional networking may be able to eliminate elected officials’ concern about the risks of power-sharing with the city manager. The literature indicates that more experienced, highly educated, and ICMA-affiliated managers are more likely to reflect the priorities of the city council and develop greater insights on how to make government more efficient and effective (Kearney, Feldman, & Scavo, 2000; Morgan & Watson, 1992; Schneider & Teske, 1992). Thus, they are more likely to be awarded opportunities by elected officials to participate in policy-making. However, city managers may earn encouragement from their professional networks and may intentionally choose cities that allow them to exert influence in policy-making. Therefore, we propose that city managers’ policy-making power will be positively associated with the level of their professional experience, education, and networking.

We also control for municipal characteristics in our model. The literature suggests that city size may influence mayoral and managers’ involvement in policy-making. One idea is that large cities may value political leadership more than small cities and to a certain degree undermine their city managers’ role in policy-making. As Yates (1977) vividly describes, in large “ungovernable” cities, a heterogeneous population contributes to group conflict and chaotic conditions, which generate great pressures for big-city mayors, who must then exercise extremely skillful political and policy leadership just to enable the city to survive. Thus, the city manager’s role in policy-making may be perceived as controversial in large cities. Another explanation is that in small cities, the relatively small size of the voting public may allow voters to hold the administration team directly accountable (Teske & Schneider, 1994), and thus it is less risky for voters and elected officials to share policy-making power with the city manager. Therefore, city size is expected to have a negative relationship with the city manager’s degree of leadership in policy-making.

Existing studies have consistently found that managers are more likely to shape policy-making in smaller cities (Morgan & Watson, 1992; Mouritzen & Svara, 2002). In addition, Zhang and Yang (2008) suggest that a city manager is likely to gain respect and autonomy in a community struggling from poor economic status because the poor community acutely needs professional help to change its status quo, and the elected officials cannot provide it. Therefore, we may expect a negative relationship between the city manager’s role in policy-making and municipal economic status; in other words, city managers in less wealthy communities will have a stronger influence on local policy-making.

To summarize our hypotheses, we provide Figure 2 to specify the variables and relationship directions in the model.

Figure 2.

The hypotheses in the model of city manager’s power in local policy-making.

Research Methods

Sample

The primary data source for this study is a mail survey of city managers in Georgia conducted from June through November 2007. We use data collected 5 years ago because no major political and legislative changes have taken place since then in Georgia municipalities. Moreover, this study investigates the general patterns of power relationships in local government. It requires that the cross-sectional data are collected during the same time period, but it is not sensitive to the overall age of the data.

The survey included questions about city managers’ participation in policy-making, interactions with council members, and managers’ demographic information. We designed the survey questions and conducted a presurvey with 15 city managers for improvement and corrections before sending out to all 192 city managers in Georgia. In Georgia, there are 505 municipalities, but only 192 (38%) of them have city manager positions. In some cases, the position is called “city administrator” or “chief administrative officer (CAO).” One hundred sixty city managers (83%) responded to the survey. Due to some missing data, our working sample for this study consists of 142 cities (74% of the 192 cities with managers). The sample is relatively representative of city managers in Georgia in terms of the distribution of city size. Of the 192 targeted municipalities, 11% are small towns with a population less than 1,000, 44% are small cities with a population between 1,000 and 6,000, 37% are medium-sized cities with a population between 6,000 and 30,000, and 8% are large cities with a population of more than 30,000. In our working sample, the four categories represent 7%, 42%, 42%, and 9% of the sample, respectively. The sample contains slightly more medium-sized and large cities but fewer small cities and towns, probably because small municipalities have fewer staff members to assist their city managers, making it less likely that the managers would respond to the survey.

Measurement

The city manager’s policy-making power was measured by their evaluation of two statements in the survey: (1) I, as the manager, have great influence on policy decisions; and (2) I have sufficient discretion to interpret the policy in policy implementation. City managers responded to both statements using a 7-point scale where 1 corresponded to “strongly disagree,” 4 was neutral, and 7 corresponded to “strongly agree.” An index was constructed by averaging the values of these two items (mean = 5.4; SD = 1.0; Cronbach’s α = 0.69).

Council power-sharing was measured by the city managers’ evaluations on two other statements in the survey: (1) the council prefers that I lead policy process; and (2) the council defers to my recommendations while making policies. These two statements used the same 7-point response scale as the previous statements. An index was constructed to average the values of these two items (mean = 4.8; SD = 1.2; Cronbach’s α = 0.64).

The council’s policy-making ability was evaluated by city managers with the following three statements, each using the previously described 7-point scale: (1) the council members well understand policy issues; (2) the council accurately defines public problems; and (3) the council develops feasible public policies. The index variable “council’s policy ability” takes the average values of the city managers’ responses to these three statements (mean = 4.9; SD = 1.0; Cronbach’s α = 0.81). The council’s desire to lead was measured by two survey statements using the same measurement scale: (1) The council promptly addresses public problems and (2) the council gives clear directions. This variable takes the average values of the two items (mean = 5.0;SD = 1.2; Cronbach’s α = 0.80).

The council–manager relationship is an index variable measured by respondents’ evaluation of the following two survey statements using the previously described 7-point scale: (1) the council generally trusts me and (2) my professional knowledge and skills are appreciated by the council. This variable takes the average values of managers’ responses to the two statements (mean = 5.9; SD = 0.9; Cronbach’s α = 0.78).

The Georgia Municipal Association provided the data regarding form of government. The original data set included strong mayor–council, weak mayor–council and council–manager forms. We recoded the variable into dichotomous values, assigning 1 to strong mayor–council and 0 to all other forms of government.¹ Of the 142 cities in the sample, 29 (20%) have a strong mayor–council system, and 113 (80%) are either weak mayor–council or council–manager cities.

Mayoral policy leadership was measured by managers’ evaluation of the following statement using the previously described 7-point scale: The mayor has more influence on policy decisions than (other) council members do. The mean response is 4.8 with a standard deviation of 1.8. One may argue that mayors in strong mayor–council municipalities should be more likely to have stronger leadership in policy-making than their counterparts in municipalities with weak mayor–council or council–manager systems. This appears to be true, because the average levels of policy influence for the two groups of mayors were 5.5 for the cities with strong mayors and 4.6 for other cities, generating an average difference of 0.9 in a 7-point scale. This difference is significant at the 0.05 level in the t-test. In addition, of 29 city managers in strong mayor–council cities, 62% agree or strongly agree that their mayors exert greater influence on policy decisions than (other) council members do, while this figure in non--strong-mayor municipalities was 41%. However, the data also show that not all mayors in strong mayor–council systems have a strong influence on policy-making; some of them do not sufficiently utilize their institutional authority to exert a strong policy influence.

Meanwhile, 41% of mayors in weak mayor–council or council–manager systems exert more influence than do other council members, indicating that mayors with institutional constraints may utilize their informal political leverage to exert greater influence in local policy-making than might be expected based strictly on their authority. Thus, the two variables of strong mayor–council system and mayoral policy leadership are not necessarily highly correlated with each other.

The measurement of city managers’ professionalism follows existing studies (Zhang & Feiock, 2010; Zhang & Yang, 2009). Professional experience was measured by the number of years a manager had served as a management professional in local governments. In our data, this number ranged from 1 to 41 years (mean = 15.4; SD = 11.3). Professional education was measured by whether the manager held an MPA degree, with 1 denoting possession of an MPA degree and 0 denoting no such degree. Of the 142 city managers in the sample, 43 (30%) had an MPA degree and 99 (70%) did not. Professional networking was measured by two questions: the frequency with which respondents participate in the ICMA network and the frequency with which respondents participate in local professional networks such as local municipal associations, councils of government, regional councils, and regional partnerships. We asked how often city managers attended meetings and activities of the two networks in the last 3 years using a 4-point ordered scale for responses, with 1 representing “never” and 4 corresponding to “very often.”

City size was measured by city’s population in 2006. Community economic status was measured by the city’s median household income in 2006. Both population and income data were obtained from census data available at city-data.com. Since the survey was conducted in 2007, using census data from 2006 should be appropriate because community characteristics did not dramatically change in the intervening year.

Test Procedure

Since the dependent variables exhibit quasi-continuous values due to the aggregation for the index and the values follow an almost normal distribution, ordinary least squares (OLS) regressions are appropriate for analysis. Given the structure of the theoretical framework, we performed path analysis with OLS (see Figure 3 for results). Prior to running the models, a correlation matrix was produced and variance inflation factors (VIFs) were evaluated to diagnose the possible problem of multicollinearity. (The VIFs are not presented in this article due to limited space.) The only high correlation (>0.5) is between council’s policy ability and council’s desire to make policy, which is expected in the model. To ensure unbiased coefficients, the interaction term of these two variables was added in Path 2 of the model. In addition, all the VIF values were below 3.7, suggesting that multicollinearity is not a significant issue in the empirical examination.

Figure 3.

The results of path analysis.

Most municipalities in Georgia are small and medium-sized cities, and only 9% of them have populations greater than 30,000, so we suspected that the results might contain biases caused by clusters of cities of certain population sizes. We therefore conducted sensitivity tests by operating two separate subsampling models with one excluding the large cities and the other excluding small towns (not presented here due to limited space). The two subsampling models provide very consistent results to those shown in Figure 3. We are therefore confident that the findings apply to municipalities with various population sizes.

Findings

Figure 3 presents the results of path analysis. The R² is 49% in Path 1 and 29% in Path 2, suggesting that the model fits the data well and has fairly strong explanation power. Comparing Figure 3 to Figure 2, we learn that most of the hypotheses are supported by the data with only one exception: The indicators of managers’ professionalism do not show significant influence as anticipated. As control variables, the municipal characteristics do not make a difference in the power relationship in local policy-making.

As shown in Figure 3, all political leadership variables have a significant impact on the city manager’s power in policy-making. A strong mayor system has a negative influence, confirming that formal political authority can systematically constrain the city manager’s role in local policy-making. City managers in weak mayor–council or council–manager cities are more likely to lead policy-making. As hypothesized, the impact of mayoral policy-making leadership is positive. A city manager is more likely to play an important role in policy-making when the mayor is more influential in shaping policy decisions than other elected officials. If mayoral policy-making leadership is weak, the manager will have less chance to gain policy-making power. The findings imply that given a certain institutional arrangement (either strong mayor–council or other forms), mayors with stronger facilitative leadership tend to endorse city managers’ leadership in policy-making. This is consistent with the proposition that a strong mayor and a strong city manager may enhance each other’s leadership roles by working together synergistically (Klase, 2008; Svara, 2008; Wheeland, 2008; Wikstrom, 1979; Zhang & Yang, 2008). Within such collaborative leadership, the mayor thoroughly understands the needs of the community, and the manager possesses professional expertise on policy issues. Together, they can make substantial changes in their community.

The manager’s policy-making power is largely determined by the extent to which the council shares its policy-making power with the manager. What leads a council to share power with the manager? According to the path analysis, a council will be more likely to do so when the manager frequently participates in local government networks, when the council has a good relationship with the manager, or when the council is in a council–manager or weak mayor city. Nevertheless, the most significant determinant is the council’s leadership—its ability and desire to make policy. Controlling for the interaction term, both the council’s policy ability and the council’s leadership desire are negatively associated with the city manager’s role in policy-making. Normally, a strong council will diminish the city manager’s role while a weak council may leave more room for the manager to exercise policy-making leadership. This illustrates a substitution relationship between council leadership and manager leadership in policy-making. However, the significant, positive coefficient of the interaction term provides an exception to the council-manager interaction pattern: When the council’s policy ability is extremely high (>6.08 out of 7),² the manager’s policy-making power becomes greater as the council’s leadership desire increases. In other words, a highly knowledgeable council will proactively lead policy-making while it shares leadership with the manager, although an average council will diminish the city manager’s role in policy-making as it becomes stronger.

Regarding administrative professionalism, the analysis results do not support the hypotheses that more professional experience and professional education will enhance a city manager’s leadership role in policy-making. Neither managers’ experience nor professional education has a significant influence. Even their participation in ICMA does not make a significant difference in the city manager’s policy-making role; only networking with regional management associations helps city managers to earn a share of policy-making power from their councils. These findings may generate two implications: First, the manager’s role in policy-making is mainly a choice of elected officials and largely depends on political leadership. Second, elected officials value regional networking more highly than national networking and will delegate more power to managers who have closer relationships with regional management associations.

Conclusion

The substitution and collaboration models have been described in the public administration literature to explain how political leadership shapes the city manager’s role in policy-making. While these models provide useful perspectives for thinking about the distribution of policy-making power between elected officials and professional administrators in local government, they might be too simplistic to comprehensively depict the complexity of local politics. Building upon existing research, we propose a new model to investigate the power relationships in local policy-making. In the new model, we differentiate between council leadership and mayoral leadership and argue that the manager–mayor interaction pattern may differ from the manager–council interaction pattern in local policy-making. At the same time, we not only examine formal authority but also take informal power into account. We propose that while the share of formal authority shifts within a fixed sum in a municipal government, informal power possessed by each official may add up to a variable sum of overall leadership in a city, and that the collaborative relationship is more likely to take place between the mayor and the manager than it is between the council and the manager. The empirical results generally support the new theoretical framework. In particular, the present study verifies the following hypotheses.

First, strong mayor–council cities will provide city managers with less chance to lead policy-making than council–manager or weak mayor–council cities. Second, a mayor who develops facilitative leadership in policy-making will help nurture the city manager’s leadership in policy-making. This confirms the pattern of mayor–manager team governance found in previous case studies (Klase, 2008; Svara, 2008; Wheeland, 2008; Wikstrom, 1979; Zhang & Yang, 2008). Third, a city manager is more likely to lead policy-making when the council’s leadership is weak, which indicates a substitutive relationship; managers may be compelled to play an active policy role by necessity when the elected body has difficulty exercising policy leadership (Adrian & Press, 1977). However, this relationship is not linear; when the council is extremely knowledgeable and active in policy-making, it is more likely to appreciate the manager’s leadership and team up with the manager to make policies. Finally, a good council–manager relationship in local government creates more opportunities for the city manager to exert substantial influence in policy-making.

Nevertheless, the research does not empirically support the hypothesis that more professional managers will have more power in policy-making. City managers may not rely on their expertise to decide the extent to which they participate in policy-making. It is mainly a choice of elected officials, who generally value city managers’ regional networking more highly than managers’ experience, professional education, and national networking. This may indicate that city managers generally comply with the political control of the governing body in the hierarchical structure (Zhang & Feiock, 2010).

This study has some limitations. First, the sample is from Georgia, so caution must be taken in generalizing the results beyond the sample. Moreover, this study may contain measurement bias. For example, most variables were not measured in an objective way. They are city managers’ perceptions instead. Although it is widely common in the public administration literature to use survey data and perceptual measurements, one cannot treat the findings as highly conclusive, considering the possible measurement errors. Moreover, the variables of political leadership were measured by city managers’ responses to survey questions. We acknowledge that collecting data directly from mayors and council members would be ideal. However, constrained by financial ability, we were not able to conduct a more comprehensive survey. Nevertheless, as professional administrators, city managers are in a relatively good position to evaluate political factors and government operations within their communities. Many scholars have used a similar approach in their studies (e.g., Browne, 1985; Feiock, Jeong, & Kim, 2003; Zhang & Feiock, 2010; Zhang & Liao, 2011). Another problematic measurement might be of city managers’ professional networking, which was measured by the frequency with which managers participate in the ICMA and regional management networks. The number of times that the manager participates in networks would be a more accurate measurement than the Likert-type scale measurement (from “never” to “very often”) that is currently used in this project. Future research is expected to fix the problems above and further refine the theoretical framework. Despite its limitations, the present study fairly serves its research purpose, which is to explore the interaction patterns of administrative leadership and political leadership in local policy-making.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Notes

Author Biography

Yahong Zhang is an Assistant Professor of Public Affairs and Administration at Rutgers University in Newark, NJ. Her research focuses on local government policy and management as well as anti-corruption studies. She can be reach at yahongzh@rutgers.edu.

References

Adrian

C. R.

Press

(1977). Governing urban America (5th ed.). New York, NY: McGraw-Hill.

Alkadry

M. G.

(2003). Deliberative discourse between citizens and administrators: If citizens talk, will administrators listen? Administration & Society, 35, 184-209.

Ammons

D. N.

Newell

(1989). City executives: Leadership roles, work characteristics, and time management. Albany, NY: SUNY Press.

Booth

D. A.

(1968). Are elected mayors a threat to managers? Administrative Science Quarterly, 12, 572-589.

Bosworth

K. A.

(1958). The manager is a politician. Public Administration Review, 18, 216-222.

Box

R. C.

(1992). The administrator as trustee of the public interest: Normative ideals and daily practice. Administration and Society, 24, 323-345.

Browne

(1985). Municipal managers and policy: A partial test of the Svara dichotomy-duality model. Public Administration Review, 45, 620-622.

Cox

R. W.

III (2004). The profession of local government manager: Evolution and leadership styles. In Newell

(Ed.), The effective local government manager (pp. 1-19, 3rd ed.). Washington, DC: ICMA.

DiMaggio

Powell

(1989). The new institutions in organizational analysis. Chicago, IL: University of Chicago Press.

10.

Fannin

W. R.

(1983). City manager policy roles as a source of city council/city manager conflict. International Journal of Public Administration, 5, 379-399.

11.

Feiock

R. C.

Jeong

Kim

(2003). Credible commitment and council-manager government: Implications for policy instrument choices. Public Administration Review, 63, 616-625.

12.

Frederickson

H. G.

Johnson

G. A.

Wood

(2004). The adapted city. Armonk, NY: M. E. Sharpe.

13.

French

P. E.

Folz

D. H.

(2004). Executive behavior and decision making in small U.S. cities. American Review of Public Administration, 34(1), 52-66.

14.

Hassett

W. L.

Watson

(2002). Long-serving city managers: Practical application of the academic literature. Public Administration Review, 62, 622-629.

15.

Kammerer

(1964). Role diversity of city managers. Administrative Science Quarterly, 8, 421-442.

16.

Kearney

R. C.

Feldman

Scavo

(2000). Reinventing government: City manager attitudes and actions. Public Administration Review, 60, 535-548.

17.

Keene

Nalbandian

O’Neil

Portillo

Svara

(2007, March). How professionals can add value to their communities and organizations. Public Management, 89(2), 32-39.

18.

Klase

K. A.

(2008). Switching roles from administrator to mayor: Winston-Salem, North Carolina. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions(pp. 109-124). Boca Raton, FL: Taylor & Francis.

19.

Loveridge

R. O.

(1968). The city manager in legislative politics: A collision of role conceptions. Polity, 1, 213-236.

20.

Morgan

D. R.

Watson

S. S.

(1992). Policy leadership in council-manager cities: Comparing mayor and manager. Public Administration Review, 52, 438-446.

21.

Mouritzen

P. E.

Svara

(2002). Leadership at the apex: Politicians and administrators in western local governments. Pittsburgh, PA: University of Pittsburgh Press.

22.

Nalbandian

(1989). The contemporary role of city managers. American Review of Public Administration, 19, 261-277.

23.

Nalbandian

(1991). Professionalism in local government: Transformations in the roles, responsibilities, and values of city managers. San Francisco, CA: Jossey-Bass.

24.

Nalbandian

(2006). Politics and administration in local government. International Journal of Public Administration, 29, 1049-1063.

25.

Nalbandian

Negron

(2008). Defining facilitative leadership: A view from inside the mayor’s office in Lawrence, Kansas. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 37-54). Boca Raton, FL: Taylor & Francis.

26.

Newell

Ammons

D. N.

(1987). Role emphases of city managers and other municipal executives. Public Administration Review, 47(3), 246-253.

27.

Newell

Ammons

D. N.

(1995). Role emphases of city managers and other municipal executives. In Fredrickson

H. G.

(Ed.), Ideal and practice in councilmanager government (pp. 97-107). Washington, DC: International City County Management Association.

28.

Oakerson

R. J.

Parks

R. B.

(1988). Citizen voice and public entrepreneurship: The organizational dynamic of a complex metropolitan county. Publius: The Journal of Federalism, 18(1), 91-112.

29.

Protasel

G. J.

(1988). Abandonments of the council-manager plan: A new institutionalist perspective. Public Administration Review, 48, 807-812.

30.

Schneider

Teske

(1992). Toward a theory of the political entrepreneur: Evidence from local government. The American Political Science Review, 86, 737-747.

31.

Selden

S. C.

Brewer

G. A.

Brudney

J. L.

(1999). The role of city managers: Are they principals, agents, or both? American Review of Public Administration, 29, 124-148.

32.

Sparrow

(1984). The emerging chief executive: The San Diego experience. National Civic Review, 74, 538-547.

33.

Svara

J. H.

(1998). The politics-administration dichotomy model as aberration. Public Administration Review, 58(1), 51-58.

34.

Svara

J. H.

(1999a). The shifting boundary between elected officials and city managers in large council-manager cities. Public Administration Review, 59(6), 44-53.

35.

Svara

J. H.

(1999b). Complementarity of politics and administration as a legitimate alternative to the dichotomy model. Administration & Society, 30, 676-705.

36.

Svara

J. H.

(2001). The myth of the dichotomy: Complementarity of politics and administration in the past and future of public administration. Public Administration Review, 61, 164-171.

37.

Svara

J. H.

(2002). City council’s roles, performance, and the form of government. In Frederickson

H. G.

(Ed.), The future of local government administration (pp. 213-226). Washington, DC: ICMA.

38.

Svara

J. H.

(2006). The search for meaning in political-administrative relations in local government. International Journal of Public Administration, 29, 1065-1090.

39.

Svara

J. H.

(2008). Reexamining models of mayoral leadership. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 3-34). Boca Raton, FL: Taylor & Francis.

40.

Teske

Schneider

(1994). The bureaucratic entrepreneur: the case of city managers. Public Administration Review, 54, 331-340.

41.

Tesluk

P. E.

Jacobs

R. R.

(1998). Toward an integrated model of work experience. Personnel Psychology, 51, 321-355.

42.

Watson

D. J.

Schortgen

(2008). Guiding change in a first ring suburb: Plano, Texas. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 125-141). Boca Raton, FL: Taylor & Francis.

43.

Wheeland

C. M.

(2000). City management in the 1990s: Responsibilities, roles, and practices. Administration & Society, 32, 255-281.

44.

Wheeland

C. M.

(2004). Enhancing the governing body’s effectiveness. In Newell

(Ed.), The effective local government manager (3rd ed., pp. 57-82). Washington, DC: ICMA.

45.

Wheeland

C. M.

(2008). The power to persuade: Philadelphia, Pennsylvania. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 327-349). Boca Raton, FL: Taylor & Francis.

46.

Wikstrom

(1979). The mayor as a policy leader in the council-manager form of government. Public Administration Review, 39, 270-276.

47.

Wood

Gabris

Olson

(2008). Mixing models of leadership in a mayor-council city: A study of Yorkville, Illinois. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 73-105). Boca Raton, FL: Taylor & Francis.

48.

Yates

(1977). The ungovernable city. Cambridge, MA: MIT Press.

49.

Zhang

Yang

(2008). Building a town and its institutions: Midway, Florida. In Svara

J. H.

(Ed.), The facilitative leader in city hall: Reexamining the scope and contributions (pp. 169-186). Boca Raton, FL: Taylor & Francis.

50.

Zhang

Yang

(2009). Citizen participation in the budget process: The effect of city managers. Journal of Public Budgeting, Accounting and Financial Management, 2, 289-317.

51.

Zhang

Feiock

R. C.

(2010). City managers’ policy leadership in council-manager cities. Journal of Public Administration Research and Theory, 20, 461-476.

52.

Zhang

Liao

(2011). Participatory budgeting in local government: Evidence from New Jersey municipalities. Public Performance and Management Review, 35, 281-301.