The federal public sector has been shaped by two easily identifiable democratic forces – the views of the people we elect about the role of the state in society and the economy as well as the federal government’s role within the federation. Federal institutions, direct programs and transfers to other levels of government have waxed and waned in response to these two forces and the public service has constantly adapted.

There is a third force that get far less attention but has driven fierce debates and waves of change initiatives within the public service itself. This third force is the ongoing tension between two perspectives. One sees the federal public service as a coherent entity that requires consistency, mobility and portability. The other argues for a public service that has more autonomy and flexibility for both the managers and their organizations. You can always find proponents of both camps and often it’s seen through the lens of “centralizing” or “decentralizing.” The debate is likely to go on forever.

Since 1970 the federal government has had a central management board – Treasury Board ministers and the Treasury Board Secretariat (TBS). It is the guardian of a wide swath of policies governing financial management, internal controls, risk management, human resources, information management, asset management, contracting, real property, transfer payments, and more. It makes it the vortex where both centralizing and decentralizing viewpoints meet. I have been part of countless committees and task forces over the years where they clashed.

Recently this tension has been revealed in heated discussions of post-pandemic workplaces. Should the “centre” impose consistency on hybrid-work arrangements or leave the discretion to individual deputy heads who could in turn delegate decisions further down in their organizations? The policy that came out tries to have it both ways, creating a common framework but leaving a lot of flexibility within it.

This debate about workplaces will continue in collective bargaining. That’s a centralizing process for drafting common rules and standards to apply across multiple organizations. The approach to collective bargaining in the Canadian public service is a choice to centralize bargaining and put it in the hands of a few specialists on each side, while other countries may let each department bargain by itself.

For many years there have been regular updates from the TBS to guide externally facing services. The 2000 policy was updated in 2014 and again in 2020. Service Canada was created in 2005 to create a single point of access for a range of key programs. Norms have been applied across all federal entities to ensure bilingual programming and more recently to ensure services meet the needs of persons with disabilities.

Government orders public service back to the office

Public service will swell to 409,000 in five years, PBO says

The drive toward coherence built upon the Federal Identity Program has evolved since the 1970s to bring greater order to signage and other visual identifiers. Successive governments have brought ever greater central control on paid advertising by federal entities. By now you are familiar with the Canada wordmark and jingle.

The most recent update of service policies includes a heavy emphasis on “digital.” The ongoing shift to digital platforms regularly triggers a fresh wave of debate along the age-old centralist/decentralist axis. Shared Services Canada was created in 2011 to upgrade information technology infrastructure and keep ahead of the rising threats of cybersecurity breaches. It was overtly centralist in intention.

At the time it was resisted, openly or passive-aggressively, by some managers in the largest organizations. They argued that they needed to retain control of their IT to be able to innovate. Frankly, I was never persuaded how hundreds of organizations could manage the transition to digital separately – including cloud computing, cyberhacking by foreign actors and the shift to hybrid work during the pandemic. How would it ever work in practice? This is one area where a centralist approach makes sense.

Indeed, I have argued elsewhere that the failure to be as rigorous on information management behind the digital agenda is starting to show up elsewhere. The TBS should pay more attention to the disparate state of information and records management across the public service.

The landmark Federal Accountability Act of 2006 subtly strengthened the decentralist camp. By clarifying the “buck stops here” accountability of deputy heads, it bolstered the hand of those who would argue some version of “if I am accountable, I have to have full decision making authority over…”

Another line of argument used by the decentralist camp was the need for flexibility and customization, or the need to innovate. They argued that decentralizing was more conducive to innovation. The centralist camp, of which I was usually a member, argued that the friction costs were adding to costs, slowing down government, impeding internal mobility, leaving smaller organizations behind while the big departments looked out for themselves. In my view, decentralization often served the interests of vendors and consultants, not public servants.

Treasury Board has reached different landing spots between the two camps over the years, as have individual departments and organizations. Over the past decade there have been the creation of common service hubs and the standardization of basic work processes for human resources, financial and accounting practices and linking management information. Standardizing and centralizing pension services to public servants has gone well, but pay services? Not so much.

There are still battles being fought in many departments about who regional staff should report to and how much autonomy their leaders should have. And the tides go in and out.

From the author: Difficult issues in public sector reform

More battles are to come. One is about how much autonomy departments and agencies should have over buildings and real estate. Another is about how much autonomy and decentralization there should be in the areas of contracting and procurement. Yet another is about how much autonomy line managers should have over recruitment and hiring processes. “Let the managers manage” is an old slogan that sounds good but in practice the outcome of highly decentralized staffing has been far from optimal. Middle managers and HR shops continue to take infuriatingly long to perform basic staffing transactions.

Interestingly, major spending reviews can work both ways. The centralist camp uses them to argue for rationalization and efficiency by bringing things together while the decentralist camp uses them to argue for getting rid of administrative burden and oversight. There is a very rough analogy here to the private sector and its ever-shifting fashions about unlocking value by breaking things up versus creating value by bringing things together.

Anyone serving on a hypothetical Royal Commission would bring conscious or unconscious bias and preferences to this debate about centralization vs decentralization. They would have to declare on the future of staffing, procurement, real estate and information management. In the real world of practitioners, the public service is pulled back and forth between impulses to standardize and centralize versus arguments for autonomy by departments, agencies and for line and regional managers within larger organizations. Each camp argues its case fiercely convinced of the rightness of their views, fuelled by the ever-shifting fashions in management literature and private sector practice.

This is the second in a planned series of articles by former clerk of the Privy Council Michael Wernick on problems facing the public service. The first article can be found here.

Do you have something to say about the article you just read? Be part of the Policy Options discussion, and send in your own submission, or a letter to the editor. 
Michael Wernick
Michael Wernick served as 23rd Clerk of the Privy Council from 2016 to 2019 after many years as a federal deputy minister. He is a consultant at MNP Digital, the Jarislowsky Chair in Public Sector Management at the University of Ottawa and the author of Governing Canada.

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License