Author:

Abstract:

This dissertation focuses on interorganisational trust in the Flemish public administration, and is the result of a four-year research project in the framework of the Flemish Policy Research Centre – Governmental Organisation – Decisive Governance (2012-2015). The objectives of our research were threefold: 1) to define and identify interorganisational trust in the empirical reality of public administration, 2) to explain the mechanisms of interorganisational trust in the empirical reality of public administration, and 3) to contribute to the development of strategies to optimise interorganisational trust in the Flemish administration where such is possible and functional. We focused our research on interorganisational trust in the horizontal departments of the Flemish administration. On the basis of these objectives, the following research questions were specified for the project: 1) What is the role of interorganisational trust in a public administration context? (Exploratory) 2) What is the distribution of interorganisational trust in the Flemish administration? (Descriptive). 3) Which mechanisms explain interorganisational trust in the Flemish administration? (Explanatory) We defined interorganisational trust in public administration (administrational trust), as “a subjective evaluation made by boundary spanners in interorganisational interactions in public administration characterised by risk, dependency and uncertainty, comprising the intentional and behavioural willingness to suspend vulnerability, on the basis of positive expectations held about the counterpart organisation”. We developed a four-phase ‘funnel’ mixed-method research design in which qualitative and quantitative exploratory, descriptive and explanatory methods were logically sequenced and integrated in order to answer these three questions. The first phase comprised desk research in order to construct the conceptual definitions and frameworks and to address the research questions in this project. In the second phase, an exploration of all three research questions was conducted in order to answer the exploratory research question, and get more acquainted with the empirical reality of interorganisational trust in the Flemish administration, in particular regarding prevalent trust problems in the Flemish administration, which highlighted interorganisational interactions with horizontal departments. In the third phase, an online survey was conducted, which focused on boundary spanners in interorganisational interactions with horizontal departments within the Flemish administration and allowed us to ‘map’ interorganisational trust and distrust in the horizontal departments and to compare quantitative data about boundary spanners’ Reference Interactions of trust with quantitative data about their Reference Interactions of distrust. In the fourth phase, twenty survey respondents were invited for qualitative interviews, which allowed us to compare qualitative data about the most frequently selected Reference Interaction for trust with qualitative data about the most frequently selected Reference Interaction for distrust. The contribution of the dissertation toward the first objective lies in the development and application of frameworks to study and map interorganisational trust. In particular, our study suggests that research in and beyond the discipline of public administration should be critical toward dominant perspectives of the following three assumptions in trust research. 1) An optimistic bias toward trust. Researchers should at the very least acknowledge that trust can also be dysfunctional while distrust can be functional. 2) A trade-off perspective which assumes a trust–distrust continuum. Researchers should at the very least avoid automatically equating ‘high trust’ to ‘low distrust’, and vice versa. 3) One-dimensional measures/discussions of trust. Researchers should at the very least clearly define which dimension of trust is referred to in empirical studies and theoretical discussions. Regarding the second objective this dissertation makes a contribution to existing theory of (interorganisational) trust. Currently, trust is theorized as a ‘universal’ process comprising an assessment of perceptions about the other party’s trustworthiness (perceived ability, benevolence and integrity). If this assessment is positive, trustors are more ‘willing to suspend vulnerability’. Trust only becomes a ‘social reality’ if this willingness is then be followed by a behavioural manifestation of risk-taking in dealings with the other party. Finally, it is argued that the outcome from such risk-taking behaviour updates the trustors’ perceptions of the counterpart’s trustworthiness, rendering trust a cyclically dynamic process. Our research expands this theory of (interorganisational) trust with insights about the internal dynamics of the trust process, and with insights about how this trust process is embedded in specific characteristics of interorganisational interactions. Regarding the internal dynamics of the universal trust process, the following insights are particularly relevant: 1) Perceived trustworthiness affects the willingness to suspend vulnerability, but the effect is moderated by the ratio between perceived trustee-dependent / trustee-independent vulnerability. 2) Willingness to suspend vulnerability affects risk-taking behaviour, but the effect is moderated by the extent of boundary spanners’ behavioural discretion. 3) The feedback loop from risk-taking behaviour strengthens perceived trustworthiness, but only until a certain level of perceived trustworthiness is achieved, after which its self-reinforcing dynamic becomes a self-sustaining dynamic. Regarding the embeddedness of the universal trust process in interaction-specific characteristics, the following insights are important: 1) Macro-level formal (institutional rules and roles) and informal (institutional routines and norms) interaction aspects directly affect all three dimensions of the interorganisational trust process. 2) Meso-level calculative interaction aspects directly affect the willingness to suspend vulnerability, and to lesser extent, perceived trustworthiness. 3) Meso-level relational interaction aspects directly affect perceived trustworthiness, and to lesser extent, the willingness to suspend vulnerability. The contribution of the dissertation toward the third objective lies in the translation of these insights into a generic management model for administrational trust. This management model for interorganisational trust specifies seven groups of recommendations in three clusters, and was applied to propose specific recommendations to the Flemish administration. First, recommendation group A focuses on breaking cycles of distrust by coercing risk-taking behaviour and removing the norm of reciprocity in interorganisational interactions. Second, recommendation groups B, C and D focus on building dimensions of trust. Group B aims at shaping macro-level aspects of interorganisational interactions to support perceived trustworthiness, willingness to suspend vulnerability and risk-taking behaviour. Group C aims at shaping meso-level calculative interaction aspects to further stimulate perceived trustworthiness and willingness to suspend vulnerability. Group D focuses on meso-level relational interaction aspects to further strengthen perceptions of trustworthiness. Third, recommendation groups E, F and G seek to empower the internal dynamics of the interorganisational trust process after trust has been established in the individual dimensions, thus establishing self-sustaining virtuous cycles of trust. Group E focuses on strengthening the interdimensional relationship between perceived trustworthiness and willingness to suspend vulnerability. Group F focuses on strengthening the interdimensional relationship between willingness to suspend vulnerability and risk-taking behaviour. Group G focuses on experience-based learning in interorganisational interactions to support a feedback loop between risk-taking behaviour and perceived trustworthiness. Finally, the dissertation sets out news paths for future research, which may test the extent to which the revised model is statistically generalisable or explore its dynamics in further depth by submitting it to rigorous empirical testing using a variety of methodological designs.