Performance management in the local public sector in France: an administrative rather than a political model

Abstract

The purpose of this research is to analyse and characterize the practices of French local authorities in the field of performance management. It sets out to do so, as a first step, by putting forward a theoretical analytical framework of the different models of local performance management. It then goes on to draw on this framework and test it empirically in order to understand the nature of the French model of local performance management. The results of our research demonstrate the existence of a French, ‘performance administration’ type model evident in the trends common to French local authorities with regard to the instrumental and methodological content and the methods of steering public performance.

Points for practitioners

The managerial and practical contributions of this research on local performance initiatives are twofold. First of all, it will provide an analytical and evaluative framework for this type of initiative, not only structured around planning, accountability and evaluation tools but also around steering methods. Second, this study makes it possible to take stock of and characterize French practices in the field and their limitations compared to more mature international practices. Thus, a local authority adopting this path can therefore draw on this study to determine the scope and structure of the tools to be rolled out, depending on their purpose, be it administrative and/or political.

Keywords

local authorities performance management public performance

Since the 1990s, public performance and, more generally, performance management, have been a major challenge for central and local governments (Newcomer, 2007), generating a great deal of academic research, both empirical and theoretical. The influence of movements such as New Public Management (NPM) and the reforms undertaken by countries such as England, Canada and New Zealand, often presented as benchmarks for public sector regeneration, tend to reflect the picture of a homogeneous and universal performance management model within the public sector. The reforms undertaken are depicted as being organized around similar principles, objectives, goals and methodological frameworks. However, some studies (Bouckaert and Halligan, 2008; Curristine, 2005; OECD, 2007; Pollitt, 2006) tend to underline national differences not only in terms of the objectives and goals, but also as regards the steering methods, methodological devices and instrumental components mobilized.¹ Despite their interest, they seem to suffer from several limitations. First, they are relatively rare and almost never address the French case (Ongaro, 2008) and the local public sector. Second, this research, based mostly on documentary studies consisting of speeches, reports and programmes, is more likely to highlight discursive or decisional convergences than convergences of practice (De Bruijn, 2009; Guenoun and Salery, 2009).

Thus, the purpose of this article is to use descriptive research to define and test an interpretative and analytical framework making it possible to characterize local public performance initiatives in France. This framework is intended to be more comprehensive than those generally used, taking into account all aspects of performance initiatives, namely the technical aspects related to the tools and methodologies used, as well as managerial aspects related to the steering method and the allocation of responsibilities.

To this end, this article will start with a conceptual and empirical literature review of practices identified in the field of public performance management to provide an interpretative framework of the typical features of a local performance initiative. We will then test the realism of our theoretical framework by means of quantitative analyses conducted on the features of the performance initiatives of French local authorities. Drawing on these findings, we highlight the significant management methods and tools of local performance management in the French local public sector and discuss, in the light of the literature, the features of the French model.

Presentation of the theoretical and conceptual framework for analysing local public performance management practices

We begin by outlining the limitations of previous studies in relation to our research objective, and then then go on to propose an analytical framework in response to the existing limitations.

The methodological and geographical limitations of previous research

For a long time confined to performance measurement and evaluation activities, public performance management has evolved into a more systematic and holistic conception and definition (Moynihan, 2008). Defined as a management system or model, it first and foremost manifests itself through the definition of performance indicators and targets, and the collection of information about them. It goes on to use this information for strategic and budgetary decision-making (Moynihan, 2006) as well as for resource management, particularly human resources (Den Hartog et al., 2004) or public programmes and policies (Bouckaert and Van Dooren, 2002).² These approaches aim to improve internal and external performance (Moynihan, 2005; Yang and Pandey, 2009) and claim, implicitly or explicitly, a relationship of subordination and causality between these two levels of performance (Boyle, 2009; Walker et al., 2012). According to Heinrich (2003), improving organizational efficiency thus only makes sense if it is accompanied by and contributes to an improvement in the external performance of public actions and policies and, more generally, the ability of public organizations to create public value within populations and social groups. Along the same lines, Bouckaert and Halligan (2008) and Halligan et al. (2012) illustrate, for their part, that the impacts and effects of public action are the key concepts of the current performance management systems and therefore represent a fundamental criterion for evaluating public performance. In France, this type of management is prompted, in part, for the local authorities who voluntarily decided to adopt it in the mid-2000s, by a local and operational transcription of the general principles of Public Finance Legislation (LOLF)³ (Kuhlmann, 2010a; Rouban 2008), reflecting the top-down influence from the central to the local.

Research addressing, directly and indirectly, the theme of national performance management models or systems can be classified into two categories. The first group includes national studies or studies on so-called homogeneous regional areas. The cases studied in the United States (Chan, 2004; Folz et al., 2009), New Zealand (Northcott and Taulapapa, 2012), Canada (Chan, 2004), England (Sanderson, 2001) and the Nordic countries (Johnsen and Vakkuri, 2006) in particular have been the subject of quantitative analyses identifying the practices in this area. However, compared to our research focus, these studies have several limitations. First of all, they concern mainly Anglo-Saxon or Nordic countries (Ongaro, 2008), none addressing the French case. Second, the characterization of practices is carried out on the basis of aspects that are either relatively restricted,⁴ or, on the contrary, abstract or general, revolving mainly around the dissemination of the general principles and pillars of NPM. Finally, these studies more often than not aim to assess the degree of penetration and effective use of this type of management by organizations and therefore do not look at the detail and characterization of the actual practices. However, and in view of our research question, several lessons can be drawn from this research. Anglo-Saxon countries thus seem to tend towards more complex and formal models, and a greater focus on external evaluation, in terms of the impacts of public policies and the measurement of service quality. It is also interesting to note an increasing integration of evaluation in the strategic, operational but also and especially budgetary decision-making processes. The Nordic countries for their part also deserve a mention for the importance awarded to the principles of transparency and ‘accountability’.⁵

The second group of studies comprises international comparative studies (Boyle, 2009; Council of Europe, 2005; OECD, 2007; Pollitt, 2006; Sterck and Scheers, 2006). This body of research, predominantly prescriptive or normative, generally compares the issues, challenges and key factors associated with the implementation of performance management in public sectors. It also avoids entering into the details of the practices and does not make it possible to determine the strong national influencess or trends in this area. In addition, the French context is not addressed by most of the research (Boyle, 2009; Curristine 2007; Sterck and Scheers, 2006) or only very briefly (Curristine, 2005). Within this category, the research covering comparative public administration (Kuhlmann, 2010a, 2010b; Wollmann, 2004, 2008) differs not only by characterizing national or regional models, but also by studying the influence of the political-administrative structures and legacies on the dynamics of the managerial modernization of public organizations. However, these focus more on the reform programmes of the central government than on the actual practices of public organizations. In addition, performance management is often discussed only indirectly and is only one aspect of the more general study of the dissemination of the managerial reforms and principles of NPM within public governments. For the French case, this research is nevertheless of interest to us in showing that while the key factors of NPM have taken root within French government, they have not entirely supplanted reflexes and mechanisms that make up the legalistic and hierarchical type bureaucratic model, leading to a mixed or hybrid system (Ongaro, 2008).

In this context, the study by Bouckaert and Halligan (2008) deserves further analysis as it is the only one to propose a formalized analytical framework making it possible to characterize and compare public performance management practices. To do so, the authors determine standard models or ideals, which they compare with the experiences and initiatives taken in the field of performance management in a number of OECD countries. They build their analytical framework by breaking down the performance management process into three generic steps or activities: (1) performance measurement, with respect to the collection and processing of data into information, (2) the integration of the information with the procedures, documents and discourses of the different actors, and (3) use, i.e. the effective inclusion of this information in the decision-making process (whether budgetary, operational or strategic). The differences observed with respect to these methods give rise to the establishment of four standard models of management, which correspond roughly to four stages of development and sophistication of performance management. These models are characterized by increasing complexity (measured by the number of steps and methodological components mobilized) and the degree of organizational integration of performance management in the internal decision-making and operational processes. The four models are, successively, ‘performance administration’ limited to a performance measurement that is not followed by a real performance management, ‘performance management’ that applies to specific functions within the organization, ‘performance management’ that reflects a comprehensive, integrated and systematic approach, and finally ‘performance governance’ which is characterized by a strong integration of external stakeholders in the initiative. According to the authors, France is at the level of performance administration which would reflect, in the light of what is observed by a part of the literature cited above, a change in the administrative system towards a neo-Weberian type model.

However, despite its many contributions (Rhodes et al., 2012),⁶ this classification grid, as well as the resulting typology of performance management models, has a number of limitations with regard to our research. On the one hand, only the instrumental and procedural dimensions are actually taken into account. Yet, the managerial aspects related to the management methods and styles associated with the implementation of these initiatives strike us as equally important to characterize the practices of public organizations. On the other hand, while this classification grid appears particularly adapted to national agencies and central governments, it is much less operational for local authorities, failing as it does to take into account the different categories of actors behind the implementation and steering of managerial innovations. However, according to the local authorities, significant differences may occur between a predominantly political steering of innovation processes and a purely administrative management (Demir, 2009).⁷ A final limitation of Bouckaert and Halligan’s (2008) model relates to its empirical validation, or at least its application in the field. As they currently stand, the analytical dimensions are not empirically tested and validated among a sample of public organizations. In addition, and this is one of the limits outlined by the authors themselves, the object of the analysis does not relate so much to the actual practices, as to the change programmes and the reforms officially announced.

The limitations of previous research in this area thus lead us to propose a specific analytical framework of the performance management models in the local public context.

The proposal of a specific analytical framework of the performance management models of local authorities

To overcome these limitations, we propose here a specific framework for analysing the public performance management models linking the four standard ideals put forward by Bouckaert and Halligan (2008) with the instrumental and managerial characteristics of local performance initiatives. The objectives of this analytical framework are numerous. First of all, conceptually, the aim is to adapt and complete Bouckaert and Halligan’s model (2008) by looking at the four standard ideals according to the steering practices of local performance initiatives. Second, on a theoretical level, the interest of this analytical framework is to approach the components and dimensions of public performance, by integrating its multidimensional (Scott, 1987), antinomian (Quinn, 1988) and hierarchical (Morin et al., 1994) characteristics. To meet these objectives, we draw on the work of Bouckaert and Halligan (2008), as well as on case studies (Favoreu et al., 2011) characterizing performance initiatives in the French local public context⁸ around tools and methods of implementation.

A framework making it possible to first analyse the tools mobilized

As far as this first point is concerned, Favoreu et al. (2011) highlight three instrumental components of the performance management initiatives, namely the planning, accountability and evaluation tools. Planning is first guided by strategic and operational aspects. The strategic formulation may also be associated with an adaptation of the allocation of resources, for example by the reorganization of the budgetary architecture around the authority’s public policies (Carassus and Gardey, 2009). Operational planning consists, for its part, of the definition or declension of strategic pathways into service projects leading to the implementation of public policies (Beauchemin et al., 2009).

Then, the tools of accountability, whether collective or individual, allow, according to Toulouse (2007), the exercise of good governance by actors who are accountable for their choices, their decisions, their resources and their results. At a more operational level, service projects can thus be seen to emerge in local authorities, making it possible to drive more global strategic orientations, at structural or individual levels. Finally, evaluation tools are the third instrumental component and can be characterized by their objects, their dimensions and their criteria. Thus, on the first point, the evaluation may look at the public policies to drive the organization towards achieving its strategic objectives, or can also focus, in an approach more internal to the authority, on the performance of the departments and agents (Carassus and Gardey, 2009). In turn, the evaluation tools can rely on several dimensions, not only endogenous and specific, mainly concerning resources and modes of operation of the authority, but also exogenous, relating, for their part, to the public service produced or the solution to the needs of the territory concerned (Huteau et al., 2008). Finally, evaluation tools are also characterized by their criteria, whether they are related to the relevance, quality, effectiveness or efficiency of public action. These different criteria, resulting from many theoretical models of public performance management (Demeestere, 2005; Hood, 1995; Pollitt and Bouckaert, 2004), can also be mobilized by the authorities involved in managerial innovation initiatives, for example, through Annual Steering Plans and Annual Steering Reports that put into perspective the performance measurements of their public policies and actions.

An analytical framework then taking into account the terms of implementation of the local performance initiative

Beyond tools, Favoreu et al. (2011) also characterize the local performance initiatives by their methods of implementation, through an analysis of the level of integration of local stakeholders in the steering of the authority. In this context, the system of integration of the actors can first be of a political and administrative nature, a process where political support is associated with the involvement of directors-general and administrative officials, determining the commitment of the intermediate levels and operational staff levels (Yang and Hsieh, 2007). The integration can also be exclusively administrative, a situation where the leadership, exercised only by officials, concerns in particular the management of the interactions and relationships between different hierarchical levels, and the reduction of resistance to change (Fernandez and Rainey, 2006). Finally, integration can also be broader by opening the public policy formation process and, more generally, the decision allocation system to all civil society to mobilize its components (Le Gales, 2006). Integration then follows a democratic logic, with a wide stakeholder involvement in the initiative, including citizens.

An analytical framework defining a total of four performance management models adapted to the local context

Based on the analytical framework developed by Bouckaert and Halligan (2008), we can thus propose a renewed analytical framework of the different local public performance management models (Table 1). This framework shows, in the rows, the instrumental and methodological characteristics identified by the study described above by Favoreu et al. (2011). These first elements are linked with the four performance management models pinpointed by Bouckaert and Halligan (2008), in the columns, to complement and adapt their analyses to the local context:

the first model, ‘performance administration’, can be defined by the absence of tools (planning and/or evaluation) and by management methods limited only to administrative actors.

The second model, ‘managements of performances’, by complementing the features of the first, broadens the scope of the tools and methods mobilized, but remains an internal logic with managerial implications. The level of integration of the initiative is both administrative and political, but with a partitioned management of steering methods: only administrative actors appear accountable for the operational steering of their actions and activities.

Then, the third model, ‘performance management’, is characterized by a governance that is co-produced by political and administrative actors. Thus, this model highlights a political and operational steering of the public action and structures by all the players of the organization, from elected representatives to officials, empowered and evaluated on the basis of their activities in an inclusive sense.

Finally, the last model, ‘performance governance’, includes all the features of the previous model, opening it further to the territory and the environment of the authority. The level of integration of the actors in the performance initiative is therefore democratic, by co-building and sharing the strategy of the authority, by involving and empowering all actors in the organization, both internal and external.

Table 1.

Proposed framework for the characterization of the different local public performance management models

Features of the local performance initiatives				Models
Features of the local performance initiatives				Perf. admin.	Perf. manag.	Managements of Perfomances	Perf. gov.
Steering tools	Planning		Strategic and budgetary			x	X
	Planning		Operational		x	x	X
	Accountability		Individual			x	X
			Political-administrative		x	x	X
			Administrative	x
	Evaluation	Objects	Public policies			x	X
		Objects	Departments, officials		x	x	X
		Dimens.	Territorial				x
			Public services			x	x
			Organizational		x	x	x
			Human		x	x	x
			Financial	x	x	x	x
		Criteria	Relevance				x
			Quality			x	x
			Effectiveness		x	x	x
			Efficiency	x	x	x	x
Steering methods	Level of integration of the local actors in the steering		Democratic				x
			Political-administrative		x	x	x
			Administrative	x

The first two models, ‘performance administration’ and ‘managements of performances’, thus appear limited to the administrative actors in the authority. In contrast, the other two models, of ‘performance management’ and ‘performance governance’, differentiated by openness to wider stakeholders, including users/citizens/taxpayers, move beyond the involvement of the administrative actors alone. Thus, for the sake of simplification, to avoid categorization at four levels, we will be able to refer to a categorization at only two levels, depending on the model, either administrative or political.

Having proposed this renewed framework, our study goes on to test its relevance based on a quantitative analysis conducted among French local authorities.

Analysis of the French public performance management model in local authorities

We start by presenting the research methodology used, before moving on to the results obtained. Finally, these results are discussed in the light of our original theoretical framework.

An exploratory and quantitative research methodology

Thanks to the support of the AFIGESE-CT (Association Finance, Gestion et Evaluation des collectivités territoriales – Financial, management and evaluation association of local authorities), a questionnaire was sent in 2009 to the executive teams of the member authorities of this network of around 400 authorities.⁹ We obtained 60 usable responses from authorities who had already initiated a local public performance initiative. This sample, broken down into 4.6 percent regional councils, 29.5 percent general councils, 21 percent EPCI (intercommunal bodies) and 45 percent municipalities, has few authorities but it is relatively close to the reference population, that is to say nearly 70 performance initiatives launched at the time of the distribution of the questionnaire.¹⁰ The study questionnaire is divided into two parts. The first, descriptive part employs dichotomous yes/no answers on the methods and tools of the performance initiatives implemented in the authority concerned (see the Appendix for excerpts from the questionnaire items). The second, more analytical part, with responses from 1 to 7 on a Likert scale, addresses the actual and perceived impacts of the implementation of these new managerial principles and tools.

After checking the quality of the data collected, we used logistic regression to determine the predictors for each subject of our analytical framework. In fact, this methodological approach allows us to highlight the statistically significant variables, because logistic regression sets out to assess the contribution of different variables on the probability of occurrence of an event (the binary nature of the answers allowed the use of this method). The interpretation of regression models is based on pseudo-R² and the ability of the logistic regression model to account for reality. Thus, the percentage of correct classification is given for each regression. The results include descriptive statistics of the answers and some regressions completing the analysis.

Results characterizing the practice of French authorities through an administrative model

We describe here the results obtained for each steering characteristic included in our analytical framework: the planning, accountability and evaluation tools along with the integration methods.

Planning tools with more of an operational nature

Of the 60 answers, 36 authorities reported actually using planning tools. We identified that this planning could be geared towards the operational or strategic dimension, or cover both of these aspects (Table 2).

Table 2.

Descriptive statistics for the planning tools

	Yes	No	Total
Planning tools	36	24	60
– including strategic and budgetary planning	21	39	60
– including operational planning	32	28	60

Chi-square between the two methods: 9.91, ddl 1, signif. 0.002.

On this point, the authorities report less use of planning described as strategic and budgetary than of operational planning. Moreover, according to the dependence test, it appears that authorities reporting use of planning tools frequently mobilize both methods identified. According to our analytical framework, this seems to point to a more administrative than political model, methods of planning being focused on the daily activity of the services.

Accountability tools of a political and administrative nature

Of the 60 responding authorities, 47 reported having implemented accountability tools (Table 3). Our scale offers three possible methods taken from the literature review: accountability tools of a political-administrative nature, of an administrative nature or of an individual nature.

Table 3.

Descriptive statistics for accountability tools

	Yes	No	Total
Accountability tools	47	13	60
– of which political-administrative accountability	36	24	60
– of which administrative accountability	31	29	60
– of which individual accountability	9	51	60

Chi-square between the methods: not significant.

The distribution of the answers thus focuses on accountability of a political-administrative nature, relatively close to administrative accountability. This result corresponds rather to a performance model combining political and administrative implications, as the elected representatives may play more or less of an instigating role and be more or less involved in the initiative. Also, this position could be analysed in addition to other results.

Evaluation methods, rather endogenous in nature

The evaluation conducted within the framework of the public performance initiatives is characterized, in our interpretative framework, by its objectives, its dimensions and criteria.

Based on three binary questions that make it possible to ascertain whether the objectives of the evaluations concern the directorates and officials, we obtained results from 33 authorities (Table 4). The result is lower for the method relating to the evaluation of public policies (24 authorities). We can thus note that these results indicate a low level of development of the evaluations and, where available, missions concerning more the directorates and officials, which corresponds in our framework to a model of an administrative nature.

Table 4.

Descriptive statistics on the objectives pursued during the evaluations conducted

	Yes	No	Total
Presence of evaluation tools	41	19	60
– including evaluation of the directorates and agents (three items)	33	27	60
– including evaluation of the public policies (six items)	24	36	60

Chi-square between the evaluation object methods; 0.181, not significant.

Among these assessments conducted within the authorities, we studied the dimensions favoured among the five variables in our interpretative framework: political-environmental or territorial, public service quality, managerial or organizational, human capital development and economic and legal (Table 5a).

Table 5a.

Statistics for the answers relating to the evaluation dimensions

	Yes	No	Total	Results of the chi-square
Political-env/territorial	17	43	60	–	–
Quality of public services	33	27	60	14.66	–
Managerial	35	25	60	8.82	32.01	–
Development of human capital	16	44	60	12.54	13.23	16.21	–
Economic-legal	42	18	60	10.66	31.42	36.01	9.36	–

The chi-squares between the five methods are significant at 1%.

Table 5b.

Logistic regression on the variables measuring the dimensions of the evaluation

	B	E.S.	Wald	ddl	Sig.	Exp(B)
Political-env/terr.	35,112	8,942,816	0.010	1	.997	1.8e11
Quality of pub. serv.	3044	2635	12,000	1	.091	1.2e9
Managerial	6724	1439	21,844	1	.004	12,904
Dev. of human capital	37,162	9,510,591	0.19	1	.997	1.3^e14
Economic-legal	14,904	1911	0.1	1	.008	9273
Constant	−9982	2277	19,217	1	.000	000

R² of Nagelkerke: 0.84; correct classification: 0.91%.

The results obtained show that dependencies appear among the five dimensions identified for evaluation. In addition, the most commonly used dimensions (‘managerial’ and ‘economic-legal’) correspond to a model described as administrative, as guided by more of an internal than a political logic. Logistic regression (Table 5b) confirms the pervasiveness of these two dimensions that are statistically significant at 1 percent.

Regarding the last element of characterization of the evaluation methods, we have identified four items for the scale of evaluation criteria:¹¹ effectiveness, efficiency, quality and relevance (Table 6a).

Table 6a.

Descriptive statistics on the evaluation criteria

	Yes	No	Total	Results of the chi-square between the methods
Effectiveness	37	23	60	–
Efficiency	36	24	60	34.95	–
Quality	33	27	60	22.28	21.45	–
Relevance	32	28	60	40.5	26.06	25.31	–

The chi-squares between the four methods are significant at 1%.

On this last point, the results highlight not only the consistency of the criteria in terms of scale, but also a degree of uniformity in the use of the criteria questioned. However, logistic regression (Table 6b) shows that effectiveness and efficiency are the most explanatory variables, at the threshold of 10 percent, of the use of evaluation criteria. This corresponds in our framework to more of an administrative model, insofar as these criteria are focused on internal objects, namely the objectives, means and achievements of the authority.

Levels of integration of the local performance initiative, of a political-administrative nature

Finally, we identified in our interpretative framework three types of methods of steering local performance initiatives: integration of an administrative, political and administrative, and democratic nature (Table 7).

Table 6b.

Logistic regression on the variables measuring the evaluation criteria

	B	E.S.	Wald	ddl	Sig.	Exp(B)
Effectiveness	5481	1.188	1128	1	.008	.004
Efficiency	4.382	1.517	5.349	1	.011	.013
Relevance	21,327	171,473	0.144	1	.997	.000
Quality	19,412	101,451	.000	1	.984	1397^e14
Constant	3.584	1014	12,495	1	.000	0.0

R² of Nagelkerke: 0.782; percentage classification correct: 92.3%.

Table 7.

Answers on the three methods of integration of the local performance initiatives

	Yes	No	Total
Political-administrative integration (eight items)	36	24	60
Democratic integration (eight items)	3	57	60
Administrative integration (nine items)	8	52	60

Of the three methods of integrating local performance initiatives, that of a political-administrative nature – combining the involvement of elected representatives and local officials – appears to be the one that is most frequently used. Integration of a democratic nature, involving users/citizens, and of an administrative nature, where only the territorial officials are involved, is only rarely used.

Tools and steering methods characterizing the French practices through a model that is more administrative than political

Overall, taking into account all the features of the performance initiatives undertaken by the French authorities, the results of our quantitative analyses show that French practices are mainly influenced by an approach and steering of a significantly administrative nature, qualified on our framework by the ‘performance management’ model (see Table 1). Planning within this model is more operational in nature with accountability of the administrative actors. The elected representatives are integrated into the initiative, but passively, without being at the heart of it. The main purpose remains the control of resources, particularly financial, incurred by the authority. The evaluation mechanisms, where they exist, are for their part focused on endogenous performance, revolving around the organizational and financial aspects of the authority. In order to both complement and synthesize these initial analyses, a comprehensive exploratory factor analysis includes all the variables that make up our analytical framework to position the respondent authorities. In this sense, the principal component analysis (PCA) revealed four factorial axes, or components, having an individual value greater than 1, whose factorial structure is illustrated in Table 8 and represents 72 percent of the explained variance. The first component (denoted comp. 1), where more than half of the authorities are present, is characterized by a system of administrative accountability, an evaluation of an economic-legal and organizational nature, a low level of integration of the other actors, elected representatives and users/citizens, and endogenous evaluation criteria prioritizing efficiency and effectiveness of resources. This axis corresponds to the two models of an administrative nature in our framework (see Table 1). The other factorial axes are, in turn, more influenced by the other two models of a political nature, in that they embrace methods of strategic planning, political or even democratic actors, accountability tools that are also focused on the agents and exogenous evaluation methods, covering the public service and territorial dimensions.

Table 8.

Matrix of the components

	Comp. 1	Comp. 2	Comp. 3	Comp. 4
Strategic planning tools	.075	.196	.762	.345
Operational planning tools	.402	321	.600	−.048
Political-administrative accountability tool	.032	.791	.041	.008
Administrative accountability tool	.492	.311	.136	−.149
Individual accountability tool	.211	.009	.911	.408
Evaluation of the territorial dimension	.452	.591	.205	.654
Evaluation of the public service quality dimension	.465	.554	.422	.687
Evaluation of the managerial dimension	.871	.498	.072	.283
Evaluation of the human capital dimension	.119	.656	.337	.492
Evaluation of the economic-legal dimension	.810	.382	.246	.368
Political-administrative integration	.137	.850	.516	.221
Democratic integration	.016	.013	.751	.382
Administrative integration	.381	.204	.118	.036
Evaluation of the public service quality criterion	.280	.190	.840	.210
Evaluation of the relevance criterion	.370	.620	.440	.700
Evaluation of the effectiveness criterion	.890	.610	.130	.380
Evaluation of the efficiency criterion	.830	.570	.580	.290
% variance explained	44.5%	11.5%	10%	7%
Spread of the 60 authorities by classification into dynamic clusters¹²	31	20	5	4

With k-means clustering, the 60 respondent authorities are distributed among the four axes, and 31 authorities belong to axis 1. Given our previous observation, the majority of current practices of French local authorities can thus be characterized by a model of an administrative nature. Practices arising from a political model may exist but remain, all in all, marginal. However, they do show the willingness and ability of some authorities to introduce innovative initiatives beyond traditional models and that make it possible to respond to their new constraints, in particular financial.

Discussion of the results

The results of our research show, first, the existence of a French performance management model resulting in trends shared by French local authorities with regard to the instrumental and methodological content, and the methods of steering public performance. It turns out in fact that the tools and methods that constitute performance initiatives, as they currently stand, continue to be guided by administrative and internal logics. Indeed, in general, we highlight the weak political dimension in the impetus, backing, implementation, monitoring of the initiatives adopted voluntarily by French local authorities. This result confirms the work of Carassus and Gardey (2009) highlighting the apolitical nature of performance initiatives rolled out in France. Nevertheless, our study shows that performance management has gained a strong foothold and can be compared to the analyses of Rouban (2008), Kuhlmann (2010a) and Ongaro (2008) on the progressive, incremental, slow and difficult dissemination of this mode of management in the French public sector. In contrast, and contrary to the view put forward by Kuhlmann (2010a), the current initiatives, because of their characteristics and their methodological and instrumental components, appear limited in their ability to improve clarity and transparency, and to promote a management geared towards the users, the impacts and the quality of public action.

In line with Carassus and Gardey (2009) or Huteau et al. (2008), and based on responses from 60 authorities,¹³ our results show, more specifically, that these initiatives are, in large part, driven by territorial civil servants. The weak political dimension is also highlighted (1) by the nature of the planning, predominantly operational, (2) the disconnection between public policies and evaluation mechanisms, (3) by the budgetary and quantitative logics of the indicators implemented, but also (4) by the focus on the optimization of resources and internal needs of the organization. In the light of the concept of performance management and its objectives, as highlighted in the literature and detailed above, the French initiatives appear incomplete or partial in their purpose and scope.

Thus, qualified as administrative rather than political, the performance management model of the French local authorities appears to be marked by the predominance of endogenous and basic control practices, as shown by Batac et al. (2009), Kuhlmann (2010b) and Busson-Villa (1999).

In addition, the French model differs, as it currently stands, from the international practices already analysed. Thus, the initiatives launched in the Anglo-Saxon or Nordic authorities seem to be characterized not only by the close involvement of elected representatives, but also by a desire to put the citizen at the centre of the tools and methods used (Council of Europe, 2005; Johnsen and Vakkuri, 2006). Through the emphasis on analysing the effects of public policy and on the quality indices, they seem to pay more attention to the evaluation of public policies and their impacts on their external environment (Ammons and Rivenbark, 2008; Folz et al., 2009). Our study thus confirms, at the local level, the existing implementation gap between a country with a Napoleonic type administrative model such as France and the Anglo-Saxon countries, a difference that was observed in some of the literature on comparative public administration.

This discrepancy could be explained by the novelty of these holistic initiatives in French local authorities, which have been around for less than ten years, especially when compared to the Anglo-Saxon authorities. The research conducted by Bouckaert and Halligan (2008) thus suggests the existence of a life cycle of performance initiatives that evolve during their development and their degree of maturation from a relatively simple and ad hoc model, limited only to the monitoring of activity indicators and resource consumption, to a more complex model including a variety of issues, actors and evaluation dimensions and integrating all internal decision-making processes (political-strategic, budgetary, operational).

In contrast to this interpretation, which may appear normative or prescriptive, there is a more contingent or institutional analysis which highlights the influence of national political-administrative and cultural contexts on the characteristics and dynamics of development of public performance management approaches. The idea here is that there is no such thing as an ideal model or typical development pathways but approaches whose effectiveness, legitimacy and sustainability are the result of their adequacy and compliance in the face of national politico-administrative systems and traditions (Ongaro, 2008; Rhodes et al., 2012). And well beyond this time factor, the administrative nature of the French local model is believed to reveal itself and affirm itself not only in the original planning, accountability and evaluation features, but also in the way in which these initiatives are implemented and steered. This deviation from Anglo-Saxon practices would thus reveal dimensions that are more cultural and institutional (Rhodes et al., 2012) than temporal. This would then confirm the analyses of Wollmann (2008), Kuhlmann (2010b), and Rhodes et al. (2012) who argue that, especially in the case of the French public sector, the importance of institutional traditions and legacies leads to the persistence of bureaucratic behaviour and modes of operation of a legalistic and hierarchical nature, despite the effective diffusion of NPM. We are therefore seeing the coexistence of these two modes of public management and the emergence of a mixed or hybrid neo-Weberian type system (Kuhlmann, 2010a).

However, the differences observed, especially the highlighting in our sample of initiatives that are relatively open to their environment and geared towards the evaluation of the effectiveness and impact of public action, show, in our view, that national political and administrative factors, while important, do not alone explain the dynamics of managerial innovation in public organizations. More local and individual factors related to organizational and managerial characteristics and learning effects appear to explain the differences noted between public organizations in the same country. We believe, like Ongaro (2008), that a combination of macro and micro factors determines the performance management dynamics of public organizations and that, despite the importance of national political and administrative institutional determinisms, the local authorities have leeway and potential for evolution.

Conclusion: a French model of local performance management currently guided by an administrative logic, which risks generating only weak lessons

The definition and testing of an analytical framework of local public performance initiatives as well as the quantitative characterization of the practices of French authorities are the main empirical and methodological contributions of our research. Our observations thus make it possible to highlight, at the instrumental and steering levels, the gap and differing pathways observed with the Anglo-Saxon and Nordic countries in some of the literature. The administrative nature of the performance management model in French local authorities, highlighted by our study, could eventually, if the current model does not change, generate only very few lessons. Indeed, the weak political dimension of French local initiatives could, first of all, have both managerial and democratic consequences. Managerial, on the one hand, due to a lack of relevance in determining public policies with environmental requirements or in the allocation of resources to the authority’s goals. Democratic, on the other hand, due to the poor quality of reporting outside the authority.

It therefore strikes us as important to analyse the terms and conditions of the transition from an administrative model characterizing the French practices to a more political model. The democratic and managerial challenges of these local performance initiatives could then be activated in a way that is complementary to current practices, in particular to meet the new demands and requirements of all local stakeholders. Highlighting national models is fundamental because it paves the way for the identification and analysis of local determinisms, whether historical, institutional, political or economic in nature, that affect the adoption process or innovation in the field of public performance management. This type of research makes it possible, on the other hand, to shed light on the international or inter-regional transferability of managerial methods. Nevertheless, we believe that the analysis of local and micro determinants may be relevant for understanding the dynamics and features of local public performance management.

One of the extensions of this research could thus consist of a study of local organizational and managerial factors influencing the degree of adoption and use of these initiatives by French local authorities. Longitudinal analyses would thus be interesting to throw light on the dissemination, ownership and institutionalization mechanisms of these practices, at the micro and organizational levels, and in particular to analyse the potential influence of time, individual and managerial factors and learning effects in the change process.

Footnotes

Notes

Christophe Favoreu is a professor of strategy at Toulouse Business School. He holds a PhD in Strategic Management. His main research interests cover public management, performance management, public strategic management as well as public policy clusters.

David Carassus is an assistant professor at the université de Pau et des Pays de l’Adour in France. He holds a PhD in Management Sciences. He supervises the researches of a dozen PhD students. His main research interests cover the evaluation of public politics and management of local institutions.

Damien Gardey is a professor of management at Pau Business School. He holds a PhD and a master’s in Public Management from the IAE of Pau. His main research interests cover public management, performance management, public performance and governance.

Christophe Maurel is an assistant professor at Le Mans University (France). He holds a PhD in Management Sciences. He supervises the researches of PhD students and his main research interest is the management control and evaluation of public politics.

Appendix

References

Ammons

Rivenbark

(2008) Factors influencing the use of performance data to improve municipal services: Evidence from the North Carolina Benchmarking Project. Public Administration Review 6(2): 304–318.

Batac

Carassus

Maurel

(2009) Evolution normative du contrôle organisationnel: application au contexte public local. Finance Contrôle Stratégie 12(1): 155–180.

Beauchemin M, Gardey D, Leuret C and Lion B (2009) Regards croisés France-québec en matière de gestion de la performance publique locale. 5ème Workshop Ville-Management, en partenariat avec le Conseil de l’Europe, Pau.

Bouckaert

Halligan

(2008) Managing Performance: International Comparisons, London: Routledge.

Bouckaert

Van Dooren

(2002) Performance measurement: Getting results. Public Performance & Management Review 25(3): 329–335.

Boyle R (2009) Performance Reporting: Insights from International Practice. Washington, DC: IBM Center for the Business of Government.

Busson-Villa

(1999) L’émergence d’une logique évaluative dans la gestion publique: le cas des organisations communales. Finance Contrôle Stratégie 2(1): 5–25.

Carassus

Gardey

(2009) Une analyse de la gestion de la performance par les collectivités locales françaises: un modèle administratif ou politique? Revue Française de Finances Publiques 107: 101–129.

Chan

YC-L

(2004) Performance measurement and adoption of balanced scorecards: A survey of municipal governments in the USA and Canada. International Journal of Public Sector Management 17(3): 204–221.

10.

Council of Europe (2005) Rapport sur la gestion des performances au niveau local, Résultats et conclusions d’une enquête auprès des membres du CDLR.

11.

Curristine

(2005) Performance information in the budget process: Results of the OECD 2005 questionnaire. OECD Journal on Budgeting 5(2): 87–131.

12.

Curristine

(2007) Performance Budgeting in OECD Countries, Paris: OECD.

13.

De Bruijn

(2009) Compte rendu critique de livres. Revue internationale de Sciences administratives 75(1): 225–231.

14.

Demeestere

(2005) Le contrôle de gestion dans les organisations publiques, Paris: LGDJ.

15.

Demir

(2009) The complementarity view: Exploring a continuum in political–administrative relations. Public Administration Review 69(5): 876–888.

16.

Den Hartog

Boselie

Paauwe

(2004) Performance management: A model research agenda. Applied Psychology: An International Review 53(4): 556–569.

17.

Favoreu C, Carassus D, Gardey D and Maurel C (2011) La caractérisation et la définition de la performance publique: une application aux collectivités locales. Actes colloque AIRMAP, Versailles-Saint Quentin.

18.

Fernandez

Rainey

(2006) Managing successful organizational change in the public sector. Public Administration Review 66(2): 168–176.

19.

Folz

Abdelrazek

Chung

(2009) The adoption, use and impacts of performance measures in medium-size cities: Progress toward performance management. Public Performance & Management Review 33(1): 63–87.

20.

Guenoun

Salery

(2009) La mesure de performance publique locale: variations françaises et anglo-saxonnes. Gestion & Finances Publiques 11: 893–911.

21.

Halligan

Sarrica

Rhodes

(2012) On the road to performance governance in the public domain? International Journal of Productivity and Performance Management 61(3): 224–234.

22.

Heinrich

(2003) Measuring public sector performance and effectiveness. In: Peters

Pierre

(eds) Handbook of Public Administration, London: Sage, pp. 25–37.

23.

Hood

(1995) Contemporary public management: A new global paradigm? Public Policy and Administration 10(2): 104–117.

24.

Hood

Lodge

(2006) The Politics of Public Service Bargains: Reward, Competency, Loyalty and Blame, New York: Oxford University Press.

25.

Huteau S, Carassus D, Diaz S and Gardey D (2008) Evolution de la gestion locale vers l’amélioration de la performance: le cas des départements. In: l’Annuaire du GRALE, 43–59.

26.

Johnsen

Vakkuri

(2006) Is there a Nordic perspective on public sector performance management? Financial Accountability & Management 22(3): 291–308.

27.

Kuhlmann

(2010a) New Public Management for the classical continental European administration: Modernization at the local level in Germany, France and Italy. Public Administration 88(4): 1116–1130.

28.

Kuhlmann

(2010b) Les démarches pour mesurer la performance des administrations européennes locales: une analyse comparative des expériences de réforme en Grande-Bretagne, en France, en Suède et en Allemagne. Revue internationale de Sciences administratives 76(2): 367–372.

29.

Le Gales

(2006) Gouvernement et gouvernance des territoires. Problèmes politiques et sociaux, no. 922, mars, Paris: La Documentation française.

30.

Morin

Savoie

Beaudin

(1994) L’efficacité de l’organisation – Théories, représentations et mesures, Boucherville, Québec: Gaëtan Morin édition.

31.

Moynihan

(2005) Managing for results in an impossible job: Solution or symbol? International Journal of Public Administration 28(3–4): 213–231.

32.

Moynihan

(2006) Managing for results in state government: Evaluating a decade of reform. Public Administration Review 66(1): 78–90.

33.

Moynihan

(2008) The Dynamics of Performance Management: Constructing Information and Reform, Washington, DC: Georgetown University Press.

34.

Newcomer

(2007) Measuring government performance. International Journal of Public Administration 30: 307–329.

35.

Northcott

Taulapapa

(2012) Using the balanced scorecard to manage performance in public sector organizations. International Journal of Public Sector Management 25(3): 166–191.

36.

OECD (2007) Performance Budgeting in OECD Countries, Paris: OECD.

37.

Ongaro

(2008) Introduction: The reform of public management in France, Greece, Italy, Portugal and Spain. International Journal of Public Sector Management 21(2): 101–117.

38.

Osborne

Gaebler

(1992) Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector, Reading, MA: Addison-Wesley.

39.

Pollitt

(2006) Performance management in practice: A comparative study of executive agencies. Journal of Public Administration Research and Theory 16(1): 25–44.

40.

Pollitt

Bouckaert

(2004) Public Management Reform: A Comparative Analysis, Oxford: Oxford University Press.

41.

Quinn

(1988) Beyond Rational Management: Mastering the Paradoxes and Competing Demands of High Performance, San Francisco, CA: Jossey-Bass.

42.

Rhodes

Biondi

Gomes

Melo

Ohemeng

Rossi

Sutiyono

(2012) Current state of public sector performance management in seven selected countries. International Journal of Productivity and Performance Management 61(3): 235–271.

43.

Rouban

(2008) Reform without doctrine: Public management in France. International Journal of Public Sector Management 21(2): 133–149.

44.

Sanderson

(2001) Performance management, evaluation and learning in ‘modern’ local government. Public Administration 79(2): 297–313.

45.

Scott

(1987) The adolescence of institutional theory. Administrative Science Quarterly 32(4): 493–511.

46.

Sterck

Scheers

(2006) Trends in performance budgeting in seven OECD countries. Public Performance & Management Review 30(1): 47–72.

47.

Toulouse J-M (2007) Rapport sur la gouvernance d’organisations privées et publiques. Rapport de recherche sur la gouvernance des institutions universitaires.

48.

Walker

Boyne

Brewer

(2012) Public Management and Performance: Research Directions, Cambridge: Cambridge University Press.

49.

Wollmann

(2004) Local government reforms in GB, Sweden, Germany and France. Local Government Studies 30(4): 639–665.

50.

Wollmann H (2008) Comparing local government Reforms in England, Sweden, France and Germany. Available at: www.wuestenrot-stiftung.de/download/local-government.

51.

Yang

Hsieh

(2007) Managerial effectiveness of governance performance measurement: Testing a middle-range model. Public Administration Review 67(5): 861–878.

52.

Yang

Pandey

(2009) How do perceived political environment and administrative reform affect employee commitment? Journal of Public Administration Research and Theory 19(2): 335–360.