1. Introduction and research question Resorting to semi-autonomous organisations and non departmental public bodies for the execution of public functions is far from being a novelty. In many countries, in fact, some activities are carried out by organisations which are in some way disaggregated from the ‘core’ government structure, organised along departmental lines. The governance problems of these bodies have been traditionally a recurrent issues in public administration studies (Wettenhall, 2005). In the last decade, however, this issue has been again on the top of research and political agendas (Pollitt and Talbot, 2004; Pollitt, 2005). Partly under the influence of New Public Management ideas (Hood, 1991), there has been a remarkable global trend toward specialisation and autonomization in the public sector (Pollitt and Bouckaert, 2004).Governments in many countries decided to establish semi- autonomous single purpose organizations (Oecd, 2002). These bodies have been variously named: quangos, non departmental public bodies, agencies. Ministries and other departmental bodies have been split up into several entities and these entities enjoyed some freedom to manage. As a result we often observe a landscape of a highly differentiated public sector with several thousands of “satellites” which operate quite independently and which are often only loosely coupled with their parent and core governments. Many problems are associated to this trend to disaggregation and specialisation: accountability of agencies to legitimized political bodies, divergent goals, information asymmetries, lacking incentives, etc. Consequently, the issue of the “autonomy” of these bodies has become a central one in public management and public administration research. On the other side, the conceptualisation of autonomy is difficult task since it is not only a legal-formal matter (Verhoest et al,2004). Under the organisational research point of view becomes thus relevant to analyse what are the “determinants” and the “antecedent factors” of autonomy of agencies. At this stage, consequently, the main research question can be broadly formulated as follows: What factors do influence the autonomy of agencies? The analysis will focus on the Italian public sector; there is, in fact, still a remarkable lack of empirical and analytical knowledge concerning Napoleonic, Rechtsstaat, south Mediterranean countries in this field. 2. Brief theoretical background Many theoretical perspectives have been adopted to analyse and explain the autonomy of agencies under different lights. We start by sketching a brief reference framework of the factors, drawing on a review of literature. First of all, the agencies’ shape and degree of autonomy can be influenced by the rational choice of the principal (governments and political officials) who delegate powers and functions to agencies. In this mainly functional logic, principals are willing to pay the costs of delegation, if they expect benefits to outweigh those costs (Thatcher e Stone Sweet, 2002). Consequently, they will grant agency some discretion to achieve the benefits of delegation, trying to minimize the risk of agency loss. According to this approach, the main explanatory factors of agency discretion and autonomy are elements like policy complexity, credible commitment (Pollack, 2003) policy conflict and uncertainty (Moe, 1995). Hypotheses related to these factors have been often tested (Elgie, 2006; Yesilkagit, Christensen, 2006; Elgie and McMenamin, 2005; Gilardi, 2002). It could be argued, for example, that agencies which operate in economic policy field are more enjoy a higher degree of autonomy since there is more need to reassure relevant stakeholders about politicians’ credible commitment. Functional logic and the Principal-Agent framework are not the only available analytical tools. It is, in fact, reasonable to assume that institutional and organisational structures and shapes are affected by some cultural, functional and structural features of the public system (Flynn and Strehl, 1996, Kickert, 1997, Olsen and Peters, 1996, Pollitt and Bouckaert, 2004). Such features can be referred to as the politico-administrative context of a country – the differentiated houses in which administrative reforms occur. According to this approach, agency shape and autonomy are highly path-dependent from deeply rooted structures of a given national context. This perspective gives the chance to empirically test the relevance of historical legacies and factors like agency age, as well as the degree of penetration of NPM-inspired designs in different polities (Verhoest et al, 2004). A third set of factors influencing agencies’ autonomy is rooted in the sociological institutionalism approach. Agencies’ autonomy and, more in general, agencies’ establishment and shape can be analysed as the result of external pressures (Olsen, 1992). Isomorphic trends are likely to take place for agencies which are embedded in similar environments where norms and beliefs are shared. For example, agencies that operate in a market or quasi-market setting (Lægreid, Roness, Rubecksen, 2005) or agencies embedded in European and international networks (Pollitt and Talbott, 2004) may be expected to have convergent patterns of autonomy. Finally, explanatory factors can operate at a meso- or micro-level of governance (Pollitt, 2006;Pollitt et al.,2004). Agencies’ autonomy, in fact, may be influenced, first of all, by some features of the primary task they perform (like being observable, in Wilson’s terms, 1989). On the other side, it has been debated the extent to which budget size affects agencies’ discretion (Lægreid, Roness, Rubecksen, 2005; Pollitt, 2005; Gains, 1999). The policy sector, now considered as regards its political salience, can represent a further determinant of autonomy. 3. Major Methodological choices Research strategy The research has been developed through the survey based methodology used in the international research network “COBRA” (Comparative Public Organization Data Base for Research and Analysis) . Cobra is an academic research network in the field of public management. The aim of COBRA is to analyse the patterns of autonomy and control of agencies and NDPBs in different European and extra-European countries through strongly empirically based comparative research. Actually Università Bocconi is the only Italian associated university. Population The analysis has focused on Agencies and Non-Departmental Public Bodies (NDPBs) at the central (national) level of government . The unit of analysis has been only those NDPBs having the feature of being single organisations in the landscape of the Italian public sector. In other words, we have identified the population only in those NDPBs that present the characteristic of being the only one to exercise a given (range of) public function(s). The procedure for identifying the list of bodies to be included has been as follows: on the one hand, we have considered the list of public bodies included in the “consolidated” state budget, and excluded all those that are not single entities (research entities, etc.). Contemporaneously, the web page of all ministries (and some related links) have been surveyed, in order to identify other bodies that, though not included in the consolidated state budget, were detected as “exercising a public function at arm’s length”. It should be added that no juridical classification has been adopted: the list includes departmental bodies, public law agencies, as well as private law entities like state-owned companies. Institutional-organisational features have formed the basis for the definition of the subject of analysis. The final list included 57 entities; the narrow definition of agencies here adopted allowed to analyse the whole population of organisations of the selected kind. In the end, respondents were 41, out of the identified population of 57. Questionnaire Data has been gathered through a questionnaire consistent with the guidelines elaborated by COBRA. The questionnaire has been sent to respondents (CEO of all agencies included in the sample). Statistical analysis The statistical methods used in our research will include first the computation of the univariate frequencies of the dependent variables; the calculation of the bivariate correlations of all relevant relationships; the multivariate analysis of independent variables having significative bivariate correlations. In the first step of the analysis all variables will be included, and then indexes of the dependent variables defining the different dimensions of autonomy will be created.

Autonomy of agencies and semi-Autonomous organisations in Italy

BARBIERI, DARIO;FEDELE, PAOLO;GALLI, DAVIDE;ONGARO, EDOARDO
2007

Abstract

1. Introduction and research question Resorting to semi-autonomous organisations and non departmental public bodies for the execution of public functions is far from being a novelty. In many countries, in fact, some activities are carried out by organisations which are in some way disaggregated from the ‘core’ government structure, organised along departmental lines. The governance problems of these bodies have been traditionally a recurrent issues in public administration studies (Wettenhall, 2005). In the last decade, however, this issue has been again on the top of research and political agendas (Pollitt and Talbot, 2004; Pollitt, 2005). Partly under the influence of New Public Management ideas (Hood, 1991), there has been a remarkable global trend toward specialisation and autonomization in the public sector (Pollitt and Bouckaert, 2004).Governments in many countries decided to establish semi- autonomous single purpose organizations (Oecd, 2002). These bodies have been variously named: quangos, non departmental public bodies, agencies. Ministries and other departmental bodies have been split up into several entities and these entities enjoyed some freedom to manage. As a result we often observe a landscape of a highly differentiated public sector with several thousands of “satellites” which operate quite independently and which are often only loosely coupled with their parent and core governments. Many problems are associated to this trend to disaggregation and specialisation: accountability of agencies to legitimized political bodies, divergent goals, information asymmetries, lacking incentives, etc. Consequently, the issue of the “autonomy” of these bodies has become a central one in public management and public administration research. On the other side, the conceptualisation of autonomy is difficult task since it is not only a legal-formal matter (Verhoest et al,2004). Under the organisational research point of view becomes thus relevant to analyse what are the “determinants” and the “antecedent factors” of autonomy of agencies. At this stage, consequently, the main research question can be broadly formulated as follows: What factors do influence the autonomy of agencies? The analysis will focus on the Italian public sector; there is, in fact, still a remarkable lack of empirical and analytical knowledge concerning Napoleonic, Rechtsstaat, south Mediterranean countries in this field. 2. Brief theoretical background Many theoretical perspectives have been adopted to analyse and explain the autonomy of agencies under different lights. We start by sketching a brief reference framework of the factors, drawing on a review of literature. First of all, the agencies’ shape and degree of autonomy can be influenced by the rational choice of the principal (governments and political officials) who delegate powers and functions to agencies. In this mainly functional logic, principals are willing to pay the costs of delegation, if they expect benefits to outweigh those costs (Thatcher e Stone Sweet, 2002). Consequently, they will grant agency some discretion to achieve the benefits of delegation, trying to minimize the risk of agency loss. According to this approach, the main explanatory factors of agency discretion and autonomy are elements like policy complexity, credible commitment (Pollack, 2003) policy conflict and uncertainty (Moe, 1995). Hypotheses related to these factors have been often tested (Elgie, 2006; Yesilkagit, Christensen, 2006; Elgie and McMenamin, 2005; Gilardi, 2002). It could be argued, for example, that agencies which operate in economic policy field are more enjoy a higher degree of autonomy since there is more need to reassure relevant stakeholders about politicians’ credible commitment. Functional logic and the Principal-Agent framework are not the only available analytical tools. It is, in fact, reasonable to assume that institutional and organisational structures and shapes are affected by some cultural, functional and structural features of the public system (Flynn and Strehl, 1996, Kickert, 1997, Olsen and Peters, 1996, Pollitt and Bouckaert, 2004). Such features can be referred to as the politico-administrative context of a country – the differentiated houses in which administrative reforms occur. According to this approach, agency shape and autonomy are highly path-dependent from deeply rooted structures of a given national context. This perspective gives the chance to empirically test the relevance of historical legacies and factors like agency age, as well as the degree of penetration of NPM-inspired designs in different polities (Verhoest et al, 2004). A third set of factors influencing agencies’ autonomy is rooted in the sociological institutionalism approach. Agencies’ autonomy and, more in general, agencies’ establishment and shape can be analysed as the result of external pressures (Olsen, 1992). Isomorphic trends are likely to take place for agencies which are embedded in similar environments where norms and beliefs are shared. For example, agencies that operate in a market or quasi-market setting (Lægreid, Roness, Rubecksen, 2005) or agencies embedded in European and international networks (Pollitt and Talbott, 2004) may be expected to have convergent patterns of autonomy. Finally, explanatory factors can operate at a meso- or micro-level of governance (Pollitt, 2006;Pollitt et al.,2004). Agencies’ autonomy, in fact, may be influenced, first of all, by some features of the primary task they perform (like being observable, in Wilson’s terms, 1989). On the other side, it has been debated the extent to which budget size affects agencies’ discretion (Lægreid, Roness, Rubecksen, 2005; Pollitt, 2005; Gains, 1999). The policy sector, now considered as regards its political salience, can represent a further determinant of autonomy. 3. Major Methodological choices Research strategy The research has been developed through the survey based methodology used in the international research network “COBRA” (Comparative Public Organization Data Base for Research and Analysis) . Cobra is an academic research network in the field of public management. The aim of COBRA is to analyse the patterns of autonomy and control of agencies and NDPBs in different European and extra-European countries through strongly empirically based comparative research. Actually Università Bocconi is the only Italian associated university. Population The analysis has focused on Agencies and Non-Departmental Public Bodies (NDPBs) at the central (national) level of government . The unit of analysis has been only those NDPBs having the feature of being single organisations in the landscape of the Italian public sector. In other words, we have identified the population only in those NDPBs that present the characteristic of being the only one to exercise a given (range of) public function(s). The procedure for identifying the list of bodies to be included has been as follows: on the one hand, we have considered the list of public bodies included in the “consolidated” state budget, and excluded all those that are not single entities (research entities, etc.). Contemporaneously, the web page of all ministries (and some related links) have been surveyed, in order to identify other bodies that, though not included in the consolidated state budget, were detected as “exercising a public function at arm’s length”. It should be added that no juridical classification has been adopted: the list includes departmental bodies, public law agencies, as well as private law entities like state-owned companies. Institutional-organisational features have formed the basis for the definition of the subject of analysis. The final list included 57 entities; the narrow definition of agencies here adopted allowed to analyse the whole population of organisations of the selected kind. In the end, respondents were 41, out of the identified population of 57. Questionnaire Data has been gathered through a questionnaire consistent with the guidelines elaborated by COBRA. The questionnaire has been sent to respondents (CEO of all agencies included in the sample). Statistical analysis The statistical methods used in our research will include first the computation of the univariate frequencies of the dependent variables; the calculation of the bivariate correlations of all relevant relationships; the multivariate analysis of independent variables having significative bivariate correlations. In the first step of the analysis all variables will be included, and then indexes of the dependent variables defining the different dimensions of autonomy will be created.
2007
EGPA CONFERENCE
Barbieri, Dario; Fedele, Paolo; Galli, Davide; Ongaro, Edoardo
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11565/40755
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact