The public sector faces an issue that threatens its historic role and future utility. If the issue is not addressed effectively, we will witness an ever more rapid deterioration in relations between senior members of the civil service and their political masters in Cabinet.
The issue is no less than the appropriate policy relationship between the political and bureaucratic tiers of government. Canadian Cabinet government, similar to all Westminster parliamentary systems in the world, involves a delicate balance between these two tiers.
It’s a bit like riding a bicycle. The political tier is the rider who provides the motive force but it is the mechanism of the bicycle that allows it to glide forward with minimal exertion. The bureaucracy is the bicycle, the highly specialized “machinery of government” whose purpose is to allow the rider to reach his or her destination.
The operating assumption of Westminster government is that the bureaucracy provides policy advice in response to Cabinet direction. This advice generally comes in the form of policy options including implementation, administrative and financial details to achieve an objective desired by Cabinet. Weighing the political and policy tradeoffs, the Cabinet then selects an option, although it can always send the matter back to the bureaucracy to generate new ones. Though sensitive to the “large P” political priorities and directions of Cabinet, the civil servants give policy advice unencumbered by “small P” partisan political considerations.
The assumptions and operating principles of Westminster parliamentary systems are based on hundreds of years of history and accumulated convention that differ in important ways from non-Westminster forms of government. Unlike congressional governments such as the United States, there is an overlap between the so-called executive and legislative branches in Westminster systems. The members of the executive council (Cabinet) are drawn from the members elected to the legislative branch. This means that Cabinet ministers play a dual role, one as an executive branch decision maker and another as legislator. But legislator in this context is more limited than a legislator in congressional systems because it is the executive that sets the overall policy direction as well as the programs, services, laws and regulation that give life to policy.
This means that the provincial legislative assemblies and the national House of Commons are venues for the executive, supported by governing party backbenchers, to propose actions already decided upon. Except for the exceptional agency that is a creature of the legislative branch, civil servants work for the executive branch – the Cabinet. By convention, the bureaucracy is professional and non-partisan, and in principle accountable only to the executive. Also by convention, civil servants are anonymous. Ministers are the public face of government and its policies.
I would argue that the Westminster system of government has served us well. In particular, it has sustained a professional corps of civil servants for generations, the more talented members of which have been in high demand beyond Canada in international, regional and non-governmental organizations.
Unlike many congressional systems where a new group of senior officials – recruited from the private sector, academia and other non-governmental sources – replace the old after every election which incumbents lose, Westminster civil servants are generally treated as permanent. This gives them years of experience to hone their skills and knowledge.
At the most senior levels, there may be the occasional firing or a shuffle, but generally those below the deputy minister level are safe. After all, the political tier of government needs to know that the hired help is sensitive enough to the political objectives of the new government to achieve its aims. In most provinces, deputy ministers are appointed by order-in-council (by Cabinet, in other words) rather than through the public service commission. This means that they serve at the pleasure of Cabinet.
Though the majority are likely career public servants, they understand that, unlike the classified service, they require the full confidence of Cabinet to function effectively. Most should not be fired on a change of government but all should be prepared to resign if they do not have the confidence of Cabinet. The rest of the “classified” civil service – if hired and promoted on a profession, non-partisan basis – should not be fired on a change of government.
Generally, when governments behave contrary to this convention they do as much harm to themselves as to our system of government. In 1964, when the Liberals came to office in Saskatchewan, they pushed out some of that generation’s most accomplished policy practitioners, and the Saskatchewan Mafia (as the diaspora were known) contributed their considerable talents to the federal government, headed up by a different crew of Liberals. Subsequent changes in government in Saskatchewan, especially those in 1982 and most recently in 2007, led to over a hundred civil servants in the classified civil service being fired.
Firings such as this are bad enough, but what comes after is even worse – filling these positions with individuals deemed friendly to the government. Beyond perpetuating a vicious cycle, this behaviour undermines the long-term and professional nature of the civil service. In particular, it damages what should be a constructive policy interface between senior public servants and their political masters. You can hardly expect civil servants to give the best policy advice to Cabinet if they are constantly subject to a partisan political litmus test.
In August of this year we saw a public illustration of this point. The political tier of the federal government decided to do away with the compulsory long-form census despite the advice of the hired help – the most senior officials within Statistics Canada, including the Chief Statistician.
As is typical in these situations, the minister responsible for Statistics Canada, Tony Clement, expected and received detailed implementation options from his department despite the department’s deep concerns about the elimination of the long form census. When presented with the options, the federal Cabinet no doubt chose the one that it felt achieved its political objective at a policy cost it was willing to shoulder.
Putting aside any discussion of the merits and demerits of this major policy shift, the process described is exactly what we should expect of a constructive relationship between the political and bureaucratic tiers of government. So what precipitated the resignation of Chief Statistician, Munir Sheikh? And what prompted Sheikh’s public denial that the voluntary survey could not be a substitute for a mandatory census as had been implied by his minister?
The answer is not what some assumed at the time, that, as Clement stated, Statistics Canada is not independent of the political tier of government. The problem was different.
Under immense public and media pressure for a decision that many in the country argued was ill-advised, Clement tried to defend the decision by saying the government was following the advice of Statistics Canada. Why? Because, Clement was trying to find shelter behind the well-established expertise of Statistics Canada, in effect arguing that if Statistics Canada made the recommendation, it must be a solid decision. This prompted him to identify Sheikh by name and to state that he was “entitled to believe that when a deputy minister – in this case the Chief Statistician – gives me a set of options, he is comfortable with those options.”
Although I was not privy to the discussions between Sheikh and Clement, based on my own experience as a senior public serv
