The aims of this thesis were threefold: to explore and increase knowledge of managerial working conditions in Swedish municipalities, to extend the job demands-resources (JD-R) model, and to provide information that public sector decision-makers and human resources departments can use and apply in promoting sustainable managerial working conditions. The argument in this thesis is that a contextual perspective on managerial work is needed, entailing a shift in focus from managers and their individual differences, to the conditions that public sector managers work under. In order to address these questions, a strategic sample of organizations and managers that allowed for comparison and examined systematic differences and similarities among them was used, accompanied by contextual quantitative and qualitative methods that included both subjective appraisals and more objective assessments and data on several levels. The main theoretical framework was based on the latest development in the work stress field, where theoretical and empirical insights from several decades have been developed into the JD-R model. The model takes a balanced approach in explaining negative as well as positive aspects of occupational well-being. Unlike previous models, the JD-R model can be adapted and tailored to the specific context and study group. This thesis comprises four empirical studies with specific aims. In the first study, the main aim was to explore different types of naturally occurring psychosocial work situations for municipal managers, based on the combination of job demands and job resources, by means of cluster analysis. Certain focus was also given to investigating how the work situations could be differentiated regarding sustainability indicators for managerial health, motivation, and performance. In the second study, these psychosocial work situations were followed up 2 years later, and the main aim was to investigate how the situations longitudinally predict managerial turnover using logistic regression and thereby identify groups that are at a high risk or a low risk for turnover, actual as well as intended. The main aim of the third study was to establish how the span of control – that is, the number of subordinates per manager as one of several possible organizational determinants of psychosocial working conditions – affects operational public sector managers’ job demands, using multilevel regression analysis. In the fourth and final study, the main aim was to deepen the understanding of first-line human service managers’ work assignment and psychosocial working conditions by qualitatively and externally assessing the job demands and job resources as well as the balance between them, through work content analysis, in order to provide explanations of the current work strain of this group of managers. The first overall conclusion of this thesis is that public sector managers work in a wide array of diverse situations, including balanced as well as unbalanced psychosocial working conditions in terms of job demands and job resources. An unbalanced work situation, characterized by a lack of correspondence between the job demands posed and the job resources provided, was found to be a reality for a large number of the managers, especially within human services. The psychosocial working conditions were found to be related to consequences for managerial sustainability in terms of health, performance, motivation, and turnover. In addition, the contextual approach applied contributed to identifying organizational and structuring factors – in other words, the type of service the manager works in and the manager’s span of control, managerial position, gender, age, and managerial experience – that provide reasons for the variation in psychosocial working conditions and their consequences. Taken together, the results can provide guidance for actions to be taken in order to promote sustainable psychosocial working conditions for public sector managers, thus reducing both significant individual and organizational costs.