Skip to main content

Thompson’s Three Models of Public Sector Accountability

Go Search
New Atlas
Atlas, A-Z
Atlas Maps
MPP/MPA Programs
Core Topics
Illustrative Courses
Topic Encyclopedia
Concept Dictionary
Career Tips
Best Practices Project
PPGPortal > Home > Concept Dictionary > T, U, V > Thompson’s Three Models of Public Sector Accountability


PPGPortal > Home > Concept Dictionary > T, U, V > Thompson’s Three Models of Public Sector Accountability
Thompson’s Three Models of Public Sector Accountability

Thompson identifies three models of public sector accountability: hierarchical, professional and participatory.

(Heath, Joseph. Lecture. University of Toronto, Fall 2008.)


A strong theoretical conception of where accountability lies and what constitutes ethical behavior can have a significant impact on behavior, enforcing the good while limiting the ability of individuals to rationalize bad behavior. In the public sector however, the complexity of the modern state and the broad range of interests that interact in decision-making has made it increasingly difficult to propose a clear framework.

In the Westminster tradition, the public service has traditionally relied upon the hierarchical model. It establishes a chain of command up through a department with accountability ultimately resting with the elected Cabinet Minister. Through a principal-agent lens, public servants individually and collectively act as the agent for the Minister, with a moral obligation to faithfully follow orders and implement the government’s decisions. This can create a challenging division of moral labour. On the one hand, the public servant’s use of discretion can be problematic as it is the elected official that is ultimately accountable. Moreover, the size and complexity of today’s departments make this centuries old conception of ministerial accountability increasingly unrealistic. On the other, the public servant can only really decline to perform a duty when ordered when they feel it to be unethical, and their only real recourse is resignation. As Ministers increasingly refuse to accept blame for public service mistakes while public servants remain largely anonymous and accountable only to the Minister, the perception is increasingly that the hierarchical model leaves nobody on the hook.

The professional – or ‘Mandarin’ – model instead seeks to constrain the exercise of state power by unelected public servants through their adherence to a code of conduct. As such, civil servants have obligations not just to their Minister but also to an independent conception of professional ethics, much as a lawyer or doctor does. This model clearly privileges the abilities of technocrats who pass judgement on what constitutes the public interest. It allows them to make more independent decisions at a more of a distance from elected officials and the public (like the judiciary or a central bank). The challenge with this model, however, is the diminishment or removal of democratic controls it requires, as well as the paternalism in entails. It is dogged by the ultimate fact that there is no one clear understanding of the public good.

The participatory model is the third Thomson outlines. Rather than upwards accountability to elected officials or reliance on a professional code of conduct, public servants are downwardly accountable to the public. The public interest is directly articulated by the public, and the state is responsive in meeting public demands directly. This participatory model is reflective both of New Public Management thinking that emphasizes a service ethic, and deliberative theory that focuses on identifying and articulating public choice. The inherent challenges in this model are also profound. How to actually aggregate the public view, or to ensure that public servants are not captured by the strongest interests in the public realm? Moreover, the fundamental moral risk is that the public cannot determine with any degree of rationality what is actually in the public interest.

The fundamental challenge in contemporary public sector ethics is that all three models are currently at play in one form or another, and that they contradict each other in many ways. The question is, is one of them preferable to the others, or is there a way to balance them and have them work in unison?

Approved for glossaryposting by Ben Eisen on January 28, 2011

Important Notices
© University of Toronto 2008
School of Public Policy and Governance