Principal-Agent Analysis in the Public Sector
Also known as: Public Principal-Agent Analysis, Agency Theory in Government, Political Control Delegation Analysis
Principal-agent analysis in the public sector applies agency theory to the chains of delegation that run through government — from voters to legislators, legislators to executives, and executives to bureaucracies. Terry Moe's 1984 article The New Economics of Organization brought this institutional-economics lens into the study of public bureaucracy, asking how political principals can control agents who have their own interests and superior information. The method identifies the principal and agent, specifies how their goals diverge, characterises the information asymmetry between them, and examines the control mechanisms principals use to limit agency losses. Its purpose is to explain bureaucratic behaviour and the design of oversight as the predictable result of delegation under conflicting incentives.
Key highlights
- Provides a rigorous, deductive logic that explains bureaucratic behaviour and oversight design as predictable responses to delegation under conflict.
- Distinguishes adverse selection from moral hazard and ex ante from ex post controls, giving precise diagnostic categories.
- Generates testable hypotheses about when control will be tight or loose, linking institutional design to information costs.
- Bridges economics and political science, connecting the study of bureaucracy to contract theory and the economics of organisation.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use principal-agent analysis in the public sector when behaviour you observe in a bureaucracy or a delegation relationship can be explained by conflicting interests and unequal information between those who delegate and those who perform. It suits the study of political control of agencies, regulatory oversight, contracting out, intergovernmental grants and the design of accountability and incentive systems. The approach assumes that principals and agents have divergent goals, that the agent holds an informational advantage, and that the relationship can be framed as delegation with monitoring. It is less appropriate where motivation is intrinsic and mission-driven rather than self-interested, where multiple principals make "the principal's" interest indeterminate, or where cooperative co-production better describes the relationship. It complements transaction-cost and accountability analyses, which examine related but distinct facets of governance.
Strengths & limitations
- Provides a rigorous, deductive logic that explains bureaucratic behaviour and oversight design as predictable responses to delegation under conflict.
- Distinguishes adverse selection from moral hazard and ex ante from ex post controls, giving precise diagnostic categories.
- Generates testable hypotheses about when control will be tight or loose, linking institutional design to information costs.
- Bridges economics and political science, connecting the study of bureaucracy to contract theory and the economics of organisation.
- Its assumption of self-interested agents underplays intrinsic public-service motivation and professional norms that drive much bureaucratic behaviour.
- Public agencies typically serve multiple principals with conflicting goals, so the singular "principal's interest" is often indeterminate.
- Public objectives are frequently ambiguous and outputs hard to measure, weakening the contractual framing the theory presumes.
- The model can be hard to operationalise empirically, since information asymmetry and agency loss are difficult to observe directly.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How does principal-agent analysis in government differ from its use in firms?
The core logic is the same — delegation under goal conflict and information asymmetry — but the public sector adds complications. Government agencies often face multiple principals, such as a legislature and an executive, with conflicting goals, so there is no single residual claimant or clear bottom line. Public objectives are ambiguous and outputs hard to measure, and many bureaucrats are driven by mission and professional norms rather than material incentives. These features make the standard contractual solutions weaker and the analysis more about institutional design than about pay-for-performance.
What is the difference between police-patrol and fire-alarm oversight?
McCubbins and Schwartz distinguished two strategies principals use to monitor agents. Police-patrol oversight is active, continuous and centralised: the principal directly samples and inspects agency behaviour, which is thorough but costly. Fire-alarm oversight is reactive and decentralised: the principal establishes rules and channels so that affected citizens and groups raise the alarm when an agency misbehaves, shifting monitoring costs onto third parties. Legislators often prefer fire-alarm oversight because it economises on their time while still constraining agencies.
Is bureaucratic slack always a problem?
Not necessarily. Some divergence between agent behaviour and principal preferences — slack or drift — is unavoidable because perfect monitoring is impossible and information asymmetry is the very reason authority was delegated. Eliminating all slack would require monitoring so intensive that its cost exceeds the benefit. The analysis therefore treats a degree of agency loss as efficient and focuses on whether control mechanisms are well matched to the information problem rather than on driving slack to zero, which is both impossible and wasteful.
Sources
- 1.Moe, T. M. (1984). The New Economics of Organization. American Journal of Political Science, 28(4), 739–777.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Principal-Agent Analysis in the Public Sector. ScholarGate. https://scholargate.app/public-administration/principal-agent-analysis-public