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Warning This page has been archived.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.

2. History of Compensation Comparability and Collective Bargaining

In this chapter, we summarize how the federal government has expressed its intentions with regard to policy on comparability over the years. We note that practical considerations have frequently taken precedence over policy. As the Report of the Royal Commission on Government Organization (Glassco Commission) remarked in 1962: "While market prices set the limits of wage policy, changing political, administrative and social influences affect the determination of a particular wage rate."[6] 

With various formulations over the years, successive federal governments have sought to position compensation policy as seeking comparability to the external labour market. However, the formulations were at such a level of generality that the results obtained were quite varied. Despite this variation, there has been a persistent pattern of paying above market at lower levels of the public service, below market in the higher ranks, and in the middle higher or lower depending on the circumstances. This pattern applies to wages and even more so to total compensation.[7]

Comparability and its application from 1913 to the early 1960s

The beginning: a merit-based pay system

Between 1913 and 1920, public service salary increases were equivalent only to about half of the national cost-of-living growth.

The First World War led to a rapid expansion of the federal public service, and to a growing recognition that to be effective the public service should be staffed on the basis of merit, not patronage. In 1918, the Civil Service Act provided for appointments to be regulated by the Civil Service Commission according to the principle of merit. The Commission was also mandated to recommend revisions in rates of pay. In pursuit of this responsibility, the Commission sought the advice in 1919 of Arthur Young & Company on how to structure a system of job classification and pay setting. The following general principles[8] guided their work.

Uniform compensation

Rates of compensation should be uniform for the same rank.

Right pay for different "classes" of work

Rates of compensation should be relatively "right" for different classes. In the case of classes of positions in different fields, this meant that they should bear the same relation to classes of positions in fields established in the "business world" among respective vocations, trades, professions and lines of work. Within the same vocation, trade or profession, this relativity was to be measured by differences in duties, responsibilities, experience, knowledge and skill.


The pay for each class should be equitable: fair to the employee and fair to the taxpaying public.

Fairness to employees

Fairness to the employee was defined to require that the compensation should permit the maintenance of a standard of living that would make for the good of society and posterity. In the case of the lowest ranks, the compensation was to be adequate to attract into the service young men and women without "family responsibilities," but of a training and capacity that would enable them to become of future value to the service and to themselves.

Fairness to the taxpaying public

Compensation should not materially exceed that paid for similar services by enlightened employers in the general industrial and commercial world. Any excess over such a prevailing average was said to be in the nature of a special subsidy with which no group should be favoured.

In comparing the compensation paid in government and in business for similar services, the relative advantages and disadvantages of employment in the two sectors should be taken into account, including permanence and continuity of tenure, hours of work, and holidays and sick leave.

This first policy statement contains several themes that recurred throughout the ensuing decades. Most fundamental is the concept of fairness to both the employee and the taxpaying public. Quite explicit is the idea that the way to achieve that balance is to emulate the business world as the standard for appropriate compensation for the various professions, trades and lines of work. The compensation limit should be that amount paid for similar services by enlightened employers. Finally, the comparison is to take account of all forms of remuneration, not just salaries. The increases implemented as a result of the Arthur Young & Companyreport were lump sums that were much larger in percentage terms at lower income levels.

In applying the report's policy recommendations, the Government created two classes of employees. The first comprised workers appointed under the Civil Service Act, whose pay rates were recommended by the Civil Service Commission. By 1922, there were about 50,000 civil servants organized into some 2,200 classes for pay determination purposes. Another 13,000 employees, mostly skilled tradesmen and ships' crews, were exempted from the application of the Act. These eventually became known as "prevailing rate" employees because they were paid at the rates applicable in the geographic locality of their employment.[9] Treasury Board set the rates for the excluded tradesmen, with the assistance of regional prevailing wage surveys by the Department of Labour.

The Beatty Commission

By the late 1920s there was evidently a consciousness of a need to improve the relative compensation of what would now be called "knowledge workers" in the public service. The Royal Commission on Technical and Professional Services, chaired by E. W. Beattyreported in 1930. It is evident from the Beatty Report that the major driver of salary change was not comparability. At the same time, Beatty observed a growing complexity in the task of public administration, as well as an "intrusion" of government into many new fields. The Beatty Commission observed that:

... so far as the scale of salaries is concerned, we find that the technical, scientific and professional workers in the junior ranks of the service are at no pronounced disadvantage as compared with other similar employees in outside employment. Indeed, it is evident that beginners' salaries in the service are not infrequently somewhat larger than beginners' salaries elsewhere.[10]

The problem lay with the difficulty of advancement within the public service compared with private employment. The Commission also expressed concern that the classification classes arising from the Young system had become "unnecessarily cumbrous." By the time the Commission reported, however, the Great Depression was taking firm hold and the Commission's recommendations lay fallow for over a decade and a half.

Depression and war

The next major policy event was the 1932 Salary Deduction Act, which reduced salaries by 10% generally. With wage cuts substantially greater in the private sector, and the gradual restoration of the 10% cut in the public service, salaries in the public service on the eve of World War Two were favourable compared to those in the private sector. Comparability appears to have been only loosely applied in determining these outcomes.

During the war, pragmatism ruled once again. Salary increases were generally restricted. To recruit needed staff and avert serious injustice, the Government used such expedients as a cost-of-living bonus, promotions in war units, war duties supplements and a loosening of the classification system.

Postwar changes

The Second World War resulted in the public service growing to 117,000 by 1946. The 1946Royal Commission on Administrative Classifications in the Public Service, chaired by W.L. Gordon, took up in effect where Beatty had left off. The Gordon Commission lamented that with the steady growth in the responsibilities of government, there were not enough "men" of high calibre in the senior and intermediate grades. They agreed with Beatty that "overlapping of salary scales, in successive positions on the ladder of promotion [is] not consistent with the maintenance of maximum efficiency."[11] The Commission criticized the conflicting roles of the Civil Service Commission and the Treasury Board in regard to wage determination, and favoured giving that role entirely to the latter organization.

In his recommendations, Gordon proposed the following principles for setting remuneration in the upper ranges of the public service:

The general level of these scales for the administrative [now executive] and for the scientific, technical and professional grades should be such that at the bottom they will attract for the public service the necessary proportion of the highest quality products of the universities; in the intermediate ranges they will recognize the increasingly important duties being undertaken and meet the rising family responsibilities which have usually to be faced by men in their early and middle thirties; and at the top will enable senior officials to perform their duties free from financial worries and distractions.

As a result of the Gordon Report there were selective increases in the salaries of various deputy ministers and other senior officials. More generally, the wartime cost-of-living bonus was incorporated into the public service salary structure.

Relatively rapid postwar increases in private sector pay rates, and continuing growth in the size of the public service put great pressure on the level and structure of public service rates of pay. The tendency to resort to across-the-board increases highlighted the weakness of the existing machinery for revising pay rates. In 1948, then Prime Minister Louis St. Laurent re-stated the principles guiding the Government's compensation policy, as follows:

The Government's policy on salaries in the Public Service has long been based on two main principles. First that they should be sufficient to attract to, and retain in, the Civil Service persons of the right type and that, considering all relevant factors such as conditions of employment, salaries for each class of work should be generally in line with those paid for comparable work by good private employers.

The greater emphasis on external comparability in this relatively succinct formulation emphasized the need for better information on private sector compensation levels and practices. In 1957, inspired by the 1955 British Royal Commission on the Civil Service, the Government established the Pay Research Bureau. Its mission was "to provide objective information on compensation and working conditions in government, business and industry, and to assemble and analyze factual evidence of trends in outside employment." This represented an important innovation in at least two respects: first, it reflected a commitment to empirical evidence in salary determination; second, its governance provided for active involvement by staff associations in shaping the Bureau's priorities.

In 1958, Prime Minister Diefenbaker made the following statement on public service compensation policy:

First, the salaries must be enough to do the job, that is, to attract enough of the right kind of men and women into the Service and keep them in it; second, they must be fair as between civil servants and people outside the Service, the taxpayers if you will, which means that the salaries we pay for any class of work should be comparable with those paid by private employers for similar classes of work, taking into account the other terms of employment that are necessary to make a fair comparison.

It is interesting to note the changes from "enlightened employers" (Arthur Young & Company, 1919), to "good employers" (Louis St. Laurent, 1948) to simply "private employers" (John Diefenbaker, 1957). We can only speculate whether this apparent watering down over the years of the description of the proper private sector comparator employers represents a conscious retrenchment of view.

Consistent with this policy statement, the new Civil Service Act (1961) included in subsection 10(2) the first legislative expression of compensation policy:

The Commission in making recommendations on remuneration shall consider the requirements of the civil service, and shall also take into account the rates of pay and other terms and conditions of employment prevailing in Canada for similar work outside the public service, the relationship of the duties of the various classes within the civil service and any other considerations that the Commission considers to be in the public interest.

By the early 1960s, therefore, comparability in public service compensation with the external labour market had become more prominent in policy statements but not notably central to the actual determination of salaries and benefits.

The Glassco Commission

Today, more than 40 years after its publication, the section on "Personnel Management" in the Report of the Royal Commission on Government Organization (Glassco Commission)[12] captures many of the issues that still challenge the public service. In the specific areas of public service compensation and its comparability to the general economy, Glassco offers a solid examination of policy and practice.

Based on work by the new Pay Research Bureau, the Glassco Commission provided an assessment of how public service salaries and benefits related to comparable jobs in the private sector. Among the Commission's general observations on comparability are these points:

Wage and salary rates for the lower grade positions in the civil service are in general equal to, or better than, those for comparable jobs in private industry. Some disparities appear in salary rates for jobs above those levels, most markedly in senior administrative and professional posts, where the government is at a marked monetary disadvantage in competition with private industry.

Lacking flexibility to adjust to local pay levels, the civil service has in many cases pay rates that are above community rates in some centres, below community rates in others.

Benefit plans for the public service are, in general, more favourable to government employees than those found in most private industries.... Some large employers offer a benefit package almost as good as that offered in the public service, and a few provide some benefits that are more attractive.... The biggest attraction of the public service plans is the provision for pensions.... Employee benefit plans are so complicated that many public servants, as well as potential recruits, fail to recognize their value.[13]

With this broad understanding of the relative place of public service compensation in the Canadian economy of the day, the Commission turned a critical eye to the policy pronouncements made to that time. In its view:

There is still no comprehensive statement of compensation principles for the public service. Past statements have been so generalized as to be of little practical value as policy guideposts.... Only in recent years has there been much attention to market forces in government pay determination.[14]

The Commission's own prescription was the following:

Pay policy should, first, facilitate the staffing of the service with competent personnel by attracting suitable recruits and retaining effective employees; second, compensate employees in the public service fairly; and third, achieve these two aims at a cost which is as reasonable as possible to the taxpayer.[15]

It might be thought that this formulation is no less generalized than the statements of compensation principles that the Commission itself criticized. In fact, the Commission proceeded to deconstruct key terms such as "market" or "going" rate. It emphasized the need to specify, for example, that wage surveys should not be limited to the "best employers" but should cover "a representative sample of those employers with whom the public service is in competition for recruits."[16]

The Commission also highlighted the importance of ensuring equitable treatment for government workers such as meteorologists, for whom there is no outside labour market equivalent. It delivered a strong critique of the existing classification system as a barrier to fair compensation[17] and declared in its conclusion that "the present remuneration system is essentially negative," in that it does not provide for "positive recognition of superior performance."[18]

The Glassco Commission Report in general marked a watershed between a traditional public service rooted in prescription and process, and a newer public service characterized by a larger role for managerial delegation and accountability for decisions tailored to local conditions. In the area of compensation and comparability, the Commission offered the most complete assessment of the relative position of the public service to that time and perhaps since. Its commentary on compensation principles and their interpretation was trenchant and suggestive of how to improve their clarity and application.

But the Report was submitted in an era when employer paternalism was losing ground to a widening social acceptance of a more collaborative approach to setting wages. Almost immediately after the Report appeared, the minority Pearson Government decided to shift salary determination in the public service into the world of collective bargaining.

Public service compensation policy and comparability in the early years of collective bargaining

Preparatory Committee on collective bargaining

In 1965, the Preparatory Committee on Collective Bargaining in the Public Service, chaired by A.D.P. Heeney, reported.[19]To prepare the ground formally for collective bargaining, two basic issues needed to be addressed. First, how would the public service workforce be structured for collective bargaining? Second, what machinery and rules would be put in place to regulate the new system?

On the first point, the Preparatory Committee noted that: "the systems of classification and pay applying to civil servants and prevailing rate employees have not undergone a complete overhaul since 1919." Recognizing (as had Glassco) that there was a "lack of clear demarcation between the two systems," and that employees of the two types performed the same kinds of work, the Committee recommended that the two be amalgamated into a single system. For this they proposed a radical simplification of the classification of positions, as illustrated in Figure 1001. The existing 700 classes and 1,700 grades for the 138,000 workers employed under the Civil Service Act, were to be replaced by a two-level structure. The "occupational category" would be a broad horizontal division of the public service, useful for planning and the development of personnel policy. The second level, the "occupational group" would be a sub-division on which to base the process of pay determination. Ability to compare with "an identifiable outside labour market wherever possible" was emphasized.

Figure 1001
A new structure for the public service with the introduction of collective bargaining

Display full size graphic

A new structure for the public service with the introduction of collective bargaining

On the second point, two observations merit attention. Overall, the Preparatory Committee concluded that: "as far as possible, the system of collective bargaining and arbitration in the Public Service of Canada should be rooted in the principles and practices governing employer-employee relations in the Canadian community at large." The Committee was clear that the Government should accept the limitation on "the historic right of the Crown to determine unilaterally the terms and conditions of employment of those in its service" inherent in accepting a regime of collective bargaining and arbitration. Nevertheless, the Commission proposed that the "third-party dispute resolution" mechanism be a permanent Arbitration Tribunal, headed by a permanent Chairman. The members were envisaged as being "men and women of considerable stature in the community at large," who would "rarely if ever bring to their semi-judicial role a narrowly partisan point of view."

The Preparatory Committee also commented on the balancing act–between external comparability and internal equity–that is essential to sound salary determination:

The demands of the labour market and of internal relativity cannot be easily reconciled in the design of a classification system. Ability to respond to market pressures calls for a flexible structure consisting of many parts, each of which may move independently. The demands of internal relativity, on the other hand, call for a rigid and unified structure in which each job can be placed and remain in a fixed relationship to all others. Optimum results can probably be achieved only by striking a reasonable balance between the two extremes.

The market, in the view of the Commission, is most pertinent in periods of economic change and development such as that which had prevailed in Canada since the Second World War. A system too inflexible to adjust will be poorly equipped to recruit and retain "quality" personnel, particularly in the "high-skill" occupations.

The Committee made a particular point of stating its view that "the system should be designed in such a way as to provide strong incentives to superior performance," including extending such practices beyond the "Senior Officer class."

The Committee expressed its conviction that in many parts of the public service, a strike would be indefensible and a lockout unthinkable. Although it decided against prohibiting strikes, they clearly expected that arbitration would be the normal means for resolving disputes.

Influenced by the experience of the 1965 postal workers' strike, and no doubt by its position as a minority in Parliament, the Pearson Government decided "to accommodate the views of those who were opposed to arbitration in principle" by providing a conciliation-strike option in the legislation.

The Public Service Staff Relations Act

A legislative framework for collective bargaining in the federal public service was enacted for implementation beginning in 1967. The Public Service Staff Relations Act (PSSRA) set the rules for union certification, dispute resolution, and the resolution of grievances. The Act incorporated the gist of the Commission recommendations. The Financial Administration Act conferred on the Treasury Board the responsibility of the employer for most of the public service, including negotiations of the terms and conditions of employment for unionized workers, and the authority to establish a classification system along the lines proposed by the Preparatory Commission.

The PSSRAdid not contain a direct statement of the federal government's compensation philosophy. However, the matters that the Arbitration Tribunal must consider in determining an arbitral award (section 68) expressed indirectly the relevant policy goals. In view of the expectation that arbitration would be the normal means of dispute resolution, these considerations took on additional importance:

In the conduct of proceedings before it and in rendering an arbitral award in respect of the matter in dispute, the Arbitration Board shall consider:

  1. the needs of the Public Service for qualified employees;
  2. the conditions of employment in similar occupations outside the Public Service, including such geographic, industrial or other variations as the Arbitration Tribunal may consider relevant;
  3. the need to maintain appropriate relationships in the conditions of employment as between different grade levels within an occupation and as between occupations in the Public Service;
  4. the need to establish terms and conditions of employment that are fair and reasonable in relation to the qualifications required, the work performed, the responsibility assumed, and the nature of the services rendered; and,
  5. any other factor that to it appears to be relevant to the matter in dispute.

First steps

As collective bargaining was applied in the early years from 1967 to 1975, both parties were learning how to operate within the new system. Arbitration was the preferred choice for most disputes. In the first fiscal year of operation (1967–68), 26 of 30 disputes went the arbitration route, although the number of employees covered by the conciliation processes was larger than those whose bargaining agents opted for arbitration.

During these eight years, it is unclear to what extent comparability with external salary levels motivated arbitrators in fashioning arbitral awards, or negotiators in arriving at settlements through the conciliation route. According to a monograph by Finkelman and Goldenberg: "No decision of the Tribunal or of the Board has discussed the relative weight to be given to each of the factors [that arbitration boards should consider] listed in section 68 of the PSSRA."

The inflation-control and post-control eras

In 1975, the Anti-Inflation Act imposed limits on salary increases across the Canadian economy, including the federal public service, lasting until 1978. During this period, federal public service arbitration was brought clearly under the broader regime of salary determination in Canada, in effect imposing a form of comparability in setting salary levels.[20]

Post-inflation-control measures–Agenda for Cooperation

In May 1977, leading up to the end of the period of anti-inflation controls, the government published Agenda for Cooperation: A Discussion Paper on Decontrol and Post-Control Issues. The context was a strong view that inflation represented a serious threat to economic well-being and could not be permitted to rage out of control. In this regard, the Discussion Paper stresses that:

Governments have important responsibilities as employers. They must set wage policies in a manner that is fair both to their own employees and to all employees in the country. Governments should not be seen to be responsible for inflationary pressures, yet they must be fair to their own employees, and in particular to employees at the lower end of the income scale.[21]

In Chapter 8 of the Discussion Paper, which was entirely devoted to public sector compensation, the Government characterized the context as follows:

Even where similar jobs exist in both [the public and private] sectors and a policy of comparability has been adopted, its achievement is rarely straightforward or complete... As a result, public sector compensation policies tend to be influenced both by comparability and other principles, including the maintenance of relativities within the public sector, ability to pay, and broad social policy objectives.

The legitimate demand for continuity in the provision of public services, coupled with the absence of a "bottom line" or market test of appropriate compensation, often make it difficult for governments to resist settlements that might be unwarranted.

Looking to the post-control period, the Paper reiterates the longstanding principles of fairness to both employee and taxpayer, and reasonable comparability to the private sector, as well as protection for the rights of public employees to responsible collective bargaining. Then the Paper breaks important new ground by emphasizing "total compensation" (i.e. the combined value of all salary and benefits provided to employees) as the basis for comparability with the external labour market, as follows:

The approach the Government intends to pursue in compensating federal public servants is one of average comparability of total compensation with a representative sample of private sector employers. It will do so in a manner that ensures that federal public service compensation will continue to follow compensation in the private sector.

This does not mean that the federal government is abrogating its responsibility to set an example in certain situations. In fact the Government will continue to set the pace when social policy concerns indicate that it is appropriate. One can envisage a need for government leadership in areas such as working conditions, employer-employee relations, or non-wage benefits. It does mean, however, that compensation provisions resulting from initiatives in these areas will be explicitly recognized as part of the total compensation of federal public servants.[22]

The Discussion Paper acknowledged there was a need to develop the analytical means to determine total compensation and comparability. There was also speculation on how difficult it could be to phase in resulting adjustments, up or down, to implement comparability. Finally, the paper recognized that there would be a concern about the degree to which collective bargaining would remain meaningful in the context of the rigour implied by a total compensation policy. In answer, the paper asserted that "collective bargaining will continue to play a dominant and meaningful role in settling the terms and conditions of employment, in establishing the appropriate mix of total compensation, and in finding solutions to ... problems."

Post-control attempts to implement total compensation comparability

To implement this direction, the Government undertook three tasks:

  • first, developing the concepts and methodology for applying a total compensation comparability approach from the employer's perspective;
  • second, introducing legislation requiring arbitrators to apply this philosophy; and
  • third, arranging for the Pay Research Bureau to collect and report the data needed to underpin the analysis.

Although progress on defining the concept of total compensation was substantial, the other two tasks were partially unsuccessful, rendering the new approach largely unimplementable in practice.

There were five basic concepts underpinning total compensation comparability:

  1. The federal government, as an employer, would not be a pace-setter, i.e. would not lead the private sector on the total value of all elements of compensation for comparable work.
  2. Security of tenure would be included in the evaluation; regional rates would be adopted "as appropriate."
  3. Compensation for very low paid jobs would be set somewhat above comparability "as an instrument of social policy."
  4. Any catch-up would be phased in over a period of two to three years.
  5. Consideration would be given to "maintaining appropriate relativities within and between groups," and to "the relative contribution and performance of individual employees."[23]

From an analytical point of view, it was urgent to decide exactly what to include in aggregate comparisons. After canvassing well over 100 potential elements, including such intriguing items as the canoe allowance and the horse allowance, Treasury Board officials concluded that 12 components plus salary amounted to at least 95% of total compensation for both the public service and for Canadian employers generally. The 12 non-salary components were:

Conditions of work, extra pay and time off

  • overtime and other premium pay
  • severance pay
  • hours of work
  • vacation leave
  • sick leave
  • holidays
  • rest periods


  • disability insurance
  • pension
  • Supplementary Death Benefit (life insurance)
  • employer's share of provincial health premiums
  • employer's share of supplementary health benefits

Explicitly excluded from consideration were items such as the cost of training and career development, which were seen as a management investment in improved performance and productivity; reimbursement of travel and other expenses incurred to perform the employee's duties; legislated social program benefits such as unemployment insurance and worker's compensation; and the cost of administering compensation elements.

Considerable attention was devoted to calculating the value of security of tenure, which was seen as a significant benefit of public service employment. Consultant advice yielded a definition comprising two parts: probability of loss of employment, and expected monetary losses from lay off. It was then proposed that the value of job security could be translated into a hypothetical premium that could be paid to insure against monetary losses likely to occur if the job ceases. This premium could then be included in the calculation of total compensation.

The Government introduced Bill C-28[24] in March 1978, which proposed amendments to the PSSRA to complement this theoretical work. The Bill defined compensation broadly as follows:

In relation to hours worked, compensation means the aggregate of pay and the monetary equivalent of those allowances, benefits and other compensation-related terms and conditions of employment.

Included also were new sub-sections in section 68 of the Public Service Staff Relations Act, requiring arbitrators to base awards "on a comparison of the aggregate of compensation for similar or analogous work." A hierarchy of comparison was also mandated, first with private profit-oriented organizations, then with other entities outside the public service, or as a last resort with the public service. Arbitrators would have been required to give written justification for their awards. The Pay Research Bureau was also to be given a statutory mandate to support the process. In the face of union opposition, Bill C-28 did not proceed past First Reading.

The plan to use the Pay Research Bureau as the organization to generate the quantitative data needed to calculate total compensation comparability met with determined union opposition. As a result of the union position, the Public Service Staff Relations Board (PSSRB) and the Pay Research Bureau (which was part of the PSSRB) were reportedly unwilling even to entertain total compensation as a concept, or to share new data that the employer would have needed to undertake its own calculations.

Overall, then, the Treasury Board found itself in a difficult position in trying to implement total compensation comparability as the basic approach to negotiating salaries and benefits in the post-control era of the late 1970s and early 1980s. Treasury Board officials expressed scepticism that it would be feasible, given "practical considerations and the realities of collective bargaining and the dispute resolution process," to alter significantly current settlements to either raise or lower salaries for particular groups because of total compensation comparability findings:

This has been borne out over ten years of experience with collective bargaining and dispute resolution in the Public Service. There have never been settlements or arbitral awards that have not provided increases to rates of pay regardless of the comparability situation. With very rare exceptions, most increases have been the "going rate." Employer proposals to award increases in lump sum form, to avoid increases in rates believed to be excessive, have never been accepted either by unions or arbitration boards.[25] 

With the failure of the Government either to enlist the unions' cooperation in implementing the total compensation comparability approach, or to give itself the legislative tools to apply it directly, the state of affairs in the real world of salary determination described in the above quotation was to persist.

It must be noted, however, that even though total compensation comparability floundered in practice, Pay Research Bureau data on salary comparisons for specific occupations based on job-matching did play a prominent role during collective bargaining negotiations.

Fiscal restraint and the evolution of comparability

During the 10 to 15 years following the Anti-Inflation Board in the late 1970s, the struggle to restrain inflation continued, and the conviction emerged that government spending needed to shrink. In 1981, as in the earliest years of collective bargaining, 75 bargaining units chose arbitration, versus 47 using conciliation, but the groups choosing the conciliation option represented a larger number of public servants.

The 1982 Public Service Compensation Restraint Act limited raises in the federal public service to 6% and 5% respectively for the following two years.

Between 1991 and 1996, the Public Service Compensation Act and successive Budget Implementation Acts froze salaries for five of six years, and imposed a 3% raise in the other year. These legislative interventions were not driven principally by a concern to ensure comparability with private sector compensation but by broader macroeconomic goals.

Throughout this period, the Treasury Board continued to deepen its understanding of total compensation comparability and to refine its analytical techniques for applying the concept. In September 1992, Wyatt Consultants reported to the Treasury Board Secretariat on how to develop further the methodology for total compensation comparisons. Recommendations included the use of actuarial valuation for all compensation elements, comparing compensation over an employee's working career, refining further the model for calculating the value of job security (distinguishing, for example, between universal and differential risk aversion measurement), improving the degree to which the private sector comparators' populations reflected that of the public service, and extending the analysis to regional rates of pay.[26]

Attempting to apply total compensation comparability in the real world

During the 1980s and early 1990s, the Treasury Board Secretariat insisted on applying total compensation comparability concepts within the sphere of salary determination that it controlled. For example, over several years, a complex system of benchmarking military occupations to analogous public service groups was built. In practice, the results of this analysis have been applied inconsistently in setting the levels of salary increases over the years, leaving a sense of frustration and misunderstanding between the Canadian Forces and the Treasury Board Secretariat over the utility of calculating total compensation comparisons.

Senior executives and RCMP uniformed members represent two generally successful applications of the total compensation comparison approach. After a false start in the 1980s, total compensation comparisons with eight other large police forces across Canada began in 1993 and have since played an indispensable role in the determination of pay increases for the regular members of the Royal Canadian Mounted Police. Currently, Hay Associates provides total compensation comparisons for Executive, Deputy Minister and other Governor-in-Council positions. This research has provided the analytical foundation for compensation recommendations for these groups by the Advisory Committee on Senior Level Retention and Compensation, chaired in turn by Lawrence Strong and Carol Stephenson.

Not applying total compensation comparability in the collective bargaining context

Still, however, the impact of the total compensation comparability policy on the actual results of collective bargaining or arbitration boards was contextual at best. A 1987 Discussion Paper on Compensation Comparability observed as follows:

While the policy intended that the monetary disparity... between Public Service and external element values be utilized as the basis and justification for raising mandates for collective bargaining purposes, in actual practice, this is not the case. With the exception of salary comparability, total compensation values are shown internally but seldom used, if ever, in the raising of mandates, at the negotiating table, or in third party proceedings.

Many of the total compensation elements are not the subject of collective bargaining, since they are determined through legislation, at the National Joint Council or some other vehicle or forum.

There are certain problems inherent in the "valuation method" which was selected to achieve the stated compensation policy. The main disadvantages cited are its conceptual and methodological complexity and the possibility of attributing artificial values to compensation elements. Such sophisticated methodology, the number of assumptions made as well as the artificial values are not conducive to or supportive of the practicalities of collective bargaining.[27]

The apparent cognitive dissonance between the ever more sophisticated refinements of methodology and the simultaneous recognition of its limited utility in actual compensation determination begs the question of why work on the total compensation comparability approach persisted. In essence, it appears that the policy expressed what the Government wanted its policy to be, and officials accordingly continued trying to develop and apply it. But this intention was not complemented by changes to the formal mechanisms for setting compensation levels or by a determination to make the idea stick. For instance, the 1992 Wyatt Consultants study was doubly ironic, in that not only was it elaborating ideas no one could figure out how to apply, but by then the public service was into the second year of what turned out to be six years of wage controls, five of them freezes.

Pay equity and internal relativity

A second example of policy dissonance emerged during the period from the late 1970s through the 1990s in the concept of pay equity. The 1978 Canadian Human Rights Act (CHRA)required employers in the federal labour jurisdiction (including the public service itself) to ensure "equal pay for work of equal value." This introduced a policy disconnect between, on one hand, the Government's overall commitment to compensation based on comparability with a representative sample of private sector employers, and on the other hand, the priority, inherent in the principle of "equal pay for work of equal value," in favour of internal relativity between positions.

To a large degree, policy thinking on comparability and pay equity appears to have evolved along distinct tracks. The various documents on comparability devoted relatively little space to the requirements of the CHRA. The Treasury Board Staff Relations Branch Compensation Briefing Notes of March 1984, for example, utilized no more than 2 of 55 pages for the latter issue. The dichotomy between the two outlooks was acknowledged as follows:

The principle of market determination for Public Service compensation may be affected substantially by ... the Canadian Human Rights Act. It stipulates that, regardless of compensation relativities in the market place, jobs performed by men and women that are judged to be equivalent in value on the basis of specified job evaluation factors must be paid equivalent compensation.[28]

A more extensive Treasury Board Secretariat paper on Compensation Determination for Represented Employees: Future Directions dated December 1992 observed along similar lines:

In general terms ... internal relativities are the prime and only consideration in compensation determination under the CHRA. This is in marked contrast to the PSSRA, where the concept of internal relativity is only one among several factors to be considered in compensation determination.[29]

While, in practice, the government has dealt with pay equity complaints separately from collective bargaining, on some occasions, the two concepts came together. In 1998, a special pay adjustment (SPA) was included in collective agreements covering various female-dominated groups involved in the Public Service Alliance of Canada pay equity complaint. Although the Treasury Board Secretariat avoided describing the SPA payments as linked to pay equity, they were in fact included in the eventual settlement. In general, the two concepts of external comparability and internal equal pay for work of equal value have co-existed, rather than being reconciled.

Reduced capacity

During the early and mid-1990s the policy under Program Review was to shrink the public service and to hold the line on salaries and benefits as part of the Government's commitment to eliminating the federal deficit and regain control of the nation's finances. The Treasury Board Secretariat reduced sharply its capacity in the area of salary determination and collective bargaining.

The Government decided to abolish the Pay Research Bureau (PRB) as part of a broader program in the 1992 Budget to eliminate agencies. Over its 35-year history, the Bureau built up considerable expertise and gained credibility in many circles. However, the management side apparently considered that PRB samples were not sufficiently representative of Canadian employers, as they focused only on large, unionized establishments only.

Just as creating the PRB in 1957 was a step forward in strengthening the empirical base for determining comparability, the closing of the PRB ended the possibility of collecting detailed data on which to assess comparability. Perhaps the fact that the Government was acting at about that time to freeze wages made the PRB irrelevant in any case, at least in the short run.

Comparability as collective bargaining resumed

When collective bargaining resumed in 1997,[30] both the unions and the Treasury Board began with what amounted to pattern bargaining. If nothing else, this reflected the reduced negotiating capacity on both sides after years without collective bargaining. Later on, settlements became more differentiated between groups but external comparability was only one factor in shaping agreements.

Terminable allowances

The main example of comparability influencing salary outcomes was the introduction in 1997 of recruitment and retention allowances (which have generally been known as terminable allowances) for the Computer Systems (CS) classification group. Over the following few years such allowances were extended to at least a dozen other groups. The decision to establish them was based on data relating to compensation paid by competing employers in circumstances where there was evidence of difficulty for the federal government to recruit or retain sufficient staff in particular high-demand occupations.

The use of the terminable allowance approach to comparability was driven by two considerations. The first was to reduce the risk of increasing pay equity pressures by staying within the exceptions to equal pay for work of equal value provided for in the Equal Wages Guidelines, 1986, specifically 16 (h): "the existence of an internal labour shortage in a particular job classification." The second was to avoid making such additional salary increases permanent, since the labour market pressures giving rise to recruitment and retention problems could turn out to be transitory.

Joint studies

Another manifestation of external comparability were the various joint studies commissioned by the Treasury Board, the relevant bargaining agent, and often the main employing line department, to examine what other Canadian employers were paying for such groups as the Correctional Services (CX), Foreign Service (FS), Aircraft Operations (AO) and Operational Services (SV) classification groups. From the employer point of view such studies often tended to lack objectivity and balance, with the result that they were normally not accepted as a basis for negotiation.

Research capacity

At a broader level, joint employer-union work began in 2002 on rebuilding a capacity like that provided by the Pay Research Bureau between 1957 and 1992. Developmental work and pilot studies for selected jobs have been undertaken by Statistics Canada under the direction of the National Joint Council's Joint Compensation Advisory Committee. The Public Service Modernization Act, passed by Parliament in November 2003, provides for the establishment of a permanent compensation research capacity as a wing of the new Public Service Labour Relations Board.

Pay equity

1999 saw the conclusion of the longstanding pay equity complaint by the Public Service Alliance of Canada in regard to various female-dominated clerical, secretarial, and educational groups. While resolving an important dispute on the application of the Canadian Human Rights Act, the settlement undoubtedly moved federal compensation further away from overall comparability to the broader Canadian labour market.

Regional rates of pay

Another trend away from comparability has been what amounts to the systematic phasing out of regional rates of pay in the federal public service. This process began well before Program Review. In 1985 and 1986, after pressure on Members of Parliament led by the Public Service Alliance of Canada, the number of pay zones affecting the General Services (GS) and General Labour (GL) classification groups was reduced. Collective agreements in 1997 and 2001 reduced the number of zones for the GS/GL employees from 10 to 7 to 3. By now the zones are so large as to bear no meaningful relation to local labour markets.

Universal Classification System

On the other hand, in 2002 the Treasury Board decided not to pursue further the so-called "Universal Classification System" solution to reforming the classification standards for the core public service. A principal reason for this decision was a concern to avoid creating too rigid a framework for evaluating public service jobs, which would have made it virtually impossible to adjust to changes in external compensation that affected only some occupations.

Compensation framework

The Treasury Board has not formally adopted a compensation policy following the resumption of collective bargaining in 1997. On two occasions, in mid-1999 and again in mid-2003, draft policies were under active internal consideration. However, changes in key personnel, and the shifting labour relations climate have so far precluded finalizing such a policy. In both cases, the drafts were recognizably similar to the policies enunciated in earlier times, emphasizing the need to balance in particular external comparability and internal equity.

The 2003 draft policy framework highlights four overarching principles. Public Service compensation shall:

  • be competitive with that provided for similar work in relevant labour markets;
  • reflect the relative value to the employer of the work performed;
  • reward performance, where appropriate and practicable, based on individual or group contributions to business results; and
  • be affordable within the context of the Government's commitments to provide services to Canadians, its fiscal circumstances and the state of the Canadian economy.

The draft further notes that the federal government is responsible for general public policy as well as being Canada's largest employer. Accordingly, compensation goals will be conditioned by public policy considerations, including: economic policy objectives such as controlling inflation, relevant laws such as the Canadian Human Rights Act, social policy objectives such as the extension of parental leave under the Employment Insurance Act, and public expectations and pressures.

Appendix B[31] provides the 2003 draft policy as it was circulated for comment.

The Canada Customs and Revenue Agency (CCRA) Board of Management did adopt such a policy in early 2001. The policy includes these points:

Compensating our employees at a fair and reasonable level makes the CCRA an attractive employer.... To ensure a better working environment, and to sustain our high standards of service, the CCRA will respond to labour market realities.

The CCRA's Board of Management and Management Team are committed to a compensation policy that recognises and addresses the realities of competitive labour market forces. If studies demonstrate that a significant gap exists between the CCRA's compensation and competitive labour market's compensation, and it can be demonstrated that this gap affects our ability to attract and retain employees, the CCRA is committed to addressing the gap.

Overall during the period following Program Review, federal public service compensation continued to be shaped by an amalgam of forces. While comparability was certainly an important consideration, it could not be said to have been the dominant factor in shaping salaries and benefits.

The post-Program Review experience broadly aligns with what has occurred since the merit principle was adopted in the public service in 1918. Comparability with the external labour market has always figured prominently in statements by the Government of the day about its philosophy of compensation for public servants. But through the years a series of other factors have tended to override comparability as a determinant of salaries and benefits. These have varied in importance over time, ranging from the relentless pressure to respect internal relativities, to pragmatic considerations such as finding a saleable deal, to the statutory requirement for equal pay for work of equal value, to the desire to set an example as an enlightened employer, to the imperative to control inflation or reduce government spending.