Many of the Canadian government’s public services to Canadian citizens come in the form of agencies, through which these services are organized and provided. This trend of “agencification” in liberal nation-states, though swift and sweeping, has not been thoroughly assessed for performance. UBC Political Science Associate Professor Dr. Carey Doberstein (PhD, UofT), in his newest publication “Assessing the Promise and Performance of Agencies in the Government of Canada,” takes a deeper dive into the efficacy and functioning of agencies in Canada, as a prime example of a liberal democracy turning to agencies for public goods.
By examining microdata from the Government of Canada’s Public Service Employee Survey (PSES) in 2017, the work asks—and answers—whether agencies are more innovative, autonomous and efficient public organizations. Read Dr. Carey Doberstein answer some of our questions about his newest research about agencies and government.
What should the regular citizen know about the “agencification” of government, and how would it affect their daily lives?
Many of the most routine and important contacts we have with “government” in Canada are in fact with agencies or semi-autonomous organizations of the state. Getting your passport renewed, obtaining health care services, crossing the border, regulating of food products on grocery shelves, getting around the city via transit, and adjudicating immigration and refugee claims, are just a handful of examples of service delivery or regulation conducted via agencies or boards in Canada. They are designed to be arms-length or shielded from direct day-to-day political control, typically theorized as a mechanism to promote enhanced, and frequently expert-informed, public value. Yet regular citizens should care about agencification trends because the other side of the arms-length coin is the observed tendency for governments to evade accountability when things go wrong in these areas.
What prompted you to investigate the experiences of employees in Canadian agencies, as opposed to the outward performance of the agencies themselves? What significance does this hold towards the functioning of these agencies in Canadian government?
This article is part of a larger project in which I am examining trends in the creation and termination of agencies in governments across Canada, and test theory that holds that there is a performance advantage to agencification—whether it be measured in terms of efficiency, innovation, or responsiveness. There is no central database of agencies in Canada, so this work requires considerable time to collect and validate the data, which is ongoing. But a complementary way in which to measure agency performance is from the perceptions of employees in them, and how they compare to public service employees in traditional departments, to see if there is an “agency bonus”. And fortunately the Government of Canada makes available the micro data from a large internal survey of public service employees from which I am able to conduct the analysis in this article.
This article is certainly not definitive as an assessment of trends in agency performance, but various prior studies have shown that perceived organizational performance to be correlated positively with objective measures of organizational performance. The methods of analysis used in this paper—which match survey respondents who are otherwise similar in most respects, but differ in terms of whether they work for an agency or a department—allow us to isolate any potential agency bonus, and I generally find that the effect is the opposite of that which has been theorized.
In your paper, you write that generally, agencies provide less innovation and work autonomy compared to departments but also more efficiency. Based on your findings, would you recommend that the Canadian government implement more/fewer agencies, or change the way they approach agencies vs. departments?
This article sought to identify trends by agency type compared to departments in Canada as a means to set the stage for further analysis of agencification, an area of research that has largely been ignored by Canadian scholars. These findings should not be interpreted as though all agencies are not innovative or allow for greater autonomy in their mission—we know some are!—but that the purported positive agency effects are more variable than many may assume. This is probably a function of the rapid growth of agencies in Canada and elsewhere in recent decades; we create agencies for various purposes these days that extend far beyond what the pioneers of agencification ever intended, thus muting their observed effects (if they ever existed). In short, the apparent rush to create agencies in this country assuming a performance advantage needs to be more critically examined.
As mentioned in your research, what is it about enforcement agencies specifically makes their dependent variable responses different from that of other Canadian agencies? What implications does this have?
The findings that agencies that have enforcement mandates—Canada Border Services Agency, Correctional Service of Canada, the RCMP, etc.—consistently underperform on these metrics was clear. This is backed up by Auditor General reports and other parliamentary investigations that these agencies suffer from some serious internal and external challenges. To be fair, they have difficult mandates, but much more research needs to be conducted to identify how the culture of these authorities affects their ability to deliver services and other government functions to Canadians.
How would you like agencies themselves to take your work and findings into account?
Agencies themselves are not, in my view, the primary audience for these findings. Rather, it is the political masters who create, expand, or merge them, and hold ultimate accountability for their mandate. We know from the limited other research into agencies in Canada that they are increasingly squeezed by their governments in terms of the autonomy they are afforded in law or policy, but which in practice is subject to enormous meddling by Ministers when it may suit the government of day. Governments must never be truly hands-off with agencies, but they do need a more principled theory for what, where and when government functions ought to be arms-length versus need direct political control.
