Public international organizations (PIOs), such as the World Health Organization (WHO) and the Organization for Economic Cooperation and Development (OECD), play an increasingly important role in global governance and policymaking, particularly through undertaking policy design and evaluation functions. Central to the ability of such organisations to properly undertake such functions is how well they systematically engage in policy learning, a process in which policy actors track and process knowledge about emerging policy problem, aiming to update their understandings and beliefs regarding viable solutions (Dunlop & Radaelli, 2013; Zaki, 2022). While policy learning within national public organisations is already complex, it is even more challenging and substantively different in PIOs. This can be attributed to three reasons inherent to the nature of PIOs and the context in which they operate. First, PIOs operate in complex environments with a broad range of external stakeholders, leading to exacerbated heterogeneity of actor interests and limited consensus, which complicates learning. Second, PIOs’ internal structures, often constrained by compliance mandates and special statutes, can limit their capacity for reflexivity and adaptation, fostering conditions that amplify barriers to learning, such as joint decision traps, increased affinity for bargaining, political deadlocks, the instrumentalization of knowledge, and power imbalances (Powell,2002; Putnam,1988; Zaki,2022). Third, PIOs face significant legitimacy challenges, particularly in terms of democratic deficit (Montpetit, 2009; Buchana, 2009). This renders them highly dependent on learning to achieve consensus, making the need for robust learning processes even more critical. Within these contexts, effective learning can help mitigate such challenges, whereas inadequate learning risk undermining the very foundation of PIOs and threaten their long-term viability (Tallberg et al., 2014; Kamkhaji, 2022; Zaki, 2024).
These factors have stimulated research on policy learning within PIOs, whereby literature has grown considerably over the years. However, several challenges persist: First, literature on learning in PIOs is highly fragmented and remains in need of conceptual, theoretical, methodological synthesis. Second, very little is known as the PIO understandings of policy learning, specifically their policy design and evaluation functions. Namely, how factors inherent to PIO arrangements influence learning processes and outcomes. Accordingly, this article addresses these challenges by conducting a systematic review of 118 studies on policy learning in PIOs. We are guided by the following key questions:
- What is the current conceptual, theoretical, methodological, and empirical landscape of policy learning in PIOs?
- How do PIOs contribute to policy design and evaluation through policy learning processes?
- What factors shape learning processes and outcomes in PIOs?
- What analytical frameworks have been used to assess policy learning in PIOs, and how can these be adapted to compare learning within and across PIOs, considering variations in their structures, mechanisms, political contexts, and values?
In doing so, this article makes three main contributions. First, it advances theoretical understanding of policy learning in PIOs by providing a comprehensive synthesis of the factors that shape learning processes and outcomes therein. Second, it offers a framework for analyzing learning in PIOs, and third, it develops an evidence-based agenda for future research.
