This blog post examines why competition and uncertainty within the political sphere lead to entrusting authority to independent agencies, and how political transaction costs become the key factor determining policy stability and controllability in this process.
Representative democracy operates on a chain of delegation: voters entrust part of their sovereignty to representatives, who then re-delegate some authority to bureaucrats or other independent bodies. However, this latter delegation risks creating unelected power, potentially conflicting with representative democracy. Why does this form of delegation occur? Functionalist theory explains it based on the principal-agent model. Specifically, it views the act of principals—political actors—delegating authority to agents as a rational choice to reduce transaction costs arising from imperfect information and the collective action dilemma. Since transaction costs encompass both information costs and trust costs, functionalist theory divides into two approaches. The logic of information views delegation as a choice to overcome expertise and information gaps, while the logic of trust sees it as an alternative to resolve the principal-agent dilemma—where mutual distrust leads principals to pursue individual short-term gains over collective long-term interests by avoiding agreement implementation.
However, delegation always carries the risk that the agent may act against the principal’s interests. Consequently, the delegation problem boils down to designing institutions that maximize the expected utility from the agent while minimizing the possibility of the agent’s betrayal. Here, the two logics propose different solutions. The logic of information posits that institutions should be designed to delegate more authority to agents as they possess greater expertise and information, and as the preferences of principals and agents align. Conversely, the logic of trust presupposes that delegating more authority to agents with preferences independent of their principals is desirable. Here, delegation is understood as an institutional means to resolve the collective action problem of principals.
However, the following criticisms can be raised against these two logics. The logic of information has the limitation that it is difficult to apply at the design stage of delegation, as the agent’s preferences and betrayal can only be observed ex post. Conversely, the logic of trust assumes that principals can appoint agents by sacrificing their short-term preferences; if so, the collective action problem never existed in the first place, leading to a logical contradiction. Therefore, to address the delegation problem precisely, one must approach it from the perspective of political transaction cost theory rather than functionalist theory.
Political transaction cost theory focuses on the competition and conflict arising during the delegation design process, understanding delegation as a product of political competition based on political uncertainty and distributional conflict. A core feature of democracy is that no political actor can stably monopolize the authority to establish and execute public policy. Due to this political uncertainty, current policies always carry the possibility of being legitimately altered by future political powers. The most effective way to ensure policy continuity under political uncertainty is to separate and insulate the policy from direct interference and power competition among political actors. Delegation is performed for this purpose, and in this process, a new form of transaction cost emerges: ‘political transaction costs’. Political transaction costs refer to all costs incurred in altering the direction or content of policies delegated to an agent, or in monitoring and controlling the agent. The higher these costs, the freer the agent becomes from political interference, and the lower the likelihood of policy reversal.
The institutional design of delegation mediated by political transaction costs requires political actors to make an inevitable trade-off between policy stability and the controllability of the agent. The forces designing the delegation inevitably increase political transaction costs to exclude political interference with the agent and secure policy stability, thereby self-limiting their own ability to control the agent. The greater the policy preference differences among competing political forces vying for political power, and the more frequent the changes in political power or the greater the number of competing forces, the higher the likelihood of policy change. Consequently, political transaction costs must inevitably increase further to ensure policy stability. By conceptualizing delegation as a process that separates policy-making from interference and control by political actors to secure policy stability, the theory of political transaction costs enabled a theoretical analysis of the conditions driving policy decisions and the new problems arising from them.
