The senior leadership of the public service is a cornerstone element of governance in our country. The professional minefields a deputy minister must navigate are many, and few would argue that these are not difficult and highly specialized jobs.

The Public Policy Forum (PPF) takes a keen interest in examining the senior leadership from across all sectors, and our continuing focus on convening these leaders around a single table has given rise to an interesting observation: the leadership of federal departments at the deputy minister level seems to change quite frequently.

The most noticeable change in the existing cadre of deputies is in the amount of time they have been in their current positions. The current deputy ministers have been in their jobs for a shorter tenure than one would expect. They are all eminently qualified and well respected public servants, and many have held multiple assignments at the deputy level. However, in their current positions, deputy ministers today have a surprisingly short average term of service compared to previous averages and compared to senior leadership in other sectors, even sectors where one might expect much higher turnover.

Exploring the issue of tenure gives rise to another question. If the most senior public servants are today serving shorter terms, then what has changed about their jobs to make this so? This change is challenging to pinpoint, and perhaps the result of the confluence of multiple factors relating to the role of the public service, the mobility of talent within government and the changing nature of the role of department heads specifically.

Examination of these issues leads to further questions. The most pertinent is this: If the knowledge required to lead a major department in the federal government is specialized, requiring the management of tasks that are unique to senior public service positions, then is it beneficial or detrimental for good governance in Canada for deputies to be spending less time in their posts? In addition, we believe it’s important to examine the broader implications this may have for governance in Canada.

Calculating the tenure of deputy ministers and heads of key agencies is complicated by the fact that there is a continuous churn in the leadership ranks of the public service. It’s impossible to know the full tenure of deputies in specific portfolios until their departure. It would be impossible to calculate the average tenure of a given set of deputies until they all left their posts. However, we are able to ascertain with relative ease the length of time deputies have been in their current position at a specific point in time — their current tenure.

We have completed work like this in the past. In November 2007 the PPF published Is Deputy “Churn” Myth or Reality? This report tallied the average tenure of deputies and agency heads between 1997 and 2007, compared to historical averages. For this article, we have selected a list of 28 deputy ministers and presidents of regional economic development agencies that roughly approximates the sample of organizations from 2007, with a few substitutions. The methodologies are similar. In the 2007 study the cut-off date was October 31, 2007, meaning that the time deputies served in their positions after that date was not included in the calculation of the average. In this article, the cut-off date is December 31, 2010, the end of last year. Thus, this article provides an approximate, though inexact, parallel to our previous examinations of tenure.

The current deputy ministers have been in their jobs for a shorter tenure than one would expect. They are all eminently qualified and well respected public servants, and many have held multiple assignments at the deputy level. However, in their current positions, deputy ministers today have a surprisingly short average term of service compared to previous averages and compared to senior leadership in other sectors.

The amount of time deputies spend in their positions throughout history has fluctuated, though it has been generally declining since Confederation. Average tenure ranges from 12.2 years (1867-1917) to a low of 2.3 years (1977-87). The tenure of deputies rebounded to an average of 4 years in the late 1980s and mid-1990s, but the 1997-2007 average had again dropped to 2.7 years.

From our current analysis of the 28 selected departments and agencies, we can see that, as of last December 31, the length of time deputy ministers have spent in their current positions has not increased (see figure 1). In fact, the average length of time for current deputies or heads of regional economic development agencies is only 19.4 months.

While this average will increase as deputies continue to hold their current positions, it is nonetheless surprising. The conventional wisdom in Ottawa is that it takes deputies two years to fully learn their department. If this is the case, then the average of 19.4 months in the job is concerning, as is the fact that only 7 deputies of the sample set of 28 had been in their jobs for two years at the time of calculation. Currently the longest-serving deputy is the Deputy Minister of Industry, Richard Dicerni, at 4.7 years. A total of nine deputies have been in their current position for less than a year.

It is clear that deputy ministers in Ottawa are spending less time leading a specific department than in previous times. While this follows a historical trend, we are forced to ask: What is the optimum length of time a deputy should serve in a portfolio? One might assume at a certain point the average will bottom out, or reverse as was seen in the 1980s. However, the tenure of the current cadre of deputies does not seem destined to do so.

While deputies have spent less time in their current positions than in the past, we must also ask, given the very demanding work they undertake, if this is really surprising. Deputies, after all, lead enormous departments, with billions of dollars in budgets and high pressure to meet shifting political objectives and public demands. They are strictly scrutinized by Parliament and the auditor general, and required to meet high standards of transparency and accountability.

To know if the tenure of deputies is uncharacteristically short, we believe, it is instructive to compare them to the senior leaders of other sectors (figure 2). The only jobs that could be conceivably compared to those of deputies are leadership positions that meet these unique criteria: high pressure to meet objectives, public scrutiny, large but constrained budgets and stringent accountability. Fortunately, we have just such comparators available.

Though there are clearly many differences between corporate executives and public sector leaders, the two jobs do share several prominent similarities. Like deputies, corporate leaders are similarly encumbered with the need to meet the performance expectations of their shareholders and boards. They manage large organizations with many employees and substantial revenue, and high pressure to produce results. These leaders are the most visible elements of a company, and one would expect that in the highly competitive world of corporate enterprise, they would be subject to similarly short turns at the top. This is, however, not so.

A sample of the top 28 companies on the FP 500 list yields 30 CEOs (two of the top 28 have co-CEOs) and presidents whose current tenure is a matter of record. The average current tenure for these top corporate leaders is 121 months — slightly over 10 years, and more than 100 months longer than the average for a deputy. The longest-serving corporate president has served in that position for more than 30 years!

University presidents, like federal deputies, are assigned to lead large organizations with substantial but constantly constrained budgets. Universities receive government funding, and are therefore subject to onerous accountability and transparency standards. And university presidents have many challenging objectives they are tasked to accomplish (e.g., enrolment targets, research funding, donations).

Drawing on the top 30 universities in Canada, as ranked by research funding, it is clear that, while having a shorter average tenure than CEOs, university presidents still greatly outlast deputies.

The average for a university president is 48.4 months — nearly two and a half years longer than for deputies. It should be noted that the executive heads of universities are usually appointed for five-year terms. The longest-serving university president has held office for more than 15 years.

No examination of governance in Canada could truly be complete without including a hockey analogy or metaphor. Fortunately, given the size of the National Hockey League (30 franchises), these organizations happen to fit well into this study as a comparator. And NHL coaches also serve as good comparators based upon the level of public scrutiny to which they are subjected and the pressure they face to produce strong results. Indeed, when thinking of the job of NHL coaches one almost instinctively considers the position to be short-term and tenuous. Reports of coaches losing their jobs after a poor season are expected, even demanded by fans and the media.

However, even under the difficult leadership conditions they face, NHL coaches still have a longer average current tenure than deputy ministers. The average for an NHL head coach is 32.5 months — more than a year longer than for deputies. With more than 13 years in the job, the longest-serving NHL head coach has a full 9 years of service on his counterpart in the deputies’ ranks.

Granted, there are tremendous organizational differences between federal government departments, private sector corporations, universities and NHL franchises. However, looking broadly at the issues that can conceivably be linked to the time leaders spend in the top positions in their respective organizations, some comparisons can be drawn. Based on these comparisons, deputies seem to wither rather quickly. Deputy ministers, plainly stated, spend much less time in their positions than other executive leaders.

To assess the deputies, comparator positions were selected based upon similar job characteristics. However, there may be elements of the comparator jobs that lead to longer terms of service, such as higher compensation. Some of the characteristics on which the comparators were selected (pressure, scrutiny) can be viewed as negatives — circumstances that might push an individual out of the job more quickly. It may simply be that these conditions are more intense in the public service, explaining the shorter terms of service. It has been shown, for example, that due to the expansion of the number of House of Commons committees, deputy ministers today are more frequently subject to bruising appearances before parliamentarians — an experience one may not be keen on repeating year after year.

We must also note that the comparisons to other professions are not entirely apt. Deputy ministers, for instance, all share a common employer — the Government of Canada. This is not true of any of the other sectors surveyed. Indeed, the other comparator positions may even be in direct competition with each other rather than working together for the same broader organizational goals. Under such circumstances, the movement of deputies may simply be a means of maximizing the efficiency within a common organization, and not an indicator of dysfunction. Under such a premise, perhaps the only appropriate comparison is between the Clerk of the Privy Council, as the head of the public service, and a corporate CEO.

The comparison to CEOs in particular brings other issues to the fore. The companies included in this review do not all share the same organization and management structure. For example, some are led by founding entrepreneurs, or the families who have controlled the corporations for generations. Such factors may tend to extend the tenure of corporate leaders, while no similar circumstances can exist in the public service.

Despite all the intrinsic differences between the comparator positions, however, we cannot fail to notice the fact that the current tenure of deputy ministers is brief and, based upon historical averages, becoming even briefer. Other factors must be at play specifically in the public service to explain why deputies are prone to such a short shelf life.

Even under the difficult leadership conditions they face, NHL coaches still have a longer average current tenure than deputy ministers. The average for an NHL head coach is 32.5 months — more than a year longer than for deputies. With more than 13 years in the job, the longest-serving NHL head coach has a full 9 years of service on his counterpart in the deputies’ ranks.

There is some debate about whether the role of deputy minister is that of executive leader, one whose skills and expertise are transferable among departments, or that of subject matter expert, one who must possess and maintain the full spectrum of knowledge related to a particular government department and its policy sphere. In reality, the job requires a combination of skills. Managing a federal department gives a deputy minister knowledge of government human resources practices, accountability standards, financial management practices and a host of other skills that are transferable. However, a deputy is still expected to possess a high level of subject matter expertise in a particular portfolio and must develop strong working relationships within their department and across all sectors — hence the proverbial minimum two-year period for deputies simply to learn their new post.

Given that the tenure of deputies is decreasing, and the average current tenure of deputies is so short, it seems reasonable to assume that the notion of transferable management skills is today a dominant consideration in the leadership ranks of the public service.

In 2007 the PPF also examined the mobility of individuals in the public service below the deputy rank. The report found that appointments to the associate deputy level are even more frequent, and the tenure even shorter, than at the deputy level. Mobility of personnel in the ranks below this is also very high, with the EX (executive) group as a whole showing rates of staffing activity, implying job movements, considerably higher than the public service average.

This mobility within the public service is itself an interesting point of discussion. Given the available data, it seems that deputies may be only the most readily identifiable individuals in a very highly mobile workforce.

One must also bear in mind that, unlike a CEO, university president or NHL coach, a deputy minister does not take on a new post entirely through their own design or ambition. The central political leadership of the government determines who holds deputy posts. In this sense, the deputy position is truly unique and perhaps more prone to change.

The change in the length of tenure in a deputy position may also be linked to the role that a deputy minister plays as a public service leader. The average time in the post may be shrinking because the role of the departments they lead may be changing. Canada has long held the professional, nonpartisan, permanent public service to be a defining feature of its government and policy development. While this article may have shown that the “permanent” element of this age-old definition is certainly open for debate, the role the public service plays in policy development has also markedly changed.

While the public service was once the central source of government policy making, this function has atrophied over the last generation. Powerful ministers’ offices with political objectives and especially the central political authority of the Prime Minister’s Office are now the leading voices in directing policy development, often with the input of the government caucus and voices external to government also carrying substantial weight. For these reasons, deputies may be highly mobile in part because they are more valuable for their ability to implement policy than for extensive subject matter expertise in developing policy in a particular portfolio. If the public service is shifting to become primarily the agent of implementation of politically determined policy, then the ideal of long-serving deputies with substantial policy knowledge in specific areas may have outlived its usefulness.

There are several conclusions that can be drawn. Primary among them is that current federal deputy ministers and heads of key agencies have shorter times in office than in previous eras, and that, benchmarked against other sectors, the time in office of federal deputy ministers is notably short. However, perhaps the more interesting outcome of this analysis is what we cannot so easily conclude — namely, why this is so.

It may not be possible to pinpoint a single specific reason why deputies are now in office for less time. It could be the changing role of the public service, the changing role of deputy ministers or simply the increasing mobility of public servants in general. Likely it is a combination of a number of issues related to the ongoing evolution of the federal public service.

However, such factors don’t by themselves provide a full explanation of why deputies spend less time in office than leaders in other sectors. The aforementioned common employer theory, for instance, makes little sense when one considers that deputies do require advanced and specialized knowledge and skills in order to lead any particular department. The comparison of a CEO solely to the Clerk of the Privy Council also fails to explain tenure issues. The Clerk does serve longer than the average deputy minister — the three most recent Clerks have had an average tenure of 42 months, and the current Clerk is now approaching the end of his second year in office. However, this tenure still reflects higher turnover than for those in a corporate CEO position. There are clearly factors at work beyond easily recognizable organizational differences.

To address the issue of deputy minster churn the PPF has previously proposed that corrective measures be considered, such as setting terms for deputy ministers. This would be akin to the practice for university presidents, who are normally appointed for a term of five years, with the possibility of renewal. To date, little has been done to advance this recommendation.

The actual impact of shorter tenure of deputies in particular departments is difficult to assess. The impact of departments being increasingly led by management generalists is also difficult to determine. However, we can be sure that the specialized knowledge of long-serving departmental leaders is difficult to reclaim once lost. Armed with the knowledge that the senior leadership of the public service changes so frequently, we must consider the impact this is having upon our system of government as a whole.

The Canadian public service, in accordance with Westminster tradition, has been defined by its status as professional, nonpartisan and permanent. The last element, it seems, should be questioned. If the federal public service can no longer boast of permanence as a defining feature, then perhaps it is appropriate to question whether adherence to the other traditions are also necessary.

Since we have tacitly accepted the rapid turnover of public service leadership, perhaps other methods of appointing senior leaders in the public service should be considered. The American system of “to the victor go the spoils,” for instance, is based upon the top ranks of the public service being filled by government appointees, with significant change and renewal with each new administration. If we are willing to concede that the leadership of Canada’s public service can be relatively fluid, and yet still effective, should we also perhaps explore other deviations from our previously hallowed traditions?

The nature of leadership in our public service is clearly changing. Rather than allowing this to occur incrementally or by happenstance, it would be preferable if it was by design. And since we are living in an age of rethinking the role of government, perhaps it would be timely and useful to openly discuss the role of public service leadership as well.

Photo: Shutterstock

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

Creative Commons License