Table of Content I. INTRODUCTION II. UNDERSTANDING INTEGRITY AGENCY INDEPENDENCE A. Defining "Integrity Agencies" B. Two Dimensions of Agency Independence 1. Formal Independence 2. De facto Independence B. The Role of Courts III. THE POLITICAL ECONOMY OF DELEGATION A. The Classic Principal-Agent Model & Critique 1. Credible Commitments 2. Avoiding Blame 3. Reducing Information Costs B. Costs of Delegation IV. ACCOUNTABILITY NETWORKS AS ADMINISTRATIVE JUSTICE ARCHITECTURE A. Network Components 1. Informal Coordination 2. Shared Services Agreements & Memoranda of Understanding 3. Organizational Initiatives 4. Colocation 5. Mutual Monitoring 6. "External" Networking B. Network Architecture and Agency Independence 1. Reputation 2. Accessibility 3. Information Cost-reduction 4. Burden Reduction & Blame Diffusion V. CONCLUSIONS I. INTRODUCTION
The federal integrity agencies that are delegated collective responsibility for public sector oversight in Canada face a common challenge to stabilize their ongoing independence from political control. (1) Strong independence serves key functional goals of accountability in public administration, guarantees fairness and access for individuals and reinforces the rule of law. But these agencies share with their brethren in regulatory policymaking and administrative adjudication the ambiguous middle position of "structural heretics" in Canada's institutional landscape, being neither government, with its internal logics of patterned hierarchy, nor courts, with their constitutional protections of judicial independence. (2) What limited warranties of independence administrative agencies enjoy--or rather, what warranties their users enjoy--are underwritten by statute or common law, and consequently vulnerable to shifting political preferences and an increasingly partisan national politics. (3) Integrity agencies thus share in the "puzzle" of agency independence in Canada, a challenge thrown into stark relief by politicians' continued attempts to influence agency decision-makers and to punish those who pursue divergent policy paths: formally, by removing them from office, or informally, by exerting public pressure to conform. (4)
What options are available to address this problem? Administrative law scholars have mainly concluded that long-term stability in arm's length relationships between administrative agencies and government ultimately depends on the will of political leadership. (5) Given the unwillingness of courts to extend to administrative agencies the constitutional protections afforded to the judiciary (6) and the dynamics of party government in Westminster-style parliamentary systems, these scholars have noted that, without strong political commitment, lasting agency independence is likely to remain an elusive goal. But they have generally declined to take the next important step of asking about: (1) what objectives actually structure politicians' preferences for agency independence in practice; and (2) what structural reforms or innovations might be available to better stabilize those political preferences over time.
In this article I aim to accomplish three goals. First, I identify the relevant dimensions of independence and control that govern relationships between federal integrity agencies and their political "principals," and I describe some recent events that underscore the challenges of maintaining long-term agency independence in a volatile political climate. Second, I introduce the PrincipalAgent framework as a model of political economy into the Canadian context and reflect on the power of this model to describe some of the institutional conditions that make the delegation of oversight powers in Canada systematically uncertain. Finally, I evaluate the possibilities for "accountability networks", populated by interconnected agencies, to function as alternative institutional arrangements that reinforce agency independence. In short, a network model relies on horizontal linkages between agencies to counterbalance the shifting political preferences that can make independence in existing Principal-Agent relationships so unstable. In order to formulate an early account of how these accountability networks might take shape in Canada, I build from the several possible network components that are emerging in practice to facilitate collaboration and mutual oversight between agencies. While I remain cautious about the complexities and pitfalls of a formal approach that binds independent agencies too closely together, I argue that network models provide a useful starting point for thinking about new architectures of agency independence beyond conventional emphases on the malleable rules and norms of agency appointments, membership tenure, budgets and administration.
Understanding the role of agency networks in this context also connects to broader questions about how to approach badly needed systemic reforms of administrative justice systems as a whole. This article builds from an incremental and bottom-up perspective that takes seriously the role of agencies as autonomous actors--not only as policy implementers, enforcers and dispute resolvers, but also as professional entities that develop new competencies and organizational cultures over time. It views agencies as valuable participants in mapping out new ways to make administrative justice more effective, efficient and accessible. (7) The need for reform results from Canada's chequered history of establishing administrative agencies in response to the ad hoc demands of the moment, leading to "kaleidoscopic" systems of administrative justice in which claimants increasingly find that institutional resources, expertise, their own knowledge of the system and their legal rights are all fragmented between disconnected entities with diverse norms and mandates. (8) This reality has been equally true for adjudicative tribunals, for conventional regulatory bodies and for the federal integrity agencies. (9) When developing new architectures for administrative justice to meet these challenges, this study suggests that reformers should consider networks as a plausible--if understudied--conceptual framework going forward.
In Part II, I define "integrity agencies" and describe the main functions and features of the nine agencies currently active at the federal level. I then outline both formal and informal dimensions of integrity agency independence, drawing on recent case studies that have attracted broad public attention to illustrate these dimensions. In Part III, I turn to a more detailed description of the political economy of agency independence, with the goal of characterizing politicians' choices to delegate public sector oversight authority based on their predicted benefits and costs. Transaction cost models developed in the American context provide a good counterpoint to understand the logics of delegation in Canada, as these models rest on key assumptions about a system of political checks and balances that translate poorly in the Canadian context. This theoretical inquiry underscores the need to refocus research on alternative mechanisms that might anchor agency independence, which leads, in Part IV, to an overview of the strategies that integrity agencies use to form network linkages with each other. I describe how these linkages might influence politicians' incentives to maintain agency independence over the long term and conclude with some reflections on network models that reformers might consider in future work.
UNDERSTANDING AGENCY INDEPENDENCE
Most integrity agencies have traditionally been peripheral actors in Canada's administrative state. But with growing interest in accountability discourse and recent efforts by Parliament to clothe federal agencies in stronger oversight powers, several of these bodies have risen to public prominence and now stand at the centre of many contemporary controversies about public sector regulation. This section begins with a working definition of integrity agencies at the federal level and then describes the relevant indicators or metrics of agency independence in both formal and informal dimensions.
Defining "Integrity Agencies"
Canada's constitutional doctrine of ministerial responsibility requires that Ministers of government answer to Parliament for their own actions and for those of their subordinates, in line with broad public values of transparency, accountability and governance. (10) Historically, the oversight roles played by elected parliamentarians included information-gathering, reporting and sanctioning functions that sought to shed light on the activities of public administration and to hold executive actors, including their administrative delegates, publically accountable for their work. But with the growing complexity of modern governance, these oversight functions have themselves been delegated to administrative bodies. Federal integrity agencies are entities that have been created by Parliament and delegated the power to perform certain aspects of that institution's oversight and regulatory functions at an arm's length from the political and bureaucratic actors traditionally subject to parliamentary scrutiny.
It is not always easy to identify exactly which federal entities should be grouped together under the "integrity agency" umbrella--a term that I use in preference to others such as "Officer of Parliament" or "Agent of Parliament." The term "Officer of Parliament" includes a class of agencies that is broader than those I am concerned with here and encompasses a disparate collection of bodies and offices including the Speakers of the two Houses of Parliament, parliamentary clerks and an array of administrators such as the Parliamentary Librarian. (11) In some recent literature, the label "Agent of Parliament" has been widely used, but I avoid this term mainly for reasons of conceptual clarity when discussing the Principal-Agent model...