Calitatea.docx

Noul management public a susținut o viziune a managerilor publici ca antreprenori ai unui guvern nou, mai ieftin și din ce în ce mai privatizat, emulând nu numai practicile, ci și valorile afacerilor. Susținătorii noului management public și-au dezvoltat argumentele în mare parte prin contraste cu vechea administrație publică. În această comparație, Noul Management Public va câștiga, bineînțeles, întotdeauna. Aici susținem că contrastul este mai mare cu ceea ce noi numim "Serviciul public nou", o mișcare construită pe munca în cetățenia democratică, comunitate și societatea civilă, umanismul organizațional și teoria discursului. Propunem șapte principii ale noului serviciu public, în special faptul că rolul primordial al funcționarului public este acela de a ajuta cetățenii să articuleze și să-și atingă interesele comune, mai degrabă decât să încerce să controleze sau să orienteze societatea. Un punct de plecare pentru înțelegerea PSM este construirea motivației, o preocupare cheie a cercetării științifice sociale și comportamentale moderne. Motivația se referă în mare măsură la forțele care energizează, direcționează și susțin comportamentul (Perry and Porter 1982). Deși motivația este deseori studiată în contextul muncii, nu limităm sfera revizuirii noastre la motivația muncii, parțial pentru că forțele nu sunt limitate doar de sarcini de lucru, ci implică forțe instituționale și de mediu, munca în sine și individual nevoile și motivele. A starting point for understanding PSM is the motivation construct, a key concern of modern social and behavioral science research. Motivation refers broadly to the forces that energize, direct, and sustain behavior (Perry and Porter 1982). Although motivation is often studied in the context of work, we do not limit the scope of our review to work motivation, in part because the forces are not bounded by work tasks alone, but involve institutional and environmental forces, the work itself, and individual needs and motives. In the last two decades, research about PSM and related constructs in other fi elds has grown signifi cantly. We describe and synthesize the development of three streams of related research, involving PSM, altruism, and prosocial motivation. Public service motivation. PSM originates from beliefs that unique motives are found among public servants that are diff erent from those of their private sector counterparts. In public administration, PSM has been defi ned in several diff erent but compatible ways. Perry and Wise defi ned PSM as “an individual’s predisposition to respond to motives grounded primarily or uniquely in public institutions and organizations” (1990, 368). Th e defi nition clearly sought to emphasize motives, such as civic duty and compassion, that are commonly associated with public organizations.1 In a subsequent analysis of PSM and government eff ectiveness, Rainey and Steinbauer off ered a more general defi nition of PSM. Th ey associated the construct with altruism in referring to PSM as a “general, altruistic motivation to serve the interests of a community of people, a state, a nation or humankind” (1999, 20). Th e Rainey and Steinbauer defi nition is similar to that of Brewer and Selden, who defi ned the concept as “the motivational force that induces individuals to perform meaningful … public, community, and social service” (1998, 417), emphasizing its behavioral implications and applicability beyond the public sector. Th e most recent variation of the defi nition within public administration emanates from research in Europe by Vandenabeele. He defi ned PSM as “the beliefs, values and attitudes that go beyond self- interest and organizational interest, that concern the interest

of a larger political entity and that motivate individuals to act accordingly whenever appropriate” (2007, 547). Altruism. Public administration defi nitions of PSM invoke the concepts of both self-sacrifi ce (Perry 1996; Perry and Wise 1990) and altruism (Rainey and Steinbauer 1999). Th us, research about altruism has direct relevance for research about PSM. Piliavin and Charng observed that altruism traditionally has been defi ned in terms of costs, but they argued that motives should be central to its defi nition and focus on acts that appear “to be motivated mainly out of a consideration of another’s needs rather than one’s own” (1990, 30). Economists have linked PSM directly to altruism. Francois referred to PSM as employees providing “eff ort out of concern for the impact of that eff ort on a valued social service” (2000, 275). In his research on public servant motivation and policy design, LeGrand concluded that “it is hard to dispute the view that altruistic motivations are prevalent among the providers of public services” (2003, 35). Prosocial motivation. A third line of research, most closely identifi ed with the fi eld of organizational behavior, is prosocial behavior, which encompasses a broad category of other-regarding behaviors (Brief and Motowidlo 1986). Some have argued that the meaning of prosocial behavior can be tied to an actor’s motives. Walster and Piliavin (1972), for instance, suggested that the defi nition should specify that the act is voluntary and without expectations for return, which places their defi nition very close to the meaning of altruism. Grant defi ned prosocial motivation as simply “the desire to expend eff ort to benefi t other people” (2008a, 49). Synthesis. Th is brief summary of PSM and related constructs suggests both convergence and divergence in the phenomena. Th e most prominent area of convergence is the emphasis on other orientation—represented by notions of self-sacrifi ce, altruism, and prosocial—across the motivation defi nitions. Th is convergence accords with our long-held understandings of the public service ethic. It also accords with recent research in organizational psychology and organizational behavior that identifi es a fundamental role for other orientation in explaining organizational behavior (De Dreu 2006; Meglino and Korsgaard 2004). With respect to divergence, the conceptions of PSM are more particular regarding objects of motivation than altruism and prosocial motivation, which are cast in general terms. “Consideration of another’s needs rather than one’s own” (Piliavin and Charng 1990, 30) and “desire to expend eff ort to benefi t other people” (Grant 2008a, 49) are quite broad. All of the PSM defi nitions draw boundaries: “motives grounded primarily or uniquely in public institutions and organizations” (Perry and Wise 1990, 368) “interests of a community of people, a state, a nation” (Rainey and Steinbauer 1999, 20) “public, community, and social service” (Brewer and Selden 1998, 417) “belief, values and attitudes that go beyond self-interest and organizational interest, that concern the interest of a larger political entity” (Vandenabeele 2007, 547) Th e boundaries placed on the scope of PSM suggest that it is a particular form of altruism or prosocial motivation that is animated by specifi c dispositions and values arising from public institutions and missions. Given the blurring of boundaries between sectors and diff erences in the location of the functions of government, we do not assert that PSM is uniquely found in government organizations

(Wise 2000). But we maintain it is grounded in the tasks of public service provision, and is more prevalent in government than other sectors (Rainey and Steinbauer 1999; Wise 2000). Because PSM has its roots in other orientation, it is conceptually distinct from self-interest, which is rooted in self-concern (De Dreu 2006), and from intrinsic motivation (Grant 2008a). De Dreu argues that self-concern and other orientation are orthogonal and unipolar, meaning that they are independent and vary from low to high. Grant (2008a) observes that prosocial motivation and intrinsic motivation diff er in that intrinsic motivation emphasizes pleasure and enjoyment as drivers of eff ort, but prosocial motivation emphasizes meaning and purpose as drivers of eff ort. Th e pursuit of public service motives is not contingent on feelings of pleasure or enjoyment. Measurement Although several fi elds have developed common ground around motivational constructs that emphasize other orientation, we still confront the diffi cult challenge of measuring PSM both for scholarly research and for practical applications. A detailed discussion of measurement issues is beyond the scope of this article, but we identify here the variety of measurements and some of their implications. At least four diff erent approaches have been used to measure PSM. Th ey include (1) single survey items about public service (e.g., Rainey 1982), (2) unidimensional scales (e.g., Naff and Crum 1999), (3) multidimensional scales (e.g., Perry 1996), and (4) behavioral proxies, such as whistle-blowing (e.g., Brewer and Selden 1998). • • • • With respect to divergence, the conceptions of PSM are more particular regarding objects of motivation than are altruism and prosocial motivation, which are cast in general terms. Revisiting the Motivational Bases of Public Service 683 Rainey’s (1982) initial eff ort to measure PSM singled out one reward preference item, “engaging in meaningful public service,” as an indication of PSM. Th e source for many of the measures of PSM is Perry’s (1996) 24-item scale, which is composed of four subscales: attraction to public policy making, commitment to civic duty and the public interest, compassion. and self-sacrifi ce. In some instances (Taylor 2007), the aggregate of the scale has been used; in other research, the four subscales have been used individually (Perry 1997) and in various combinations (Moynihan and Pandey 2007a, 2007b). Th e self-sacrifi ce subscale calls attention to the roots of PSM in altruism and prosocial motivation. At the same time, the other subscales point to PSM’s focus on public institutions, especially the attraction to policy making and commitment to civic duty and the public interest subscales. Th e development of a survey-based measure for PSM (Perry 1996) has been useful for facilitating comparisons across disparate services and national settings (see, e.g., Kim 2009; Liu, Tang, and Zhu 2008; Taylor 2008; Vandenabeele et al. 2009), informing research in other disciplines (Francois 2000;

Georgellis, Iossa, and Tabvuma 2008; Grant 2008b), and creating foundations for cumulating results. At the same time, the variety of measures used has several consequences (Wright 2008). One is that the results of empirical analyses are not fully comparable across studies. Another is that the meaning of PSM has the potential to shift from study to study. We acknowledge that cultural and language diff erences make such shifts in the meaning of any construct likely and limit comparison, but urge scholars to pursue eff orts to achieve converging meanings and recognize diff erences in measurement and defi nition when interpreting fi ndings. Summary During the past 20 years, scholarship about PSM and related constructs has fl ourished, bringing important advances across several fi elds. A critical mass of research about other-oriented motivation creates prospects for signifi cant improvements in our understanding of PSM during the next decade. Within public administration, the increase in research activity has produced a proliferation of methods used to measure PSM. Although this proliferation has helped expand research more rapidly than might otherwise have occurred, the variety of measures could impede cumulating fi ndings. D espite the variety of constructs and measures, research across several disciplines using diff erent methods and measures leads to similar inferences. We now turn to a review of that research. What Have We Learned from Public Service Motivation Research? We examine three propositions originally off ered by Perry and Wise (1990) in light of subsequent research. We assess the validity of the propositions given subsequent empirical research, what we now know, and what gaps exist in our knowledge related to each of the three propositions. Attraction-Selection-Attrition In the original formulation, the fi rst relationship Perry and Wise (1990) proposed was that between PSM and the likelihood of an individual selecting a public organization. Th ey posited, Proposition 1: Th e greater an individual’s public service motivation, the more likely the individual will seek membership in a public organization. Th e rationale for the prediction was straightforward: Perry and Wise reasoned that individual behavior would be infl uenced by the magnetic eff ects of individual identity and organization characteristics. Individuals who are high in PSM would seek out contexts compatible with their dispositions, and the contexts to which they likely would be attracted were organizations that satisfi ed their prosocial and altruistic orientations. Review of public administration research. Public administration research on this proposition is limited but generally supportive. In a formative study associated with the beginnings of PSM research, Rainey (1982) found that public managers valued meaningful public service more highly than private managers and the preference for public service was signifi cantly related to job satisfaction. More recent research reinforces Rainey’s (1982) fi ndings. In an analysis of U.S. General Social Survey (GSS) data from 1989 and 1998, Lewis and Frank (2002) found signifi cant positive associations between a desire to help others and to be useful to society and preferences for government jobs. Th ey suggested that these relationships might be stronger for college graduates, younger employees, and specifi c employment classifi cations such as education. Vandenabeele (2008), using a sample of 1,714 advanced master’s degree students at Flemish universities, found that PSM was positively correlated with student preferences for prospective public employers. Th e association of student preferences was stronger for government organizations classifi ed as high, in contrast to low, publicness. An analysis of a large Dutch

data set (Steijn 2008) showed that public sector workers had higher levels of PSM than private sector workers. Interestingly, private sector workers with high levels of PSM were more likely to be looking for public sector jobs. Research on attrition from government organizations also supports the original proposition. Crewson (1997) linked PSM empirically to higher organizational commitment and lower turnover. Using data from the 1997 Merit Principles Survey (MPS), Naff and Crum (1999) found a positive association between PSM and intent to remain. Steijn’s (2008) Dutch study showed that workers with high PSM fi t were more satisfi ed and less inclined to leave their jobs and the organization they work for than workers without such a fi t. A recent study (Wright and Christensen 2010), using a panel data set of employment information for attorneys, produced mixed fi ndings and off ers insights into why the PSM and attraction-s electionattrition relationships are more nuanced than Perry and Wise (1990) originally projected. Wright and Christensen (2010) found that a strong interest in social service and helping others did not predict the employment sector of a lawyer’s fi rst legal job, but it did increase the likelihood of holding subsequent jobs in the public sector. Th e authors suggested that both initial sector choices and retention were aff ected by other factors that moderated the infl uence of PSM. Th ey concluded that the key question for research is when and under what conditions PSM aff ects employee attraction and retention. Other studies have also addressed this question. Steijn (2008) found that PSM’s eff ect 684 Public Administration Review • September | October 2010 on public employee job satisfaction and intention to stay in their jobs was stronger when employees felt that their work was useful to society. Similarly, Taylor (2008) found that employee PSM did not predict organizational commitment (or job satisfaction) unless employees also felt that their job provided opportunities to satisfy their PSM. Th ese studies point to the i mportance of understanding task or work role choices for motivation (Rainey and Steinbauer 1999; Wise 2004). Social and behavioral science evidence. Economists have begun to contribute important research on PSM’s relationships to attraction-selection-attrition. Gregg et al. (2008) studied donated labor (unpaid overtime) in caring industries in the United Kingdom. Th ey found that people with high PSM in caring industries were more likely to move to fi rms in the public or nonprofi t sector. A series of studies by Delfgaauw and Dur (2007, 2008a, 2008b) modeled self-selection decisions in a perfectly competitive economy. Th ey concluded that when prosocial motivation is not a revealed attribute of applicants, increasing the wage raises the probability of attracting workers with lower prosocial motivations. A study by Georgellis, Iossa, and Tabvuma (2008) used data from the fi rst 14 waves of the British Household Panel Survey covering the period 1991–2004. Th eir analysis focused on 747 transitions from the private to the public sector within the data. Th ey concluded that a signifi cant share of individuals move to the public sector because of the higher likelihood of fulfi lling their PSM. Utility of the research and knowledge gaps. As we noted earlier, the research related to attractionselection-retention is beginning to demonstrate that the eff ects of PSM are more nuanced than Perry and Wise (1990) projected. Th e results are promising, however, with respect to PSM as a factor in attraction and retention. Research from public administration and economics holds out the near-term prospect of creating usable knowledge for improving recruitment, selection, and retention in public organizations.

A line of research that bears on attraction-selection-attrition that recently has received attention but has not been fully assimilated into PSM research is person–environment fi t models. Th e person– environment fi t models refer to several diff erent ways of assessing the congruence between individuals and the settings in which they work (Kristof-Brown, Zimmerman, and Johnson 2005). Th e logic of the person–environment fi t (i.e., how congruent an individual is within a job or social unit) models is that individuals are likely to behave according to their essence if they are well adapted. Several recent studies have used various forms of person–environment fi t models to analyze the eff ects of PSM (Bright 2008; Steijn 2008; Vandenabeele 2008). Bright (2008) found a strong, signifi cant relationship between PSM and person–organization fi t. Given the theoretical case and the prospects that this research will help identify more completely factors infl uencing attraction and retention, we believe this line of research merits continuing attention. Performance Th e second proposition in Perry and Wise (1990) focused on the relationship between PSM and individual performance. It reads, Proposition 2: In public organizations, PSM is positively related to individual performance. Perry and Wise noted at the time, “Systematic empirical evidence about the relationship between PSM and performance does not exist” (1990, 371). Th eir arguments for the proposition rested on two premises. First, public jobs would be intrinsically motivating for individuals with high PSM because these individuals would embrace work characterized by attributes such as high task signifi cance. Second, PSM is likely to aff ect positively organizational commitment, which infl uences prospects for reliable role behaviors and innovative activities, both of which are critical for high individual performance. Review of public administration research. Several studies have tested the proposition by using selfreported measures of individual performance (Alonso and Lewis 2001; Bright 2007; Leisink and Steijn 2009; Naff and Crum 1999; Vandenabeele 2009). In an early study, Naff and Crum (1999) found a positive relationship between PSM and self-reported performance appraisals among U.S. federal employees. In a subsequent study of federal employees with data from the 1991 Survey of Federal Employees and the 1996 MPS, Alonso and Lewis (2001) used job performance ratings and grade to measure performance. Using a diff erent model specifi cation for the 1996 data than Naff and Crum (1999), Alonso and Lewis were able to replicate Naff and Crum’s fi ndings that PSM had a positive impact on performance ratings; however, they were unable to fi nd a relationship between valuing service to others and higher ratings in the 1991 data set. PSM also had no association with grade level in 1996, and valuing service to others was negatively related to grade level in 1991. Although Alonso and Lewis acknowledged that differences across the data sets in key measures made multiple interpretations of their fi ndings plausible, their analysis raises legitimate questions about the original proposition. Th ree of the studies that used self-reports of performance also incorporated mediators into their models. Bright (2007) used selfr eported performance as the dependent variable in a model that tested person–organization fi t as a mediator of the relationship between PSM and performance. He found indirect eff ects of PSM mediated by the fi t measure, but no independent direct eff ects in a sample of 205 public health care employees representing three states and three levels of government. In a study of a large sample of Flemish civil servants, Vandenabeele (2009) concluded that there was a positive and signifi cant relationship between PSM and self-reported p erformance. In contrast to Bright, Vandenabeele found support for both direct and indirect eff ects on performance. Th e indirect eff ects on the motivation–performance relationships were mediated by job satisfaction and normative and aff

ective c ommitment. Leisink and S teijn (2009) analyzed the eff ects of PSM on three performancer elated outcome variables—commitment, willingness to exert eff ort, and perceived job performance— among 4,130 Dutch public employees from all levels and varieties of public services. Th e hypothesis that person–organization fi t mediates the relationship between PSM and the outcome variables was rejected. Th e eff ects of PSM and fi t on the outcome variables were independent. Th e reasons underlying variations in the role of mediators and direct and indirect eff ects of PSM on performance merit attention in future research. Two other studies tested a model proposed by Rainey and Steinbauer (1999) in which PSM is one of three types of motivation (public service, task, and mission) that are the proximate determinants of organizational eff ectiveness, rather than individual performance, in Revisiting the Motivational Bases of Public Service 685 government. Th e fi rst of these studies (Brewer and Selden 2000), using data from the 1996 MPS, found a positive and signifi cant relationship between PSM and perceived organizational eff ectiveness. Based on responses from 1,739 full-time public employees in nine central government, fi ve provincial, and 26 lower-level local government agencies in Korea, Kim (2005) found that PSM was a signifi cant positive infl uence on perceived organizational eff ectiveness, replicating the results of Brewer and Selden (2000). Switching from organizational eff ectiveness to effi ciency, Ritz (2009) used a three-item survey-based measure of internal effi ciency as the dependent variable in a sample of 13,532 Swiss federal employees from seven ministries. He regressed seven independent variables, including two dimensions of PSM— attraction to public policy making and commitment to the public interest—against the dependent variable. Commitment to the public interest was signifi cant, but attraction to public policy making was not. In addition to studies of individual and organizational performance, scholars have also looked at discrete variables that represent facets of performance or could mediate the motivation–performance relationship. Brewer and Selden (1998) concluded that PSM was positively related to propensity to “blow the whistle.” Two more recent studies (Kim 2005; Pandey, Wright, and Moynihan 2008), one based in Korea and the other in the United States, found positive associations between PSM and organizational citizenship behavior. Andersen (2009) investigated the infl uence of inferred PSM and professional norms on the performance of a sample of 24 public and private health professionals in Denmark. Specifi c, discrete behaviors were used as performance measures. PSM was at the same high level for public and private health professionals so that it did not aff ect performance, which did vary with professional norms and economic incentives. Social and behavioral science evidence. Research in organizational behavior and economics supports a positive relationship between PSM and performance. Grant (2007) proposed that the social architecture of jobs can be used to reinforce task signifi cance and thereby prosocial motivation. Subsequent empirical research (Grant 2008a, 2008b) on public university fund-raisers and fi refi ghters supports his theoretical arguments. Th e eff ects of experimental manipulations of task signifi cance aff ected not only performance, but also persistence and productivity.

In economics, research by Francois (2000) and LeGrand (2003) is noteworthy. Francois’s essay “Public Service Motivation as an Argument for Government Provision” (2000) acknowledged that economists have not taken seriously the claims of public administration scholars about a public service ethic. Using formal mathematical modeling, Francois demonstrated that when PSM exists, conditions can be created for government bureaucracy to better obtain eff ort from employees than a standard profi t-maximizing fi rm. LeGrand’s book Motivation, Agency, and Public Policy (2003) developed a complex argument grounded in two constructs: motivation and agency. LeGrand argued that the motivation of public servants is complex, but includes both self-interested and altruistic motives. He coupled his analysis of motivation with agency—that is, the capacity to take desired action. He concluded that public policies should be designed to create quasi-markets in which public agencies compete with for-profi t fi rms in the delivery of services. Th e consequence of such quasi-markets is that they serve to harness altruistic motivations of all providers, public and private. Although Francois and LeGrand support a relationship between PSM and performance, a limitation of their research is that it relies on formal models rather than empirical analysis. Utility of the research and knowledge gaps. At this juncture, the research points to the conclusion that PSM matters for performance, but a good many questions remain unanswered about the degree to which it matters and whether its eff ects are collective rather than individual. Indeed, an interesting aspect of the evolution of the research is that scholars have developed cases at both the individual and collective levels for the contribution of PSM. Th is is a departure from Perry and Wise’s (1990) focus on individual performance, but opens up both research and institutional design options that were not considered 20 years ago. Th e existing research also points to the complexity of motivation, the need to consider competing explanations of bureaucratic behavior, and the importance of contextual factors (Wise 2004). A research gap highlighted by the contributions of diff erent disciplines is that public administration research has focused primarily on the individual level of analysis, while economics research has emphasized institutional design. It would be advantageous for these two streams of research to intersect. Public administration research could promote intersection were it to give more attention to the institutional environment—such as the implications of ownership of the fi rm (public, commercial, nonprofi t) or the incentive structure in the workplace. Two signifi cant but less fundamental issues that need attention in future research are the heavy reliance on self-reports in performance studies (Hondeghem and Perry 2009) and the role of mediators such as person–organization fi t in the PSM–performance relationship (Bright 2007; Brewer 2008; Wright and Pandey 2008). Wright and Pandey (2008) proposed a model that helps explain inconsistencies in motivation–performance research. Th ey suggested that the eff ect of PSM on job satisfaction is mediated by employee–organization value congruence, a form of person– organization fi t. Th eir model off ers promising avenues for future research. Another interesting question is how PSM and performance interact over time. Research to date views PSM as a driver of performance, but we also need research that investigates how eff ective organizations might stimulate or inculcate public service motives among their employees (Moynihan and Pandey 2007b; Rainey and Steinbauer 1999) and whether PSM is a static or dynamic attribute (Wise 2004). Th e motivational framework outlined in Shamir, House, and Arthur (1993) supports the proposition that PSM is higher in high-performing public services. Research on the causal relationships

between PSM and performance could be highly informative for both theory and practice. Finally, research clarifying the relationships between PSM and diff erent types of performance (individual/ collective, effi ciency/eff ectiveness) would advance understanding. Research in organizational behavior and economics supports a positive relationship between PSM and performance. 686 Public Administration Review • September | October 2010 Organizational Incentive Structures Th e third and fi nal proposition in Perry and Wise (1990) focused on the relationship between PSM and the structure of organizational incentives. It reads, Proposition 3: Public organizations that attract members with high levels of public service motivation are likely to be less dependent on utilitarian incentives to manage individual performance eff ectively. Th e rationale for the proposition was based on Knoke and Wright-Isak’s predisposition-opportunity model, which conceptualized motives as “predispositions to act under appropriate external cues” (1982, 210). If PSM represents an individual’s predispositions to act, then the incentives (i.e., external cues) that organizations are prepared to off er members for their commitments represent the opportunity side of the model. Knoke and Wright-Isak developed a typology composed of eight diff erent types of organizational incentive systems representing familiar schemes, including pure utilitarian and service incentive systems. Th e proposition does not imply that public employees are devoid of utilitarian motives and unconcerned with their level of compensation as sometimes interpreted by subsequent studies (Wise 2000); rather, it is about the degree of dependence on utilitarian and monetary rewards within organizations for stimulating individual performance. Review of public administration research. Th is proposition has empirical support in the public administration literature, but much of the evidence focuses on how highly public employees value fi nancial rewards relative to their private sector counterparts or relative to some portfolio of rewards. Reviews prior to 1990 (Perry and Porter 1982; Rainey, Backoff , and Levine 1976) summarized this evidence, indicating that fi nancial rewards were less important than nonpecuniary rewards. Relatively little empirical research in public administration has looked at incentive systems either in the precise form expressed in the proposition or as conceived by Knoke and Wright-Isak. Regardless of the consensus in reviews prior to 1990, more recent research is mixed, so that, like propositions 1 and 2, we have reason to believe the incentive relationships are more nuanced than originally thought. One study that illustrates the diffi culty of interpreting research fi ndings with respect to proposition 3 is Crewson’s (1997) analysis of PSM. Crewson used data from the 1989 GSS and a 1994 survey conducted by the Institute of Electrical and Electronics Engineers (IEEE). He compared private and public sector respondents on the importance of pay and did not fi nd the diff erence to be statistically signifi cant in either the 1989 GSS sample or the 1994 IEEE sample (Crewson 1997, 504). Despite the absence of a diff erence on pay, he identifi ed diff erences on other items. Crewson concluded that “although there is no signifi cant diff erence between sectors in the importance placed on high pay, public employees rate other extrinsic rewards lower in importance than do employees from the private sector. In turn, intrinsic rewards are more important to public employees than to those employed in the private sector” (503–4). Th us, Crewson’s conclusion is consistent with the proposition that public organizations are less

dependent on utilitarian incentives, even if he did not sustain fi ndings from previous reviews that fi nancial incentives are less important. Karl and Sutton (1998) and Bright (2005, 2009) reported inverse relationships between PSM and preferences for monetary rewards. In contrast, Alonso and Lewis (2001), analyzing responses from two large-n surveys of federal employees in 1991 and 1996, found no evidence that the link between material rewards and performance mattered any less to those with high PSM. Frank and Lewis (2004), using 1989 and 1998 GSS data, found that interaction terms intended to capture public–private diff erences in a range of extrinsic and intrinsic rewards did not approach statistical signifi cance. Th is led them to conclude that “[i]n both sectors, an interesting job that allows one to help others and a strong desire for job security appeared to increase the probability that one will put in extra eff ort, and the size of the eff ect appeared to be about the same in both sectors” (2004, 46). Th ese fi ndings suggest that prosocial tasks m otivate performance regardless of sector and undermine the notion that people with high PSM discount monetary rewards. In a study of Flemish university graduates, Vandenabeele et al. (2002) found that those who wanted to work for government attached more importance to the social signifi cance of a job as well as to job characteristics related to quality of life, but this did not mean that they attached less importance to extrinsic job factors. Salaries and fringe benefi ts and promotion opportunities were of equal importance for those who wanted to work for government and those who did not. Th e results can be interpreted as both supporting the social signifi cance of work tasks as a motivator while refuting the notion that concern for utilitarian incentives is absent among those who seek public service. A review of all the public administration research on reward preferences and incentives is beyond the scope of this article, but Wright’s (2007) analysis suggests a promising path to reconcile the ambiguous results reported here regarding proposition 3. He used goal theory to assess the infl uence of organizational mission and extrinsic rewards on work motivation among 807 employees in a New York State agency. Wright concluded that “the intrinsic rewards provided by the nature or function of the organization may be more important to public sector employees than—or compensate for the limited availability of—performance-related extrinsic rewards” (2007, 60). Wright’s approach using goal theory and similar theoretical eff orts to study incentives in more integrated ways merits attention in future research. Social and behavioral science evidence. Scholarship in psychology, economics, and political science is making signifi cant contributions toward illuminating the relationship between motivation and incentive structure. Th e longest-standing line of research, originating with Deci and his collaborators (Deci, Koestner, and Ryan 1999; Deci and Ryan 2004), posits that a variety of types of tangible contingent rewards undermine intrinsic motivation, but unexpected and task-non-contingent rewards have no eff ect on intrinsic motivation. Th e experimental evidence suggests that recipients experience positive feedback that can amplify intrinsic motivation diff erently from tangible rewards, such as contingent pay. Th e Deci and Ryan research has infl uenced motivation crowding theory (Frey 1997; Frey and Jegen 2001), which originated in economics. Frey and Jegen summarized the two main premises of motivation crowding theory: (1) all interventions originating from Revisiting the Motivational Bases of Public Service 687

outside the person under consideration—both positive monetary rewards and regulations accompanied by negative sanctions—may aff ect intrinsic motivation, and (2) external interventions may crowd out or crowd in intrinsic motivation (or leave it unaff ected) (2001, 592). One inference that some economists have drawn from this line of research is that lower-powered incentives (e.g., nonpecuniary rewards) are optimal in public organizations (Francois and Vlassopoulos 2008). Scholarship in economics (Delfgaauw and Dur 2008a, 2008b; Georgellis, Iossa, and Tabvuma 2008) puts forward the notion that PSM advances the interest of a cost-minimizing government because it provides an argument for an employer to off er weaker fi nancial incentives than private fi rms do. Proposition 3 indicates that public service organizations could off set costs for fi nancial rewards by relying on nonutilitarian incentives if they are populated by employees with strong PSM. Whether public service organizations should off er lower fi nancial rewards, however, is a normative question. A principal–agent analysis of public bureaucratic incentives by two political scientists came to conclusions compatible with the research from psychology and economics. Miller and Whitford (2007) argued that principal–agent theory posits that it is in the principal’s interest to fi nd incentives that channel the agent’s self-interest toward effi cient levels of eff ort from the agent. Th ey demonstrated, however, that the principal’s self-interest in public settings militates against off ering such incentives because bonuses large enough to produce the effi cient incentive eff ect in public organizations are prohibitively expensive for the principal, creating what they called the “principal’s moral hazard constraint.” Th ey suggest that the solution for the principal’s dilemma is to revert to using motives underlying PSM. Utility of the research and knowledge gaps. Although research has produced mixed results, some observers (Moynihan 2008) have contended that the substitution of economic rewards for the service ethic has damaged public service missions. Th e growing body of research from across several disciplines led the author of a recent review to conclude, “Rebuilding public sector motivation is viewed as a way to improve public service quality and volume without incurring the transaction/monitoring costs associated with ‘higher powered’ incentives such as performance-related pay” (Myers 2008, 6). A strategic question for future research is how to create balance between utilitarian and service incentive systems. Both LeGrand (2003) and Wise (2004) have contended that one issue that should be high on the future agenda is the composite of motivations that aff ect behavior. LeGrand wrote, “Altruism exists alongside more self-interested motivations, and is combined with them to aff ect behaviour in diff erent ways” (2003, 35). Wise was more direct about the need for holistic models: “If we look only for evidence to support the existence of public service motives…we cannot obtain a picture of the complexity of human behavior in a given organization” (2004, 670). Let us off er one example where more encompassing research— research addressing multiple motives and situations—could off er strategic payoff s. Th is more encompassing research could involve at least two individual motivational components, PSM and the need for security and job security rights, which is an organizational policy variable. Th is set of variables has been at the center of debates about American public administration for more than a century. Job security has been a central tenet of civil service since the Pendleton Act of 1883. Th e attack on the job security of government employees continues unabated today, with states, such as Georgia and Florida, and federal agencies, such as the Department of Homeland Security, promoting at-will employment as an alternative to traditional job tenure systems (Kellough and Nigro 2006). We also know that the need for security and public service

are important motivators, particularly in traditional civil service systems. How do variations in these two factors infl uence the motivation of civil servants? What are the behavioral consequences of relaxing property rights in government jobs? How do diff erent combinations of individual PSM and need for security infl uence the behavior of civil servants? Conclusion Th is article reviewed empirical research about public service motivation conducted during the last 20 years. Th e anchor for the review was three propositions proposed by Perry and Wise (1990). We conclude here with some general observations about the research, the state of our knowledge, and additional issues that deserve attention in future research. Th e review refl ects that public service motivation, broadly defi ned, is a construct that has attracted attention in several disciplines. At the heart of the construct is the idea that individuals are oriented to act in the public domain for the purpose of doing good for others and society. Diff erences are evident across disciplines, particularly with respect to the drivers for the other orientation. Th e commonality, however, is that human behavior is driven by other-regarding motives, not only by self-concern and selfinterest. Rational choice theories, grounded in assumptions of self-interested behavior, have had an important infl uence on public management theory and practice in recent decades. PSM off ers an alternative perspective for public management research and practice. A second conclusion is that research on PSM has diff used widely. Although PSM research was initially concentrated in the United States, it is now conducted in Europe, Asia, Australia, and South America. Th e diff usion has created challenges for the conceptualization and operational measurement of PSM. Th e values associated with public service are quite diff erent around the globe. It is likely that the meanings of PSM vary as well and are less institutionalized in some countries than in others. Th is makes comparative, crossnational research diffi cult, but a necessary focus in itself. As the internationalization of research proceeds, scholars and professionals need to be attentive to linguistic, contextual, and cultural considerations. With regard to Perry and Wise’s (1990) proposition 1, the research supports the view that PSM is important in attraction-selectionretention processes. At the same time, it is one of a constellation of factors. One has to be aware that public service motives are not the only—perhaps not even the most important—factors in the attraction-selection-retention process. Th e question thus is: what is A strategic question for future research is how to create balance between utilitarian and service incentive systems. 688 Public Administration Review • September | October 2010 the relative importance of PSM compared to other motives, and are there diff erences among applicants and incumbents related to work roles and tasks (Leisink and Steijn 2008; Wise 2004). With regard to proposition 2 on individual performance, empirical research appears to support a relationship, but the scholarship to date begs for some important refi nements. Th e role of intermediate variables, mediating the relationship between PSM and performance, is still unclear. Researchers should also examine the assumptions and linkages for how PSM contributes to higher mission performance and public organization eff ectiveness, as Rainey and Steinbauer (1999) and others (LeGrand 2003) propose. Th e original Perry-Wise proposition focused only on individual performance.

Th e relationship between PSM and individual and organizational performance is a complex issue that merits careful attention, including the direction of causality and the roles of intervening variables. With regard to proposition 3 on organizational incentives, the empirical evidence shows that researchers vary in their understanding of the proposition. More nuance is needed in both the research conducted and how fi ndings are interpreted. Research is confounding whether public employees attach less importance to material rewards than their private sector colleagues overall. Public service organizations appear to have a broader range of incentives because they can call upon the social signifi cance of public jobs. Intrinsic rewards might, therefore, be more important than performancerelated extrinsic rewards. Th is also means that public organizations are more apt to use lower-powered incentive structures. Motivation crowding theory suggests that this is a rational strategy that avoids undermining intrinsic motivation. Th e conclusions with regard to the three propositions argue strongly for more holistic research in the future. Although a relatively large body of research has been published on PSM during the last two decades, much of it has focused on defi nition, measurement, and incidence. Th ese issues continue to merit attention in future research, but the foundation has been laid across disciplines so that scholars can take up other, more substantive and complex issues. Wise (2004) has indicated several avenues that might be pursued in future research that looks more holistically at PSM. One avenue is to look at other motives and human needs besides public service motives, as illustrated earlier, where we suggest joining research on PSM with individual needs for security. Another avenue is to look at situational factors that play a role in explaining when public service motives surface and dominate individual behavior and when behavior occurs as a consequence of other motives. A third avenue is to look at contextual factors related to individual actors, situations, events, and organizations in which behavior occurs that infl uences the strength of PSM. Notable strides have been made in studying the motivational bases of public service since 1990. It is worth highlighting that much of the research directly focused on PSM has appeared since 2000, so progress has been especially rapid during the last decade (Perry and Hondeghem 2008). It is also noteworthy that progress spans disciplines and countries. Beyond progress, future research needs to explore new directions that entail closer integration with other disciplines, measurement advances, and new methodological strategies for advancing knowledge. Important work also needs to be done in translating PSM research into managerial practice. Th e research reviewed here indicates signifi cant progress, but now new challenges lay ahead for scholars and public professionals. Acknowledgments Th e fi rst author’s participation in this research was supported by the World Class University program of the Korea Science and Engineering Foundation, funded by the Ministry of Education, Science and Technology (R32-20002). Notes 1. Among the organizations we would include as public organizations are both government organizations and nonprofi t organizations, which confer public benefi ts. References Alonso, Pablo, and Gregory B. Lewis. 2001. Public Service Motivation and Job Performance: Evidence from the Federal Sector. American Review of Public Administration 31(4): 363–80. Andersen, Lotte Bøgh. 2009. What Determines the Behaviour and Performance of Health Professionals? Public

Service Motivation, Professional Norms and/or Economic Incentives. International Review of Administrative Sciences 75(1): 79–97. Brewer, Gene A. 2008. Employee and Organizational Performance. In Motivation in Public Management: Th e Call of Public Service, edited by James L. Perry and Annie Hondeghem, 136–56. Oxford: Oxford University Press. Brewer, Gene A., and Sally Coleman Selden. 1998. Whistle Blowers in the Federal Civil Service: New Evidence of the Public Service Ethic. Journal of Public Administration Research and Th eory 8(3): 413–39. ———. 2000. Why Elephants Gallop: Assessing and Predicting Performance in Federal Agencies. Journal of Public Administration Research and Th eory 10(4): 685–711. Brief, Arthur P., and Stephan J. Motowidlo. 1986. Prosocial Organizational Behaviors. Academy of Management Review 11(4): 710–25. Bright, Leonard. 2005. Public Employees with High Levels of Public Service Motivation: Who Are Th ey, Where Are Th ey and What Do Th ey Want? Review of Public Personnel Administration 25(2): 138–55. ———. 2007. Does Person–Organization Fit Mediate the Relationship between Public Service Motivation and the Job Performance of Public Employees? Review of Public Personnel Administration 27(4): 361–79. ———. 2008. Does Public Service Motivation Really Make a Diff erence on the Job Satisfaction and Turnover Intentions of Public Employees? American Review of Public Administration 38(2): 149–66. ———. 2009. Why Do Public Employees Desire Intrinsic Workplace Opportunities? Public Personnel Management 38(3): 15–37. Crewson, Philip E. 1997. Public-Service Motivation: Building Empirical Evidence of Incidence and Eff ect. Journal of Public Administration Research and Th eory 7(4): 499–518. De Dreu, Carsten K. W. 2006. Rational Self-Interest and Other Orientation in Organizational Behavior: A Critical Appraisal and Extension of Meglino and Korsgaard (2004). Journal of Applied Psychology 91(6): 1245–52. Deci, Edward L., Richard Koestner, and Richard M. Ryan. 1999. A Meta-Analytic Review of Experiments Examining the Eff ects of Extrinsic Rewards on Intrinsic Motivation. Psychological Bulletin 125(6): 627–68. Deci, Edward L., and Richard M. Ryan. 2004. Handbook of Self-Determination Research. Rochester, NY: University of Rochester Press. Delfgaauw, Josse, and Robert Dur. 2007. Signaling and Screening of Workers’ Motivation. Journal of Economic Behaviour and Organization 62(4): 605–24.

Background As it is used here, the "New Public Management" re- fers to a cluster of ideas and practices (including reinven- tion and neomanagerialism) that seek, at their core, to use private-sector and business approaches in the public sec- tor. While there have long been calls to "run government like a business," the contemporary version of this debate in this country was sparked in the 1990s by President Clinton's and Vice President Gore's initiative to "make government work better and cost less." Modeled after con- cepts and ideas promoted in Osborne and Gaebler's 1992 book Reinventing Government (as well as managerialist efforts in a variety of other countries, especially Great Brit- ain and New Zealand), the Clinton administration cham- pioned a variety of reforms and projects under the mantle of the National Performance Review. In part, what has dis- tinguished these reforms and similar efforts at the state and local level, from older versions of the run-government- like-a-business movement is that they involve more than just using the techniques of business. Rather, the New Pub- lic Management has become a normative model, one sig- naling a profound shift in how we think about the role of public administrators, the nature of the profession, and how and why we do what we do. Yet many scholars and practitioners have continued to express concerns about the New Public Management and the role for public managers this model suggests. For ex- ample, in a recent Public Administration Review sympo- sium on leadership, democracy, and public management, a number of authors thoughtfully considered the opportu- nities and challenges presented by the New Public Man- agement. Those challenging the New Public Management in the symposium and elsewhere ask questions about the inherent contradictions in the movement (Fox 1996), the values promoted by it (deLeon and Denhardt 2000; Frederickson 1996; Schachter 1997); the tensions between the emphasis on decentralization promoted in the market model and the need for coordination in the public sector (Peters and Savoie 1996); the implied roles and relation- ships of the executive and legislative branches (Carroll and Lynn 1996); and the implications of the privatization move- ment for democratic values and the public interest (McCabe and Vinzant 1999). Others have suggested that public en- trepreneurship and what Terry (1993, 1998) has called "neomanagerialism" threaten to undermine democratic and constitutional values such as fairness, justice, representa- tion, and participation. We would like to suggest that, beyond these separate critiques, what is missing is a set of organizing principles for an alternative to the New Public Management. We re- ject the notion that the reinvented, market-oriented New Public Management should only be compared to the old public administration, which, despite its many important contributions, has come to be seen as synonymous with bureaucracy, hierarchy, and control. If that is the compari- son, the New Public Management will always win. We would like to suggest instead that the New Public Man- agement should be contrasted with what we term the "New Public Service," a set of ideas about the role of public ad- ministration in the governance system that places citizens at the center. While there have been many challenges to the New Public Management and many alternative ideas promi- nently advanced by scholars and practitioners, there have been no attempts to organize these efforts and underscore their common themes. This article is an effort to do so. First, it briefly summarizes the foundations and major ar- guments of the new public management as it contrasts with the old public administration. It then describes an alterna- tive normative model we call the "New Public Service." This new model further clarifies the debate by suggesting new ways of thinking about the strengths and weaknesses of all three approaches. We conclude by considering the implications of placing citizens, citizenship, and the pub- lic interest at the forefront of a New Public Service.

The New Public Management and the Old Public Administration Over the past decade and a half, the New Public Man- agement (again, including the reinvention movement and the new managerialism) has literally swept the nation and the world. The common theme in the myriad applications of these ideas has been the use of market mechanisms and terminology, in which the relationship between public agen- cies and their customers is understood as based on self- interest, involving transactions similar to those occurring in the marketplace. Public managers are urged to "steer, not row" their organizations, and they are challenged to find new and innovative ways to achieve results or to priva- tize functions previously provided by government. In the past two decades, many public jurisdictions and agencies have initiated efforts to increase productivity and to find alternative service-delivery mechanisms based on public-choice assumptions and perspectives. Public managers have concentrated on accountability and high performance and have sought to restructure bureaucratic agencies, redefine organizational missions, streamline agency processes, and decentralize decision making. In many cases, governments and government agencies have succeeded in privatizing previously public functions, holding top executives accountable for performance goals, establishing new processes for measuring productivity and effectiveness, and reengineering departmental sys- tems to reflect a strengthened commitment to account550 Public Administration Review * November/December 2000, Vol. 60, No. 6 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions ability (Aristigueta 1999; Barzelay 1992; Boston et al. 1996; Kearns 1996). The effectiveness of this reform agenda in the United States, as well as in a number of other countries, has put governments around the world on notice that new standards are being sought and new roles established. These ideas were crystallized and popularized by Osborne and Gaebler's book, Reinventing Government (1992; see also Osborne and Plastrik 1997). Osborne and Gaebler provided a number of now-familiar principles through which "public entrepreneurs" might bring about massive governmental reform-ideas that remain at the core of the New Public Management. Osborne and Gaebler intended these principles to serve as a new conceptual or normative framework for public administration, an ana- lytical checklist to transform the actions of government: "What we are describing is nothing less than a shift in the basic model of governance used in America. This shift is under way all around us, but because we are not looking for it, because we assume that all governments have to be big, centralized, and bureaucratic, we seldom see it. We are blind to the new realities, because they do not fit our preconceptions" (1992, 321). Other intellectual justifications for the New Public Man- agement evolved as well. These justifications, as Lynn (1996) notes, largely came from the "public policy" schools that developed in the 1970s and from the "managerialist" movement around the world (Pollitt 1990). Kaboolian notes that the New Public Management relies on "market-like arrangements such as competition within units of govern- ment and across government boundaries to the non-profit and for-profit sectors, performance bonuses, and penalties (to) loosen the inefficient monopoly franchise of public agencies and public employees" (1998, 190). Elaborating this point, Hood writes that the New Public Management moves away from traditional modes of legitimizing the public bureaucracy, such as procedural safeguards on ad- ministrative discretion, in favor of "trust in the market and private business methods ... ideas ... couched in the lan- guage of economic rationalism" (1995, 94). As such, the New Public Management is clearly linked to the public choice perspective in public administration. In its simplest form, public choice views the government from the standpoint of markets and customers. Public choice not only affords an

elegant and, to some, compel- ling model of government, it also serves as a kind of intel- lectual road map for practical efforts to reduce govern- ment and make it less costly. And it does so unabashedly. John Kamensky, one of the architects of the National Per- formance Review, comments that the New Public Man- agement is clearly related to the public choice movement, the central tenet of which is that "all human behavior is dominated by self-interest" (1996, 251). The New Public Management is not just the implemen- tation of new techniques, it carries with it a new set of values, specifically a set of values largely drawn from the private sector. As we have already noted, there is a long- standing tradition in public administration supporting the idea that "government should be run like a business." For the most part, this recommendation has meant that gov- ernment agencies should adopt practices, ranging from "scientific management" to "total quality management," that have been found useful in the private sector. The New Public Management takes this idea one step further, argu- ing that government should not only adopt the techniques of business administration, but should adopt certain busi- ness values as well. The New Public Management thus becomes a normative model for public administration and public management. In making their case, proponents of New Public Man- agement have often used the old public administration as a foil, against which the principles of entrepreneurship can be seen as clearly superior. For example, Osborne and Gaebler contrast their principles with an alternative of formal bureaucracies plagued with excessive rules, bound by rigid budgeting and personnel systems, and pre- occupied with control. These traditional bureaucracies are described as ignoring citizens, shunning innovation, and serving their own needs. According to Osborne and Gaebler, "The kind of governments that developed dur- ing the industrial era, with their sluggish, centralized bu- reaucracies, their preoccupation with rules and regula- tions, and their hierarchical chains of command, no longer work very well" (1992, 11-12). In fact, while they served their earlier purposes, "bureaucratic institutions ... in- creasingly fail us" (15). What are the tenets of this bureaucratic old public ad- ministration, and is it reasonable to characterize any con- temporary thinking which falls outside New Public Man- agement as evidence of the old public administration? Certainly there is not a single set of ideas agreed to by all those who contributed over the decades to the old public administration (just as there is not a single set of ideas that all associated with the New Public Management would agree to). But there are elements of public administration theory and practice that seem to constitute a guiding set of ideas or a normative model that we now generally associ- ate with the old public administration. We suggest this model includes the following tenets: * Public administration is politically neutral, valuing the idea of neutral competence. * The focus of government is the direct delivery of ser- vices. The best organizational structure is a centralized bureaucracy. * Programs are implemented through top-down control mechanisms, limiting discretion as much as possible. The New Public Service: Serving Rather than Steering 551 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions * Bureaucracies seek to be closed systems to the extent possible, thus limiting citizen involvement. * Efficiency and rationality are the most important values in public organizations. * Public administrators do not play a central role in policy making and governance; rather, they are charged with the efficient implementation of public objectives. * The job of public administrators is described by Gulick's POSDCORB (1937, 13). If we compare the principles of New Public Manage- ment with these principles,

the New Public Management clearly looks like a preferred alternative. But even a cur- sory examination of the literature of public administration demonstrates that these traditional ideas do not fully embrace contemporary government theory or practice (Box 1998; Bryson and Crosby 1992; Carnavale 1995; Cook 1996; Cooper 1991; deLeon 1997; Denhardt 1993; Farmer 1995; Fox and Miller 1995; Frederickson 1997; Gawthrop 1998; Goodsell 1994; Harmon 1995; Hummel 1994; Ingraham et al. 1994; Light 1997; Luke 1998; McSwite 1997; Miller and Fox 1997; Perry 1996; Rabin, Hildreth, and Miller 1998; Rohr 1998; Stivers 1993; Terry 1995, 1998; Thomas 1995; Vinzant and Crothers 1998; Wamsley et al. 1990; Wamsley and Wolf 1996). The field of public administration, of course, has not been stuck in progres- sive reform rhetoric for the last 100 years. Instead, there has been a rich and vibrant evolution in thought and prac- tice, with important and substantial developments that can- not be subsumed under the title "the New Public Manage- ment." So there are more than two choices. We will now explore a third alternative based on recent intellectual and practical developments in public administration, one that we call the New Public Service. Roots of the New Public Service Like the New Public Management and the old public administration, the New Public Service consists of many diverse elements, and many different scholars and practi- tioners have contributed, often in disagreement with one another. Yet certain general ideas seem to characterize this approach as a normative model and to distinguish it from others. While the New Public Service has emerged both in theory and in the innovative and advanced practices of many exemplary public managers (Denhardt 1993; Denhardt and Denhardt 1999), in this section we will ex- amine the conceptual foundations of the New Public Ser- vice. Certainly the New Public Service can lay claim to an impressive intellectual heritage, including, in public ad- ministration, the work of Dwight Waldo (1948), and in political theory, the work of Sheldon Wolin (1960). How- ever, here we will focus on more contemporary precursors of the New Public Service, including (1) theories of democratic citizenship; (2) models of community and civil so- ciety; and (3) organizational humanism and discourse theory. We will then outline what we see as the main te- nets of the New Public Service. Theories of Democratic Citizenship Concerns about citizenship and democracy are particu- larly important and visible in recent political and social theory, both of which call for a reinvigorated and more active and involved citizenship (Barber 1984; Mansbridge 1990; Mansbridge 1992; Pateman 1970; Sandel 1996). Of particular relevance to our discussion is Sandel's sugges- tion that the prevailing model of the relationship between state and citizens is based on the idea that government ex- ists to ensure citizens can make choices consistent with their self-interest by guaranteeing certain procedures (such as voting) and individual rights. Obviously, this perspec- tive is consistent with public choice economics and the New Public Management (see Kamensky 1996). But Sandel offers an alternative view of democratic citizen- ship, one in which individuals are much more actively en- gaged in governance. In this view, citizens look beyond self-interest to the larger public interest, adopting a broader and longer-term perspective that requires a knowledge of public affairs and also a sense of belonging, a concern for the whole, and a moral bond with the community whose fate is at stake (Sandel 1996,5-6; see also Schubert 1957). Consistent with this perspective, King and Stivers (1998) assert that administrators should see citizens as citizens (rather than merely as voters, clients, or customers); they should share authority and reduce control, and they should trust in the efficacy of collaboration. Moreover, in con- trast to managerialist calls for greater efficiency, King and Stivers suggest that public managers seek greater respon- siveness and a corresponding increase in citizen trust. This perspective directly undergirds the New Public Service. Models of Community and Civil Society Recently, there has been a rebirth of interest

in the idea of community and civility in America. Political leaders of both major political parties, scholars of different camps, best-selling writers and popular commentators not only agree that community in America has deteriorated, but ac- knowledge that we desperately need a renewed sense of community. Despite increasing diversity in America, or perhaps because of it, community is seen as a way of bringing about unity and synthesis (Bellah et al. 1985, 1991; Etzioni 1988, 1995; Gardner 1991; Selznick 1992). In pub- lic administration, the quest for community has been re- flected in the view that the role of government, especially local government, is indeed to help create and support "community." In part, this effort depends on building a healthy and active set of "mediating institutions" that simultaneously 552 Public Administration Review a November/December 2000, Vol. 60, No. 6 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions give focus to the desires and interests of citizens and pro- vide experiences that will better prepare those citizens for action in the larger political system. As Putnam (1995) ar- gues, America's democratic tradition depends on the ex- istence of engaged citizens, active in all sorts of groups, associations, and governmental units. Collectively, these small groups constitute a "civil society" in which people need to work out their personal interests in the context of community concerns. Only here can citizens engage one another in the kind of personal dialogue and deliberation that is the essence of community building and of democ- racy itself. Again, as King and Stivers (1998) point out, government can play an important and critical role in cre- ating, facilitating, and supporting these connections be- tween citizens and their communities. Organizational Humanism and Discourse Theory Over the past 25 years, public administration theorists, in- cluding those associated with the radical public administrationists of the late 1960s and early 1970s (Marini 1971), have joined colleagues in other disciplines in suggest- ing that traditional hierarchical approaches to social organi- zation and positivist approaches to social science are mutu- ally reinforcing. Consequently, they have joined in a critique of bureaucracy and positivism, leading, in turn, to a search for alternative approaches to management and organization and an exploration of new approaches to knowledge acquisi- tion-including interpretive theory (for example, Harmon 1981), critical theory (Denhardt 1981), and postmodernism (Farmer 1995; Fox and Miller 1995; McSwite 1997; Miller and Fox 1997). Collectively, these approaches have sought to fashion public organizations less dominated by issues of au- thority and control and more attentive to the needs and con- cerns of employees inside public organizations as well as those outside, especially clients and citizens. These trends have been central to interpretive and criti- cal analyses of bureaucracy and society, but they have been even further extended in recent efforts to employ the per- spectives of postmodern thinking, especially discourse theory, in understanding public organizations. While there are significant differences among the various postmodern theorists, they seem to arrive at a similar conclusion-be- cause we depend on one another in the postmodern world, governance must be based on sincere and open discourse among all parties, including citizens and administrators. And while postmodern public administration theorists are skeptical of traditional approaches to public participation, there seems to be considerable agreement that enhanced public dialogue is required to reinvigorate the public bu- reaucracy and restore a sense of legitimacy to the field of public administration. In other words, there is a need to reconceptualize the field and, both practically and intel- lectually, so as to build a New Public Service.

The New Public Service Theorists of citizenship, community and civil society, organizational humanists, and postmodernist public administrationists have helped to establish a climate in which it makes sense today to talk about a New Public Service. Though we acknowledge that differences exist in these viewpoints, we suggest there are also similari- ties that distinguish the cluster of ideas we call the New Public Service from those associated with the New Pub- lic Management and the old public administration. More- over, there are a number of practical lessons that the New Public Service suggests for those in public administra- tion. These lessons are not mutually exclusive, rather they are mutually reinforcing. Among these, we find the fol- lowing most compelling. 1. Serve, rather than steer. An increasingly important role of the public servant is to help citizens articulate and meet their shared interests, rather than to attempt to control or steer society in new directions. While in the past, government played a central role in what has been called the "steering of society" (Nelissen et al. 1999), the complexity of modem life sometimes makes such a role not only inappropriate, but impossible. Those policies and programs that give structure and direction to social and political life today are the result of the interac- tion of many different groups and organizations, the mix- ture of many different opinions and interests. In many ar- eas, it no longer makes sense to think of public policies as the result of governmental decision-making processes. Government is indeed a player-and in most cases a very substantial player. But public policies today, the policies that guide society, are the outcome of a complex set of interactions involving multiple groups and multiple inter- ests ultimately combining in fascinating and unpredictable ways. Government is no longer in charge. In this new world, the primary role of government is not merely to direct the actions of the public through regu- lation and decree (though that may sometimes be appro- priate), nor is it to simply establish a set of rules and in- centives (sticks or carrots) through which people will be guided in the "proper" direction. Rather, government becomes another player, albeit an important player in the pro- cess of moving society in one direction or another. Gov- ernment acts, in concert with private and nonprofit groups and organizations, to seek solutions to the problems that communities face. In this process, the role of government is transformed from one of controlling to one of agenda setting, bringing the proper players to the table and facilitating, negotiating, or brokering solutions to public prob- lems (often through coalitions of public, private, and non- profit agencies). Where traditionally government has responded to needs by saying "yes, we can provide that The New Public Service: Serving Rather than Steering 553 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions Table 1 Comparing Perspectives: Old Public Administration, New Public Management, and New Public Service Old Public Administration New Public Management New Public Service Primary theoretical and Political theory, social and political Economic theory, more sophisticated Democratic theory, varied epistemological foundations commentary augmented by naive dialogue based on positivist social approaches to knowledge social science science including positive, interpretive, critical, and postmodern Prevailing rationality and Synoptic rationality, "administrative Technical and economic rationality, Strategic rationality, multiple tests associated models of human man" "economic man," or the self- of rationality (political, economic, behavior interested decision maker organizational) Conception of the public interest Politically defined and expressed in Represents the aggregation of Result of a dialogue

about shared law individual interests values To whom are public servants Clients and constituents Customers Citizens responsive? Role of government Rowing (designing and Steering (acting as a catalyst to Serving (negotiating and implementing policies focusing on unleash market forces) brokering interests among citizens a single, politically defined and community groups, creating objective) shared values) Mechanisms for achieving policy Administering programs through Creating mechanisms and incentive Building coalitions of public, objectives existing government agencies structures to achieve policy nonprofit, and private agencies to objectives through private and meet mutually agreed upon needs nonprofit agencies Approach to accountability Hierarchical-administrators are Market-driven-the accumulation of Multifaceted-public servants must responsible to democratically self-interests will result in outcomes attend to law, community values, elected political leaders desired by broad groups of citizens political norms, professional (or customers) standards, and citizen interests Administrative discretion Limited discretion allowed Wide latitude to meet Discretion needed but constrained administrative officials entrepreneurial goals and accountable Assumed organizational structure Bureaucratic organizations marked Decentralized public organizations Collaborative structures with by top-down authority within with primary control remaining leadership shared internally and a encies and control or regulation within the agency externally of clients Assumed motivational basis of Pay and benefits, civil-service Entrepreneurial spirit, ideological Public service, desire to contribute public servants and protections desire to reduce size of government to society. administrators service," or "no, we can't," the New Public Service sug- gests that elected officials and public managers should re- spond to the requests of citizens not just by saying yes or no, but by saying, "let's work together to figure out what we're going to do, then make it happen." In a world of active citizenship, public officials will increasingly play more than a service delivery role-they will play a con- ciliating, a mediating, or even an adjudicating role. (Inci- dentally, these new roles will require new skills-not the old skills of management control, but new skills of brokering, negotiating, and conflict resolution.) 2. The public interest is the aim, not the by-product. Public administrators must contribute to building a collective, shared notion of the public interest. The goal is not to find quick solutions driven by individual choices. Rather, it is the creation of shared interests and shared responsibility. The New Public Service demands that the process of establishing a vision for society is not something merely left to elected political leaders or appointed public admin- istrators. Instead, the activity of establishing a vision or direction is something in which widespread public dialogue and deliberation are central (Bryson and Crosby 1992; Luke 1998; Stone 1988). The role of government will increas- ingly be to bring people together in settings that allow for unconstrained and authentic discourse concerning the di- rection society should take. Based on these deliberations, a broad-based vision for the community, the state, or the nation can be established and provide a guiding set of ideas (or ideals) for the future. It is less important for this pro- cess to result in a single set of goals than it is for it to engage administrators, politicians, and citizens in a pro- cess of thinking about a desired future for their commu- nity and their nation. In addition to its facilitating role, government also has a moral obligation to assure solutions that are generated through such processes are fully consistent with norms of justice and fairness. Government will act to facilitate the solutions to public problems, but it will also be respon- sible for assuring those solutions are consistent with the public interest-both in substance and in process. In other words, the role of government will become one of assur- ing that the public interest predominates, that both the so- lutions themselves and the process by which solutions to public problems are developed are consistent with demo- cratic norms of

justice, fairness, and equity (Ingraham and Ban 1988; Ingraham and Rosenbloom 1989). In short, the public servant will take an active role in creating arenas in which citizens, through discourse, can articulate shared values and develop a collective sense of 554 Public Administration Review * November/December 2000, Vol. 60, No. 6 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions the public interest. Rather than simply responding to dis- parate voices by forming a compromise, public adminis- trators will engage citizens with one another so that they come to understand each other's interests and adopt a longer range and broader sense of community and soci- etal interests. 3. Think strategically, act democratically. Policies and programs meeting public needs can be most effectively and responsibly achieved through collective efforts and collaborative processes. To realize a collective vision, the next step is establish- ing roles and responsibilities and developing specific ac- tion steps to move toward the desired goals. Again, the idea is not merely to establish a vision and then leave the implementation to those in government; rather, it is to join all parties together in the process of carrying out programs that will move in the desired direction. Through involve- ment in programs of civic education and by developing a broad range of civic leaders, government can stimulate a renewed sense of civic pride and civic responsibility. We expect such a sense of pride and responsibility to evolve into a greater willingness to be involved at many levels, as all parties work together to create opportunities for par- ticipation, collaboration, and community. How might this be done? To begin with, there is an obvious and important role for political leadership-to ar- ticulate and encourage a strengthening of citizen responsi- bility and, in turn, to support groups and individuals involved in building the bonds of community. Government can't create community. But government and, more spe- cifically, political leadership, can lay the groundwork for effective and responsible citizen action. People must come to recognize that government is open and accessible-and that won't happen unless government is open and acces- sible. People must come to recognize that government is responsive-and that won't happen unless government is responsive. People must come to recognize that govern- ment exists to meet their needs-and that won't happen unless it does. The aim, then, is to make sure that govern- ment is open and accessible, that it is responsive, and that it operates to serve citizens and create opportunities for citizenship. 4. Serve citizens, not customers. The public interest re- sults from a dialogue about shared values, rather than the aggregation of individual self-interests. Therefore, public servants do not merely respond to the demands of "customers," but focus on building relationships of trust and collaboration with and among citizens. The New Public Service recognizes that the relation- ship between government and its citizens is not the same as that between a business and its customers. In the public sector, it is problematic to even determine who the cus- tomer is, because government serves more than just the immediate client. Government also serves those who may be waiting for service, those who may need the service even though they are not actively seeking it, future gen- erations of service recipients, relatives and friends of the immediate recipient, and on and on. There may even be customers who don't want to be customers-such as those receiving a speeding ticket. Moreover, some customers of government have greater resources and greater skill in bringing their demands for- ward than others. Does this justify, as it would in the pri- vate sector, that they be treated better? Of course not. In government, considerations of fairness and equity play an important role in service delivery; indeed, in many cases, these are much more important considerations than the desires of the immediate

customer. Despite the obvious importance of constantly improv- ing the quality of public-sector service delivery, the New Public Service suggests that government should not first or exclusively respond to the selfish, short-term interests of "customers." Instead, it suggests that people acting as citizens must demonstrate their concern for the larger community, their commitment to matters that go beyond short-term interests, and their willingness to assume per- sonal responsibility for what happens in their neighbor- hoods and the community. After all, these are among the defining elements of effective and responsible citizen- ship. In turn, government must respond to the needs and interests of citizens. Moreover, government must respond to citizens defined broadly rather than simply in a legal- istic sense. Individuals who are not legal citizens not only are often served by government programs, they can also be encouraged to participate and engage with their com- munities. In any case, the New Public Service seeks to encourage more and more people to fulfill their respon- sibilities as citizens and for government to be especially sensitive to the voices of citizens. 5. Accountability isn't simple. Public servants should be attentive to more than the market; they should also at- tend to statutory and constitutional law, community values, political norms, professional standards, and citi- zen interests. The matter of accountability is extremely complex. Yet both the old public administration and the New Public Management tend to oversimplify the issue. For instance, in the classic version of the old public administration, public administrators were simply and directly responsible to political officials. As Wilson wrote, "[P]olicy will have no taint of officialism about it. It will not be the creation of permanent officials, but of statesmen whose responsibility to public opinion will be direct and inevitable" (1887, 22). Beyond this, accountability was not really an issue; politi- cians were expected to make decisions while bureaucrats carried them out. Obviously, over time, public administraThe New Public Service: Serving Rather than Steering 555 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions tors assumed great capacities for influencing the policy process. So, at the other end of the spectrum, in the ver- nacular of the New Public Management, the focus is on giving administrators great latitude to act as entrepreneurs. In their entrepreneurial role, the new public managers are called to account primarily in terms of efficiency, cost ef- fectiveness, and responsiveness to market forces. In our view, such models do not reflect the demands and realities of public service today. Rather, public administrators are and should be influenced by and held account- able to complex constellations of institutions and standards, including the public interest, statutory and constitutional law, other agencies, other levels of government, the me- dia, professional standards, community values and stan- dards, situational factors, democratic norms, and of course, citizens. Further, the institutions and standards which in- fluence public servants and to which they are held account- able interact in complex ways. For example, citizen needs and expectations influence public servants, but the actions of public servants also influence citizen expectations. Laws create the parameters for public administrators' actions, but the manner in which public servants apply the law in- fluences not only its actual implementation, but also may influence lawmakers to modify the law. In other words, public administrators influence and are influenced by all of the competing norms, values, and preferences of our complex governance system. These variables not only in- fluence and are influenced by public administrators, they also represent points of accountability. The New Public Service recognizes the reality and complexity of these responsibilities. It recognizes that public administrators are involved in complex value con- flicts in situations of conflicting and overlapping norms. It accepts these realities and speaks to how public ad-

ministrators can and should serve citizens and the public interest in this context. First and foremost, the New Pub- lic Service demands that public administrators not make these decisions alone. It is through the process of dia- logue, brokerage, citizen empowerment, and broad-based citizen engagement that these issues must be resolved. While public servants remain responsible for assuring that solutions to public problems are consistent with laws, democratic norms, and other constraints, it is not a mat- ter of their simply judging the appropriateness of com- munity-generated ideas and proposals after the fact. Rather, it is the role of public administrators to make these conflicts and parameters known to citizens, so that these realities become a part of the process of discourse. Do- ing so not only makes for realistic solutions, it builds citi- zenship and accountability. 6. Value people, not just productivity. Public organiza- tions and the networks in which they participate are more likely to succeed in the long run if they are operated through processes of collaboration and shared leadership based on respect for all people. In its approach to management and organization, the New Public Service emphasizes the importance of "man- aging through people." Systems of productivity improve- ment, process reengineering, and performance measure- ment are seen as important tools in designing management systems. But the New Public Service suggests that such rational attempts to control human behavior are likely to fail in the long term if, at the same time, insufficient atten- tion is paid to the values and interests of individual mem- bers of an organization. Moreover, while these approaches may get results, they do not build responsible, engaged, and civic-minded employees or citizens. If public servants are expected to treat citizens with re- spect, they must be treated with respect by those who man- age public agencies. In the New Public Service, the enor- mous challenges and complexities of the work of public administrators are recognized. They are viewed not just as employees who crave the security and structure of a bureaucratic job (old public administration), nor as partici- pants in a market (New Public Management); rather, pub- lic servants are people whose motivations and rewards are more than simply a matter of pay or security. They want to make a difference in the lives of others (Denhardt 1993; Perry and Wise 1990; Vinzant 1998). The notion of shared leadership is critical in providing opportunities for employees and citizens to affirm and act on their public service motives and values. In the New Public Service, shared leadership, collaboration, and em- powerment become the norm both inside and outside the organization. Shared leadership focuses on the goals, val- ues, and ideals that the organization and community want to advance; it must be characterized by mutual respect, accommodation, and support. As Bums (1978) would say, leadership exercised by working through and with people transforms the participants and shifts their focus to higher level values. In the process, the public service motives of citizens and employees alike can be recognized, supported, and rewarded. 7. Value citizenship and public service above entrepre- neurship. The public interest is better advanced by pub- lic servants and citizens committed to making mean- ingful contributions to society rather than by entrepreneurial managers acting as if public money were their own. The New Public Management encourages public admin- istrators to act and think as entrepreneurs of a business enterprise. This creates a rather narrow view of the objec- tives to be sought-to maximize productivity and satisfy customers, and to accept risks and to take advantage of opportunities as they arise. In the New Public Service, there is an explicit recognition that public administrators are not 556 Public Administration Review * November/December 2000, Vol. 60, No. 6 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions

the business owners of their agencies and programs. Again, as King and Stivers (1998) remind us, government is owned by the citizens. Accordingly, in the New Public Service, the mindset of public administrators is that public programs and resources do not belong to them. Rather, public administrators have accepted the responsibility to serve citizens by acting as stewards of public resources (Kass 1990), conservators of public organizations (Terry 1995), facilitators of citizen- ship and democratic dialogue (Chapin and Denhardt 1995; King and Stivers 1998; Box 1998), catalysts for community engagement (Denhardt and Gray 1998; Lappe and Du Bois 1994), and street-level leaders (Vinzant and Crothers 1998). This is a very different perspective than that of a business owner focused on profit and efficiency. Accord- ingly, the New Public Service suggests that public admin- istrators must not only share power, work through people, and broker solutions, they must reconceptualize their role in the governance process as responsible participant, not entrepreneur. This change in the public administrator's role has pro- found implications for the types of challenges and respon- sibilities faced by public servants. First, public adminis- trators must know and manage more than the requirements and resources of their programs. This sort of narrow view is not very helpful to a citizen whose world is not conve- niently divided up by programmatic departments and of- fices. The problems that citizens face are often, if not usu- ally, multifaceted, fluid, and dynamic-they do not easily fall within the confines of a particular office or a narrow job description of an individual. To serve citizens, public administrators not only must know and manage their own agency's resources, they must also be aware of and connected to other sources of support and assistance, engag- ing citizens and the community in the process. Second, when public administrators take risks, they are not entrepreneurs of their own businesses who can make such decisions knowing the consequences of failure will fall largely on their own shoulders. Risk in the public sec- tor is different. In the New Public Service, risks and op- portunities reside within the larger framework of demo- cratic citizenship and shared responsibility. Because the consequences of success and failure are not limited to a single private business, public administrators do not singlehandedly decide what is best for a community. This need not mean that all short-term opportunities are lost. If dia- logue and citizen engagement is ongoing, opportunities and potential risks can be explored in a timely manner. The important factor to consider is whether the benefits of a public administrator taking immediate and risky action in response to an opportunity outweighs the costs to trust, collaboration, and the sense of shared responsibility. Implications and Conclusions From a theoretical perspective, the New Public Service offers an important and viable alternative to both the tradi- tional and the now-dominant managerialist models. It is an alternative that has been built on the basis of theoretical explorations and practical innovations. The result is a nor- mative model, comparable to other such models. While debates among theorists will continue, and administrative practitioners will test and explore new possibilities, the commitments that emerge will have significant implica- tions for practice. The actions that public administrators take will differ markedly depending on the types of as- sumptions and principles upon which those actions are based. If we assume the responsibility of government is to facilitate individual self-interest, we will take one set of actions. If, on the other hand, we assume the responsibil- ity of government is to promote citizenship, public dis- course, and the public interest, we will take an entirely different set of actions. Decades ago, Herbert Kaufman (1956) suggested that while administrative institutions are organized and oper- ated in pursuit of different values at different times, during the period in which one idea is dominant, others are never totally neglected. Building on this idea, it makes sense to think of one normative model as prevailing at any point in time, with the other (or others) playing a somewhat lesser role within the context of the prevailing view. Currently, the New Public Management and its surrogates

have been established as the dominant paradigm in the field of gov- ernance and public administration. Certainly a concern for democratic citizenship and the public interest has not been fully lost, but rather has been subordinated. We argue, however, that in a democratic society, a con- cern for democratic values should be paramount in the way we think about systems of governance. Values such as efficiency and productivity should not be lost, but should be placed in the larger context of democracy, community, and the public interest. In terms of the normative models we examine here, the New Public Service clearly seems most consistent with the basic foundations of democracy in this country and, therefore, provides a framework within which other valuable techniques and values, such as the best ideas of the old public administration or the New Public Man- agement, might be played out. While this debate will surely continue for many years, for the time being, the New Pub- lic Service provides a rallying point around which we might envision a public service based on and fully integrated with citizen discourse and the public interest. The New Public Service: Serving Rather than Steering 557 This content downloaded from 14.139.232.50 on Sun, 7 Sep 2014 05:48:33 AM All use subject to JSTOR Terms and Conditions References Aristigueta, Maria P. 1999. Managing for Results in State Gov- ernment. Westport, CT: Quorum. Barber, Benjamin. 1984. Strong Democracy: Participatory Poli- tics for a New Age. Berkeley, CA: University of California Press. Barzelay, Michael. 1992. Breaking through Bureaucracy. Ber- keley, CA: University of California Press. Bellah, Robert, et al. 1985. Habits of the Heart. Berkeley, CA: University of California Press. . 1991. The Good Society. New York: Knopf. Boston, Jonathan et al. 1996. Public Management: The New Zealand Model. New York: Oxford University Press. Box, Richard. 1998. Citizen Governance. Thousand Oaks, CA: Sage Publications. Bryson, John, and Barbara Crosby. 1992. Leadership for the Common Good. San Francisco, CA: Jossey-Bass. Burns, James MacGregor. 1978. Leadership. New York: Harper and Row. Carnavale, David. 1995. Trustworthy Government. San Fran- cisco, CA: Jossey-Bass. Carroll, James, and Dahlia Bradshaw Lynn. 1996. The Future of Federal Reinvention: Congressional Perspectives. Public Ad- ministration Review 56(3): 299-304. Chapin, Linda W., and Robert B. Denhardt. 1995. Putting "Citi- zens First!" in Orange County, Florida. National Civic Re- view 84(3): 210-215. Cook, Brian J. 1996. Bureaucracy and Self-Government. Balti- more, MD: Johns Hopkins University Press. Cooper, Terry. 1991. An Ethic of Citizenship for Public Admin- istration. Englewood Cliffs, NJ: PrenticeHall. deLeon, Linda, and Robert B. Denhardt. 2000. The Political Theory of Reinvention. Public Administration Review 60(2): 89-97. deLeon, Peter. 1997. Democracy and the Policy Sciences. Al- bany, NY: State University of New York Press. Denhardt, Robert B. 1981. In the Shadow of Organization. Lawrence, KS: Regents Press of Kansas. . 1993. The Pursuit of Significance. Pacific Grove, CA: Wadsworth. Denhardt, Robert B., and Joseph E. Gray. 1998. Targeting Community Development in Orange County, Florida. Na- tional Civic Review 87(3): 227-35. Denhardt, Robert B., and Janet Vinzant Denhardt. 1999. Lead- ership for Change: Case Studies in American Local Govern- ment. Arlington, VA: Price Waterhouse Coopers Endowment for the Business of Government. Etzioni, Amitai. 1988. The Moral Dimension. New York: The Free Press. . 1995. The New Communitarian Thinking. Charlottes- ville, VA: University of Virginia Press. Farmer, John David. 1995. The Language of Public Administra- tion. Tuscaloosa, AL: University of Alabama Press. Fox, Charles. 1996. Reinventing Government As Postmodern Symbolic Politics. Public Administration Review 56(3): 256-

61. Fox, Charles, and Hugh Miller. 1995. Postmodern Public Ad- ministration. Thousand Oaks, CA: Sage Publications. Frederickson, H. George. 1996. Comparing the Reinventing Government Movement with the New Public Administration. Public Administration Review 56(3): 263-9. . 1997. The Spirit of Public Administration. San Fran- cisco, CA: Jossey-Bass. Gardner, John. 1991. Building Community. Washington, DC: Independent Sector. Gawthrop, Louis C. 1998. Public Service and Democracy. New York: Chandler. Goodsell, Charles T. 1994. The Case for Bureaucracy. Chatham, NJ: Chatham House Publishers. Gulick, Luther. 1937. Notes on the Theory of Organization. In Papers on the Science of Administration, edited by L. Gulick and L. Urwick, 1-46. New York: Institute of Government. Harmon, Michael. 1981. Action Theory for Public Administra- tion. New York: Longman. . 1995. Responsibility as Paradox. Thousand Oaks, CA: Sage Publications. Hood, Christopher. 1995. The "New Pu

2. Conceptualizing QoG: A Critique Our opening story from St. Lucia serves to highlight a strong recent tenet within comparative political economy: It is not necessarily the lack of entrepreneurship or resources in human or physical capital that hinders economic development but the low quality of government institutions that exercise and implement laws and policies (Acemoglu, Johnson, and Robinson2001,2002;Clagueet al.1999;Easterly2001;EasterlyandLevine 2003; Evans and Rauch 1999; Hall and Jones 1999; Knack and Keefer 1997; Mauro 1995; Rodrik, Subramanian, and Trebbi 2004; RoseAckerman and Kornai2004).Notbeingabletopredictgovernmentactionwhenitreaches you and the lack of accurate information about what government bureaucrats can and cannot do to you are central ingredients of this problem (cf. Evans 2005; Lange 2005). Therapidgrowthinresearchon“goodgovernance”inrecentyearshas, however, not only been concerned with growth and economic development. The “quality of government” factor has also been argued to have substantial effects on a number of important noneconomic phenomena, both at the individual level—such as subjective happiness (Frey and Stutzer 2000; Helliwell 2003) and citizen support for government (Anderson and Tverdova 2003)—and at the societal level—such as the incidence of civil war (Fearon and Laitin 2003; Öberg and Melander 2005) and democraticstability(MungiuPippidi2006;R.RoseandShin2001;Zakaria 2003). This perspective has shifted the focus away from variables such as physical capital, natural resources, and human capital to matters directly related to the sphere of government and politics. The emphasis is also differentfromsomepreviousstudiesthatpointtolong-termculturaltraits asbeingrelatedtotheimportanceofsocialcapital(Putnam,Leonardi,and Nanetti 1993). Social capital, defined as access to social networks and generalizedtrustinotherpeople,seemstobedeterminedbythequalityof government institutions rather than the other way around (Delhey and Newton 2005; Rothstein 2005). Weshallargue,however,thatthisresearchagendahasfailedtoaddress a key issue: the question of how QoG should be defined conceptually. At least three problems may be identified with existing definitions: Either they are extremely broad, or they suffer from a functionalist slant, or they WHAT IS QUALITY OF GOVERNMENT? 167 deal only with corruption. Especially the first two problems seriously weaken the usefulness of existing conceptions. To illustrate the first problem, consider Kaufmann and associates at the World Bank, responsible for providing the most widely used empirical governance indicators. They define governance as “the traditions and institutions by which authority in a country is exercised.” More specifically, this includes: (1)theprocessbywhichgovernmentsareselected,monitored,andreplaced,(2) the capacity of the government to effectively formulate and implement sound policies, and (3) the respect of citizens and the state for the institutions that govern economic and social interactions among them. (Kaufmann, Kraay, and Mastruzzi 2004, 3) The problem is that such a definition is just about as broad as any definitionof“politics.”Forexample,itdoesnotdistinguishbetweenissuesthat concern the access to power and those that are related to the exercise of power. Moreover, it fails to distinguish between the content of specific policy programs on the one hand and the governing procedures on the other.InthewordsofKeefer(2004,5),“Ifthestudyofgovernanceextends to all questions related to how groups of people govern themselves..., thentherearefewsubjectsinallofpoliticalscienceandpoliticaleconomy that do not fall within the governance domain.” Or put differently, if QoG is everything, then maybe it is nothing. Yet, clearly some political institutions or aspects of “politics” must matter more than others for what should count as

QoG. We thus agree with the critique launched by Grindle (2004, 530) that “the good governanceagendaisoverwhelming”and,inparticular,withherargumentthat it fails to distinguish between various institutional particularities and more basic principles. The second problem is well illustrated by other economists who have tried to be more specific by defining “good governance” as what can be shown to be “good-for-economic-development” (La Porta et al. 1999, 223). One problem with this definition is that many important noneconomic consequences of QoG referred to above, such as high social trust and subjective well-being, are left out. Another problem is that, as with all functionalist definitions, it comes at the expense of being unable to define a country’s level of QoG without first having to measure the effects of QoG. This makes it impossible at the conceptual level to generalize what QoG is because the type of institutional arrangements that cause growth in one country may be very different from that of other counties. The functionalist approach thus makes it impossible to work out a general theory of QoG. As is well-known, functionalist approaches also border on tautologies. As argued by The Economist (June 4, 2005), defining “good governance” as “good-for-economicdevelopment” may generate the following infinite regress: “What is required for growth? Good governance. And what counts as good governance? That which promotes growth. And what is required for growth. . . .” 168 BO ROTHSTEIN AND JAN TEORELL Although it avoids omitting noneconomic outcomes, the same criticism can be launched at the definition provided by Huther and Shah (2005, 40): Governanceisamultifacetedconceptencompassingallaspectsoftheexerciseof authority through formal and informal institutions in the management of the resourceendowmentofastate.Thequalityofgovernanceisthusdeterminedby the impact of this exercise of power on the quality of life enjoyed by its citizens. Toparaphrase:“Whatisrequiredforthequalityoflifeenjoyedbycitizens? Quality of governance. What is quality of governance? That which promotes the quality of life. . . .” Lastly, QoG cannot be defined solely as the absence of corruption. The reason is that while a high degree of corruption is clearly an antithesis to QoG, the latter encompasses more than merely the absence of corruption. Many other practices that are usually not seen as corruption, such as clientilism, nepotism, cronyism, patronage, discrimination, and cases where administrative agencies are “captured” by the interest groups that they are set out to regulate and control, should be included (RoseAckerman 2004). A major problem in the literature is that the efforts by most researchers to define what should count as “QoG” have been detached from analysis of modern political philosophy, especially works that have been carried out on democratic theory and theories of justice. We shall argue that what should count as “quality” concerning how public policies and institutions are designed and operate should primarily be treated as a normative problem and thus cannot be solved without taking the questions raised in political philosophy into account. At the same time, what counts as QoG from a normative perspective should not be at odds with features that have been shown to have beneficial social and economic consequences. In what follows, we will present such a conceptualization of QoG that incorporates insights from political philosophy accompanied by empirical research about the effects of different institutional arrangements. 3. QoG as Impartiality A state regulates relations to its citizens on two dimensions. One is the “input”sidewhichrelatestotheaccesstopublicauthority.Theotheristhe “output”sideandreferstothewayinwhichthatauthorityisexercised.On the input side, where the access to power and thus the content of policies are determined, the most widely accepted basic regulatory

principle has been formulated by Robert Dahl (1989) as that of political equality.As we shall further stress in a subsequent section, political equality certainly implies impartial treatment on the input side of the system. This makes equality and impartiality partially overlapping concepts. Our argument for a more precise definition of QoG is, however, based on the idea that WHAT IS QUALITY OF GOVERNMENT? 169 democracy in the form of political equality on the input side must be complementedwithimpartialityontheoutputsideofthepoliticalsystem, that is, in the exercise of public authority. Our definition of impartiality in the exercise of public power is the following: When implementing laws and policies, government officials shall not take into consideration anything about the citizen/case that is not beforehand stipulated in the policy or the law (Strömberg 2000).1 As Cupit writes: “To act impartially is to be unmoved by certain sorts of considerations—such as special relationships and personal preferences. It is to treat people alike irrespective of personal relationships and personal likes and dislikes” (Cupit 2000). As these references make clear, impartialityisfirstandforemostanattributeoftheactionstakenbyjudges,civil servants, politicians, and the like. In order to effectuate this ideal it may, however, also be necessary to inscribe impartiality as an ideal into the mindset of these actors, an issue that we shall return to in the concluding section. Becausepoliticalequalityisabasicnormforlegitimatizingdemocracyin general, we want to stress that the way a particular democracy can constitutionally organize access to power can vary a lot (presidentialism vs. parliamentarism, unicameralism vs. bicameralism, proportional vs. majoritarian electoral systems, variation in the power of the courts, etc.). As long as the principle of equality in the access to power is not violated (e.g., by giving one specific political party the right to rule, or by refusing to give some specific group of citizens the right to stand for office or take part in the public debate) we call such differing political systems as in Denmark and in the United States democracies. Impartiality as the parallel legitimatizing principle for the “output” sidecan,inasimilarway,alsoencompassvariousadministrativepractices. For example, impartiality applied to decisions about recruitment to the civil service implies that selection should be based on the merits and qualifications that beforehand are stated as necessary for the position but exactly which merits or qualifications that should count can vary significantly. However, if recruitment were to be based on clientilistic personal contacts, political leanings, bribes, or ethnic belongings, the impartiality principlewouldbeviolated.Asabasicnorm,impartialitycanalsoencompass very different policies and does not rule out support for specific groups or interests. One example is different social policies, for example, support to poor families with children. The enactment of such policies would not break the principle of impartiality, while denying such allowances for families from a certain ethnic group or parents with a certain sexual orientation when implementing the policy would. How does QoG as impartiality relate to established definitions of corruption? The established view defines corruption as the “abuse (or misuse)ofpublicofficeforprivategain,”orsomeclosevariantalongthose lines (see, e.g., Alt and Dreyer Lassen 2003, 345; Treisman 2000, 399). This definition has some virtues when compared to earlier alternatives such as 170 BO ROTHSTEIN AND JAN TEORELL the “public opinion” (corruption is that which the public perceives to be corrupt)and“publicinterest”(corruptionisthatwhichviolatesthepublic interest) conceptions of corruption.Yet it suffers from a crucial weakness: It makes no reference to the kinds of acts that

constitute the “misuse” or “abuse” of public office, which makes the definition of corruption culturally relativistic. We therefore concur with Kurer (2005, 230) in stating that “corruption involves a holder of public office violating the impartiality principle in order to achieve private gain.” As Kurer argues, the advantage with this definition of corruption is that what counts as a breach of impartiality is fairly universally understood and is thus not related to how things like “abuse” or “misuse” of public power are viewed in different cultures. The norm that is violated when corruption occurs is instead the impartiality principle governing the exercise of public power, the core component of which is the notion of nondiscrimination.2 We would add, however, that the norm of impartiality does not only rule out all forms of corruption but also practices such as clientilism, patronage, nepotism, political favoritism, discrimination, and other forms of “particularisms” (Mungiu-Pippidi 2006). In other words, whereas impartiality implies the absence of corruption, the reverse is not necessarily true: The absence of corruption does not preclude all forms of partial exercise of government power. Equallyimportant,however,arethethingsthatthenormofimpartiality does not rule out. Since impartiality is a procedural norm, one important field that is not affected by our conception is the content of policies. This builds on the idea that noncorruption implies that “a state ought to treat equally those who deserve equally” (Kurer 2005, 223). This is in line with ourargumentthatthecontentofpublicpoliciesshouldnotbeincludedin the definition of QoG. Instead, it is impartiality in the exercise of power (the “ought to treat equally” principle) that is the central component of QoG.3 In political philosophy, this distinction between which norms should guide the content versus the procedural sides of the political system is readily seen in Brian Barry’s important book, Justice as Impartiality (1995). Barry argues that impartiality should be a normative criterion not only in the exercise of political power but also in terms of the content of public policy. However, a close reading of Barry reveals that when it comes to decisions about the content of the policies that governments should pursue, it is in fact not impartiality but “reasonableness” that is his main criterion (Lundström 2004). By this he means that people engaged in the political process should give sound arguments based on a secular understandingofknowledgeforwhytheyprefercertainpoliciesoverothers.In Barry’s words: “What is required is as far as possible a polity in which arguments are weighed and the best arguments win, rather than one in which all that can be said is that votes are counted and the side with the most votes wins” (Barry 1995, 103). WHAT IS QUALITY OF GOVERNMENT? 171 To this we want to object that reasonableness is not the same as impartiality. For example, in a given situation there may be good reasons for lowering pensions and increasing support to families with children. This is, however, not the same as being impartial between these two groups because there is no such thing as an impartial way to decide in a case like this (Arneson 1998). A large part of the critique from within political philosophy that has been directed against Barry’s theory is, to our mind, due to his conflation of the input and output sides of the democratic system and his idea that impartiality should not only guide implementation but also the content of public policies (cf. de Jasay 1996). Wealsodisagreewiththe“smallisbeautiful”ideaputforwardbysome economists. A case in point is Alesina and Angeletos (2005) who, from a strict deductive logic, conclude that “a large government increases corruption and rent-seeking” (18). The clear policy implication of this statement is that if one wants to improve QoG, one should lower the policy ambitions, especially when it comes to redistributive and interventionist policies.Ourfirstargumentagainstthisideaisthatinmoststudieswhere QoG and/or corruption have been measured, the four Nordic countries

rankatthetop(i.e.,theyhavelowcorruptionandhighQoG).4 Onecansay many things about these countries but not that they are characterized by low levels of public spending or ambitions when it comes to government intervention (cf. Lindert 2004; Rothstein 1996). That the size of government or how extensive its policies are have little or nothing to do with QoG was actually discovered in one of the first articlesinthislineofresearch(LaPortaet al.1999).Usingawealthofdata from between 49 and 212 countries in their search for what determines QoG, they arrived at the following conclusion: “Finally, we have consistentlyfoundthatthebetterperforminggovernmentsarelargerandcollect highertaxes.Poorlyperforminggovernments,incontrast,aresmallerand collect fewer taxes” (266). However, we cannot but point out that the authorsofthisarticlewerequicktoaddthat“thisresultdoesnotofcourse implythatitisoften,orever,sociallydesirabletoexpandagovernmentof a given quality, but it tells us that identifying big government with bad government can be highly misleading” (266). We agree with Alesina and Angeletos (2005) that (big) governments sometimes enact policies that are counterproductive to economic growth. However, defining exactly which policies benefit economic growth turns out to be a difficult task. Should pensions, health care, or education be a private or a public matter, or any mix thereof? Should the government be engaged in helping small firms grow? Is an active labor market policy a “sound policy” or not?As the Nordic cases indicate, there is very little to speak for the notion that governments characterized by high levels of public spending and intervention have low QoG. ThesecondargumentforreservingthedefinitionofQoGtoprocedures andnottocontentiswhatcouldbecalledthe“Platonian–Leninist”riskin the “good governance” discussion, namely, that the democratic process 172 BO ROTHSTEIN AND JAN TEORELL will be emptied of most substantial issues if experts from various international organization such as the World Bank5 prescribe almost all public policies.Afterall,whatshouldpoliticalpartiesdo,andwhatisthepointof having an ideological debate and elections if the content of most of the important policies has beforehand been decided by international experts? If QoG becomes a way for experts to define what are to be understood as “soundpolicies”(thetermusedbytheWorldBank),thereisnotmuchleft forpoliticalpartiesandpoliticianstodecideontherepresentationalsideof the democratic system. The argument against the “Platonian–Leninist” alternativetodemocracyhasbeeneloquentlyputforwardbyRobertDahl, and we accept his conclusion that “its extraordinary demands on the knowledge and virtue of the guardians are all but impossible to satisfy in practice” (Dahl 1989, 65). Yet when a policy has been decided upon, QoG implies that it has to be implemented in accordance with the principle of impartiality. In the following section, we will specify what we mean by impartialityasQoGbyrespondingtofourdifferenttypesofcriticismsthat have been leveled against this concept. 4. The Scope of Impartiality Contrarytoourpriorbeliefs,theconceptofimpartialityhasnotbeenhigh on the agenda in research on bureaucracy and public administration. For example, the newly issued 664-page Handbook of Public Administration has no index entry on “impartiality” (Peters and Pierre 2002). A search in the Social Science Citation Index also gives a surprisingly meagre result.6 This lack of interest is all the more surprising because the idea about the impartial civil servant goes back to the most central figure in bureaucratic theory, namely Max Weber. Moreover, it seems as if most research in public administration has argued either for ineffectiveness in the impartial mode of operating the state

machinery or that this is an ideal that is impossible to achieve (cf. Du Gay 2000). The bureaucratic mode of operation has been said to be too rigid for the active modern policy-oriented state,whichrequiresflexibilityandcommitmentbyitsofficials.Inthisline of reasoning, the ideal of the impartial civil servants has been accused of being insensitive to the complexities and the special needs of different cases (Olsen 2006). Another critique has come from the public choice approach.According to this theory, civil servants are driven more by a self-interest to promote their own interests rather than by ethics related to some public interest, such as impartiality (Dunleavy 1991). This critique is in line with a more general idea that politics is foremost to be seen as a partisan interest struggle,inwhichcasetherewouldbenoroomforimpartialagents(such as judges or civil servants) within the state. A third critique has come from within the field of political philosophy, not least the various multiculturalist and feminist approaches.7 The argument has been that impartiality is in fact impossible to achieve because WHAT IS QUALITY OF GOVERNMENT? 173 individuals, be they civil servants or otherwise, cannot step outside their particularistic personalities. Instead, their actions will always be impregnatedbythingslikeideologicalcommitments,sex/gender,cultural-ethnic belongings, class background, etc. Moreover,ithasbeenstatedthatpartialityistheideaoflifeitself—tobe deeply attached to other persons and causes is what life is really about. Impartiality is thus an offense against this inner meaning of life (Mendus 2002, 2f). The idea of “justice as impartiality” launched by Brian Barry (1995) is according to this line of critique a nonstarter simply because impartiality is both an impossible and an undesirable ideal. As the feminist political philosopher Iris Marion Young states, “no one can adopt a pointofviewthatiscompletelyimpersonalanddispassionate,completely separated from any particular context and commitments” (Young 1993, 127f). To summarize, impartiality is a truly contested concept. Several intellectual discourses argue that it is either an impossible or an unwanted ideal. To address these objections we draw upon an idea launched by Michael Walzer (1983), namely, that society consists of different moral sphereswithdifferentmodesofdominationthatnormativelyshouldnotbe transferable. Impartiality and Societal Spheres Walzer’s (1983) theory, in short, is that we have different normative ideas about what should count as fairness in different societal spheres. For example, while it is normatively fair to use money to get one’s way in market transactions, it is not so when it comes to politics (for instance, we do not allow the buying and selling of votes). What is important for us in Walzer’s approach is not only the idea that norms should be different in different societal spheres but that we should recognize that the same individuals often operate simultaneously in different spheres. For example, a judgecanalsobeadevotedmother,apersonthatoperatesonthehousing market, and an activist in an animal rights group. This simultaneity has two very important implications. First, QoG implies that those who exercise public power need to know the borderlines between the different moral spheres. Second, they need to know which norms are appropriate (andinappropriate)inthe“state”sphereandhowthesenormsdifferfrom what is legitimate in the other spheres. One problem with Walzer’s (1983) theory is that the number of moral spheres seems to be both infinite and arbitrary. He acknowledges that his theory is not meant to be universal. Second, there is no basic concept or theory behind his idea of the historical existence of different moral spheres that explains why he ends up with the ones that he presents. Walzer’s moral spheres are thus defined by

reference only to historical particularities (20). The lack of universality in his theory of “spheres” is problematic because a theory of how QoG should be defined ought to be 174 BO ROTHSTEIN AND JAN TEORELL universal. Otherwise, we may very well end up with one theory of QoG per country.8 A lack of universality implies that not only a comparative approach but also the search for a common theory becomes futile. Our solution to these problems starts from an idea of the two dimensions of interests. One is the type of interest, the other is its scope. Type refers to the distinction between self-regarding and other-regarding interests. Scope refers to the “demos” question: How many are to be included. Here, the question is whether the type of interest that dominates is to be for “everyone” or if it is restricted to one’s friends, family, clan, tribe, or other such groups. If we combine these two dimensions, we get the following four spheres of conduct that relate to impartiality (see Figure 1). The logic of this model is the following: In the “state” sphere, the norm isthattheexerciseofpowershouldbeinaccordancewiththeenactedlaws and polices and that they apply equally to “all,” as stated in the two principles of political equality and equality before the law. In the “market” sphere, the accepted norm is that behavior according to self-interest is justifiable, but the scope dimension “all” implies that everyone should have equal access to the market (exemplified in, e.g., laws against monopolies/trusts). We should simply sell to and buy from anyone, regardless of his/her family background, ethnic belonging, or religion. However, the accepted norm in the private sphere is that we should not behave according to self-interest against our family members or friends but to pursue what we from some other-regarding notion deem as good forallthemembersofthissmallgroup.Thepointhereisthatsuchgroups donot,aswiththemarket,havefreeentrancebutarerestrictedtoitsgiven “members.” Lastly, special interest groups are driven by the idea of making things better for their members (self-interest), and they are also restricted to members. In general, members or leaders of such groups cannot be expected to act according to some “public” interest. On the contrary, their actions are based on some “logic of exchange.” This is FIGURE 1 Dimensions of Interest and Societal Spheres Type of interest Other-regarding Self-regarding “All” The state The market Scope of interest “Few” The family/clan The interest group WHAT IS QUALITY OF GOVERNMENT? 175 where we find policies and practices related to what usually is labeled “neo-corporatist” or “interestgroup” politics (for a somewhat related idea, see Lange 2005). The point we want to make from this application of Walzer’s (1983) normative theory, which is also our response to the critics of impartiality as a basic norm for the exercise of public power, is that social science shouldnotbebasedontheideaofsocietybeingdominatedbyagentswith only one script of human behavior or a single set of moral norms, be that self-interest, the principle of care, rent-seeking, bureaucratic

ethics, feelings of community, or altruism. According to our model, humans have a greater repertoire than being only self-regarding, etc., and they understand that what is appropriate in one sphere is fundamentally wrong in another sphere (March and Olsen 1989). From a normative perspective, we can also see that while self-interest is justifiable in some spheres, it is unacceptable in others. For example, agents on a market that would use “other-regarding interest” as their main template of behavior would simplybedeemedstupidandprobablysoongooutofbusiness,whilecivil servants or parents who act according to pure self-interest would by most people be seen as morally deplorable. In this respect, a large part of the discussion of Brian Barry’s Justice as Impartiality (1995) appears to be misguided because this distinction between “spheres of behavior” has been left out. While, for many, increased justice implies policies that contain more partiality (e.g., extra resources to underprivileged groups), they usually do not want these policies, once enacted, to be implemented in a partial way where bureaucrats are given total discretion in each and every case (Tebble 2002; Young 1990). Similarly, while impartiality is a norm to be followed in one sphere, it would be dysfunctional and/or unethical in other spheres. According to our model, QoG as impartiality implies that government power should be exercised according to a certain moral principle, namely, impartiality.9 Thisnotionstandsinsharpcontrasttothepublicchoiceidea of public officials maximizing their self-interest. For example, the impartial civil servant should not be susceptible to bribery, should not decide in cases where his/her friends and relatives are involved, and should not favor any special (ethnic, economic, or other type of organized) interest when applying laws and rules. Thus, in “the state” sphere we are concerned with the typical civil servant,policeman,orjudgeactingaccordingtotheimpartialityprinciple, implying that he or she should be guided by the public interest instead of by any self-interest. Decisions should be made according to what is stipulated and intended in the law/policy, disregarding the bureaucrats’ own interests. The special interests that are acceptable in other spheres (money fromthemarket,loyaltiestofamilyandfriends,andadherencetodifferent special interests) should not be allowed to influence his/her decisions. However,wehavenomoralobjectionswhentheverysamepersonsleave the courtroom or public office and go to the “market” and try to get the 176 BO ROTHSTEIN AND JAN TEORELL verybestdealwhensellingtheirhouseorbuyinganewcar.Likewise,we do not object if these persons in their private dealings take special care for theirfamilyandfriends(cf.Barry1995,205).10 However,weobjectstrongly if they, in their professional life, do not refrain from handling cases that concern members of their family or friends. Moreover, as private individuals, we would argue that these persons have the same rights as every other citizen to support whatever special interests or political causes that they like. But again, if these persons are strongly involved or engaged in a certain cause or interest organization, they should declare a conflict of interest and abstain from handling a case that may influence the outcome of this cause or be of importance to this interest organization. Likewise, a civil servant who handles public contractscannothaveeconomicinterestsinanyofthepotentialbidders.Thus, the demand for impartiality from civil servants is not absolute, and we do notbasetheideaofQoGasimpartialityonthepresumptionthatthereare certain persons who have it “in their nature” to be disinterested in all straits of life. What QoG demands is that people employed to exercise government powers recognize that there are clear boundaries between this sphere and other societal spheres and that these boundaries put severe restrictions on what types of behavior can be accepted.

We can illustrate this point by referring to the fact that the common language uses different words for different kinds of transgressions. When norms from the market are imported to the state sphere, it is usually referred to as bribes. When norms from the family/clan sphere enter the state sphere, the word is usually nepotism or patronage. Lastly, when normsfromtheinterestgroupsphereareemployedinthestatesphere,the problem is usually called corporatism or influence peddling (as in the “military/industrial complex”). The Feminist Challenge: Commitment, Flexibility, and Impartiality Feminist scholars have pointed to the possibility of a conflict between the principleofimpartialityandthecapacityofthestatetodeliverthekindof social services required of public-sector employees in the welfare state who must perform curative and caring work. Following Joan Tronto, Helena Stensöta has argued that we expect, for example, preschool teachers, medical professionals, and social workers to demonstrate empathy and compassion, and not to be governed by some general and abstract logic of justice as impartiality (Stensöta 2004).According to this approach, the “logic of care” leads to a more context-dependent ethic than the impartial application of universal rules. In specific terms, we do not want a nurse in a public hospital to treat all patients alike but to give more care and attention to those who need it. In this and many other similar policy areas, legitimacy in the implementation process requires that public employees are committed, engaged, and dedicated to their tasks. WHAT IS QUALITY OF GOVERNMENT? 177 The feminist discussion about “the logic of care” puts light on an importantdimensionofourtheoryaboutQoGasimpartiality,namely,that impartiality is not to be understood as implying that the implementation of public policies equals an old-style Weberian rigid rule following, personal detachment, or the lack of creativity and flexibility by the people working in the public sector. Certainly, most of us want children who attend a public preschool to be approached with empathy and concern rather than some dry-as-dust rule-following personnel. Obviously, different children need different degrees of attention, comfort, and support in different situations. However, most people would be morally upset if preschool staff deliberately directed their care and concern toward children from families that had bribed them, or who belonged to a certain ethnicgroup,andthusinpracticediscriminatedagainsttheotherchildren. As this case shows, there is no conflict between professionally distributed care and the principle of impartiality. As defined here, a traditional rulebased Weberian bureaucracy may in some areas be an incarnation of the impartiality principle but so may a public organization that is based on strong commitment to the policy goals while implementing these goals with a degree of flexibility, be they the reduction of poverty, the preservation of forests (Kaufman 1960), or an active labor market policy (Rothstein 1996). 5. Competing Conceptions of QoG Impartiality could hardly be considered the sole normative yardstick by which all aspects of a political system may be assessed. In this section we therefore relate our theory of impartiality to a tripartite set of competing conceptions of what constitutes QoG. First, we shall argue that although democracy is probably a necessary condition for QoG, it is hardly sufficient. Second, we show that the impartiality principle implies and encompasses the rule of law, while the opposite is not true. Efficiency/ effectiveness, finally, lacks an independent normative justification and is thereforealwayssecondarytoimpartiality.Nevertheless,therearereasons to believe that impartiality should enhance effectiveness/efficiency.

Democracy The reasons why democracy as a majoritarian rule does alone not suffice as a definition of QoG are both theoretical and empirical. The theoretical reason is well known: There are simply no guarantees that a majority will respect the impartiality principle when government power is to be exercised. For example, in an ethnically divided society, the ethnic majority, after having been elected, may decide that all citizens from the minority workingascivilservantsshouldbefired.Thiswouldclearlybeabreachof the impartiality principle in the exercise of government authority, albeit stemming from democratic principles in the access to authority. 178 BO ROTHSTEIN AND JAN TEORELL Empirically, there is no straightforward relationship between democracyintheaccesstopublicpowerandimpartialityintheexerciseofpublic power.Onthecontrary,democracyseemstobecurvilinearlyrelatedtothe level of corruption (Montinola and Jackman 2002; Sung 2004). Empirical research indicates that some democratization may at times be worse for impartialitythannone.Forexample,someoftheworstcasesofcorruption have appeared in newly democratized countries, such as Peru under its former president Fujimori (McMillan and Zoido 2004). Conversely, some undemocraticcountrieshaveshownimpressiveresultsincurbingcorruptionandestablishingfairlyimpartialbu reaucracies,primeexamplesbeing Hong Kong and Singapore (Root 1996). Moreover, the track record of democracy in terms of producing valued social outcomes is surprisingly uneven. The inherently ambiguous results in the empirical research on whether democracy matters for growth is perhaps the most prominent example (see, e.g., Kurzman, Werum, and Burkhart 2002; Przeworski and Limongi 1993; Sirowy and Inkeles 1990). True, democracy usually comes out as a strong predictor of human rights (Davenport and Armstrong 2004; Poe, Tate, and Keith 1999). But democracy should arguably be defined at least partly in terms of human rights suchaspersonalintegrity(Hadenius1992;HadeniusandTeorell2005),so this finding is not all that surprising. A case in point is the relationship between democracy and the probability of civil war. Empirical research shows an inverted U-curve, with strongautocraciesandfulldemocraciesbeingleastlikelytoengageincivil violence (Hegre et al. 2001). Curvilinearity is of course not tantamount to a null effect, but this does indicate that some democracy may at times be worse than none (although a lot of democracy is better than some). A related argument is the “democratic peace” theory, where the strong empirical regularity pertains to the dyadic level, that is, between pairs of states, both of which are democracies (Oneal and Russett 1999). Monadically speaking, however, democracies are not significantly less aggressive than autocracies, whereas the incidence of incomplete democratization even seems to make a country more likely to go to war (Mansfield and Snyder 2005). Finally, some recent work seriously questions the presumed positive effects of democracy on human development, arguing that this is either extremely slow and evolving over decades (Gerring, Thacker, and Alfaro 2005) or, even worse, vanishes completely once missing data bias have been corrected for (Ross 2006). Simply put, knowing the extent to which a country is democratic or not cannot help in explaining the multitude of highly valued economic and social consequences of QoG documented in the literature. Having said this, democracy and impartiality do overlap at the conceptual level in two very important areas. First, this is apparent with respect tothe“bundleofpoliticalrights”requiredtoupholdademocraticsystem. Democracy, in O’Donnell’s (2001, 18) words, presupposes “a legal system WHAT IS QUALITY OF GOVERNMENT? 179

that enacts and backs the universalistic and inclusive assignment of these rights.” Democratic legitimacy requires that political rights, such as freedomsofassociationandexpression,mustbesecuredwithinalegalframework, and this framework must, in turn, be impartially applied to all its subjects.As a consequence, democracy as political equality entails impartial government institutions in the regulation of the access to political power. Second, this overlap is also readily seen if we consider the idea of “free and fair elections.” Elections have to be administrated by the existing government,butiftheyaretobeconsideredfreeand,inparticular,fair,the ruling party must refrain from organizing them in a manner that undermines the opposition’s possibilities to obtain power. That is, free and fair elections must be administered by impartial government institutions (Choe 1997; Schedler 2002, 44). But again, the impartial organization of elections does not imply that the content or outcome of this process is impartial. On the contrary, the reason why many people take part in elections (and politics in general) is because they are motivated by very partial interests. Thejointconditionofthelegalsystemenforcingpoliticalrightsandthe election process itself being impartially administrated means that when democracy, in its basic sense, has been accomplished, certain spheres of government action must be regulated by the impartiality principle— even on the input side. In other words, democracy and impartial government institutions are partially overlapping concepts. They are only partially overlapping, however, because (1) democracy can threaten impartiality in the exercise of government power, and (2) democracy is related to many other things than impartiality, especially when it comes to the content of policies. Although we argue that democracy is not a sufficient condition for QoG,itmaywellbeanecessaryone,asitservestosetcertainlimitsonthe types of policies that may be pursued in the name of impartiality. A state thatenacts“apartheidtype”laws,forexample,cannotbeseenashavinga high QoG even if they are applied by ever so impartial bureaucrats. Such laws would come into conflict with the principle of equality that is meant toregulatetheaccesstopoliticalpower.Similarly,ourargumentisnotthat the lack of democratic responsiveness in places such as Hong Kong and Singapore—undemocratic but relatively clean from corruption—should somehow be excused on account of their relative absence of corruption. Fromanormativeperspective,QoGrequiresbothdemocracyintheaccess to power and impartiality in the exercise of this power. The Rule of Law Establishing the rule of law is usually placed high on the agenda for reformingdevelopingandtransitionalcountries.“Onecannotgetthrough a foreign policy debate these days,” writes Carothers (1998, 95), “without 180 BO ROTHSTEIN AND JAN TEORELL someone proposing the rule of law as a solution to the world’s troubles.” Butwhatdoes“theruleoflaw”mean?Althoughunequivocallyembraced as a virtue of any political system, the concept is rarely defined. One reason for this may of course be that the concept is inherently ambiguous. As a recent review article makes clear, even legal scholars argue over its exact meaning (J. Rose 2004). To begin with, they dispute whether or not the rule of law should be given a purely procedural interpretation, bearing no implications for the actual substance of promulgated laws. Defenders of a procedural notion claim that the rule of law must be distinguished from the rule of “good” law. Critics argue that this would allow morally detested regimes, such as Nazi Germany, to be classified as abiding by the rule of law. Against the procedural view, these critics seek to inscribe into the rule of law various substantive moral values of liberal democracy. Yet even among proceduralists, who adhere to a narrower conception, ambiguities remain. Usually, more attention is paid to the internal

qualities of the laws themselves—such as the need for the law to be clear, understandable, general, internally consistent, prospective, stable, etc.—rather than to define the core principles that a political system must abide by in order to be in accordance with the rule of law. Searching for these core principles, we instead turn to conceptions developedwithinpoliticalscience.Weingast(1997,245)definestheruleof law as “a set of stable political rules and rights applied impartially to all citizens.” Similarly, O’Donnell (2004, 33) states a minimal definition of the rule of law as “that whatever law exists is written down and publicly promulgated by an appropriate authority before the events meant to be regulatedbyit,andisfairlyappliedbyrelevantstateinstitutionsincluding the judiciary.” As these definitions should make clear, the rule of law is perfectly compatible with our principle of impartial government institutions. This connection is most explicitly recognized by O’Donnell: By “fairly applied” I mean that the administrative application or judicial adjudication of legal rules is consistent across equivalent cases; is made without taking into consideration the class, status, or relative amounts of power held by the parties in such cases; and applies procedures that are preestablished, knowable, and allow a fair chance for the views and interests at stake in each case to be properly voiced. (33) The rule of law thus embodies the principle “equality [of all] before the law.” It entails “a crucial principle of fairness—that like cases be treated alike” (33–34). The connection between the two concepts may be stated even stronger: Impartiality implies the rule of law. Procedural impartiality, to be accomplished in practice, requires a set of rules that regulates proper conduct—such as distinguishing which cases are “like” cases and what specific concerns are legitimate in specific cases. The most general set of such rules for governing a society are its laws. These laws must be consistently applied to everyone—including those who promulgate the laws WHAT IS QUALITY OF GOVERNMENT? 181 themselves. No one is above the law. Thus, a corollary of impartial government institutions is the notion of an impartially applied legal system—the rule of law. Impartiality, however, also applies to spheres of state action other than those directly governed by law. When public policy is to be enacted in so-called “human processing” areas, such, as for example, education, health care, welfare benefits, and active labor–market programs, widely discretionary powers usually need to be transferred to lower-level government officials responsible for implementing policy. Impartial, nondiscriminatory behavior on behalf of these policy enactments is of course a key virtue according to our theory. But it falls outside the sphere of government activity regulated by the rule of law. In other words, the impartiality principle not only entails but in addition encompasses the rule of law. Effectiveness/Efficiency It would certainly be strange to argue that a government that is very inefficient or ineffective can have a high QoG. Economists often treat the state’s capacity for action as a core element of their governance concept. Shah (2005, xxiii), for one, considers two features of government performance: responsiveness (“whether the public manager is doing the right things—i.e., delivering services consistent with citizen preferences”) and efficiency(“whetherthepublicmanagerisdoingthemright—i.e.,providing services of a given quality in the least-cost manner”). Similarly, La Porta et al. (1999) include “efficiency”—“successful provision of essential public goods”—and “effective spending” among their operational QoG indicators. As these examples suggest, there are actually (at least) two dimensions of state capacity: the degree of successful policy implementation—effectiveness—and the amount of government output delivered relative to input—efficiency. We shall, however, treat these two dimensions jointly. Our argument on this point is,

first, that impartiality is always preferable to efficiency/effectiveness. This rests on the Rawlsian antiutilitarian theory of justice which implies that utility always comes second to rights (Rawls1984).InDworkin’s(1977)extensionofthistheory,theimplication for the exercise of power is that the state should treat every citizen with “equal concern and respect” (180f), which in our terminology equals impartiality. Issues about effectiveness and efficiency simply lack an equally strong theoretical and normative underpinning as impartiality. Second, there is a case for arguing that impartiality may enhance effectiveness/efficiency. One of the organizational structures arguably underpinning impartiality is meritocratic recruitment to the civil service. Impartiality would thus imply recruitment based on merit and competence rather than political connections and/or being a member of clientilistic networks. As has been shown by Rauch and Evans (2000), in 182 BO ROTHSTEIN AND JAN TEORELL developing countries, meritocratic principles of recruitment and promotion turn out to be strongly positively related to the subjective ratings of corruption and bureaucratic efficiency, as well as to economic growth. Moreover,the“tollbooth”theoryofcorruptiondevelopedbyShleiferand Vishny (1993), supported by empirical evidence in Djankov et al. (2002), suggests that inefficient procedures are deliberately crafted by corrupt bureaucrats in order to maximize opportunities for bribery. Since impartiality implies the absences of such practices, this is another mechanism through which impartiality should promote effectiveness/efficiency.11 6. Concluding Remarks Returning to our opening story about the small, run-down coffee shop at Vigie Beach in St. Lucia, we would predict that if the women who run the places would have had a “system of beliefs” whereby they could take for granted that the government institutions responsible for the surveillance of their businesses would have been guided by the impartiality principle, they may have taken the risk to borrow capital to buy (or get a long-term lease for) the piece of land and build a place that would cater to the needs of the stream of tourists now passing by their businesses. The reason why this does not occur appears to be that the women take for granted that “people like us” will always be treated in an unfair manner and taken advantage of by the authorities. The development of the ethics and norms that underlie the principle of impartiality appears to be a long and complicated process (cf. Rueschemeyer2005).Insocietiesdominatedbypatronageandneo-patrimonialstructures and/or strong notions of tribal belongings, the idea of civil servants acting on the principle of impartiality when wielding public power may not even exist in the mindset of most people. On the contrary, if given a position in the public administration, the dominant idea is to use it to further one’s specific interests, be it clan, tribe, family, or other special interests. As stated by Dele Olowu (2000): . . . [P]olitical life in Africa as in other Third World regions is characterized by patron–client relationships. The public sector becomes an instrument for building public support for factions that are competing for power. . . . Thepublic sectoristhereforedysfunctionalinservingthepublic,butcriticaltothesurvival andsustenanceofthosewhowieldexecutivepower . . . asaresult . . . thepublic services lack even the basic meritocratic features of efficiency, productivity, and other universalistic values. (162) In a recent article, the Romanian political scientist Alina MungiuPippidi (2006) criticizes the many anticorruption policies that have been established in her country with the assistance of what can now be called the international “good governance” regime. The main result of these policies is, according to her, the establishment of many new institutions (anticorruption ombudsmen and special prosecutors, etc). The problem, she argues, is that these new institutions are quickly overtaken by the

WHAT IS QUALITY OF GOVERNMENT? 183 existing corrupt (or semi-corrupt) political networks that are dominated by “particularistic” norms while what is lacking is the existence of a universalisticpoliticalculture,whichinourterminologyequalsimpartiality(MungiuPippidi2006).Itmaythusnotbenewinstitutionsbutinstead new normative constructions that are necessary for increasing QoG. Our ambitiontolinknormative-conceptualanalyseswithempiricalresearchin thisareamaythushaveimportantimplicationsforhowpoliciesfor“good governance” should be enacted. The idea of impartiality when wielding public power has, in all likelihood, not existed for that long in most Western liberal democracies (North, Wallis, and Weingast 2006). In the case we know best, Sweden, it was not until the latter half of the nineteenth century that this principle became dominant in the civil service. Until then, many public officials viewed their positions as something similar to a feudal enfeoffment, that is, something they could use more or less as their property to extract private resources from (Rothstein 1998). This is not the place to go into a detailedanalysisofwhytheprincipleofimpartialityhascometodominate the public administration ethos in some parts of the world but not in others(butseeErtman2005).Briefly,however,wewouldliketoreflecton the possibility that impartiality is a mental construction in the same way that Hernando de Soto has argued about the development of “capital.” To give a very short recapitulation of de Soto’s (2001) wellknown argument: Capital is not the same as assets or even property. For assets/ property to become capital, they have to become a universally accepted legal construction by which ownership is generally respected. Through such a normative/legal invention, assets/property that become accepted as capital can be used, for example, as a security for loans and of course exchanged. The point is that de Soto shows that for this to happen in the Western world it took a long and very complex process of legal development that lasted for several hundred years. The feudal idea of what constituted property was, for example, very different from the modern/ capitalist idea. According to de Soto, assets cannot be transformed into and used as capital until it is recognized by “all” others as such. This process demands not only a strong legal framework but also a change of mind by both the public and those who are responsible for securing property rights. Ourargumentaboutimpartialityfollowsthesamelineofreasoning.As “capital,” the notion of civil servants behaving according to the principle of impartiality in the exercise of public power is a mindset or an established “system of beliefs” that in our part of the world may have taken a very long time to develop. The idea that when given a job in a governmental system, one should not primarily see this position as an asset that canbeusedtoserveone’sownfamily,clan,tribe,orspecialinterestgroup but that policies should be impartially implemented in accordance with what is stated in the law, may, as with the legal structures that create “capital,” have evolved over a long period of time (Rueschemeyer 2005). 184 BO ROTHSTEIN AND JAN TEORELL As with capital, this has probably required both a legal framework to make civil servants accountable and a conceptual development of the importanceofethicsinthepublicservice.Tounravelhowandunderwhat circumstances this process has historically come about would be an important task for future research.