**6. Conclusions**

With regard to the "wicked problems" in the cocoa and chocolate sector, such as poverty, child labor and deforestation, the interviews conducted revealed that GISCO is not a panacea. In the previous section, it became clear that power influences the collaboration of actors in the MSI in many ways. For example, on the one hand, the search for compromise in often multiple loops of negotiation usually produces lowest common denominator agreements among stakeholder groups. On the other hand, the case study shows that in-depth knowledge of the interactions in an MSI contributes significantly to gaining a better understanding regarding its possible impacts. The findings show that GISCO, as an MSI in global economic contexts, is an arena in which power asymmetries between actors are diminished only to a certain extent.

In order to examine power in MSIs, existing concepts of power in the literature on network governance, institutions and GPNs were drawn upon and related to each other in the context of MSIs. The case study conducted at a German MSI shows which forms of power occur at different levels in the network and how their relationships to each other can be described. In summary, it appears that power in MSIs is expressed in various forms and effects and occurs at different levels. With this knowledge, a conceptualization of power in MSIs can be formulated according to the "continuous improvement model" that does justice to this insight. It is modeled on the conceptualizations of power described at the outset. The empirical study confirmed that power operates between persons and organizations within a stakeholder group, as well as between stakeholder groups. It operates in different forms, which have their origin partly inside, partly outside the MSI. Furthermore, these power relations underlie network-based regulations, which apply to all actors. Person-based and organization-based power can be distinguished in their effects between "power over" and "power to".

Yet, this conceptualization of power in MSIs does not reflect that organizations negotiate within their stakeholder groups to determine what interests they *collectively* represent. To this end, group-based organizational power was introduced as a specific level of power within an MSI as a result of the empirical study. It is particularly this form of group-based organizational power that makes it so difficult to achieve far-reaching results in GISCO. At the same time, the specificity of the network with its institutions is conceptually considered. The observed negotiations in GISCO indicate the relationship between the specific forms of power of each stakeholder group in global economic contexts and how these are (not) negotiated. Further research could explore in more depth how organizations perceive their power in global economic contexts and what factors influence how representatives in MSIs strategically use this perceived power. Since this case study only examined one MSI of the "continuous improvement model", expanding the research to include other MSIs, especially standard-setting MSIs, is urgently needed.

**Author Contributions:** Conceptualization, D.S. and I.M.; methodology, D.S. and I.M.; software, D.S.; validation, D.S.; formal analysis, D.S.; investigation, D.S.; resources, I.M.; data curation, D.S.; writing—original draft preparation, D.S. and I.M.; writing—review and editing, D.S. and I.M.; visualization, D.S.; supervision, I.M.; project administration, D.S. and I.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This paper is partly a product of research conducted in the Collaborative Research Center 1342 "Global Dynamics of Social Policy" at the University of Bremen. The center is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—project number 374666841—SFB 1342.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper.

**Data Availability Statement:** The data presented in this study could be made available upon request from the corresponding author. The data are not publicly available due the assured anonymity of interviewees.

**Conflicts of Interest:** The authors declare no conflict of interest.
