Principal–Agent Theory

Principal–agent theory (PAT) was developed in the field of economics in the 1970s to understand the prevailing problems that appear every time Person A (the “principal”) asks Person B (the “agent”) to do something on his or her behalf for a given price. The basic assumption of PAT is the existence of a double asymmetry between the principal and the agent, since they have different preferences (e.g., the customer/principal wants healthy teeth, while the dentist/agent wants to receive good pay) as well as different levels of information (e.g., the customer/principal can observe the outcome but not the costs of the action that the dentist/agent takes). The solution to this double asymmetry is an outcome-based contract in which it is stated that the agent receives a bonus in case a certain outcome is reached. PAT explores how these contracts should be designed according to the variations in the informational asymmetries between principals and agents in different settings.

Similar to many other economic tools, PAT has been incorporated into political science with the general aim of creating more parsimonious models. Ironically, PAT has only marginally been used for understanding one of the most important principal–agent relationship in politics—the one between citizens, or constituencies (as principals), and elected politicians (as agents). The reason is that the nature of this relationship differs too much from the basic premises of PAT, in which the principal holds a dominant position and is endowed with all the bargaining power. As a consequence, the principal makes a take-it-or-leave-it offer to the agent, which becomes the main object of study for the standard PAT. On the contrary, in the central political delegation chain, the citizen/principal has a subordinate role vis-à-vis a politician/agent, who can single-handedly write or, at least, substantially modify the terms of the “contract” between them. In simple words, governments are the only agents in the world that tell their principals what to do.

PAT has nevertheless been applied extensively for understanding the second delegation chain in a polity—that of politicians as principals and bureaucrats as agents. The main reason is that bureaucracy is one of the fields in political science that has resisted more theoretical progress, and PAT offers a useful tool to develop highly analytical models with clear testable propositions. In addition, PAT seems particularly well suited to address one of the most long-lasting puzzles in bureaucratic politics: the so-called Weberian asymmetry. As Max Weber indicated, a profound problem of [Page 2128]democratic polities is that “political masters” find themselves in the position of dilettantes who stand opposite of expert bureaucrats. This asymmetry has initiated recurrent scientific and ideological debates—most notable among them being the discussion on “state failures.” For the public choice literature in general, and for Willian Niskanen in particular, bureaucracies are too big and too inefficient.

Since it was imported from economics in the early 1980s, PAT has offered creative answers to these debates surrounding public bureaucracies and, while answering them, has also undergone an important evolution. This entry surveys the main aspects of this transformation of a fundamentally economic PAT into a more political theory.

...