A Swedish Model of Administration? State level Agency Autonomy from a Multi-Dimensional Perspective
Governments today face complex problems such as migration and environmental threats, problems which cut across territorial and sectoral borders and are truly hard to handle. Politicans and scholars alike now call for innovative public management policies that can enhance the governing capacity and help solving these wicked problems of our time. State level agencies are at the hub of these processes and we aim to study the autonomy of these actors. While agency autonomy can enhance the quality of some decisions, and their legitimacy, it can also constitute an obstacle for inter-agency coordination that results in sub-optimal decisions. Also, the autonomy of state agencies must be matched with democratic control. To develop steering to enhance coordination, we need more and substantial knowledge about the actual autonomy of contemporary state agencies, and be sensitive to how this varies across policy sectors and administrative traditions. Hence, using unique large-N data in combination with interviews and a new survey, we will study formal and informal steering of agencies and their relationships with a broad range of actors in order to determine their autonomous decision-making capability. We study 1) whether agencies are autonomous or not in specific types of decision making (e.g. managerial decisions, policy-making and implementation); and 2) how levels of autonomy vary in relation to different types of actors (to whom or what are agencies autonomous).
Final report
The theme indicated in the title of this project - A Swedish Model of Administration? State Level Agency Autonomy from a Multi-Dimensional Perspective – have gained renewed relevance due to the Covid 19-pandemic. How the Swedish central government administration is organized and how Swedish governments apply steering tools are now issues on the political agenda. This project thus informs analyses and policy development that are likely to follow in the aftermath of the Covid 19-pandemic.
The main aims of this project have been to offer a thorough analysis of the autonomy of Swedish central government agencies, and to study how this autonomy relates to their interaction with other actors. The theoretical frame and the strategy to use different types of methodology have been intact throughout the project. The project has developed in terms of added precision to the initial research questions and contributions that are now represented in publications (see below). It should also be mentioned that the Covid 19-pandemic made it hard to engage in collaboration and conferences. The possibilities to complete the project made available by Riksbankens Jubileumsfond have been invaluable.
The project can be summarized in terms of its sub-studies and their respective research questions, data and analyses. Two sub-studies analyse the Swedish government’s use of formal steering tools and the agencies’ autonomy. These analyses are based on an original database containing information about 188 central government agencies for the period of 1997-2017 (Ahlbäck Öberg & Wockelberg 2022). In this part, the project describes how governments apply different types of steering in their relationship with agencies of different kinds and sizes. One of the sub-studies analyses how governments balance autonomy with control when steering the agencies. Central government agencies with an independent management model, and that are allowed high levels of managerial autonomy, are controlled to a larger degree than others in terms of performance information demands. Thus, governments are found to balance autonomy with control in some dimensions of the government-agency relationship (Ahlbäck Öberg & Wockelberg 2021). A second sub-study has it focus on ex post controls and governments’ demand for performance information. Here, the Swedish application of Management By Objectives and Results (MBOR) is studied over time, including a reform specifically targeting the number of demands for performance reporting and aimed at de-escalating the model in use. This study concludes that central government agencies’ tasks are of importance when governments decide what type of performance information to demand from agencies (Wockelberg & Ahlbäck Öberg 2021a). A third sub-study adds the perspective of informality to the picture of formal steering, control and autonomy. In this study, informal contacts across the organizationally divided national executive are studied from within constitutional and practical perspectives, respectively. The methodology used is content analysis of policy documents and interviews with public servants in the Government Office. A fourth sub-study investigates central government agencies’ interaction with others, with the aim to analyse possible links between agency autonomy and organizational interaction. A comparative design that includes Norway and Sweden is used to study how top-level managers within central government agencies describe organizational interaction, what measures that are applied and their quality. Based on survey-data, this study finds an equally positive perception of interaction in both countries. It also finds that managers at the highest levels have more positive views of interaction than those serving on lower levels within the agencies. In addition, the differences between the cases that is observed is discussed in terms of the constitutional and administrative traditions in respective country. One interesting observation is that the core features of the Swedish model of administration, that is, its strong formal and organizational safe guards for agency autonomy, can be understood as having a positive effect on organizational interaction (Wockelberg & Ahlbäck Öberg 2021b).
One of the main results of this project is 1) that agency autonomy and governmental steering of agencies have been studied as multi-dimensional phenomena and as both formal and informal relationships. Here, our unique database has enabled other types of analyses than those commonly applied in this field. In this part, the main contributions are systematic analyses of, and new knowledge about, how governments steer central government agencies, how agencies are allowed certain types of autonomy rather than other, and why. A related contribution comes with the database being published with open access (Ahlbäck Öberg & Wockelberg 2022). The project also 2) delivers an empirically well-founded, up to date, description of the Swedish model of administration. This model, with its dual executive organization, and its many and in relative terms resourceful, independent central government agencies, is still today distinct from within a comparative perspective. At the same time, this project reveals that the Swedish model of administration is a frame containing some variation in terms of how different types of agencies are controlled and over time. Based on this research, we have gained in knowledge about different factors that influence government steering of agencies. We have also created new knowledge about the Swedish model of administration in the era post New Public Management, confirming prior research describing the dynamic developments of public management models over time. This contemporary picture of the Swedish model of administration is interesting also when it comes to what it implies about organizational interaction. In this part, the project suggest that the stronger safeguards of agency autonomy that a context offers, the lower risks it will be for agencies to interact with others. This result is preliminary, and should be investigated further. An additional contribution stems from 3) the analysis of informality in the government – agency relationship. One conclusion is that even though informal contacts across the dual executive are perceived as important and necessary, they are potentially problematic from a constitutional perspective. The challenges that are identified are the ban of informal steering and the fact that ministerial rule is not allowed. In this part, the project contributes with an up to date constitutional analysis and observations of current practices applied to manage informal contacts without over-stepping constitutional or organizational boundaries. Here, it is interesting to note that some informal relationships rest on carefully designed agreements about when and why specific individuals should communicate informally.
This project offers new knowledge about the Swedish model of administration. It has also inspired new research questions. One of these concerns the suggestion that agency autonomy enables organizational interaction. This is a contribution to an on-going discussion and it would be relevant to investigate our preliminary conclusion about this this relationship further. Moreover, this project informs future research on what central government agencies do, that is, their tasks. Not all agencies carry out traditional tasks like case processing or implementation, some instead produce knowledge and provide expertise and are mandated to apply knowledge as a steering tool towards others. These so called expert agencies have become visible during the management of the Covid 19-pandemic and are present in many policy areas. Their exact role in democratic governance is debated, and should be studied empirically.
The results from the project are disseminated mainly in the form of publications and the abovementioned database being made openly accessible on-line (Ahlbäck Öberg & Wockelberg 2022). The project members have also participated in different fora where the Swedish model of administration is analysed and discussed, such as the Swedish Parliament and central government agencies. They have also served in public inquiry commissions (as member and expert).
References
Ahlbäck Öberg, S. & Wockelberg, H. (2022). The Performance Information Demand Dataset: Swedish governments request for performance information from central government agencies 1993 – 2017. [Svenska regeringars krav på återrapportering från statliga myndigheter 1993 – 2017.] Open access at Swedish National Data Service (SNDS) repository https://doi.org/10.5878/hhp6-ah69
Ahlbäck Öberg, S. & Wockelberg, H. (2021). ‘Agency Control or Autonomy?: Government Steering of Swedish Government Agencies 2003–2017’. International Public Management Journal, Volume 24, Issue 3 (2021). Open access https://www.tandfonline.com/doi/full/10.1080/10967494.2020.1799889
Wockelberg, H, & Ahlbäck Öberg, S. (2021a). ‘Explaining the Dynamics of Management by Objectives and Results Post-NPM: The case of the Swedish National Executive’. Scandinavian Journal of Public Administration (25:2). Open access http://uu.divaportal.org/smash/get/diva2:1566613/FULLTEXT01.pdf
Wockelberg, H, & Ahlbäck Öberg, S. (2021b). ‘Agency Autonomy and Organizational Interaction’. Public Organization Review. Open access https://doi.org/10.1007/s11115-021-00551-3
The main aims of this project have been to offer a thorough analysis of the autonomy of Swedish central government agencies, and to study how this autonomy relates to their interaction with other actors. The theoretical frame and the strategy to use different types of methodology have been intact throughout the project. The project has developed in terms of added precision to the initial research questions and contributions that are now represented in publications (see below). It should also be mentioned that the Covid 19-pandemic made it hard to engage in collaboration and conferences. The possibilities to complete the project made available by Riksbankens Jubileumsfond have been invaluable.
The project can be summarized in terms of its sub-studies and their respective research questions, data and analyses. Two sub-studies analyse the Swedish government’s use of formal steering tools and the agencies’ autonomy. These analyses are based on an original database containing information about 188 central government agencies for the period of 1997-2017 (Ahlbäck Öberg & Wockelberg 2022). In this part, the project describes how governments apply different types of steering in their relationship with agencies of different kinds and sizes. One of the sub-studies analyses how governments balance autonomy with control when steering the agencies. Central government agencies with an independent management model, and that are allowed high levels of managerial autonomy, are controlled to a larger degree than others in terms of performance information demands. Thus, governments are found to balance autonomy with control in some dimensions of the government-agency relationship (Ahlbäck Öberg & Wockelberg 2021). A second sub-study has it focus on ex post controls and governments’ demand for performance information. Here, the Swedish application of Management By Objectives and Results (MBOR) is studied over time, including a reform specifically targeting the number of demands for performance reporting and aimed at de-escalating the model in use. This study concludes that central government agencies’ tasks are of importance when governments decide what type of performance information to demand from agencies (Wockelberg & Ahlbäck Öberg 2021a). A third sub-study adds the perspective of informality to the picture of formal steering, control and autonomy. In this study, informal contacts across the organizationally divided national executive are studied from within constitutional and practical perspectives, respectively. The methodology used is content analysis of policy documents and interviews with public servants in the Government Office. A fourth sub-study investigates central government agencies’ interaction with others, with the aim to analyse possible links between agency autonomy and organizational interaction. A comparative design that includes Norway and Sweden is used to study how top-level managers within central government agencies describe organizational interaction, what measures that are applied and their quality. Based on survey-data, this study finds an equally positive perception of interaction in both countries. It also finds that managers at the highest levels have more positive views of interaction than those serving on lower levels within the agencies. In addition, the differences between the cases that is observed is discussed in terms of the constitutional and administrative traditions in respective country. One interesting observation is that the core features of the Swedish model of administration, that is, its strong formal and organizational safe guards for agency autonomy, can be understood as having a positive effect on organizational interaction (Wockelberg & Ahlbäck Öberg 2021b).
One of the main results of this project is 1) that agency autonomy and governmental steering of agencies have been studied as multi-dimensional phenomena and as both formal and informal relationships. Here, our unique database has enabled other types of analyses than those commonly applied in this field. In this part, the main contributions are systematic analyses of, and new knowledge about, how governments steer central government agencies, how agencies are allowed certain types of autonomy rather than other, and why. A related contribution comes with the database being published with open access (Ahlbäck Öberg & Wockelberg 2022). The project also 2) delivers an empirically well-founded, up to date, description of the Swedish model of administration. This model, with its dual executive organization, and its many and in relative terms resourceful, independent central government agencies, is still today distinct from within a comparative perspective. At the same time, this project reveals that the Swedish model of administration is a frame containing some variation in terms of how different types of agencies are controlled and over time. Based on this research, we have gained in knowledge about different factors that influence government steering of agencies. We have also created new knowledge about the Swedish model of administration in the era post New Public Management, confirming prior research describing the dynamic developments of public management models over time. This contemporary picture of the Swedish model of administration is interesting also when it comes to what it implies about organizational interaction. In this part, the project suggest that the stronger safeguards of agency autonomy that a context offers, the lower risks it will be for agencies to interact with others. This result is preliminary, and should be investigated further. An additional contribution stems from 3) the analysis of informality in the government – agency relationship. One conclusion is that even though informal contacts across the dual executive are perceived as important and necessary, they are potentially problematic from a constitutional perspective. The challenges that are identified are the ban of informal steering and the fact that ministerial rule is not allowed. In this part, the project contributes with an up to date constitutional analysis and observations of current practices applied to manage informal contacts without over-stepping constitutional or organizational boundaries. Here, it is interesting to note that some informal relationships rest on carefully designed agreements about when and why specific individuals should communicate informally.
This project offers new knowledge about the Swedish model of administration. It has also inspired new research questions. One of these concerns the suggestion that agency autonomy enables organizational interaction. This is a contribution to an on-going discussion and it would be relevant to investigate our preliminary conclusion about this this relationship further. Moreover, this project informs future research on what central government agencies do, that is, their tasks. Not all agencies carry out traditional tasks like case processing or implementation, some instead produce knowledge and provide expertise and are mandated to apply knowledge as a steering tool towards others. These so called expert agencies have become visible during the management of the Covid 19-pandemic and are present in many policy areas. Their exact role in democratic governance is debated, and should be studied empirically.
The results from the project are disseminated mainly in the form of publications and the abovementioned database being made openly accessible on-line (Ahlbäck Öberg & Wockelberg 2022). The project members have also participated in different fora where the Swedish model of administration is analysed and discussed, such as the Swedish Parliament and central government agencies. They have also served in public inquiry commissions (as member and expert).
References
Ahlbäck Öberg, S. & Wockelberg, H. (2022). The Performance Information Demand Dataset: Swedish governments request for performance information from central government agencies 1993 – 2017. [Svenska regeringars krav på återrapportering från statliga myndigheter 1993 – 2017.] Open access at Swedish National Data Service (SNDS) repository https://doi.org/10.5878/hhp6-ah69
Ahlbäck Öberg, S. & Wockelberg, H. (2021). ‘Agency Control or Autonomy?: Government Steering of Swedish Government Agencies 2003–2017’. International Public Management Journal, Volume 24, Issue 3 (2021). Open access https://www.tandfonline.com/doi/full/10.1080/10967494.2020.1799889
Wockelberg, H, & Ahlbäck Öberg, S. (2021a). ‘Explaining the Dynamics of Management by Objectives and Results Post-NPM: The case of the Swedish National Executive’. Scandinavian Journal of Public Administration (25:2). Open access http://uu.divaportal.org/smash/get/diva2:1566613/FULLTEXT01.pdf
Wockelberg, H, & Ahlbäck Öberg, S. (2021b). ‘Agency Autonomy and Organizational Interaction’. Public Organization Review. Open access https://doi.org/10.1007/s11115-021-00551-3