Abstract
The purpose of this master’s thesis is to examine the impact of cluster management in the public sector on the area of social and health-care for the elderly. This study is relevant, since management in the area of social and health-care for the elderly is inadequately eluci-dated. This is remarkable seen in the light of the fiscal size of the area in the overall local government sector. However, it is also important because cluster management is seen as one of the largest changes in the public management structure since the beginning of the year 2000. Cluster management has been extensively examined on the day care area, but since this area differ significantly from the area of social and health-care, this knowledge cannot be transferred directly. This means that there is very little documented knowledge on the field of cluster management in the area of social and health care for the elderly, particularly in re-lation to tasks and individual managerial roles.

In this thesis, we are identifying and characterizing the managerial tasks performed by the cluster manager, the manager’s managerial and leadership roles and the associated relation-ship with the levels of management above and below. In addition, we are identifying re-quirements to and expectations of a cluster manager and assessing whether cluster man-agement leads to professional boosts and improved service delivery through better quality, or just inserts an extra level of administration.
To examine the above, we are using Henry Mintzberg’s set of the manager’s ten working roles as a tool to map the managerial roles and working tasks of cluster managers. Further-more, we will introduce the terminology of Leadership Pipeline to identify and assess the ex-act leadership requirements for the specific level of management.

This thesis is based on a single case study in the area of social and health-care policy in the municipality of Randers. Cluster management was introduced in the municipality of Randers in 1991 and the extensive experience gained by Randers since then makes the case study very relevant for this study. Other municipalities will be able to use this knowledge as an in-spiration when introducing cluster management or when seeking to optimize existing man-agement structures. We will be using methodological triangulation consisting of interviews, document analysis and observations to examine the tasks, managerial roles and relations re-lated to cluster management.
To examine cluster management and the manager’s relations, we have conducted interviews with the cluster managers and with the administrative level immediately above – the Chief of Social and Health Care for the Elderly and the level immediately below – the frontline man-agers. We have conducted a total of six interviews.

We conclude that cluster management in the municipality of Randers include almost all ten of Mintzberg’s working roles, with those as Leader, Liaison, Monitor, Entrepreneur, Resource al-locator, being the primary roles performed by the cluster managers. Likewise, the results show that cluster management as applied, fully meets the expectations and requirements de-fined as required for this level of management by the Leadership Pipeline approach.

We cannot clearly conclude whether cluster management leads to professional boosts and improved service delivery through better quality or just inserts an extra level of administra-tion. While the physical distance from the frontline managers to the cluster manager has in-creased as result of the restructured arrangements and with negative consequences, it is clear that the arrangement allows the frontline manager more freedom and time to focus on the professional aspects of their area of expertise and responsibility. This has led to higher levels of quality and professionalism, and resulted in even better utilization of resources. We largely ascribe these improvements to professional management practices at the frontline level.
When researching and writing the thesis, we have identified some issues, which we believe should be reconsidered and we have made five related recommendations for practitioners who works in the area of social and health-care for the elderly. It is our hope that our rec-ommendations can serve as a source of inspiration in connection with the introduction of cluster management or in optimization of existing cluster management based management structures.

Our recommendations:
-We advise you to use differentiated management and education to benefit from the frontline managers resources.
-We recommend introducing professional management practices at the frontline level.
-Visible management is more than physical presence, which is why it is important to discuss the expectations.
-Introduce a forum where the frontline managers can interchange thoughts and ideas.
-To define the roles and responsibilities of every manager.