Turmoil, Trouble, and Turnover Among City Administrators
Askew School of Public Administration and Policy Florida State University Tallahassee FL 32306
Department of Political Science University of Alabama at Birmingham
James C. Clingermayer
Department of Government, Law, and International Affairs Murray State University
Askew School of Public Administration and Policy Florida State University
Paper to be presented at the Southern Political Science Association meeting in New Orleans, January, 2003.
Turmoil, Trouble, and Turnover Among City Administrators
Turnover in top administrative positions has been linked to the management difficulties in public, private and non-profit organizations. Much of the research on managerial turnover in the private sector indicates that turnover has effects on those who remain behind in the organization and upon organizational performance (Sheehan, 2001; Droege and Hoobler, 2003; Shaw, Gupta, and Delery, 2002). Recognition of the importance of executive turnover has led to renewed interest in turnover in the executive ranks of local government by scholars of local politics. City manager turnover is particularly important because a range of studies indicates that most policy proposals adopted by the city councils in council- manager cities originate with the manager (Martin, 1990). Several empirical studies have explored what factors account for the tenure and turnover of managers (DeSantis and Renner, 1993; Renner, 1990; DeHoog and Whitaker, 1990; Whitaker and DeHoog, 1991; Feiock and Stream, 1999). Much of this research suggests tenure and turnover patterns of local government administrators are shaped by what have been commonly referred to as “push” and “pull” factors (see, e.g., DeHoog and Whitaker, 1991). Push factors include characteristics of an administrator’s current employment situation that encourage him or her to seek and find positions elsewhere. Push factors encompass perceived gaps in governmental policies and performance, political conflict in the community, and differences between the manager and council in regard to style, orientation, or policy. Pull factors, on the other hand, refer to the marketability of managers and their opportunities for professional, financial, or personal advancement in other positions. Clearly, administrators who are perceived as successful
will feel a greater “pull” from prospective employers than those administrators who do not appear successful. Previous work has explored how demographic factors and fiscal policies such as tax, spending and borrowing influence tenure, but report only modest effects (Feiock and Stream 2002). Discussions of manager turnover invariably focus on how change s in the
composition of the council can result in the exiting of a manager, but this relationship has been tested only indirectly because most work has relied on demographic characteristics of communities to measure potential conflict between manager and council. This paper adopts a different approach by directly measuring turnover on the elected council and examining how stability or change in council composition affects a manager’s tenure. Push factors have been generally been found to be more important than pull factors, but recent work has linked pull factors to the performance of the local economy (Feiock, et al. 2001). Because managers may be given credit for income growth in the community, successful economic development efforts can create job opportunities for local government managers. Recent work has suggested that managers who can creatively use development policies to attract growth can cash in these development gains in the job market by moving to a manager position in another community (Stein 1992; Feiock and Stream 2002; Feiock, Clingermayer and Hicks-Kopp 2004). The empirical analysis reported in this paper identifies how income growth in an administrators’ community influences his or her tenure in office. High levels of income growth are expected to increase manager turnover in council- manager cities. Because low income growth could also work as a push factor, we anticipate a curvilinear relationship between income change s and turnover. Managers of communities with
declining economic growth are more likely to be pushed from their position, while managers of fast growing communities are more likely to be pulled to other more attractive positions.
City Management Careers During the late 19th and early 20th centuries, many observers argued that a professional manager might best handle the administration of municipal affairs. Virtually all successful business organizations were recognized to have some type of chief executive officer who works for the board of directors and who provides the leadership necessary to make the corporation run. The city manager seemed to be the answer to the need for such an office in the “municipal corporation”. The city manager is a professional city administrator who is also chief executive in the council- manager city. The city manager is, by virtue of his or her experience or training, particularly well-equipped to handle the administrative problems peculiar to the running of a city (Blair, 1964). The professional organization of city managers, the International City Management Association (ICMA), provided a model city charter, which outlined the ideal relation between the manager and the council and was the basis of numerous actual city charters and many state-enabling statutes (Lineberry, 1974). Even in cities with mayor-council form of government there is typically a full time chief administrative officer, usually appointed by the mayor. Under a council- manager form of government, the city manager is hired by the council and is subject to removal, at any time, by a majority vote of the council. More recently, many city managers have been granted multi- year contracts that obligate the
city to employ (or at least pay) the manager for a specified period of time. In the early days of the council- manager plan, most managers were civil engineers by training (Lineberry, 1974). Today, approximately 80 percent of city managers have been trained in graduate programs in public policy/administration (Barber, 1988). They are familiar with financial administration, personnel management, municipal law, and planning. In the absence of a city manager, responsibility for administration falls to an elected mayor, or to a mayor’s appointees, such as a city administrator serving at the pleasure of the mayor. Such administration may not be characterized by the professional values or expertise that the training of a city manager would involve. City managers tend to be highly educated, well paid, and mobile professionals (Morgan and England, 1996). Over 85 percent do not come from the community they serve. A typical career path of a city manager might include work as an assistant manager, then appointment as manager in a small town, and, finally, similar posts in larger or more prestigious communities (Paul, 1981; Barber, 1988).
Implications of City Manager Turnover A recurrent theme the literature studying city managers is the leadership role managers play in policy making (Nalbandian 1999, 2000; Svara, 1991; Almy, 1975). In the early days of the council- manager form of government, city managers were viewed as technicians with a limited role in the policy- making process (Svara, 1991). However, scholars soon came to recognize the politics-administration dichotomy is often merely symbolic. Even if this is the case, Gary Miller argues that this myth functions to enhance
both the policy and administrative role of managers by providing a credible constraint on political intervention (Miller 2000). The issues that city managers bring before the council, the information they present to support their recommendations, and the directions they provide to employees to carry out public policy all involve the city manage r in the policy-making process (Svara 1988). Recognition of the important policy making role that managers play suggests change from one city manager to another will have significant policy implications if different managers bring different preferences, skills, and backgrounds. Turnover of city managers can have powerful consequences on policy choices in arenas involving public and private sector actors, long term obligations, or future commitments (Feiock and Clingermayer, 1993; 2001). Uncertainty resulting from turnover alters the transaction costs of policy decisions and the time horizons of leaders. Feiock and Clingermayer (1993) elaborate on these ideas by examining how uncertainty caused by executive turnover affects privatization. Their empirical analysis revealed that turnover in the city manager or appointed chief administrative officer position discouraged contracting and the use of alternative service delivery mechanisms (Clingermayer and Feiock, 1993). Additionally, turnover may narrow leaders’ time horizons. When there is rapid turnover, local officials are faced with electoral threats and political uncertainty. Given local government manager’s role in policy- making, it is particularly important to understand what influences the tenure of city managers.
What Do We Know About City Manager Turnover? The existing literatures on city manager turnover and tenure indicate managers leave for a variety of reasons. Many city mana gers are fired or forced to leave after disagreements with the city council that create a situation where he or she can not manage effectively. Managers also leave voluntarily to pursue career advancement in another city management position or other opportunities. The traditional career path of a city manager involves several jurisdictions, usually from smaller to larger communities. Of course, both forces can work together. Managers may decide to look for opportunities elsewhere because they perceive political turmoil on the horizon. In a pioneering study of 39 city manager turnovers in 10 Florida cities, Kammerer et al. (1962) found that two-thirds of these exits were involuntary terminations by the council. Most terminations were a direct result of political disputes. Political disagreements also indirectly influence voluntary terminations, as managers decide to leave before the conflict escalates to the point at which the council fires them (DeHoog and Whitaker 1990). The influence of turnover among council members has captured the attention of scholars and practitioners. Kaatz (1996) found that “dysfunctional” conflict on the council led to turnover among managers in some Chicago area municipalities. Managers typically cite council turnover as a primary reason for their departures (Vasser, 2000). Case study evidence indicates that change in the composition of the city council or commission creates tremendous difficulty for incumbent managers. Turnover in elected office may signal political controversy that would put politicians and administrators at odds. The conventional wisdom extolled at professional meetings is that big change in
the council make- up adversely impacts manager tenure. Surprisingly, this proposition has not been subjected to systematic empirical tests at the national level and, with the exception of Whitaker and DeHoog’s (1991) study of Florida managers, it has not been addressed comparatively at a sub-state level either. Community composition and conflict have also been linked to turnover. In heterogeneous areas cleavages based in wealth, race or other social conditions can result in conflicts that spill over to local government. Administrators serving in more affluent and racially homogeno us communities may experience less conflict and have a lower likelihood of turnover. Attempts to explain manager tenure based on conflict or ambition also direct our attention to institutions that define the powers and authority of managers. For example, DeHoog and Whitaker (1990) indicate that cities with popularly elected mayors were more likely to oppose incumbent city managers than mayors chosen from among the council (see also Svara, 1987, 1990; Booth, 1968; DeSantis and Renner 1994). Local government fiscal choices have also been linked to manager turnover. In communities with strong fiscal capacity, low taxes and low debt, the performance of managers may be uncontroversial. Policies such as development may also enhance a manager’s marketability. One way that managers promote development is through bond supported infrastructure and development projects. Elaine Sharp (1986) demonstrates that council manager cities were more likely to incur revenue bond debt in their development efforts than were mayor-council communities. Recent work has begun to examine how government performance can create employment opportunities for city managers. Examining administrators’ actual performance while in office is problematic given the difficulty of defining and/or
measuring good and bad performance. Yet, in studies of electoral cycles and retrospective voting (see, e.g., Fiorina, 1981), researchers have used the economic performance of a politicians’ jurisdiction as a proxy for perceived performance of the incumbent administration. Changes in economic growth may also be a good indicator of city manager performance. Economic development has been a great concern of managers. A casual examination of ICMA national meeting programs illustrates the enduring presence of development issue to the profession. Bob Stein (1990) suggested that successful city managers enjoy a national market for their services. Perhaps policy or development success could lead to turnover, with managers moving to more lucrative positions in larger communities or private firms. Presumably, administrators who serve at the pleasure of a city’s mayor would not be part of this national market, and therefore turnover in their positions would not be as closely tied to economic success. Recent research that has systematically examined the economic performance of city managers while in office. Feiock et al. (2001) used the economic performance of a politicians’ jurisdiction as a proxy for perceived performance of the incumbent administration and found economically growing more likely to retain city managers. Pervious work has examined only linear relationships between income growth and manager tenure. Either decline or rapid growth could lead a manager to leave. This indicates a curvilinear relationship between income growth and turnover.
Research Design This study advances an innovative design and employs a large national sample of city managers and city administrators to more precisely estimate the effects of economic
growth and change in the composition city councils on manager turnover. This model is estimated with a pooled cross sectional time series of 269 U.S. cities with populations of at least 75,000 in 1989. Turnover may be a normal event in smaller jurisdictions with less professionalized governments run by part-time, low ambition sorts of politicians. Focusing on large cities not only provides more complete data, it allows us to examine turnover in communities where more ambitious candidates contend for office and turnover among elected leaders represents greater conflict and uncertainty. Included in this sample are cities with mayor-council and also council- manager forms of government. This allows us to estimate the determinants of turnover of top city administrators who are not actually city managers selected by the city council. These administrators, whose careers are much less frequently examined in the public administration literature, often serve at the pleasure of a mayor. Yet they may perform many of the administrative tasks that city managers perform. We examine turnover from 1987 to 1999 using a fixed effects logit model. The dependent variable is a dichotomous measure indicating whether there has been turnover in the manager position in the previous year. These data were collected from va rious volumes of the National League of Cities’ Directory of City Policy Officials and the International City and County Management Association’s Municipal Yearbook.
Independent Variables The independent variables include measures of turnover among elected council members, demographic characteristics of communities, political system structures, and economic growth. We estimated models with a one-year lag for fiscal and demographic variables. Our main explanatory “push” variable is turnover among elected council
members. We measure of the rate of council turnover by calculating the proportion of members serving four years earlier no longer serving on the council. We expect that the greater the proportion of incumbent council members that leave, the more likely the manager will exit. We identified the members of local governing bodies each year from 1983-1999 and recorded all changes to derive the council non-return rates. Our model also includes various push factors linked to manager turnover such as census measures of city population, change in population, per capita personal income, change in bond ratings in the previous four years, and the average change in per capita personal income over the previous four years. Local policy variables included are own source revenue which is an indicator of fiscal capacity, per capita property taxes and per capita long-term debt. The other political system variable we include is a dichotomous indicator of whether the city operated under council- manager form of government. We expect managers serving in council- manager cities will be more likely to exit because of career opportunities available to through the national network of city managers and the ICMA. Council ma nager government may also have non-additive effects. Because we believe the effect of income change may be curvilinear (with low or negative growth leading to involuntary turnover and high growth leading to voluntary turnover) we included a squared income change term. This squared term for income growth was estimated as an interaction with council- manager form of government because the external market for successful managers is more relevant to managers serving in council manager governments. Data regarding form of government were derived from surveys conducted by the International City Manager Association (ICMA) in 1986, 1991, 1996 and 2001
and follow up inquires. Fiscal data were drawn from published and on- line reports provided by the Geographic Division of the U.S. Bureau of the Census.
Results: What Accounts for Manager Change The results of this analysis are reported below in Table 1. The data provide a good fit to the model. We find both push and pull factors influence the career paths of city managers. The turnover rate of city council members is, as expected, a strong predictor of administrative turnover. We find that the greater the proportion of council members that do not return to office, the greater the likelihood of a manager leaving. Population levels and racial composition made little difference. The form of government had no statistically significant additive effect, nor did bond ratings, debt levels, or change in tax levels. Table 1 here The most dramatic results pertain to the effects of income levels and growth, particularly in interaction with council- manager form of government. The level of per capita income is negatively related to turnover, suggesting that in a well-off community the manager is likely to be retained. But the change in per capita income level is positively related to turnover, indicating that administrators often leave communities that are growing in wealth. The interaction of council- manager government and income
change, however, is negative, while the interactive effect of council- manager government and the square of the income change is strongly positive. This provides strong support
for the predicted curvilinear relationship between income change and turnover, yet it only applies to city managers, not to non- manager city administrators.
Conclusion The findings reported here suggest that local government manager careers are subject to the forces of both political and economic change in the communities in which they serve. When there is substantial change on the city council, there is often change in top administration. This finding is consistent with conventional wisdom and with considerable extant research. Variables that dealt with the actual fiscal condition of the city government itself did not have a dramatic effect upon turnover. What did matter was the economic circumstances of the community itself. In council- manager cities, communities that
experienced little or no income growth and those that experienced a great deal of growth were both likely to lose their city mana ger. We expect that in most instances managers
in the cities that were faring poorly were forced out while those doing well were lured away by attractive outside employment opportunities. While we provide a less than complete accounting of potential pull forces, our strong results suggest the importance of career ambition and the value of development success on the manager job market. Future research should extend the model to include additional measures of economic conditions and development activity. We hope in the future to carry out additional survey research to gain additional information about administrators’ backgrounds, education, marital status, spouse’s education and employment, along with additional information that could help us assess the tradeoffs and interactions of push and pull factors in these career choices.
Barber, D.M. 1988. “Newly promoted city managers.” Public Administration Review, 48: 694-699. Blair, George S. 1964. American Local Government. New York. Berry, William, and Stanley Feldman. 1985. Multiple regression in practice. Sage University Paper Series on Quantitative Applications in the Social Sciences, 07-050. Beverly Hills, CA: Sage. Booth, D.A. 1968. “Are elected mayors a threat to city managers?” Administrative Science Quarterly, 12: 572-589. DeHoog, Ruth, and Gordon Whitaker. 1990. “Political conflict or professional advancement: Alternative explanations of city manager turnover.” Journal of Urban Affairs, 12: 367-77. DeHoog, Ruth, and Gordon Whitaker. 1991. “City managers under fire: How conflict leads to turnover.” Public Administration Review, 51: 156-165. DeSantis, Victor, and Tari Renner. 1993. “Contemporary patterns and trends in municipal government structures.” In The Municipal Yearbook, 57-69. Washington, D.C.: ICMA. DeSantis, Victor, and Tari Renner. 1994. “The impact of political structures on public policies in American counties.” Public Administration Review, 54: 291-295. Droege, Scott B., and Jenny M. Hoobler. 2003. “Employee Turnover and Tacit Knowledge Diffusion: A Network Perspective,” Journal of Managerial Issues, 15:5064. Feiock, Richard, and James Clingermayer. 1993. “Institutions, incentives, and city debt. “Prepared for delivery at the annual meeting of the Southern Political Science Association, Savannah, GA, November. Feiock, Richard, James Clingermayer, Christopher Stream, Barbara Coyle McCabe, and Shamima Ahmed. 2001. “Political Conflict, Fiscal Stress and Administrative Turnover in American Cities.” State and Local Government Review, 33: 101-108. Feiock, Richard and Christopher Stream. 2002. “The Implications of Local Government Structure for Manager Turnover and Policy Choice” in The Future of Local Government Administration: A Tribute to William Hansel . Washington DC: ICMA.
Feiock, Richard C. Melanie Hick-Kopp, and James Clingermayer. 2004. “City Managers as Residual Claimants on Local Government Performance” in preparation. Fiorina, Morris P. 1981. Retrospective Voting in American National Elections. New Haven: Yale University Press. Hill, Melvin, Jr. 1984. State Laws Governing Local Government Structure and Administration. Athens, GA: Institute of Government, University of Georgia. Holbrook, Thomas. 1993. “Institutional strength and gubernatorial elections: An exploratory analysis.” American Politics Quarterly, 21: 267-71. International City Management Association. 1984-90. The Municipal Yearbook. Washington, DC: ICMA. Kammerer, G.M., C.D. Farris, J.M. DeGrove, and A.B. Clubok. 1962. Cit y managers in politics: an analysis of manager tenure and termination. Gainesville: University of Florida Press. Kammerer, G.M. 1964. “Role diversity of city managers.” Administrative Science Quarterly, 8: 421-42. Kiser, Larry, and Elinor Ostrom. 1982. “The three worlds of action: A metatheoretical synthesis of institutional approaches.” in Strategies of political inquiry, Elinor Ostrom, ed., Beverly Hills, CA: Sage Publications. Lineberry, Robert L., and Edmund Fowler. 1967. “Reformism and public policies in American cities.” American Political Science Review, 61: 701-716. Lineberry, Robert L., and Ira Sharkansky. 1974. Urban politics and public policy, 2nd edition. New York: Harper and Row, Publishers. Lyons, William. 1978. “Reform and response in American cities: Structure and policy reconsidered.” Social Science Quarterly, 59: 118-131. MacManus, Susan. 1999. “The Resurgent City Councils,” in American State and Local Politics: Directions for the 21 st Century, Ronald Weber and Paul Brace (eds). Chatham House Publishers. New York. Martin, David L. 1990. Running City Hall: Municipal Administration in America, The University of Alambama Press. Miller, Gary J. 2000. "Above Politics: Credible Commitment and Efficiency in the Design of Public Agencies." Journal of Public Administration Research and Theory 10 (April): 289-328.
Morgan, David, and John. Pelissero. 1980. “Urban policy: Does political structure matter?” American Political Science Review, 74: 999-1006. Morgan, David, and Jeffrey Brudney. 1985. “Urban policy and city government structure: Testing the mediating effects of reform.” Paper presented at the annual meeting of the American Political Science Association, New Orleans. Morgan, David, and Robert England. 1996. Managing Urban America, Chatham, New Jersey, Chatham House Publishers. Nalbandian, John. 1991. Professionalism in local government. San Francisco: JosseyBass Publishers. Nalbandian, John. 1999. “Facilitating Community, Enabling Democracy: New Roles for Local Government Managers.” Public Administration Review, 59: 187-197. Nalbandian, John. 2000. “The Manager as Political Leader.” Public Management, March. North, Douglass C. 1990. Institutions, institutional change, and economic performance. Cambridge: University Press. Ostrom, Elinor. 1990. Governing the commons. New York: Cambridge University Press. Ostrom, Elinor. 1992. Crafting institutions for self- governing irrigation systems. San Francisco, CA: ICS Press. Renner, Tari. 1990. “Appointed local government managers: Stability and change.” In The municipal yearbook. Washington, D.C.: ICMA. Sharp, Elaine B. 1986. "The Politics and Economics of the New City Debt," American Political Science Review 80: 1271-1287. Shaw, Jason D., and Nina Gupta, John E. Delery. 2002. “Voluntary Turnover and Organizational Performance,” Academy of Management Proceedings, HR: B1-B6. Sheehan, Eugene P. 2001. “Affective Responses to Employee Turnover,” The Journal of Social Psychology 135: 63-69. Stein, Robert M. 1990. Urban Alternatives. Pittsburgh: University of Pittsburgh Press. Stillman, R.J., II. 1982. “Local public management in transition: A report on the current state of the profession.” in Municipal Yearbook, 1982. Washington, D.C.: ICMA. Stillman, R.J., II. 1974. The rise of the city manager. Albuquerque, NM: University of New Mexico Press.
Svara, James H. 1987. “Mayoral leadership in council- manager cities: Preconditions versus preconceptions.” Journal of Politics, 49: 207-227. Svara, James H. 1990. Official leadership in the city: Patterns of conflict and cooperation. New York: Oxford University Press. Whitaker, Gordon, and Ruth DeHoog. 1991. “City managers under fire: How conflict leads to turnover.” Public Administration Review, 51: 156-657. Vasser, Jan. 2000. Lewiston, Idaho city manager. Interview with the author. Svara, James H. 1990. Official leadership in the city: Patterns of conflict and cooperation. New York: Oxford University Press. Whitaker, Gordon, and Ruth DeHoog. 1991. “City managers under fire: How conflict leads to turnover.” Public Administration Review, 51: 156-65. Vasser, Jan. 2000. Lewiston, Idaho city manager. Interview with the author.
Table 1: Estimates of Local Government Manager Turnover Variable B Std. Err t .421540 2.340361 0.18 Log of Population .000021 .000017 1.28 Population Change -.512713 2.856923 -0.18 White Population -2.840338 6.101365 -0.47 Council Manager 3.573673** 1.530321 2.34 Council Turnover -.001547** .000196 -7.86 Income .010956** .001223 8.96 Income Change -.001973* .001099 -1.79 CM-Income Change 2.07e-06** 9.93e-07 2.08 CM-Income Change 2 -.742923 .6789385 -1.09 Bond rating change -.000256 .0002084 -1.23 Long-term debt .000686 .0010263 0.67 Fiscal Capacity .001002 .0025004 0.40 Property Taxes Log likelihood -225.25745 LR chi2 (13) = 286.42 N = 1753
*p < .10 **p < .05