Treasury Board of Canada Secretariat
Symbol of the Government of Canada

ARCHIVED - Expenditure Review of Federal Public Sector - Volume One - The Analytical Report and Recommendation


Warning This page has been archived.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the "Contact Us" page.


13. Specific Substantive Compensation Issues Relating to Salaries

In addition to the broad reforms advocated in the previous two chapters, there are a dozen areas where we explore more specialized compensation issues in this chapter and the next. Change in these areas is less essential than in the areas of transparency and management. However, taken together, improvements in these more specific fields would also contribute substantially to creating a compensation regime that would favour the success of the public service in a changing world.

Occupational group structure

In the preceding chapter, we emphasized the desirability of each federal employer adopting an occupational group structure, and a corresponding bargaining unit structure, well adapted to the nature of the organization's mission. For illustration purposes, here we comment more specifically on this issue as it relates to the core public service. This of course comprises most of the main departments of government, with the Treasury Board as the formal employer, although day-to-day management of the workplace rests with deputy ministers and departmental managers.

Figure 1004 in Chapter 3 illustrates the current occupational group structure for the core public service. This structure was established in 1999[175] as a consolidation into 29 groups of the more than 70 groups that had been set up when collective bargaining was introduced in the federal public service in 1967.

Under legislative amendments approved by Parliament in 1993, the government had the authority to determine how the workforce would be structured for classification and collective bargaining purposes.[176] The context of the actual decision to proclaim a consolidated structure was the reintroduction of collective bargaining in 1997 after a hiatus of six years, as well as the expectation that a single, universal classification standard would be introduced soon. In this situation it was understandable that the government did not want to annoy the public sector unions by making group structure changes that would have cut across existing union boundaries. However, the new structure was a compromise at best, from the perspective of sound management of employees and compensation.

Among the unresolved issues carried over into, or exacerbated by the new structure, were these, affecting the specific groups noted.

The Program and Administrative Services (PA) group is amorphous to the point of rendering it unworkable. In labour relations terms, there cannot realistically be a true community of interest among parole officers employed in the Correctional Service, clerks involved in routine transactions, senior policy advisors in line departments, and managers of local offices employing hundreds of people.

Other notably heterogeneous occupational groups are the Health Sciences (SH) and the Technical Services (TC) groups. Sub-groups have limited affinities among themselves but strong links to other groups. For example, Technical Inspectors within the TC group relate best with various transportation regulatory jobs, while certain technologists in the TC group align more closely to the Applied Science group.

Other groups have a sound core but include smaller, largely unrelated occupations. A prime example is the Research (RE) group, which is dominated by research and defence scientists. There is little attention given in this group to the interests of mathematicians that generally work with members of the Economics and Social Science Services (EC) group, or to those of historical researchers.

The Computer Systems (CS) group, the fastest growing occupational group in recent years, in effect combines two distinct sets of employees: a high-end group that contributes critical knowledge in a world driven by information technology; and a more operational group that in many ways is similar to the administrative staff included in the PA group.

Conversely, changes in the larger economy would likely drive the creation of a new Information Management group that would consolidate jobs currently dispersed across several existing classification and occupational groups.

Some groups exist as separate entities only because they are represented by separate unions. Three groups, Ship Repair (East), Ship Repair (West), and Ship Repair Charge Hands, total no more than 1,400 employees with similar responsibilities serving one employing department, National Defence.

As we noted earlier in discussing separate employers, the Audit, Commerce and Purchasing (AV) group is purely a marriage of convenience, with external labour market comparators for auditors and purchasing officers having no linkage.

Further examples of structural illogic could be adduced. But these are enough to illustrate the point: the present occupational group structure works mainly because both managers and employees are used to it and work around it as necessary to carry out the government's business.

It is unreasonable to accept that such illogic is cast in stone simply because it exists or because some unions would oppose constructing groups with a tangible community of interest if that resulted in a loss of members. Only by taking on the task of redesigning the groups can we expect to bring together employees who can see the common interests they share with those in the same group, and to establish meaningful comparisons with the external labour market.

Reforming the occupational group structure is a necessary consequence as well of the 2002 Treasury Board decision to discontinue work on a single universal classification standard. If all federal public service jobs were to be assessed according to a single standard, it would matter little how they were structured by occupation.[177] Since we are continuing to recognize occupations in pay determination––and to ensure that group classification standards minimize gender bias, and that overall each employer provides equal pay for work of equal value––it is important that the occupational group structure make evident good sense.

Actually determining the best way to redesign our occupational group structure is beyond the competence and resources of this Review. So this task is one that will need to be undertaken as a follow up. In principle, this is not an excessively difficult assignment. However, it could easily bog down in conflicts between and within unions, and of course between the unions and the employer. The public service unions can be expected to be reluctant to deal with this challenge since it calls into question the distribution of union membership––and therefore of dues revenue and relative size––and perhaps even the existence of some smaller unions.

The best approach will combine expert external advice, frank mediated conversations between the employer and the unions, a decision by the employer on the most sensible occupational group structure, and an adjudication by the new Public Service Labour Relations Board regarding the bargaining units that make most sense in light of the new occupational group structure. Other federal employers whose occupational and bargaining structures are not attuned with their business needs could undertake a similar process. As a last resort only, if this approach does not yield a sensible result in a reasonable time of perhaps three years,  the federal government could consider implementing a sound structure for occupational groups and bargaining units by legislation.

Ideally, the employer would undertake the necessary analysis and consultations, and decide on a new structure that would provide a solid framework for all aspects of people management. Under the Public Service Labour Relations Act[178] the Treasury Board and the separate employers have this authority. This power is subject only to the authority of the Public Service Labour Relations Board (PSLRB) to determine the bargaining units for collective bargaining purposes. The expectation would be, however, that the PSLRB would accept a sensible occupational group structure proclaimed by the employer.

This work has proven difficult, primarily because of the inevitable controversies related to possible changes to union membership. It will require clear thinking, courage and determination. However, putting a solid structure in place is urgent, since it is essential to so much of the work relating to sound management of compensation in the federal public sector. Therefore if, for whatever reason, the Treasury Board or a separate employer feels unable to complete this work expeditiously, this matter should be referred to an expert panel that would be mandated to weigh the options and propose a realistic structure.

Specifically, then, we recommend as follows:

Recommendation 10

10.1 The Treasury Board,[179] after consultation with the public service unions and appropriate external experts, should publish a proposed occupational and bargaining structure that appropriately combines employees with common interests and working conditions, and that will facilitate compensation comparisons with the external Canadian labour market. This work should be completed within 12 months.

10.2 Within six months of publishing this proposal, after an opportunity for public debate, the Governor-in-Council should proclaim a new occupational group structure for the core public service.

10.3 If, for whatever reason, the responsible Treasury Board portfolio officials cannot complete the project expeditiously, the President of the Treasury Board should consider establishing an expert panel to develop an independent proposal as the basis for the Governor-in-Council to proclaim a new occupational group structure.

10.4 The Public Service Labour Relations Board should determine the appropriate bargaining unit structure in relation to the new occupational group structure as soon as practicable thereafter.

10.5 As a last resort, if there is no other way to move within three years to an appropriate new structure, the government should consider using legislation to implement such a framework.

10.6 The other federal separate employers should proceed with parallel but distinct processes to adapt their occupational and bargaining group structures to the nature of their businesses, also within the next three years.

Classification reform

Classification standards and structures are important because they define the framework for determining salaries. They evaluate the relative value of a particular job, compared with other jobs. Collective bargaining or executive decisions then establish the pay rate for each level of each group. A sound compensation system thus depends on a sensible classification system.

As noted above, occupational groups were consolidated for collective bargaining purposes in 1999. However, for classification purposes, essentially the same 70 or so groups exist as were established when collective bargaining was introduced in 1967. Several of the related classification standards have been revised or updated from time to time. However, many are clearly dated, and the new occupational groups that combined previous occupational groups do not themselves have a classification standard.

Classification reform has proven elusive in the core public service for most of the past two decades. This resulted in large measure from the attempt to create a single universal classification standard that would permit evaluation of all public service work according to one gender-neutral scale of value. Such a standard would have supported the negotiation of a single pay line for all jobs in the core public service. As noted earlier, however, this goal was set aside in 2002, mainly in recognition of the great diversity of work in the federal public service, and of the need to maintain reasonable links to the external labour market.[180]

Since 2002, the approach has been to modernize classification standards group by group. In doing so, care is to be taken to ensure that the new standards do not discriminate between men and women. While classification standards remain a management prerogative and are not themselves subject to collective bargaining, the eventual pay lines do need to be negotiated. So in practice, revising a classification standard entails substantial dialogue with the relevant union throughout the process.

Intensive discussions on the Foreign Service (FS) standard with the Professional Association of Foreign Service Officers (PAFSO) and the main employing departments resulted in a new four-level structure and classification standard. A new standard has also been approved for the new Border Services (FB) group that was created as a result of the recently established Canada Border Services Agency, mainly from the former Canada Customs and Revenue Agency. New standards are being developed for the Economics and Social Science Services (EC), the Law (LA), the Computer Services (CS), the three Ship Repair (SR), and the Program Administration (PA) groups. Moreover, some long-standing issues, such as how to deal with the classification of fisheries officers, have been resolved within the existing standards.

Overall, this work has been slow and difficult. It is urgent that we carry on with modernizing our classification standards. However, logically, the work on modernizing the occupational group structure described in the previous section should take precedence. So the short-term priorities for classification reform should focus on those groups that are unlikely to change through a redesign of the occupational group structure. Once the new structure has been put in place, it will be important to decide on the best sequence and approach for modernizing the related classification standards.

In any scenario, however, the business of updating occupational group structures and classification standards will take many years to complete. Thus, the more immediate issue is what to do in the meantime. As part of this, we need to challenge the basic premise of what could be called the traditional approach to classification. In this approach, a complex standard awards precise numbers of points for specified degrees of certain factors considered to be important in evaluating a job. These points are added up, and the job is slotted into the classification level corresponding to that point total. There is an implication of almost scientific precision to the process. There is also present an idea that the demands on a job are stable and well defined. In the end, however, classification is really an overall judgement about whether at a given time a given job is bigger or smaller than reference positions whose classification is well established.

In the world of rapidly changing demands described in Chapter 10, the system of classification as we know it is increasingly anachronistic. The work entailed in writing lengthy job descriptions aimed at winning points against an often outdated classification standard, in debating the minutiae of the various elements, and in arriving at a conclusion, seems increasingly misdirected as organizations are called on to form and reform themselves to deal with fluid business priorities.

What we need is a simpler approach to classification and its administration. At the same time, the results must be capable of withstanding scrutiny. Among the tactics and tools that would assist in navigating between simplicity and accountability are these:

  • reducing the length and complexity of job descriptions;
  • extensive use of generic position descriptions;
  • updating of key benchmark positions; and,
  • greater use of rationales that relate the classification of new or revised descriptions to the benchmarks with less emphasis on detailed assessments of individual elements.

There is no useful reason for perpetuating 20-page or longer job descriptions. Such "novellas" usually accumulate verbiage in the hope of convincing classification raters that the position is really important. In fact, however, a few words are usually sufficient to describe genuinely important jobs. Lengthy descriptions are thus prima facie evidence of an inflated presentation. Limiting descriptions to something like five pages should be mandatory.

More and more, the distinctions between similar jobs are unimportant from the perspective of assessing their relative value. Duties often evolve with changing departmental priorities, so detailed cataloguing of the current situation risks rapid obsolescence. Accordingly, generic job descriptions for analogous jobs can contribute greatly to simplifying not only classification, but also staffing. Some organizations have adopted this approach to great effect. Statistics Canada, for example, uses about 200 job descriptions to cover over 5,000 employees.

A more radical approach that is often discussed is "appointment to level." This term refers to a system whereby employees would be classified on the basis of their skills, knowledge and performance. Such a philosophy is already in place for certain scientific research groups, where advancement depends on careful assessments of an employee's scientific merit. It also plays a role in developmental or training programs, for example for economists or management trainees, where advancement to a certain level depends on meeting defined criteria. It would be desirable to experiment more aggressively with the application of this approach to other groups, especially those where skills, knowledge and individual performance can be objectively assessed. No wholesale program of appointment to level should be attempted, however, since the track record of applying general solutions to classification issues in the public service offers ample grounds for caution. More generally, nevertheless, generic job descriptions offer an attractive and manageable middle ground between classification based on the requirements of a position, and classification dependent on an employee's merit.

An effective investment in rendering the existing classification standards more usable pending their modernization would be in updating key benchmarks. These would normally be the most common or typical positions covered by the occupational group. If the benchmark positions are carefully described, and the rationale for their classification level clearly enunciated, it becomes fairly easy to compare them to other positions. It is in this context that classification rationales could reasonably focus on an overall assessment vis-à-vis a benchmark, rather than a detailed element-by-element rationale.

In pursuing what might be called a pragmatic approach to managing classification, while the lengthy work of modernizing classification standards unfolds, it is critical to minimize gender bias. To achieve this in the short term, we must ensure that our means for applying the unreformed standards avoid gender bias as far as possible. Our current standards do not exhibit overt gender bias; such bias tends to be subtle and rooted in conscious or ingrained expectations about the relative importance of work traditionally associated with men and women. During the next several years, while the process of modernizing our classification standards according to the revised occupational group structure unfolds, the Public Service Human Resources Management Agency should provide accessible non-technical training and tools to assist managers and others to minimize such bias. Later in this chapter we offer pragmatic advice on the work we need to undertake now to identify and address cases where there may be salary differences between male- and female-dominated groups that cannot be explained by factors other than discrimination.

In summary, our recommendations on the issue of classification reform are these:

Recommendation 11

11.1 In parallel with the design of a new occupational group structure, the Treasury Board[181] should determine, in consultation with the public service unions, the most appropriate approach and sequence for developing modern classification standards that reflect the current realities of public service work, and that minimize gender bias.

11.2 In the meantime, the Public Service Human Resources Management Agency should focus on supporting departments to simplify the management of classification by directing that job descriptions not exceed five pages, by encouraging the use of generic job descriptions, by updating the most suitable benchmark positions for the most numerous groups, and by providing guidance on how to use overall rationales for classifying positions by comparison with the benchmarks.

11.3 Consideration should be given to expanding the use of appointment to level for employees in groups or situations where their skills, knowledge, and performance can be assessed objectively in relation to well-defined standards.

11.4 As practicable, the Public Service Human Resources Management Agency, in consultation with the relevant public service unions, should give priority to completing the updating of the classification standards for priority occupational groups that are not likely to be changed as a result of the reform of the occupational group structure proposed in Recommendation 10.

11.5 Once the Treasury Board has decided on the occupational group structure for the core public service, the Public Service Human Resources Management Agency should publish its planned timetable for modernizing classification standards to correspond with the recommended occupational group structure, and report to Parliament annually through the Agency's Report on Plans and Priorities.

Aspects of general salary policy and management

There are several broad policy issues relating to pay that will not be resolved through the business of rethinking the occupational group structure and modernizing classification standards and practices. These issues include:

  • Pay for performance
  • Pay for knowledge
  • Special pay for exceptional expertise
  • Recruitment and retention (known as terminable) allowances
  • Structure of pay bands
  • Regional pay

We comment briefly on each in the following sections.

Pay for performance

Productivity improvement is a preoccupation of all modern enterprises, whether private or public. So the idea of linking some part of remuneration to performance is both logical and attractive from a management perspective. The dominant historical practice in Canadian federal public service compensation, however, has been to avoid an explicit link to performance in establishing pay. Yet it is plainly observable that some employees work harder and better than others, and could reasonably be thought to deserve higher pay as a result. The issue of whether and how to link pay to performance in the public service is especially pertinent as the pressure on the federal public service grows to innovate, to be flexible, to cope with increased workloads, to serve Canadians with distinction.

On the other hand, many employees and most unions oppose the adoption of any form of pay for performance. Among the critiques advanced is the idea that performance pay for individuals creates unhealthy tensions in the workplace, as destructive competition may emerge and some employees may be tempted to claim credit for what is in fact a group result. More fundamentally, there is mistrust of the objectivity and fairness of public service managers in assessing performance and awarding any pay for performance fairly.

The American government has decided to use the occasion of establishing the massive Department of Homeland Security (DHS) as the opportunity to reform its compensation philosophy and practices. The new department is mandated to implement a pay-for-performance program that will replace the long-established General Schedule with market-based pay bands, in which "employee pay progression is solely driven by performance and/or competency attainment–not longevity."[182] This initiative is responding to recommendations presented in January 2003, in the wake of the events of September 11, 2001, by the bipartisan National Commission on the Public Service, chaired by Paul Volcker. The new system is to be fully implemented in the Department of Homeland Security by 2009. The expectation appears to be that it will be expanded eventually to most of the United States public service.

In assessing the Final DHS Human Capital Regulations, the General Accounting Office (GAO) stated that "while GAO strongly supports federal human capital reform, how it is done, when it is done, and the basis on which it is done can be the difference between success and failure."[183] The GAO emphasized particularly the need for clear leadership of the initiative, for effective communications, and for building the details with rigour and credibility.

Perhaps these success criteria will be met in the atmosphere of crisis and reform that continues to pervade official Washington in the context of the War on Terror. But this is far from guaranteed. Under the existing General Schedule system, progression through the 10 steps of a given pay level was to be faster or slower based on performance. In practice, however, nearly all employees advance one step per year of service, whatever their performance. So the ideal of American public service managers defining performance levels clearly, and facing up to making and explaining judgements about the performance of individual employees has already failed in execution historically.

In the Canadian public service, our track record in defining and assessing performance is not encouraging. Volume Two of this Report does present the apparently successful example of the Translation (TR) group which receives extra pay based on exceptional productivity. However, as in the United States, the original idea that annual increments within the salary band for a particular classification level should be based on satisfactory performance has essentially never been applied. Even the number of new employees released while on probation is tiny, amounting to fewer than 100 per year since 1996–97.[184] (Thirty-six indeterminate employees were rejected during probation in 2002–03, representing 0.7% of total external full-time indeterminate hires for the year.)

The record in the Executive category is more mixed. For several years, members of the Executive group have been eligible for at-risk pay of up to 10% (15% for the top two levels) of salary, based on their performance. As noted in Volume Two, at the EX 1 level, up to 7% of the possible at-risk pay merely brings an executive employee to the total compensation level equivalent to private sector pay for the same level of responsibility. In practice, however, deputy ministers have found it difficult to distinguish performance levels, especially at the less-than-fully satisfactory level. Such ratings are seen as personally demeaning, rather than a legitimate message about performance issues in a particular year. In the past two or three years, under pressure from the Stephenson Committee[185] and the Public Service Human Resources Management Agency, rating managers have apparently applied performance criteria somewhat more rigorously.

A further complication in regard to executive pay-at-risk is that the amounts to be earned are small relative to regular salary. For Executives, the most that can be earned is 10% (or 15% for those at the assistant deputy minister level). By contrast, performance pay in the private sector can be equivalent to several months' salary or more. This must be combined with the reality that senior executives in the private sector often have access to long-term incentives, normally in the form of a right to purchase company shares at a fixed price at certain times. This aspect of compensation is explicitly excluded in comparing federal public service executive compensation with that offered in the private sector. This is reasonable, since there clearly is no equivalent to share prices in the public sector. In effect, then, the federal public service pay-at-risk program may be more symbolic than substantive at the level of top public service managers. Executives certainly appreciate the program as a supplement to salaries that generally lag private sector counterparts, but the amounts available are not large enough, nor is the current program rigorously enough administered to serve as an important motivator.

Overall, the federal public sector has not developed a strong culture of defining performance expectations and assessing employees accordingly. Strengthening this culture is the place to start. It might be thought that introducing a pay-for-performance plan would drive the adoption of more rigorous practices in the areas of setting goals and performance standards, and evaluating results against these criteria. Experience suggests, however, that we cannot count on this happening.

At the same time, it would be instructive to experiment with performance pay plans in particular for groups or organizations where there is already a reasonable performance management culture, and where the employees––and where applicable, the union––are open to this prospect. Likely the best places to look would be specialized or professional groups where professional standards are well defined by external bodies and well understood within the field.

Looking to the future, then, sound management reasoning suggests two principles to follow:

  1. We should focus first on strengthening our management of performance, so that we have the capacity to tie salaries to performance credibly.
  2. We should undertake selective experiments in making greater use of pay for performance in federal public sector compensation.

Pay for knowledge

In this area as well, logic might favour rewarding knowledge in federal public sector compensation. After all, we continually emphasize the importance of knowledge work as playing a greater and greater role in the public service. This idea is implicit in the selection standards used for staffing various specialized jobs, which may require a particular academic or professional qualification, or the equivalent in experience. However, there is little current use in the federal public sector of explicit incremental remuneration to encourage or reward the acquisition or possession of specific knowledge.

An unhappy exception is the bilingualism bonus. This is an allowance of $800 annually that is paid to non-executive employees who meet the bilingual standards required of their positions. The amount has remained fixed at $800 since 1979. It was introduced originally as a pragmatic response to unrest among francophone employees who felt they should be rewarded for shouldering a large share of the responsibility to serve the public in English or French. While its real value has declined from year to year with inflation, its connection to building bilingual capacity is tenuous. Employees may qualify for the bonus, but do not necessarily use their second official language in practice. In effect, this allowance has become untouchable, even though its rationale is weak.

In fact, language is an area where pay for knowledge could be especially sensible. Some employees come to the public service already bilingual, or learn the other official language essentially on their own, with little or no formal instruction at public expense. Such employees save the public purse thousands of dollars of costs. Paying some kind of one-time or continuing bonus to such employees could encourage individual initiative to learn one of our official languages. More generally, knowledge of third languages can be of tremendous value in serving our increasingly multicultural nation, both here in Canada and at our missions abroad. Where such knowledge is clearly an asset in a particular job, it could make sense to provide an allowance to attract and retain people with appropriate capacity.

Another potentially fruitful application of pay for knowledge would relate to implementing a system of certifying management competencies at various levels. Completion of critical courses could be recognized either in faster advancement through a salary band, or through a one-time or ongoing allowance.[186] A similar approach could be employed for specialized groups to keep up with developments in their field, or to advance to higher qualifications that are important in the workplace.

There are of course difficulties in applying pay for knowledge. For example, management needs to ensure that the additional qualifications are in fact pertinent to the workplace. There need to be reliable means to verify that the knowledge has been acquired. And we would need to avoid the fossilized approach that has come to characterize the existing bilingualism bonus, adapting any system to changes in what knowledge is critical and can usefully be encouraged through the pay system.

Special pay for exceptional expertise

This topic is an extension of the theme of pay for knowledge. Increasingly, there is a need in the public service for high-level expertise in particular fields or disciplines. This tendency manifests itself both at the executive level and among the ranks of senior analysts or professionals with little or no management role.

At the executive level in recent years we have needed top talent in such areas as information and technology management, financial management and other specialties. Because of our policy to match private sector executive total compensation only at the EX 1 level, there is an increasing compensation gap at higher executive ranks. We refer here to total compensation, not just salaries. In some recent cases, it has proved impossible to attract people with the requisite experience at the compensation offered.

One approach is to bring top talent in temporarily through Interchange Canada. Essentially this involves an assignment, normally not exceeding three years, whereby an employee of another organization works for the federal government, but is compensated according to the rules of his or her home organization, with reimbursement to that employer from the federal government. The program requires the participant to return to his or her home organization at the end of the assignment. In this way, exceptional executives serving the government may earn substantially more than regular public servants in positions classified at the same level.[187] In 2005, the federal government launched the Prime Minister's Fellows Program to give new emphasis to the interchange idea. This built upon an earlier informal initiative referred to as the bridging program that facilitated several temporary assignments for senior public servants and private sector executives in each other's worlds.

While interchange programs are undoubtedly useful in accessing exceptional or highly specialized talent on a temporary basis, we can expect there to be cases where the federal government wants to attract such an employee on a more permanent basis. For example, the requirement may extend well beyond the normal three-year limit on an assignment. In other instances, interchange programs may not fit well since the person sought is self-employed and does not really have an employer to return to. Or seconding an expert from an external organization could create a real or perceived conflict of interest.

Accordingly, it would make sense to use, albeit judiciously and sparingly, the existing authority of the Treasury Board to authorize an executive to be paid more than the maximum established for the position to which he or she is appointed.[188] Such a practice would need strict controls in order to withstand the inevitable pressures to make increasingly frequent use of exceptions. Formal approval of particular cases would best lie with the Treasury Board itself, on the joint recommendation of the Deputy Minister of the sponsoring department, and the President of the Public Service Human Resources Management Agency for executive positions, or the Clerk of the Privy Council for deputy minister or other order-in-council positions. In effect, the policy would permit exceptionally the public service to close, or at least partly close the gap between public service compensation and what a prime candidate is earning already.

Experience suggests that the introduction of such flexibility risks abuse. Various devices to get around the intent of the proposed policy change can easily be imagined. For this reason, it is essential that any exceptions be well documented, be recommended by the highest officials personally, and be approved explicitly by the ministers of the Treasury Board. Such a practice would, of course, be subject to audit by the Auditor General.

An analogous issue relates to the juxtaposition between senior specialist and executive compensation in the public service. Some non-executive groups, such as the Economics, Sociology and Statistics (ES) group, overlap at the top (ES 7) level significantly with Executive (EX) group salaries. Figure 1053 shows how salaries for the top levels of several professional occupational groups corresponded with EX 1 and EX 2 salaries in March 2003.

Figure 1053
Illustration of salary band overlaps between the EX Group and Professional Groups, March 2003

Display full size graphic

Illustration of salary band overlaps between the EX Group and Professional Groups, March 2003

From a traditional hierarchical perspective, having non-executives paid as much as or more than executives is offensive. However, in a world where knowledge and expertise is increasingly valuable, it could well make sense to pay specialists as much as or more than executives. In recent years, the number of EX positions has increased more quickly than the size of the public service as a whole. Determining the underlying reasons would require a specific study. However, one contributing factor is likely to be the need to pay more to highly skilled knowledge workers, even where the true executive content of a job may be modest. Building in more capacity to reward specialists without making them executives could prove advantageous. In fact, burdening top specialists with management duties can detract both from sound management and the supply of substantive expertise.

At a minimum, we cannot find an explicit rationale for the existing overlaps, either in regard to which groups have overlaps or to the extent of that overlap. The time has come to review this overlap issue systematically, likely in conjunction with the rethinking we propose later in this section concerning the extent of external comparability for executive compensation. Consideration should also be given to permitting individual cases of special pay for special expertise in cases below the executive level where interchange programs may not suit. Such cases would require the same strict controls as proposed above in relation to executives.

Recruitment and retention allowances

As we described primarily in Chapter 4 of this volume, recruitment and retention––or terminable––allowances were introduced over the past several years for various bargaining groups. These allowances provided additional compensation, in theory on a temporary basis, for groups experiencing unusually tough competition in the external labour market. Using temporary allowances addressed at least two issues from the employer's perspective. First, hot labour markets come and go as the economy waxes and wanes. Second, such allowances are a recognized exception in evaluating equal pay for work of equal value.

Like many temporary policies, however, there has been substantial pressure to make recruitment and retention allowances permanent and to roll them into the regular salary levels. In cases where the pay level for a particular group in the general labour market in fact advances relative to other groups on a sustained basis, absorbing terminable allowances into regular pay makes sense. However, in a case such as the Computer Systems (CS) group, where the overheated market of the late 1990s was succeeded by lay-offs and lower wages in the private economy in the early years of the new century, the federal government should in principle reduce or terminate the terminable allowance. At a minimum the government should constrain economic increases for the group until a reasonable parity is restored.

In practice, however, the federal government collective bargaining system makes such reductions or constraint virtually impossible to implement. The employees of such a bargaining unit, largely secure in their employment, and responsible for such critical functions as processing federal social benefit cheques, have no incentive to accept salary adjustments below the going rate for collective agreements. Unless the Government is willing to impose through legislation a collective agreement with a lower allowance, or to accept a strike in an area of great public sensitivity until the union is ready to settle on that basis, the terminable allowance is likely to persist indefinitely.

Accordingly, great care is needed before any terminable allowance is accepted. Under the present culture and practices of federal public service collective bargaining, such allowances are almost impossible to end once they are launched. A better approach is that of tracking the private sector more closely, based on a rigorous assessment of compensation in the broader Canadian labour market. Where there are rapid increases in hot markets, perhaps it would be better to use one-time bonuses to recruit or retain essential staff until it is clear whether relative salary increases are themselves temporary or structural.

Structure of pay bands

A relatively arcane but still important aspect of pay management is the structure of pay bands. Ideally, the salary structure for an occupational group will offer a progression that provides incentives to move to the next level, without unduly large gaps between levels. Figure 1054 provides an illustration of what we might expect to find. In this picture, there is a reasonable difference between the maximum pay rates for successive classification levels, normally at least 10%. Second, we observe sufficient spread within a pay range to recognize growth and development. Note that the lower levels have fairly narrow bands, whereas the higher levels offer more room to progress within a given classification level. Third, note also that the pay bands overlap slightly, to allow smooth entry on promotion without excessive costs.

Figure 1054
Illustration of a rational salary structure

Display full size graphic

Illustration of a rational salary structure

The pay bands for several federal public service occupational groups are quite different from what could be considered normal. For example, in the Program Administration (PM) classification group, there are large gaps in the pay bands between PM 4 and PM 5, and between PM 5 and PM 6. An opposite type of anomaly is illustrated by the Community Health Nursing subgroup. Top pay for the NUCHN 02 level is only slightly below the entry level for the NUCHN 06 level, suggesting that there are too many distinct levels. The Library Science classification group manages to display both problems: substantial overlap between the LS 3 and LS 4 levels, with gaps between LS 2 and LS 3, and between LS 4 and LS 5. These unusual pay band structures are illustrated in Figure 1055a, 1055b and 1055c.

Figure 1055a
Illustration of the salary structures for selected federal public service occupational groups

Display full size graphic

Illustration of the salary structures for selected federal public service occupational groups

There does not appear to be a policy or management rationale for having such pay band structures. They are the result of numerous rounds of collective bargaining over the years. In practice, most collective agreements simply increase the existing structures by a certain percentage. It is possible that the current structures can be traced in some cases back to the introduction of collective bargaining in 1967, when the occupational groups now still in use were formed. In creating them, pre-existing departmental groups were consolidated. These initial structures may have been adjusted to minimize red-circling, that is, cases where an individual's salary is capped until future negotiated economic increases bring the group salary up to the individual's protected salary level.[189]

Figure 1055b
Illustration of the salary structures for selected federal public service occupational groups

Display full size graphic

Illustration of the salary structures for selected federal public service occupational groups

It might be argued that salary structures are not important, but they are. Gaps between levels affect how both employees and managers operate. For employees, insufficient differences in the salary maximums of two levels can reduce the incentive to accept greater responsibility. Excessive gaps, on the other hand can make managers reluctant to promote employees to the next level, preferring to hire someone whose salary is closer to the new level. This type of management response could push employees to move to another occupational group in order to advance. Such movement is not bad in itself, and may in fact be desirable. But it should result from a reasoned approach to career development, rather than a desire to get around something whose only explanation is an accumulation of past decisions.

Figure 1055c
Illustration of the salary structures for selected federal public service occupational groups

Display full size graphic

Illustration of the salary structures for selected federal public service occupational groups

Another concern about irrational pay band structures is their relation to the definition of promotion. At present, a move between groups is not a promotion unless the maximum salary level for the new group and level is at least 4% greater than for the employee's existing position. Poorly structured salary bands render the application of this policy more arbitrary than desirable.

Reshaping the federal public service pay bands towards greater resemblance to the norm illustrated in Figure 1054 will be difficult. Unions will be comfortable with such changes as long as all adjustments are upwards. Such an approach would likely prove costly. This reality no doubt explains why there has been little action to remedy the existing anomalies. In the medium term, reluctance to address this issue is likely wise. Over several decades, however, failure to introduce a sensible pay band structure is worrisome. It implies that the compensation system is so rigid that desirable improvements cannot be implemented. This issue should be tackled as part of the modernization of classification standards that we propose in an earlier section of this chapter.

Regional pay

Controversy has long marked the issue of whether regional differences in external salary levels should be taken into account in setting pay levels in the federal public service. Some commentators urge that the principle of comparability with the external labour market requires the federal government to vary its compensation according to the location of a job. Failing to do so, these observers argue, distorts the local labour market by forcing private sector or other non-federal employers to offer higher than needed wages, or to accept less qualified staff. If higher-than-necessary wages are offered, economic theory suggests that as a consequence fewer jobs will be created or sustained.

The counterarguments are also forceful. First, opponents of regional pay argue that there should be the same pay for equivalent work for the same national employer, wherever the job is located. The implication is that paying less in some locations is demeaning to those who work there, and unfair. For unions, such differences are also undesirable because they foster dissension among members. Second, as a national employer the federal government moves its employees from one location to another, whether temporarily or permanently, fairly frequently. Regional pay policies could discourage employees from accepting operationally desirable moves.

In principle, the soundest policy is likely that which guided federal compensation in the decades before the introduction of collective bargaining in 1967. There were in fact two broad categories of employees: those who were seen to be recruited nationally, and those who were recruited locally. The first group were paid the same rates across the country, and the second the prevailing rates for their region. For example, if in recruiting auditors we want to hire the best available specialists from across Canada, we need to offer national pay rates. In hiring an administrative assistant, however, we would normally look to a more local pool of candidates, so local rates would likely suffice to attract qualified candidates.

The real problem with regional pay is not theory, but rather the difficulty of applying it fairly and credibly in practice. While the distinctions espoused in the pre-1967 era were compelling, their application was not. In fact, employees hired under the national rubric sometimes ended up carrying out duties similar to those undertaken by the prevailing rate employees. At the same time, the latter group included tradesmen and marine workers, but not administrative staff that would also usually be hired locally.

Defining meaningful geographic zones for each of dozens of professions, gathering reliable local labour market data on a timely basis, and adjusting pay rates accordingly would be a complex and difficult enterprise. Such an approach could be expected to engender persistent complaints and appeals, particularly if adjustments lagged labour market changes to any substantial extent. This kind of nuanced management of imprecise distinctions is simply not something the federal public service has historically handled well.

The existing situation regarding regional pay in the federal public service is unsustainable. A few regional distinctions have survived the determined opposition of the public service unions, but most make little sense. As we pointed out in Chapter 4, successive rounds of collective bargaining have collapsed 16 zones for trades-related groups as recently as 1989, to three in the 2001 settlement. The three zones are so large and diverse as to be meaningless for the purpose of tracking local labour market pay differences. In the 2004 round of bargaining, the three zones were maintained; however, this was in the context of comparatively larger increases for trades-related employees because of general evidence that they lagged private sector analogues. We can expect the union to target the remaining zones more forcefully in the next round of bargaining.

On balance, our conclusion is that the federal employer should seek to restore a rational regional pay policy for any group that meets these criteria:

  • Nearly all hiring is local and there is no need to recruit nationally to attract an adequate number of qualified candidates.
  • There are substantial differences in salary levels for the same work across the country.
  • Data on these differences can be collected in a credible way, and at a reasonable cost.
  • Mobility between regions of federal employees in the group is negligible.

In designing a modernized occupational and bargaining group structure for the core public service, consideration should also be given to the issue of whether applying regional pay would make sense for particular groups, according to the criteria stated.[190]

Cost-of-living allowances

Sometimes the issue of regional pay is confused with the related phenomenon of differences in the cost of living across Canada. There is no doubt that it costs more for housing, transportation and parking, for example, in such cities as Toronto, Vancouver, Calgary and Ottawa than in smaller centres. In effect, national pay rates mean federal employees in the latter locations can enjoy a higher standard of living than their colleagues in the major cities.

It would be possible to set a national pay rate, and then provide for a variable allowance based on the local cost of living relative to a national average. The Canadian Forces in fact has such an allowance in the Post Living Differential, implemented in 2000–01. In February 2003, the value of this allowance varied from zero in Kingston, for example, to as much as $1,138 per month in certain parts of Toronto. The United States Civil Service also includes a locality pay component in its General Schedule pay ranges. Extra payments are provided where non-federal pay exceeds federal pay by more than 5%. Different rates are set for 32 geographic areas, including a "rest-of-U.S." area. In 2003, this locality adjustment was 21.08% over base pay for federal employees living in San Francisco, the highest cost area in the United States. The national average for locality pay was approximately 15% over base pay.[191] 

The feasibility of moving to such a regime in the federal public sector generally would depend on how the national rate was set. At present, Ottawa-Gatineau appears to be the implicit reference labour market for most groups. This makes sense in that the largest proportion of the public service works in the National Capital Region (NCR). On the other hand, the public service is such a prominent employer in the NCR that the external labour market may not be sufficiently influential in shaping the NCR labour market. The most prominent private sector component of the labour market, the high technology sector, is itself a relatively well-paid industry that is unrepresentative of the overall Canadian private sector.

Our conclusion is that the cost-of-living differential issue is important but not urgent at present. In the absence of significant inflationary pressures, employees are not constantly reminded of these differences. Once some of the more pressing pay management and compensation issues covered in this chapter have been resolved, it would make sense to examine the desirability and feasibility of introducing a system of national rates with regional supplements based on substantial cost of living differentials.

In summary, then, our recommendations in the area of general salary policy and management are these:

Recommendation 12

12.1 On pay for performance, we should focus first on strengthening our culture and capacity for performance management, including developing mature skills in defining organization objectives and standards, setting individual goals, assessing performance, giving feedback, and taking necessary action.

12.2 Deputy ministers should propose to the Treasury Board selective experiments in making greater use of pay for performance in cases where the organizational capacity and culture are suitable and credible with both managers and employees.

12.3 The Treasury Board should similarly work with willing departments and the appropriate unions to experiment also with one or more well-defined pilot projects to test the utility of pay for knowledge in such areas as language skills and advanced qualifications for specialized groups.

12.4 The Public Service Human Resources Management Agency should propose for Treasury Board approval ground rules for cases where the Treasury Board could authorize a higher maximum salary for specific executives or other employees bringing exceptional expertise to the public service, in situations where interchange programs are inappropriate as a means to access the required expertise.

12.5 As part of the review of executive salaries proposed later in this section, the overlap of top level specialist pay with Executive pay ranges should be reviewed and rationalized so that high level technical expertise can be appropriately remunerated without the need to resort to Executive classifications for jobs that focus mainly on advice and relatively little on executive management.

12.6 Recruitment and retention (terminable) allowances should generally be avoided. If there is clear evidence in the external labour market of a sustained relative salary level increase for a particular group, this should be included in public service salary levels. If rapid increases occur in a hot market externally, and there is doubt whether the change is temporary or structural, one-time bonuses should be used to recruit or retain essential staff in as targeted a manner as feasible.

12.7 As part of the process of modernizing classification standards recommended earlier in this section, the Treasury Board Secretariat and the Public Service Human Resources Management Agency should work together with the relevant unions to ensure that the structure of pay bands for these groups makes sense according to accepted salary administration norms.

12.8 In recommending on how best to modernize the public service's occupational and bargaining group structure, consideration should be give to the suitability of applying regional pay to some parts of the suggested new structure. Criteria for identifying such groups would include an emphasis on local hiring, substantial salary differences across the country, availability of credible data on these differences, and negligible geographic mobility in the group.

12.9 Once the more pressing pay management and compensation issues covered in this chapter have been resolved, the Treasury Board should examine the desirability and feasibility of introducing a system of national salary rates with regional supplements based on substantial cost of living differentials.

Pay determination for special groups

The government determines salary levels and other compensation for several groups in the federal public sector through various methods other than collective bargaining. The most important among these groups[192] are these:

  • executives and deputy ministers;
  • heads of Crown corporations and others appointed by Order-in-Council;
  • the Canadian Forces;
  • regular and civilian members of the Royal Canadian Mounted Police;
  • federally appointed judges; and
  • parliamentarians.

Executives and Deputy Ministers

In Chapter 3 of Volume Two, we describe both current practice and the history in recent years of how salaries for Executives and Deputy Ministers have been determined. In essence, the Treasury Board sets these salaries on the basis of advice from an arm's length Advisory Committee on Senior Level Retention and Compensation. The Committee in turn relies on a comparison, updated each year by Hay Associates, of total compensation paid by a sample of the Canadian private sector and broader public sector organizations for jobs equivalent to those classified at the EX 1 level in the federal public service. The salary level is set to match total compensation, taking account of the relative value of benefits inside and outside the federal public service.[193] Salaries for higher Executive and Deputy Minister classification levels are set as fixed multiples of the EX 1 level rate.

Successive Governments have decided to match the external labour market in the area of executive compensation at the EX 1 level, but not at higher levels. In fact, the equivalence is at best inexact in that there are both structural and policy-related lags at the EX 1 level. The structural lag results from the fact that salary levels set effective in April of a given year reflect data reported from the external labour market as of the previous September. Moreover, policy considerations led to decisions by the Treasury Board in two recent years to approve salary increases below the levels recommended.[194]

At successively higher classification levels above EX 1, annual Hay Associate reports[195] show a progressively larger lag between federal executive salaries and compensation versus their sample of comparable jobs in the Canadian private sector. For example, the March 2004 report (giving market data as of September 2003) showed actual salary and total compensation gaps[196] as follows:

Classification 
Level

Base 
Salary Gap

Total 
Compensation 
Gap

EX 1  (893 points)

-4%

-3%

EX 2  (1134 points)

-12%

-15%

EX 3  (1447 points)

-22%

-35%

EX 4  (1847 points)

-34%

 -48%

EX 5  (2207 points)

-47%

-71%

DM 2 (3455 points)

-80%

-101%

If the principle of comparability with the external labour market is to be fundamental to federal public service compensation policy, it would be logical to reduce the existing gap at the levels of Executives and Deputy Ministers. From the data set out above, it would be reasonable to focus especially on the Assistant Deputy Minister (EX 4 and 5) and Deputy Minister (for example, DM 2) levels in moving top federal public service salaries closer to their private sector counterparts.

In considering this issue, care would be needed to confirm that the method of comparing the level of responsibility is sound. For example, it is often argued that private sector executives are much more clearly accountable than senior public officials for defined outcomes such as profits or increases to share value, that they must raise the funds needed to operate, and that they enjoy little security of tenure (although severance packages can be generous). On the other hand, well-informed private sector leaders generally acknowledge that the complexity, volatility, and transparency of public sector management generally exceed anything encountered in the private sector. Careful reflection would be required to define how best to ensure federal public service executive compensation is compared accurately to that provided in the private sector.[197]

Moving towards greater comparability for executives above the EX 1 level also implies greater rigour in defining what are executive positions in the public service. The essence of being an executive in the private sector is to be accountable for substantial results. In the federal public service in recent years, we have earlier observed on a tendency to classify high level technical or advisory positions within the EX group in order to attract suitably qualified candidates. The best way to deal with this issue, as we propose in Recommendation 12, is to provide more clearly for senior advisory or technical jobs to be paid at levels similar to lower level Executives, but without classifying them as Executives. The case for compensating public service leaders more comparably to their private sector counterparts would be more compelling to the extent that the EX group is limited to those with substantial management responsibilities.

A perennially sensitive point about remuneration for senior public servants is the comparison with the pay of Members of Parliament and ministers. There appears to have been an unwritten rule that deputy ministers should generally not be paid more than ministers. This is, of course, a purely political matter. If the Government wishes to adhere to such a view, then the existing system for setting Executive pay should probably be left unaltered. Strictly from a human resource management point of view, however, it would make sense to revise Executive––especially assistant deputy minister––and deputy minister salaries upward in order to correspond more closely with their private sector counterparts.

As an important corollary, the terms and conditions of Executive and Deputy Minister positions should provide clearly for termination, where an executive's performance has not met expectations for two years in a row.[198] To prevent cases where poor relations between one Deputy Minister or other senior executive and an individual executive might lead to unfairness in a decision to terminate, such decisions should be subject to appeal to a panel of three deputy or assistant deputy ministers, depending on the level of the person affected. In effect we are advocating a policy of "real executive pay for real executives" for the federal public service. Overall, the combination of higher compensation with less secure tenure should yield a greater commitment to performance in the management of the public service. Strengthening this culture at the top levels of the public service could be expected to permeate the broader institution as well. In this way, improved executive compensation could be expected to encourage a more results-oriented federal public service.

As part of any move to raise public service executive compensation, we would need to consider increasing the proportion of pay that is determined by performance during the year against agreed goals. At present, as we have noted earlier, maximum at-risk pay for Executives ranges from 10% at levels 1 to 3, and 15% at levels 4 and 5. For Deputy Ministers, the range is from 15% at level 1, to 20% at level 2 and 3, and 25% for level 4. At the EX 1 level, 7% is part of the total compensation equivalence with the private sector. So only up to 3% is a potential bonus, and relatively few managers earn this extra compensation. Because of the persistent public confusion over this subject, any increases to the performance pay regime should clearly distinguish re-earnable pay from bonuses for exceptional performance.

It could well be argued that most or all of the re-earnable portion of performance pay should be included in regular executive salaries. Performance pay would then be limited to a true bonus system, restricted to only the best performers––perhaps 20% of the group at most. Such a system would be more readily understood by the public and more easily defended. However, this approach would remove an element of accountability for performance among executives generally, an emerging culture that has made progress and has room to become still more rigorous. Backtracking in this way on performance pay for executives could also make it virtually impossible to contemplate expanding performance-based pay practices to other groups of employees.

Finally, implementing such a change would, in effect, reward more poorly performing executives, whose pay would increase as a result, whereas most of the better performers would see no improvement in their total pay. They could even experience an effective reduction, if only part of the re-earnable portion of performance pay were rolled into regular salaries, in order to create a larger pot for bonuses for the top 20% or so.

Heads of Crown corporations and others appointed by
Order-in-Council

Compensation for the heads of Crown corporations and for others appointed by Order-in-Council is set in a manner similar to that used for Executive and Deputy Minister compensation. The Advisory Committee on Senior Level Retention and Compensation again recommends salary range and other compensation adjustments based on a parallel Hay Associates study of private sector and broader public sector counterparts.

In 2000, the Privy Council Office (which manages these appointments) completed a thorough review of the classification of the position descriptions for the heads of Crown corporations. This review updated compensation in this area to a significant degree. Then in 2005, the Committee on Senior Level Retention and Compensation, concerned about external comparability and internal equity, recommended that Crown corporation total compensation at the Group 1 level be matched to the median (50th percentile) of comparable positions in the private and public sectors.

As for the other Governor-in-Council positions, the review of their classification and compensation structure completed in 2002 generated substantial improvements in compensation. In 2005, the Committee on Senior Level Retention and Compensation recommended that the most populous level, GC-3, be matched in terms of total compensation with similar positions at the median in the private and broader public sectors in Canada.

The Government approved the 2005 recommendations, based on the Committee's assessment that they would re-establish internal relativities among the different types of senior level positions in the federal public sector, including Executives and Deputy Ministers.

If it is decided to move public service Executive and Deputy Minister salaries closer to those of their private sector counterparts, it would be appropriate to review further the salary levels of the heads of Crown corporations and others appointed by Order-in-Council, to preserve reasonable relativity within the federal public sector.

Canadian Forces

We explain in Volume Two the complex methods used to determine compensation for the Canadian Forces. Briefly, a comparison of total compensation is calculated separately for Non-commissioned Members (NCMs) and for General Service Officers (GSOs). The comparison matches a very specific set of compensation elements. Benchmark positions are used to determine a compensation difference with the core public service, known as the "warranted percentage increase." Additional amounts are paid for the military factor, a percentage in lieu of overtime, and an amount for acting pay. General Service Officer (GSO) compensation is set based on public service comparators using a modified version of the Hay Plan. Specialist officers such as legal officers, and medical/dental officers at the rank of captain and above, are benchmarked respectively to Department of Justice lawyers and to private sector physicians and dentists. Senior Officers are paid specified multiples of core public service Executive compensation.

Volume Two also gives some sense of the history of frustrations and misunderstandings that have marked the setting of military compensation levels over the years. In practice, the system for calculating the warranted percentage increase is so complicated and so little understood––or in fact trusted––that it is seldom accepted by the Treasury Board. Typically it falls to senior Canadian Forces and Treasury Board officials to work out an acceptable number based on considerations ranging from the tempo of military operations to the current state of core public service collective bargaining negotiations. From time to time, a more thorough revision of military compensation has been approved, for example in the late 1990s.

The existing approach can fairly be described as an inefficient and non-transparent combination of complexity and arbitrariness. Canada's military personnel are increasingly called upon to undertake life-threatening missions throughout the world. Respect for their difficult and often dangerous contribution to Canada suggests the need to adopt a more transparent and balanced method for determining remuneration for the Canadian Forces.

In countries such as Denmark, France, Ireland, the Netherlands, New Zealand, and the USA, it is reported that "the defence ministry is responsible for determining pay in the Armed Forces. In most cases these decisions must be ratified by a higher level of Government or by the Treasury."[199] On the surface, the system in these countries appears similar to ours. There is, however, considerable variation in how each country's practices actually work.[200]

Australia has the most independent institutional arrangement, in the form of a tribunal called the Defence Force Remuneration Tribunal. Established in 1984, the Australian Tribunal has a mandate to "inquire into and determine the salaries and relevant allowances to be paid to members of the Australian Defence Force (ADF)."[201] While this is an interesting case, it would be unlikely to fit the Canadian scene. Australia has a long history of using quasi-judicial bodies to determine pay awards in the private sector, so extending the practice to the ADF would have been entirely consistent with general norms.[202] In Canada, no group's remuneration is entirely independent of approval by the Government, so treating the Canadian Forces as beyond the control of the Treasury Board and Parliament seems a non-starter.[203]

More in keeping with Canadian parliamentary and public service traditions and values, the United Kingdom has an independent Armed Forces Pay Review Body that advises the Prime Minister and the Secretary of State for Defence jointly on the remuneration and allowances for the Armed Forces. First established in 1971, since 1998 the Review Body's terms of reference direct it "to have regard to the following considerations" in reaching its recommendations:

  • the need to recruit, retain and motivate suitably able and qualified people, taking account of the particular circumstances of Service life;
  • Government policies for improving public services, including the requirement on the Ministry of Defence to meet the output targets for the delivery of departmental services;
  • the funds available to the ministry of Defence as set out in the Government's expenditure limits; and
  • the Government's inflation target.

The Review Body is also directed to "have regard for the need for the pay of the Armed Forces to be broadly comparable with pay levels in civilian life."[204]

Although we recognize that institutions cannot simply be imported from one political/bureaucratic context to another, a version of the UK Armed Forces Pay Review Body could serve us well in Canada. Here matters of remuneration are usually the responsibility of the Treasury Board, so a Canadian Forces Compensation Advisory Committee would best report jointly to the Minister of National Defence and the President of the Treasury Board. Members of the Committee should include persons of eminence in Canadian public life who would be expected to balance a deep concern for the effectiveness of the Canadian Forces and the well being of its members, with a prudent and frugal regard for the public purse. The expectation would be that well-supported recommendations would be implemented by the Government, although there would always be the option of altering what was proposed or of phasing in some elements.

An important difference between Canadian and British circumstances is how departments are financed. In the United Kingdom, each department must keep its spending within defined expenditure limits which presumably act as a constraint on what the UK Armed Forces Pay Review Body can recommend in terms of compensation increases. In Canada, the practice has been, in most cases, to fund compensation increases from the Treasury Board's reserves. It would be important to ensure that any Canadian Forces Compensation Advisory Committee is effectively required to weigh desirable increases against the need to ration public funds wisely, within appropriate fiscal limits.

The approach proposed here would serve several goals. First, by adopting a transparent process for determining military compensation, the Committee would reassure Canadian Forces members that remuneration decisions affecting them take full account of their circumstances and needs. Second, by emphasizing the unique characteristics of military life and compensation, the Committee's work would make it easier to distinguish pay and benefit decisions affecting the Canadian Forces from those affecting other groups in the federal public sector. Finally, if completely implemented, this approach could expect to enjoy credibility with the general public and all-party support in Parliament.

Civilian and regular members of the Royal Canadian Mounted Police

Volume Two describes the creation of the RCMP Pay Council in the mid 1990s, with a mandate to develop an "orderly, independent, transparent and professional" approach to compensation in the Royal Canadian Mounted Police. Overall, the RCMP Pay Council has worked well over the past decade. By using a consistent methodology to compare total compensation for the RCMP with the principal provincial and municipal police forces across Canada, the Council has largely succeeded in determining the compensation increases approved by the Treasury Board in recent years.

However, in 2004, the system came under strain. For the first time in several years, the Treasury Board approved a lower rate of salary increase than the RCMP Pay Council had recommended, based on its annual survey. At the root of this were an unresolved policy difference, and an exceptionally large increase for the Toronto Police Force.

The RCMP has argued consistently that its members' compensation should track the average of the top three comparator police forces in Canada. The Treasury Board applied that approach in practice for some years, but declined to adopt it as a policy, precisely because the Secretariat was concerned that one or two exceptionally large pay increases, adopted more perhaps because of local politics than labour market considerations, could lead to an unreasonably large increase for the RCMP. Such an increase could also have a disruptive impact on other pay determination processes in the public service.

More importantly, as we saw in the previous section on military compensation, the UK Armed Forces Pay Review Body is required to take account not only of comparability, but also of such issues as productivity improvement, expenditure limits, and the Government's inflation target. The Treasury Board Secretariat has maintained as well that the relative ease or difficulty of recruiting and retaining RCMP members should be considered in determining pay levels for the federal police force.

In 2004, in fact, a Toronto-specific increase pushed the average of the top three forces well beyond what was on offer to the largest public service union in collective bargaining. The resulting decision to adopt a lower increase provoked bad feelings within the RCMP.[205]

In our view, the basic approach embodied in the RCMP Pay Council is appropriate and should be maintained. However, after nearly a decade in place it would be reasonable to review the experience, including how the total compensation comparisons are calculated, and how best to deal with unusually large increases affecting only one or two forces. The review should also consider how to make explicit the role of factors other than comparability (e.g. ease or difficulty in recruiting and retaining members) in assessing pay increases. Such a review could be expected to result in a renewed system that both the RCMP and the Treasury Board Secretariat could manage with confidence for many more years to come.

Federally appointed judges

Remuneration for federally appointed judges is set out in the Judges Act, as amended from time to time. A Commission is established every four years (hence the title Quadrennial Commission) to review the suitability of judicial compensation and to make recommendations for appropriate changes. The Commission is composed of representatives named by the Minister of Justice and by the Canadian Judicial Council (which represents federal judges), and a Chairperson determined by these two members.

This approach preserves the judiciary's constitutional independence in that the Government, on whose actions the federal and provincial superior courts must rule, is not responsible for determining judges' salaries and benefits. Although Parliament is responsible for reviewing the Commission's recommendations, it is not bound by them. In amending the Judges Act, Parliament may deny or alter a Commission's recommendations, provided it substantiates this action rationally. However, the accumulating precedent of adopting these recommendations in recent decades makes such an outcome increasingly less likely.

On the other hand, it is not clear what the proper comparator is for setting judicial compensation. Historically, as described in Chapter 3, Commissions have tended to use the mid-point of the salary range for Deputy Ministers, level 3. In the most recent report, published in 2004, the Commission favoured comparison with the compensation of law firm partners in major urban centres. It seems likely that this shift in reference points was motivated by the fact that public service deputy ministers' salaries have grown only slowly in the past few years.

A broader public debate about the issue of to whose salaries judges' compensation should be compared would be warranted, especially if the Government decides to increase the comparability of public service Executive and Deputy Minister salaries with their private sector analogues. In fact, the work of judges is very different from that of the deputy heads of federal departments and agencies in content, conditions of work, and accountability. The main pool for recruiting judges is law firm partners. So experienced private sector lawyers are in principle almost certainly better comparators. The problem, of course, is the wide variability of the remuneration of such private attorneys, across regions, firms, legal specialties and individuals. This issue of identifying the appropriate comparators in recommending changes in the level of judicial compensation demands more attention, if the current Quadrennial Commission system is to retain its public support.

Parliamentarians

Volume Two explains how it was decided, pursuant to a Commission report in 2001, to raise the salaries of Members of Parliament and Senators, to include their tax-free allowance within their salaries, and to tie future increases to the same rate of increase as legislated for federally appointed judges. In 2004, however, the Government committed to de-couple parliamentary salary increases from those approved for judges. In 2005, the Parliament of Canada Act and the Salaries Act were amended to establish a new index for calculating annual salary increases for Members of Parliament and Senators.

The decision to separate the means for setting judicial pay from that for compensating parliamentarians is conceptually sound. The work and responsibilities of these two groups of critical actors in Canada's system of governance could hardly be more different. Judges serve for life and must act independently, usually taking their decisions alone. Parliamentarians serve for terms of varying length and normally operate within a system of party discipline, deciding matters collectively.

The approach adopted for parliamentary compensation in 2005 provides for annual increases that are transparent, independent of federal public sector increases, and reflective of increases awarded to average Canadians. It also removes the need for parliamentarians to vote on their own salary increases. By providing annual increments corresponding to the change in the industrial aggregate wage level, the new policy will preserve parliamentarians' purchasing power relative to the labour force as a whole.

However, it is predictable that over time the relative position of parliamentary compensation versus that of other senior federal officials, including judges and deputy ministers will raise doubts about whether simply matching average Canadian wage growth will attract strong candidates to seek election. From time to time, therefore, it would make sense to appoint a Commission, similar to that appointed every four years to review judges' compensation, to recommend whether and how to update parliamentary compensation.

It could be argued that MPs and Senators are put in a conflict of interest in having to decide on their own compensation in response to the recommendations of such a Commission. To mitigate this criticism, it could be decided that any increases beyond the annual industrial aggregate adjustment should apply only after a new Parliament is elected. Even this would not eliminate controversy, since some candidates would no doubt be tempted to promise to rescind a pre-approved increase. The fact is, however, that only parliamentarians can decide on their own level of compensation. Even adopting the annual industrial-aggregate-increase approach was such a decision. Any pressure over time to move beyond the level yielded by the new formula would best be decided transparently on the basis of independent advice.

In summary, then, the recommendations related to the area of determining salaries for particular groups in the federal public sector are as follows:

Recommendation 13

13.1 The Treasury Board President should ask the Advisory Committee on Senior Level Retention and Compensation to review and offer recommendations on the extent to which federal public sector executives should be compensated at levels equivalent to their private sector counterparts. It would likely make most sense to focus initially on the salaries of Assistant Deputy Ministers and Deputy Ministers. In doing so, the Committee should assure itself that the method of comparison is conceptually sound, and that the comparability data is reliable. It should also consider proposing a narrowing of the definition of positions that can be included in the Executive group to exclude those without substantial management responsibilities, and recommending terms and conditions of employment that acknowledge that an executive may be let go for inadequate performance.

13.2 The Committee should at the same time reconsider the appropriate role of pay for performance in the overall executive compensation package.

13.3 If the Treasury Board President decides to ask for the review proposed in recommendation 13.1, the Privy Council Office should request the Advisory Committee to undertake a similar review and offer recommendations with regard to our positioning of compensation for heads of Crown corporations and others appointed by Order-in-Council vis-à-vis private sector counterpart compensation.

13.4 The Minister of National Defence and the President of the Treasury Board should jointly appoint an eminent Canadian public figure to propose how best to establish a Canadian Forces Compensation Advisory Committee, modelled generally on the United Kingdom Armed Forces Pay Review Body.

13.5 The Minister of Public Safety and Emergency Preparedness and the President of the Treasury Board should invite a reputable consulting firm[206] specializing in compensation issues to review how the RCMP Pay Council measures total compensation, what considerations other than comparability ought to be weighed in determining RCMP compensation, and how the Treasury Board should deal with situations where compensation increases for one or more police forces are unusually high.

13.6 The Minister of Justice and the Canadian Judicial Council should ask the next Quadrennial Commission, scheduled to report in 2008, to conduct as part of its deliberations an in-depth analysis of the value and relevance of various possible comparators in determining the appropriate level of compensation for federal judges.

13.7 If, in the future, there is a sense that parliamentarians' salaries are out of step with what is paid for other senior federal public sector positions, an independent commission, similar to that appointed every four years to recommend changes to judicial compensation, could provide a transparent and independent means to advise Parliament on how best to update parliamentary compensation.

13.8 In mandating all the various advisory groups on compensation, the principle should be respected that they are independent in regard to their analysis and formulation of recommendations, but that the final decision on compensation rests with the Government or Parliament, as appropriate.