Principal-Agent Modelling and Learning

Abstract

Principal-agent modelling has become a very influential way of thinking about bureaucratic politics in a wide range of settings. Simple agency models have recently been extended and bureaucratic relationships placed in their wider temporal and socio-political contexts. By developing a conceptualization of principal-agent learning this article offers a nuanced account of the extent to which principals can learn to develop institutions that enhance their political control of bureaucratic agents and predispose those agents toward the principal's preferences, limiting adverse selection and moral hazard problems. The revised model is applied to the context of the European Commission's selection and management of scientific committees in the case of agricultural hormone growth promoters. The findings not only confirm the usefulness of more dynamic accounts of principal-agent relationships that eschew ahistoricism and acontextualism, they also suggest that extended principal-agent models should include constraints on learning by principals. In this case the principal's learning was reactive — relying heavily upon external actors and venues to do their thinking for them.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.