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

• Previous Page
• Table of Contents
• Next Page

This page has been archived.

15. Possible Areas for Legislative Change

This chapter discusses in turn two sensitive areas governed by legislation: collective bargaining in the federal public service, and pay equity, that is equal pay for work of equal value. In both cases, the issues are complex and deserving of in-depth analysis in their own right. In fact, there have been substantial reports from special Advisory Committees on both topics in recent years: the Advisory Committee on Labour-Management Relations in the Federal Public Service, chaired by John Fryer, which reported in 2000 and 2001;^[220] and the Pay Equity Task Force, chaired by Beth Bilson, which reported in 2004.^[221] In the present Report we limit ourselves to describing certain issues specific to responsible compensation management that we consider deserve further examination, notwithstanding the extensive work of both the Fryer Advisory Committee and the Pay Equity Task Force.

Collective bargaining and compensation management

To begin, it is important to affirm our view that unions play a vital role in providing employees with a common voice on important workplace issues, and that collective bargaining is the best way to set the terms and conditions of employment for unionized workers. In this section, we challenge the view that collective bargaining should be a kind of trial of strength in the public sector. However, this critique is undertaken within the framework of a fundamental commitment both to the legitimacy of unions, and to their crucial role in representing employees, especially in the process of setting the terms of public sector compensation.

There will no doubt always be controversy about the relationship between collective bargaining and compensation outcomes. Actual results in the federal public service since 1967 have on balance been positive, despite many bumps along the road. Looking to the future, however, we may reasonably ask whether similar or better results could be achieved with a system better adapted to serving the public without interruption.

An historical overview

The federal public service experience with collective bargaining over the nearly four decades since it was introduced in 1967 can best be characterized as mixed. Chapter 3 describes how periods of collective bargaining have alternated with periods of legislative wage determination. There were in fact three periods of direct wage setting by the Government:

• 1975–1978: The Anti-Inflation Act regulated wage increases for most employers in Canada, including the federal public service.
• 1983–1984: The Public Service Compensation Restraint Act applied specifically to the federal public sector, limiting salary increases to 6% and 5% in 1983 and 1984 respectively.
• 1991–1997: The Public Sector Compensation Act, and successive Budget Implementation Acts, froze federal public service salaries for five of six years, mandating a 3% increase in 1992.

For the first 30 years of collective bargaining, then, the system was only allowed to operate for 19 years in total, with uninterrupted operation for a maximum of eight years, in the period immediately following the implementation of the new regime in 1967. The present era, beginning in 1997 with the resumption of collective bargaining following the end of the freezes, is therefore already as long as the longest previous stretch of continuous collective bargaining. It must be acknowledged, however, that even the periods during which collective bargaining prevailed included cases of specific recourse to legislation to end or even prevent a particular strike.

Much public discussion of federal public service collective bargaining focuses on what is described as bad relations or mutual distrust between the Treasury Board Secretariat, which bargains in relation to most federal public servants, and the public sector unions. The implication seems to be that if somehow the parties worked harder at getting along with each other, or if the employer side were less prone to resorting to arbitrary controls, everything would be fine.

In fact, relations between Treasury Board Secretariat management and human resources professionals and their union counterparts have nearly always been civil and constructive. We note, however, that relatively frequent turnover in key players, especially on the management side, has likely interfered with nurturing sound relations of personal trust between the parties.

In any case, the instability in the collective bargaining system since 1967 has deeper roots, relating mainly to disputes over the appropriateness of public service wage and benefit increases in the context of broader economic and political events and, we would argue, the unsuitability of the available tools for resolving disputes under the applicable legislation.^[222]

Critiques of union-management relations in the federal public service often include the assertion––generally unsupported by empirical evidence––that public servants have suffered financially in comparison with private sector employees in Canada. This Report shows that while there was some truth to such a view during the period of pay freezes in the early and mid 1990s, this is now no longer the case.

Chapter 2 of this volume reports that in the period following the end of salary freezes in 1997, the federal public sector––specifically the combined core public service for which the Treasury Board is the employer and the separate employers––experienced a 14.1% increase in real average salary in the five years from 1997–98 to 2002–03. In fact there was a reduction in the real average salary (constant 2003 dollars) of about 3.7%^[223] in the mid-1990s. However, this reduction was more than reversed by 1999–2000. Chapter 5 (refer to Figure 1027, in particular) shows that the cumulative rate of increase in the federal public sector^[224] average salary indeed lagged private sector negotiated year-over-year increases between 1990–91 and 1999–2000. However, cumulative federal public sector increases since 1990–91 outstripped private sector agreements by a total of more than 8% to 2002–03.^[225]

At the same time that these wage results were achieved, federal public servants gained improved pension benefits while paying a reduced share of the cost of current service contributions. For example, Public Service Superannuation Act amendments in 1999 reduced the salary-averaging period for calculating benefits from six to five years. Other benefits remained essentially unchanged, except for the introduction of a Pensioners' Dental Service Plan in 2001, although their costs increased by about half between 1997–98 and 2002–03, nearly entirely at the employer's expense.

Overall then, it must be concluded that federal public service employees have done well relative to the Canadian private sector since the introduction of collective bargaining, even if there have been periods of arbitrary restraints, especially in the early and mid 1990s. 

An inappropriate model?

There are, in principle, only three ways to determine compensation: one party decides unilaterally; the parties agree; or a third party decides. Historically, the federal government determined public service compensation on its own, with input from employee associations, until 1967.^[226] In effect, the various interruptions in the collective bargaining system, affecting 11 of the years since 1967, were returns to the older unilateral approach. Canadian public culture would almost certainly not countenance a definitive return to such a policy. Moreover, in a world where employers need to engage the loyalty of their employees in order to promote effectiveness in the workplace, it would be counterproductive to seek to exclude employee representatives from determining the terms of compensation for the federal public service.

The option of agreeing on compensation can take individual or collective forms. As we argue earlier in this section, the need to recruit employees with exceptional skills in a world where knowledge is critical to success calls for increased flexibility in setting individual compensation packages. In Canada, however, the right to bargain salaries and benefits collectively is well entrenched, especially in large public sector workplaces. For the vast bulk of the public service, compensation will no doubt continue to be determined through collective bargaining with certified unions.

The success of collective bargaining depends on how disputes are resolved. Ideally, the parties will come to mutually agreeable terms through direct negotiation. However, the way an employer or a union bargains is shaped by their expectation of what will happen if they cannot agree. The Public Service Labour Relations Act gives unions a choice of two routes: conciliation followed by the possibility of a strike, or third-party arbitration. For now, we will focus only on the conciliation/strike option, which is in fact the route regularly chosen by the largest unions.

It has been argued that the federal government enjoys an inappropriate predominance in collective bargaining with its own employees. The criticism is that if the Government does not like how the process unfolds, it has the option of overriding collective bargaining by seeking Parliament's intervention to legislate an outcome unilaterally. In an ultimate sense this is undeniable, and there have certainly been cases where Parliament has imposed the Government's will, either on the whole public service, or in relation to a particular bargaining group.

Most of the time, however, the Government is reluctant to resort to legislation. Presenting a Bill in Parliament to resolve a public service labour dispute is normally unpopular, unless there is a clear indication that the public is somehow at risk, or that striking employees are behaving irresponsibly in a way that puts the public at risk.^[227] In more ordinary circumstances, the union may be able to derive significant bargaining power from the sensitivity of the services provided by their members. In order to avoid or minimize public inconvenience, the Government may find itself pressured into conceding salary increases or other benefits that are not justified by external labour market comparisons. Often buying peace with a generous settlement is judged the best short-term solution to a public service labour dispute.

In theory, workers who provide services essential to the public are identified and prohibited from participating in strike action. The new Public Service Labour Relations Act maintains the employer's exclusive right to determine the level at which an essential service is to be provided to the public. The employer must then negotiate with the union what number of positions is necessary for the employer to deliver an essential service. If the parties cannot agree, either may apply to the Public Service Labour Relations Board (PSLRB) to determine any unresolved matter. The resulting essential services agreement, whether determined through negotiation, or by a decision of the PSLRB, remains in effect until amended by agreement or by adjudication.

In practice, however, this protection of the government's ability to meet public needs during a public service labour conflict can be compromised in either or both of two ways. In the first case, departmental managers may fail through negotiation or in persuading the PSLRB to identify a sufficient number of employees to deliver essential services to the public. They may also fail to amend the agreement on a timely basis in the lead-up to a strike. This is of course an employer responsibility; it is not the union's fault if the designations are inadequate or untimely. Whatever the reason, when this occurs, the capacity of the employer to withstand a withdrawal of services on which the public depends is limited, especially as problems accumulate over time. This was the case, for example, in relation to the Aircraft Operations (AO) group in the labour dispute of 2002–03. As time passed, with successive targeted strike actions, the backlog of aircraft and pilot certifications came to threaten the ability of the Canadian airline industry to operate.

The second problem with designations is more corrosive of the very idea of essential services. This is the increasingly frequent refusal by union picket lines to allow those designated essential workers, or indeed even excluded employees such as Executives or Personnel Officers, to enter the workplace without waiting various arbitrary lengths of time. During rotating strikes in Ottawa in 2004, for example, pickets declared quite formally that excluded or designated employees, or even in some cases members of the public, would have to wait an hour (as determined by the picketers) before entering the workplace. At some locations this enforced delay was apparently extended to as long as four hours. There is nothing in labour law that authorizes this type of union regulation of workplace entry. In effect, this practice is a form of intimidation that risks getting out of hand. For the first time, the new Act (in Section 199) prohibits interfering with employees whose position is identified as essential when they are entering or leaving their place of work. Section 200 provides for a penalty of up to $1,000 for contravening section 199. The section is now expected to be enforceable in Federal Court, with orders applicable across Canada, if appropriate.

It is too early to know how these changes in the labour law governing federal public service collective bargaining will work in practice. However, there is reason to hope that the new provisions will assist in protecting the Government from having to choose between acceding to union demands in the face of public concern about essential services, and resorting to the blunt instrument of a legislative resolution of the dispute.

Whatever the impact of the PSLRA, however, the more fundamental issue will remain that the use of strikes to resolve public sector labour disputes is of doubtful public policy merit. As Roy Adams remarked, strikes and lockouts are akin to "the medieval method of dispute resolution known as trial by combat."^[228]

Underlying labour legislation generally is the idea that labour disputes are best resolved through the interplay of economic power or, when necessary, political power. The power struggle perspective was well expressed by Barbara Wootton, a Labour member of the United Kingdom House of Lords, as follows: "[It is] the business of a union to be anti-social; the members would have a just grievance if their officials and committees ceased to put sectional interests first."^[229]

This concept of trial by combat may suit the private sector. Excessive compensation outcomes can threaten the size or even the survival of a private company, necessarily placing a constraint on union demands. In the public sector there is no such economic brake on compensation outcomes, at least not in the short term, especially for relatively small groups of employees. One may ask whether it is reasonable to pay public employees relatively high compensation simply because they may be able to translate their privileged role in providing essential services to the public into a means to force generous collective agreements.

Yvon Tarte, Chairperson of the Public Service Staff Relations Board, commented in a speech to the 1999 National Joint Council Seminar, as follows:

The strike method of dispute resolution in the public sector does not have the same foundation as it does in the private sector. In 1967, Jake Finkelman, the father of labour relations in the federal public service, thought long and hard and hesitated before being convinced to accept and recommend conciliation/strike as a method of dispute resolution for the federal Public Service. Several years later he expressed the view that given the opportunity he would not go down that road again.^[230]

It is inherent in the power struggle model of collective bargaining that principles such as comparability are at best reference points in negotiation. As we sketched in Chapter 2, this was the experience of the federal public service as well, with market comparability appealed to or ignored, depending on what was convenient in the collective bargaining struggle.

The recourse to economic power invites a response by political power. Thus it is not really so difficult to understand the relatively frequent reliance by the federal Government over the past 38 years on unilateral wage determination through general wage controls or freezes, or back-to-work legislation for specific groups.

Both economic and political power positions fail to address a critical aspect of collective bargaining in the federal public service: the relationship of salary increases to productivity. In the private sector, bargaining is as much about work rules as it is about salaries. This is so for the simple reason that an enterprise's capacity to pay compensation increases greater than the rate of inflation generally depends on achieving higher productivity. Such productivity improvements occur as a result of the use of the best available equipment, and of the most efficient work methods.^[231]

In the federal public service there are three broad reasons that there is seldom a direct link between collective bargaining and productivity improvement. First, in much of the public service measuring productivity meaningfully and reliably is inherently difficult. Second, even where managing productivity is clearly feasible, for example, in relation to large volume processes such as issuing cheques, the bargaining relationship is one where either economic or political power prevails. Normally an outcome emerges without the need to relate the result to wage comparability with the private sector, or to productivity levels or changes. Third, federal public service bargaining normally occurs at a very aggregate level at which it is difficult, if not impossible to surface, let alone address meaningfully workplace issues that are critical to the effectiveness of an individual department or agency.^[232]

In summary, then, collective bargaining in the federal public service based mainly on the conciliation/strike approach to dispute resolution sort of works, but it is neither reliably stable, nor designed really to serve the public interest. Thus the question arises: is there an alternative way to settle labour disputes that we could reasonably expect to be capable of delivering better results both for employees and for the Canadian taxpayer?

Is there a better model?

The dilemma we have described above in relation to fair compensation determination in the federal public sector is of course largely shared across the broader Canadian public sector. If there were a model that was unquestionably superior to our present mix of industrial warfare and political unilateralism, surely it would be spreading even now through Canadian labour jurisdictions. But no such emerging consensus is evident.

Conceptually, any alternative must involve third-party resolution of labour disputes in which the government and the union cannot agree on an outcome through collective bargaining. Numerous variations on this theme have been tried over recent decades. John O'Grady offers an excellent survey of the subject in his paper Arbitration and Its Ills.^[233] Among the most pertinent points emphasized in this paper are the following:

• Evidence suggests that replacing the right to strike with compulsory arbitration results in a decline in the proportion of contract disputes that are settled through direct bargaining between the parties.
• The weight of evidence supports the view that arbitration has an upward bias in the public sector as a whole, though this is not a unanimous view.^[234]
• Studies suggest that there is significant variance in the weight that arbitrators attach to different criteria.
• Studies consistently find that productivity, ability-to-pay, and labour market disequilibria play little role in shaping arbitral decisions. The weight attached to comparability is so great that it tends to marginalize other criteria.
• Directing arbitrators to consider total compensation appears to have comparatively little impact on arbitral practice.
• Arbitration should not be viewed as simply a substitute for the right to strike. Replacing the right to strike with compulsory arbitration fundamentally alters the process of wage determination in the public sector.^[235]

Federal public service experience with the arbitration route, since this option was restored in 2002, has generally followed patterns established through direct settlements. Economic increases awarded have been at the going rate; restructures through adding pay increments were approved in five of seven awards to June 2005, somewhat more frequently than in direct negotiation (11 of 17 settlements included restructures).

The amended Public Service Labour Relations Act, which came into force in April 2005, added an additional point to the existing factors to be considered in making an arbitral award, namely "the state of the Canadian economy and the Government of Canada's fiscal circumstances." There is no guarantee how arbitrators will choose to interpret this point.^[236] In fact, with several years of surpluses in a row, it might be argued that the federal government is well placed to pay whatever compensation increases an arbitrator may feel is warranted.

In short, arbitration is unpredictable in its results, to the point sometimes of appearing to be arbitrary. Thus simply throwing federal public service labour disputes into the hands of what might be called traditional arbitration is unlikely to please anyone, including the Government of the day. So where should we go for the future?

One logical alternative would be to embrace the power struggle approach more fully. If we are to decide compensation through a trial of strength, then it would be sensible not to tie the hand of the employer to the extent now in place. In particular, in addition to the picket line tactics described earlier, federal public service unions have tended more and more to adopt the tactic of brief rotating strikes to disrupt the workplace, without incurring significant costs to either the striking employees in the form of lost wages or to the union in the form of strike pay. With the intimate knowledge of government operations possessed by union members, it is not difficult to plan these strike activities to inflict maximum disruption with minimum pain to the union or its members.

There are at least two responses that could help to even the balance. The first response is administrative. Public service managers, according to anecdotal evidence, are often not rigorous in documenting strikers' absence from the workplace.^[237] As well, the pay system is such that any lost wages affect pay cheques long after the strike action has passed. Both these tendencies reduce even further any pressure on striking employees to resolve their dispute. Systematic management action to record employee absences, and adjustments to the pay system to implement pay reductions immediately, would move rotating strikes from the realm of street theatre to that of incurring tangible costs as a result of the decision to strike, whatever the duration.

The second response would require a legislative change: giving the employing department the right to lock out employees who go out on a rotating strike, for any period up to the conclusion of a collective agreement. It has long been the general consensus that the government should not lock out its workers, since that would amount to the government itself creating a disruption in service to the public. However, with the increasingly sophisticated use of rotating strikes, unions can undermine operations to almost the extent of a full strike, but at little cost to employees or the union's strike fund. So a limited power to lock out those whose hit and run tactics approach the impact of a full strike could encourage a resolution of the matters remaining in dispute. In real life, the government will always, and properly, resist the idea of locking out employees serving the public. But fear that a lockout might result from limited strike action would act as a restraint on what is now low-risk, low-cost, targeted strike action by the unions.

Overall, however logical the option of embracing the conciliation-strike model more fully, it is unlikely to improve either labour-management relations in the federal public service or the appropriateness of compensation outcomes. So the better course almost certainly lies in seeking variations on the use of third-parties to resolve labour disputes that resist settlement through direct bargaining.

The most interesting proposal along these lines in recent years was advanced by the Advisory Committee on Labour Management Relations in the federal Public Service, chaired by John Fryer, which reported in June 2001. The Committee considered that the choice of procedures between a conciliation/strike route and an arbitration route tends to hinder the process of voluntary settlement. Particularly where the union selects the arbitration route, there is little need to make hard choices since the arbitrator will make the final decision in any case.

Instead, the Committee recommended the establishment of a Public Interest Dispute Resolution Commission (PIDRC), modelled on the Public Interest Disputes Commission suggested in 1968 by the Woods Task Force for use in the larger federal labour jurisdiction. The PIDRC would report to Parliament, and comprise members with union, management and neutral backgrounds.

The PIDRC would assist the parties in resolving bargaining impasses by working through a wide variety of established dispute-settlement techniques.^[238] It was suggested that such flexibility on the part of the Commission would keep the parties in a state of uncertainty about the remedy that the Commission could impose in the case of a deadlock, which would in turn encourage the parties to settle on their own and avoid the imposition of a less desirable solution.

The new Public Service Labour Relations Act includes the concept of Public Interest Commissions (PIC). However, such Commissions are much more limited in concept than the Public Interest Dispute Resolution Commission advocated in the Fryer Report. In essence, the PIC is a relabelled Conciliation Board as it was known in the previous Public Service Staff Relations Act. There is some possibility that in establishing an agreed list of eligible PIC Chairpersons, care may be taken to include exceptionally accomplished candidates of broad public credibility. If such individuals are in fact selected to head some early PICs, perhaps a new standard will be set in the persuasiveness of their recommendations, both with the parties and with the general public.^[239]

In this regard, willingness to pay competitive per diem fees, which is permitted under the PSLRA, would assist in recruiting some of the most accomplished neutrals experienced in resolving labour disputes, or distinguished public figures whose sound judgement is widely recognized. In the past, the policy of limiting such fees to relatively small amounts was no doubt intended to save money. In practice, this was almost certainly a false economy.

In considering whether there is a realistic and sustainable alternative to the conciliation/strike model in resolving labour disputes in the federal public sector, it is sobering to reflect on John O'Grady's paper, "Arbitration and Its Ills," and its overview of arbitration practice in Canada through to the early 1990s. The main message seems to be that arbitrators as a profession have their own ideas of what their role is: "the predominant view among arbitrators [is] that arbitration's purpose is to replicate a bargained outcome, not an outcome that proceeded from a unilateral determination by an employer."^[240]

In fact, any permanent alternative third-party process would need to be designed to apply fairly and consistently a philosophy of compensation rooted in comparability. Most observers agree that the fairest way to compensate public employees is to adhere to the principle of comparability with what is paid in the private sector for similar work. As Fogel and Lewin remarked:

The output of government does not pass through the marketplace where its relative worth can be assessed by customers. In the absence of a product market discipline imposed on pay practices ...what could be [fairer] than to pay government employees what their private industry counterparts are getting? To attract employees of at least average quality to the government, the pay offered must be comparable to that available in the private sector. For the government to pay more than the private sector, however, would be unnecessary and would waste government revenues.^[241]

A third-party dispute resolution system that did not permit strikes in the federal public service would need to meet several conditions, including these:

1. Reliable information on comparable compensation in the broader Canadian labour force would need to be available. As we noted earlier, it is difficult to design and implement surveys that are convincing to both employer and union officials. Nevertheless, we have a chance with the compensation analysis and research services that are to be established by the new Public Service Labour Relations Board, to build such a capacity. Success will depend on timely senior level attention to the careful design of the survey criteria and methods.
2. The Public Service Labour Relations Act would have to specify the compensation policy to be applied. Such a policy would not be formulated as considerations to take into account, but as criteria to be implemented. We outline below a concept for such criteria.
3. The Adjudicators^[242] mandated to resolve disputes would need to serve for a fixed term so they could maintain appropriate independence. Adjudicators would need to be personally credible with both the government employer and the public service unions. To accomplish this, appointments should be from among candidates acceptable to both the government and the unions. By serving for a non-renewable term of five years, for example, Adjudicators would be able to act with reasonable independence.
4. Adjudicators would be required to give a detailed written rationale for their decision. The only way to demonstrate that the statutory criteria for resolving disputes are applied in practice is for decisions to be transparent in their reasoning, and in the interpretation of evidence.
5. Decisions should be subject to appeal to a panel of Appeal Adjudicators on the grounds of failing to apply the statutory criteria. A third-party system with statutory criteria implies a need to be accountable for applying the criteria responsibly. These are not issues that should be referred to the court system, since judges normally lack the necessary specialized expertise.^[243] The Appeal Adjudicators would be a group of the most experienced and credible of the Adjudicators. To prevent undue delays in determining compensation outcomes, there would need to be statutory time limits for filing and ruling on any appeals.

On the matter of criteria, it would be logical to set out a framework that would promote comparability in a context of overall compensation, not just salaries. Such criteria might look something like this:

• The presumption would be that compensation is comparable or set to respect the principle of equal pay for work of equal value. So in the absence of evidence to the contrary, salary increases would track a general measure of changes in Canadian private sector compensation levels. Such a measure could be the Industrial Aggregate used to adjust judicial and parliamentary salaries each year.^[244]
• The Adjudicator could vary the annual increases up or down based on compelling evidence that the salaries for a group of employees were not comparable with appropriate private sector comparators. The evidence for this adjustment would need to be documented and would be subject to review as described above.
• The Adjudicator would take into account arguments relating to the principle of equal pay for work of equal value in determining any such adjustments. In this chapter, we set out some complementary ideas for modernizing the regime for applying pay equity in the federal public service.
• Normally, the non-salary aspects of compensation would be reserved for agreement between the government and the union. Subsequent to a determination of a salary increase by the Adjudicator, the parties could agree to direct some of that increase toward the benefits allowance proposed in Recommendation 15.2.

A critical question in establishing a third-party dispute resolution process of any type is the impact it will have on direct collective bargaining between the parties. The evidence adduced by O'Grady's paper and other literature on adjudication suggests that such a regime makes it harder to achieve an agreement through direct bargaining. Key to maintaining a fair chance of settling compensation terms and conditions directly is a reasonable uncertainty on the part of both parties about what will emerge from the third-party process.

A likely criticism of the approach outlined in this section is that it could reduce collective bargaining to the application of formulas. There is some truth to this view. However, if one accepts the primacy of the concept of comparability as the principal criterion for federal public sector compensation, then determining salary or other compensation levels does become largely a matter of defining and applying suitable methods for assessing comparability with the private sector.

However, making such determinations about comparability will never be without controversy. For many groups, there are no direct comparators in the Canadian private sector, so comparability must be established indirectly through defining internal comparability within the public service. Thus both parties must remain uncertain how an Adjudicator will in fact rule.

To augment the uncertainties that could be expected to promote settlements directly between the government and one of its unions, we could add elements of the dispute resolution toolkit approach advocated in the 2001 Fryer Report, for the Public Interest Dispute Resolution Commission (PIDRC). These could include fact-finding, mediation, and issuance of a report on recommended terms of settlement. The option of imposing a settlement should be reserved for the case where the negotiations are declared at an impasse and an Adjudicator would be assigned to determine the terms of the collective agreement according to the criteria set out in statute.

As one further possible element to encourage settlements through collective bargaining, consideration could be given to authorizing the Adjudicator to set a longer duration to the collective agreement than the parties had been negotiating.

The ideas set out in this section are not a formal proposal for reform. At most they are intended to outline elements for consideration in devising a compensation determination regime that excludes recourse to strikes. The recently adopted Public Service Labour Relations Act (PSLRA), was at most a modest revision of the previous Public Service Staff Relations Act (PSSRA). The broader Public Service Modernization Act (PSMA), of which the new PSLRA was a component, was aimed primarily at modernizing staffing and related recourse in the federal public service. So the task of rethinking the statutory framework for collective bargaining in the public service remains to be undertaken.

Any such rethinking will of course need the active engagement of the public service unions, federal employers, academic experts, and interested non-governmental organizations. At first, unions can be expected to be suspicious of such a process, fearing that it could threaten their members' interests as well as the institutional interests of the unions themselves. There will certainly be a vocal group that will denounce any attempt to put aside the strike weapon, on both ideological and practical grounds. On reflection, however, many unions may adopt a more nuanced position. Collecting fees to mount strikes and finance strike pay is a major challenge to union leaders. Moreover, although recent strikes have allowed for some blowing off of steam by disgruntled employees, and provided an opportunity to build solidarity, the impression of most observers would likely be that public service employees generally feel very uncomfortable leaving their work serving the public to emulate briefly the tactics of their private sector union brethren.

In 2010, there will be a parliamentary review of the PSLRA. That could be a reasonable time to look at possible changes to the public service collective bargaining regime. Over 40 years after the introduction of a collective bargaining regime adapted from the prevailing private sector model of trial of strength through strikes, it would not be too early to consider introducing a more suitable model. Such an approach would need to be better suited to balancing the public interest in employee participation in setting their terms of compensation, with the broader public interest of delivering services reliably to citizens and of protecting the interest of taxpayers to pay enough but not too much for public services. The intervening three or four years could be used to encourage a wide debate on possible alternative models for optimizing such a balance, and to gain experience with the modest changes in fact contained in the new Public Service Labour Relations Act.  

Our recommendations, therefore, in regard to the statutory framework for collective bargaining in the federal public service are these:

This concludes our reflections on possible changes to the statutory framework for collective bargaining in the federal public service. We now turn to our last substantive topic in this volume: proposals for improving the clarity and reasonableness of the means for applying the requirement to provide equal pay for work of equal value in the federal public service. As with the area of collective bargaining above, in this Report we cannot realistically undertake a full critique and prescription for reform. Nevertheless, a brief survey of the issues and a sketch of a possible way forward seem to us an indispensable component of a comprehensive assessment of federal public sector compensation.

Applying pay equity in the federal public service

Pay equity is the shorthand term normally used to express the principle of equal pay for work of equal value. As early as 1951, the International Labour Organization adopted a Convention (Number 100) on this subject: Convention Concerning Equal Remuneration for Men and Women Workers for Work of Equal Value. Canada ratified this convention in November 1972. The Canadian Human Rights Act (CHRA), which was passed in 1977, gave effect to this convention in the federally regulated labour jurisdiction.^[245] Specifically, section 11 of the CHRA states that it is discriminatory to establish or maintain different wages for men and women doing work of equal value in the same establishment. The Equal Wage Guidelines issued by the Canadian Human Rights Commission in 1986 provide more detail on such topics as the definition of "establishment" and permissible exceptions such as performance pay, labour shortage supplements, and regional rates of pay.

Chapter 4 of this volume includes an overview of the nature and financial impact of settlements in the federal public service of pay equity complaints under the Canadian Human Rights Act. Appendix H provides a summary of specific settlements in the core public service up to 2003. Appendix I presents the best available estimate of the costs of pay equity settlements from 1980 to 2003. About $3.4 billion was paid out in lump sum amounts (including over $900 million in interest) over this period. The cumulative total of resulting pay increases for affected employees was estimated at over $1.8 billion to March 2003, and the ongoing portion of the core public service salary mass attributable to implementing pay equity was over $200 million per year in 2002–03.

This level of annual costs amounts to between 2% and 2.5% of the relevant public service salary mass. The Pay Equity Task Force Report, citing studies by SPR Associates (1991), Canadian Facts (1992 and 1993), and the Institute for Social Research (1994) found that "the costs of pay equity to employers in the Ontario labour jurisdiction ranged from 0.5% to 2.2% of payroll for public sector organizations."^[246] It is worth noting that for private sector employers with 500 or more employees, the total payroll adjustment was reported as only 0.6%. For smaller employers, the adjustments averaged 1.4% or less. Results in the Quebec labour jurisdiction were reported in the Pay Equity Task Force Report as being a "relatively moderate percentage of the payroll, usually under 2.5%."^[247]

It is indispensable that the federal public service come to terms constructively with the Canadian Human Rights Act obligation to ensure equal pay for work of equal value. Existing uncertainties of interpretation cast a shadow on any systematic management of compensation. Until now, acrimonious litigation and the fear of litigation, as well as an unsuccessful search for a panacea through a universal classification system have distracted attention from adopting a pragmatic, fact-based strategy for addressing pay equity. Now is the time to take up this challenge as part of a broader reform of federal public service compensation management. To begin, however, we must probe some of the fundamental but not always explicitly acknowledged issues that lie at the heart of the pay equity debate.

Fundamental issues

By now, the concept of equal pay for work of equal value is well entrenched as a right protected by legislation in nearly all Canadian jurisdictions. Many commentators consider equal pay for work of equal value to be a fundamental right with a quasi-constitutional status, as an essential element of human rights prohibitions against discrimination on the basis of gender. At a common sense level, the concept is compelling, with a ready appeal to basic fairness. This is particularly true in light of the well-established fact that on average women earn considerably less than men in the Canadian workplace.

According to Statistics Canada, the 2001 Census indicated that women earners in Canada aged 15 and over made 64 cents for every $1 earned by their male counterparts. This proportion represents an improvement from 52 cents in 1980. Comparing only full-year, full-time employees, female workers earned 70 cents for every dollar earned by male workers. For young university-educated women working full time, full year, the female pay level was 81 cents for every dollar earned by their male counterparts. Within the ten most common occupations chosen by men, university-educated women aged 25 to 29 earned 89 cents compared with each dollar earned by male workers.^[248]

A 2002 Statistics Canada study, The "Who, What, When and Where of Gender Pay Differentials" addresses directly the issue of whether there are reasonable explanations for the gender pay gap other than discrimination.^[249] Whereas previous studies of this type focused on differences in the characteristics of individual workers, the data collected through the 1999 Workplace and Employee Survey (WES) allowed an assessment as well of the contribution of "workplace characteristics such as high performance workplace practices, foreign ownership, non-profit organizations, training expenditures, desirable employment contracts, and the workplace part-time rate." The author, Marie Drolet, concluded that:

Unlike other studies that estimate the explained component [i.e. that portion not attributable to gender discrimination] to be about 50% of the gap, the inclusion of workplace characteristics–in particular more accurate industry measures–increase the explained component to 61% of the gap ... Yet despite the inclusion of the new WES variables, a significant portion (38.8%) of gender pay differentials remains unexplained.^[250]

This study suggests that some element of discrimination does affect women's earnings in the Canadian workplace. 

However, the issue of how to interpret the deceptively simple phrase equal pay for work of equal value, and to apply it in practice as a guide to reducing discrimination raises difficult and complex philosophical and analytical challenges. Among these fundamental issues are the following:

• How do we measure value in comparing different types of work?
• To what extent are wage differences between male-dominated and female-dominated groups the result of discrimination?
• How does pay equity relate to collective bargaining?
• What is the appropriate scope of comparison for determining equal value?
• What is an appropriate group for undertaking comparisons?

How do we measure value in comparing different types of work?

Sub-section 11(2) of the 1977 Canadian Human Rights Act (CHRA) states:

In assessing the value of work performed by employees employed in the same establishment, the criterion to be applied is the composite of the skill, effort and responsibility required in the performance of the work and the conditions under which the work is performed.

So far, no firm method has been established for assessing equal value in the federal public service. The need to develop such a method is underlined by the approach applied by the courts during lengthy litigation on the main public service pay equity case, that begun by the Public Service Alliance of Canada (PSAC) in 1984 and concluded in 1999. The settlement in the PSAC case was based in effect, in the absence of better information or methodology, on a doubtful sample of positions and a comparison of multiple male-dominated groups to particular levels of particular female-dominated groups. Male-dominated groups that had even one position evaluated in the point range of that female-dominated group and level were included in calculating the equal value gap. However pragmatic it may have been for the courts to accept such an approach to settle a long-standing case, it is hard to argue reasonably that such a method in fact compares work of equal value.

The inherent difficulty of designing such a method is profound. For example, the language of subsection 11(2) already feels disconnected from the reality of modern Canadian workplaces. Increasingly, the most critical element of value to any employer is an employee's knowledge. Yet the term "knowledge" is not even explicitly listed among the statutory criteria. Skill is a related but hardly equivalent concept.^[251]

In any case, vast efforts were expended in the federal public service over more than two decades to develop a Universal Classification System (UCS) that was intended to evaluate all the tens of thousands of jobs in the public service according to a single standard of value without gender bias.^[252]

The premise of such an approach is that one yardstick can meaningfully measure value across diverse workplaces and types of work within the federal government. An implication is that once established, such a yardstick will remain valid over time, or at least change only slowly. Both propositions are doubtful. For example, knowledge––or skill, in the terminology of the CHRA––has for some time been increasing in importance as a critical component of public service work, and would logically then command a growing share in assessing the overall value of a job.

In fact, the Treasury Board decided in 2002 to discontinue the pursuit of such an overarching standard of value in classifying jobs in the public service. This decision was based primarily on the conclusion that such a system would be too inflexible for such a diverse workforce. To continue with the knowledge theme, an important problem with the provisional UCS evaluations of positions across the public service was that knowledge-intensive jobs had trouble achieving what managers judged to be adequate recognition through the proposed evaluation system. In any case, it was decided to modernize classification standards group by group to ensure that each deals equitably with both men and women. Until now, however, only standards for the Foreign Service group and the Border Services group have been completed and approved for implementation.

To what extent are wage differences between male-dominated and female-dominated groups the result of discrimination?

Obviously this issue can engender fierce debate. The 2002 Statistics Canada study by Marie Drolet suggests that over 60% of the observable differences between male and female wages in the Canadian labour market can be explained by factors other than discrimination. How this analysis can be applied in comparing particular occupational groups in a given establishment is far from clear, however.

Section 16 of the Equal Wages Guidelines, 1986 does set out reasonable factors that could justify differences in wages between male and female employees performing work of equal value in an establishment. These include: different performance ratings, seniority, a re-evaluation and downgrading of an employee's position, a rehabilitation assignment during recuperation following an injury or illness, demotion, temporary training, an internal labour shortage, red circling, and regional rates of pay. These exceptions make sense, but they at best indirectly acknowledge the main reasons for pay differentials across occupational groups such as differences or changes in the external labour market or differences in the bargaining power of particular groups of employees. They take no account of factors such as education, work force attachment, degree of part-time work or similar factors that figure in the type of analysis offered by Statistics Canada explaining male/female wage differentials.

How does pay equity relate to collective bargaining?

The employer obligation to ensure equal pay for work of equal value was superimposed on a wage determination system based on collective bargaining between the public service employer––normally the Treasury Board––and employee representatives organized in unions. The CHRA places no requirement on unions to include pay equity in its bargaining considerations, nor are unions responsible for achieving agreements that provide for pay equity.

In general, collective bargaining is driven by pay comparisons with other groups, both inside and outside the workplace, and by bargaining strength. While internal group comparisons might favour pay equity, comparisons to the external labour market and the strength of a union's position in bargaining are irrelevant to the factors driving equal pay for work of equal value within a given establishment such as the federal public service.^[253]

A plausible scenario would be for a union to push for salary increases for male-dominated groups, and place greater emphasis on non-cash benefits in negotiations relating to female-dominated groups.^[254] Later, the union could try to make up any salary lag affecting its female-dominated groups through whatever pay equity process is available. Such a strategy would drive up employer labour costs in the short term; in the longer term it would risk undermining employer trust in collective bargaining as a viable wage determination system.

To the extent that a universal classification system might be adopted by an employer, there would be pressure to consolidate bargaining units and unions. Once the salary level for a given set of jobs was set through one negotiation, likely with the largest union in the establishment, the employer could only negotiate different results with another union for positions of equal value at the risk of incurring a pay equity liability.^[255]

Such pressure to combine unions could be argued to infringe on the right of free association, which is at the heart of union legitimacy. Although the right of free association does not enjoy the same degree of protection in Canadian law as the right of equal pay for work of equal value, in the context of International Labour Organization (ILO) Conventions, both principles are seen as core labour rights. Presumably employees choose to belong to different unions because they prefer different approaches to expressing their interests. 

What is the appropriate scope of comparison for determining equal value?

Section 11 of the CHRA says wage comparisons are to be undertaken within the same establishment. Subsection (3) reinforces the idea of a broad interpretation of the term "establishment" by stating that separate establishments "established or maintained by an employer solely or principally for the purpose of establishing or maintaining differences in wages between male and female employees" shall be deemed to be the same establishment. Finally, section 10 of the Equal Wages Guidelines, 1986 clarifies that employees of an establishment include "all employees of the employer subject to a common personnel and wage policy, whether or not such policy is administered centrally."

In effect, the statutory and regulatory intent appears to favour larger rather than smaller establishments. This orientation makes sense if the goal is to define the broadest feasible field of comparison in determining equal value. In some labour jurisdictions, and most notably Ontario, the law provides for proxy comparisons between distinct establishments if there are no appropriate male comparator groups within an establishment. Quebec uses the approach of constructing a theoretical male comparator job in certain cases.

In the case of the federal public service, it has been accepted so far that the core public service is one establishment, and that the various separate employers are distinct establishments. This view is under challenge in the courts, however, with the Public Service Alliance of Canada arguing that in the end the Treasury Board exercises general control over much of the federal public sector, so that most or all of that entity should be treated as a single establishment.

Conceptually, however, the more diverse an establishment, the more varied its business lines and areas of operation, the less likely it is that the value of a particular set of skills would in fact be uniform. For example, as we noted in Chapter 4, for the Canada Revenue Agency auditors are of fundamental importance to the organization's mission. For most of the core public service, auditors are a much less critical group. So evaluating the group's value as equal across the two employers would not reflect their real value to one or the other, or perhaps even both. Using such a group for comparison to other groups would likely be misleading.

What is an appropriate group for undertaking comparisons?

The federal legislation does not explicitly define what constitutes a group for the purpose of assessing equal pay for work of equal value. The term "identifiable occupational group" is used several times in the Equal Wages Guidelines, 1986. In the context of the federal public service, this terminology appears to refer to one of the 70-plus occupational groups defined in 1967. However, the use of the modifier "identifiable" leaves open the possibility of focusing on some other set of employees. This issue applies separately to how we define male-dominated and female-dominated groups, and presumably leaves open the possibility of using different approaches on the two sides of the equal value equation.

A further complication is how to define gender dominance. The Equal Wages Guidelines, 1986 set out in section 13 a sliding scale based on group size, with 70% as the threshold for groups with fewer than 100 members, 60% between 100 and 500 members, and 55% above 500 members. For most groups in the federal core public service, the 55% standard applies. This seems an unreasonably small distinction in gender representation on which to ground an assessment of gender-based discrimination.^[256] As gender balances shift over time through employment equity and the simple evolution of the labour market, one could easily encounter male-dominated occupational groups shifting to female-dominated, or vice-versa.^[257] Logically, any gender-based discrimination resulting from such a shift could only creep in over many years. Yet in principle an occupational group shifting to female dominance could register a complaint immediately.

Our purpose in canvassing briefly the five issues set out above has been to illustrate that applying the concept of equal pay for work of equal value is fraught with ambiguity. Most fundamentally, pay equity seeks to redress the income gap between men and women in the Canadian economy. Ironically, those women most vulnerable to discriminatory wage practices almost certainly are concentrated in small businesses in the private sector. Yet the figures cited in the Pay Equity Task Force Report, show that on average pay equity salary increases for small private sector employers were in the order of only one half of public sector increases. And this comparison takes no account of non-wage benefits, which are generally much more generous in the public sector than among small private sector employers. It is reasonable to suspect that existing pay equity legislation benefits mainly women who are already relatively better off.

Towards a workable approach

The federal public service is of course a part––albeit the largest part by far––of the federal labour jurisdiction. As such, it is governed for pay equity purposes by the Canadian Human Rights Act, as we have described. Thus any approach to managing the obligation to ensure equal pay for work of equal value must fit within the prevailing statutory framework.

The 2004 Report of the Pay Equity Task Force has proposed legislative amendments to the existing framework. That report was commissioned by the Minister of Justice and the Minister of Labour in 1999, in the aftermath of the settlement of the lengthy pay equity dispute of the Public Service Alliance of Canada. For various reasons, the Task Force was not fully established until June 2001. The purpose of the Task Force was to conduct a broad review of the legislation, "with a view to ensuring clarity in the way pay equity is implemented in the modern workplace."

It is beyond the scope of this more general review of federal public sector compensation to offer a thorough assessment of the Pay Equity Task Force Report. In many respects this is an impressive piece of work, the fruit of extensive research and discussions with academics, practitioners, and employer and union representatives. However, despite its 500+ pages and over three years of work, in our view the Pay Equity Task Force Report does not really come to terms with the fundamental issues we highlighted earlier in this section. Considerable attention is devoted to institutional and process issues, but there is relatively little focus on such questions as:

• What outcomes is pay equity legislation intended to achieve?
• To what extent these have been accomplished?
• What has been the cost of the accomplishments to other important objectives such as cost-effectiveness, the place of women in the workplace, and the integrity of collective bargaining?

We recognize that Parliament may amend the statutory regime within which federal jurisdiction employers must meet their pay equity obligations. The rest of this section suggests possible directions for reform, as well as steps that federal public service employers should undertake in managing the obligation to ensure equal pay for work of equal value.

Legislative and policy issues

The legislative framework should be shaped, among other considerations, to respect these five premises.

1. Pay equity should aim to redress that portion of the male/female wage gap that is attributable to discrimination, and not to other factors.

As we noted earlier, the reasonable factors recognized in the Equal Wage Guidelines, 1986, do not exhaust the non-discriminatory factors that can explain wage differences between particular groups of male and female employees. Other factors include, for example, years of experience, education levels and hours worked. It might be argued that differences in these areas are themselves a result of broad societal gender-based discrimination, and therefore should not be excused in assessing pay equity. Such a view, however, would place on individual employers an unreasonable burden to undo broad social trends.

Pay equity legislation, therefore, should permit employers to make a case that some definable part of any male/female wage gap is a result of factors other than gender-based discrimination. As outlined these factors should be broader than the reasonable factors enumerated in the existing Equal Wages Guidelines, 1986.

One potential factor that probably could not be assessed with precision is that referred to as bargaining strength. This is true if only because no employer would want to admit publicly that part of a settlement was forced by union bargaining power. Such an admission could make the next round even more difficult to manage.

2. Pay equity should operate at a systemic level to detect and remedy patterns of differential pay for men and women performing work of equal value within an establishment.

Over the past 25 years, the trend in pay equity has been towards designing complex systems to quantify value for various jobs according to a supposedly gender-neutral standard. Implicitly, there would be one such standard for a given establishment. The result in the federal public service has been an expensive and distracting search for a kind of Holy Grail. In a large, complex organization such as the federal public service, any single standard for measuring value must be arbitrary, sliding rapidly out of touch with changing workplace needs and therefore relative values.

In effect, we have felt obliged to try to create measurement tools of such complexity and obscurity that few could even claim to understand them. Such an approach to job evaluation alienates employees and managers, and ensures recurrent bouts of discontent as internal relativities drift out balance. A seeming science cannot overcome the underlying reality that value is difficult to assess at the best of times, varies by organization according to the nature of their work, and changes fairly rapidly as technology and client needs evolve.

The better approach would be to analyse actual differences in salaries for men and women in an establishment, determine whether there are explanations for these differences that do not involve gender discrimination, and then design means to redress any remaining systemic gaps. It is not obvious, by the way, why redress must take the form of cash; other benefits or terms of employment might in fact appeal more to a particular affected group. Such analysis could be undertaken jointly by the employer and union representatives, with outside assistance where needed. It is in this sense that the proactive approach, as opposed to the existing complaints-based method of surfacing issues advocated by the Pay Equity Task Force report could make a lot of sense.^[258]

3. Pay equity should function within a collective bargaining wage determination system with multiple bargaining groups based on community of interest, and should disrupt that system as little as possible.

It makes no sense to maintain a legislative framework that is internally incoherent. At present, in effect, the human rights principle of equal pay for work of equal value can trump the results of collective bargaining. Perhaps it was understandable that the interconnection of these two systems was poorly appreciated in 1977 when the Canadian Human Rights Act was first adopted, and collective bargaining in the federal public service was barely a decade old. In fact, so far the interaction of the two regimes has been limited. But if we establish a proactive duty on the employer to maintain pay equity, while leaving the unions with no analogous responsibility, there is a great danger of significant salary distortions emerging. The accompanying loss of confidence in the integrity of collective bargaining could be profound. There would be no excuse now, nearly 30 years on, to amend the pay equity laws yet leave the issue of their interaction with collective bargaining to be resolved through trial and error.

Equal pay for work of equal value should be a shared objective for employer and unions, at least within the scope of workers represented by a given union. Failing to respect this principle should be grounds for finding either party to be bargaining in bad faith. The employer would retain a responsibility to see to pay equity across unions, but on a systemic basis as described above.

4. Pay equity should recognize that Canada is a market economy in which salaries generally reflect the forces of supply and demand. Federal jurisdiction employers should not be expected to stray far from market norms in implementing pay equity.

By its very nature, pay equity is at odds with the external labour market. It seeks to substitute for the impersonal workings of supply and demand as the determinant of workplace value, a competing system that essentially works within a given establishment. This could perhaps succeed if the employer had little or no need to recruit from the external labour market or to keep existing employees from defecting to other employers. Like all employers, however, the federal public service can expect to face an increasingly competitive labour market as the baby boomers retire over the coming decade, and so cannot insulate itself from the overall Canadian labour market.

Accommodating external labour market realities needs to be better accepted within a sensible pay equity regime. There should not be an expectation that just because a particular group of employees may be in high demand, this creates an obligation to adjust the salaries of other groups that may have been seen as having a similar value. The fact is that the relative value of groups changes over time, both up and down, within the Canadian labour market, and it should not be the task of pay equity to attempt to overrule this reality within a given establishment.

It may also be the case that extraordinary circumstances such as supporting military operations overseas would make the employer ready to negotiate unusual salary increases to avoid a work stoppage with a particular group. While legislation might be used to override such a need, this course would likely seem undesirable, if only to avoid leaving critical employees disgruntled.

5. For large employers with diverse workforces, pay equity should permit appropriate partitioning of jobs into groupings with broadly similar characteristics, especially required skill levels.

The value of particular skills can be quite different in distinct parts of a large, diverse establishment. Forcing homogeneity in such assessments can only distort the workplace. The better approach for large, diverse establishments is to recognize the legitimacy of defining appropriate families of jobs that share similar levels of education and responsibility within which differences in the remuneration of men and women can be assessed. Later in this section, we outline how such a method might be applied in practice. Such an approach would reduce or eliminate one of the least defensible aspects of pay equity as it has come to be interpreted, whereby the search for male-dominated comparators that will justify a pay equity increase takes on more weight than do discernible similarities in the nature of the work.

The Pay Equity Task Force Report provides ample meat for consideration by the Government and Parliament in drafting possible new or amended pay equity legislation affecting the federally regulated labour jurisdiction. However, in our view, the Pay Equity Task Force Report's recommendations do not go nearly far enough in coming to terms with the legislative principles outlined above, which are themselves rooted in the fundamental issues described earlier in this section. Nearly 30 years into our experience with equal pay for work of equal value in the federal jurisdiction, workers, taxpayers and employers deserve greater clarity on the points in contention.

Administrative practices

In the meantime, there are several important steps that federal public service employers could take to pay more attention to issues of gender equality in compensation. Broadly speaking, we advocate a pragmatic, proactive approach to identifying and remedying compensation differences between male- and female-dominated groups, where these gaps cannot be explained by non-discriminatory factors. The four most desirable practices are described below.

1. Ensure that new classification standards minimize gender bias.

The 2002 Treasury Board decision to shift from pursuing a universal classification standard to a group by group renewal of existing standards aims to adapt these standards to the current realities of the public service workplace, and to minimize any gender bias. The Public Service Human Resources Management Agency has made substantial progress in developing quantitative means to apply gender neutrality tests relating to such topics as: omitted elements or dimensions of work, biased measures, progression and element (dimension) weights, and double counting or redundancy of elements.^[259]

2. Assist managers and others involved in classification to minimize gender bias in the application of existing classification standards.

At least as important as modernizing standards is the training of classification officers and managers in the application of existing, unreformed standards. This training should include sensitization to gender issues and guidance on how to address them throughout the job evaluation process, including the way jobs are described in the first place.

3. Decide on a sensible framework for comparing groups for pay equity purposes.

One effective way to group employees in appropriate shared contexts would be to establish additional separate employers, where the business lines are sufficiently unique and self-contained. We have suggested giving separate employer status to such organizations as the Canada Border Services Agency, the new Service Canada, Correctional Service Canada and Statistics Canada. Adopting this approach, where appropriate, would avoid or at least reduce conflicts around how work is valued in the public service by allowing individual businesses to assess work according to their distinct missions and objectives.

Whether or not the Government decides to implement further separate employers, the core public service will remain a large, diverse employer engaged in a wide range of businesses. For the purpose of assessing equal pay for work of equal value, it would make sense to develop a framework for cross-group comparisons that facilitates meaningful comparison. Such a device would reduce the need persistently felt to turn to complex amalgams of male-dominated groups as a comparator for a given level of a particular female-dominated group. Such tactics might yield appealing results for those lodging complaints, but the very abstraction of such consolidated information renders these comparisons questionable measures of relative value.

A promising tool for grouping broadly similar types of work is the National Occupational Classification (NOC) structure. Developed and maintained by Human Resources and Social Development Canada and its predecessor departments, the NOC "is the nationally accepted taxonomy and organizational framework of occupations in the Canadian labour market." This structure is described as:

... of critical importance for the provision of labour market and career intelligence, skill development, occupational forecasting, labour supply and demand analysis, employment equity ... It provides a standardized framework for organizing the world of work in a manageable, understandable and coherent system and has been successfully implemented in a number of major applications over the past decade.^[260]

Interestingly, the Edition 2001 NOC also emphasizes that over the preceding decade the labour market has undergone significant changes. Technological advancement, globalization of the economy and the increasing importance of knowledge in many occupations have all contributed to occupational change.

In summary then, the NOC offers a framework that is specifically developed for use in broad human resource management. It is widely utilized including, for example, for employment equity comparisons, and it is updated regularly to evolve with the realities of the Canadian labour market.

The NOC uses a three-tiered hierarchical structure for classifying occupations, organized into 26 major groups, 140 minor groups and 520 unit groups. Skill types are distinguished into 10 broad occupational categories, as follows:

The NOC also distinguishes four skill levels, as follows:

1. University degree
2. Some post-secondary at community college, institute of technology or CEGEP; or two to five years of apprenticeship training; or three to four years of secondary school plus more than two years of on-the-job training, courses or specific work experience; and other occupations with supervisory responsibilities or significant health and safety responsibilities
3. Some secondary school; or up to two years on-the-job training, courses or specific work experience
4. Little or no formal educational requirements or on-the-job training^[261]

This tool may assist in carrying out the task proposed in Recommendation 10 to develop a more sensible occupational group structure for the core public service. Aligning the public service group structure better with the broader Canadian labour market will facilitate comparisons with the private sector. Such an approach would also aid in defining meaningful groupings of jobs of broadly equal value within which we could analyze differences between male- and female-dominated groups. 

So far, the Treasury Board Secretariat and the Public Service Human Resources Management Agency have undertaken some preliminary work on how the NOC could be used to specify better what sets of jobs are reasonably comparable for pay equity purposes. Of particular interest is the broad skill level distinction used within the NOC. A preliminary impressionistic application of these definitions to the existing core public service suggests about 10,000 positions may fall into the category of management occupations, 45,000 into skill level A, 45,000 into level B, 30,000 into level C, and 2,000 into level D. Other positions were more difficult to categorize, so were not included. Since knowledge––skill in the language of the Canadian Human Rights Act––is undoubtedly the most important source of value for an employer in the contemporary workplace––and whose importance can be expected to grow, using broad skill levels as defined by the NOC to frame occupational groups for assessing equal pay for work of equal value appears sensible. A more thorough and rigorous development of these concepts should now be undertaken.

4. Conduct an analysis of significant differences in male and female salaries in the core public service.

We have spent decades in the federal public service struggling with theory, rhetoric and complex abstractions in the area of equal pay for work of equal value. It is overdue for us to shift our energy to concrete analyses of the actual differences between male and female compensation in the federal public service, and an assessment of whether such differences can reasonably be attributed to gender discrimination. This could be seen as a case of the sort of gender analysis that is intended to condition any policy work in the public service.

Such an analysis was beyond the mandate and capacity of the present Report. However, we can offer a proposal on how to undertake such a review. The following five steps would make sense:

1. Document the proportion of male and female employees by existing classification group and level, and the related average salaries.
2. Document the work-related characteristics of male and female employees in these groups and levels, according to the criteria used by Statistics Canada and other social researchers in evaluating to what extent salary differences may be explained by factors other than discrimination.
3. Assess to what extent there are group salary differences that cannot be reasonably explained by factors other than discrimination.
4. Develop appropriate comparison domains, particularly broad skill levels, based mainly on the NOC framework, to facilitate cross-group comparisons.
5. To the extent that the above analyses yield significant discrepancies that may be the result of gender discrimination, include these as issues for priority consideration in the next round of collective bargaining for the affected groups.

Such an analytical journey will no doubt be technically complex and controversial. For example, the whole area of non-discriminatory factors will engender great debates. However, a resolutely pragmatic and transparent enquiry into the real situation of male and female employees in the federal public service is long overdue. Ironically, despite a huge effort over many years initiated in response to the issue of pay equity, there has been relatively little work done on what is actually going on. The quest for a seeming science of equal value has driven out simple social science examinations of public service compensation from a gender perspective. Now is the time to adopt a more humble but likely more fruitful strategy.

Of critical importance to any such analysis will be a careful assessment of how the relative position of men and women in the public service has evolved since the Canadian Human Rights Act was passed in 1977. Particularly striking as evidence of dramatic changes under the surface is the evolution of the ratio of men and women within the more than 70 occupational groups first established at the time collective bargaining was introduced in the public service in 1967. Appendix K^[262] provides details of the distribution of these groups by gender for 1981, 1986, 1991, 1996, and then each year from 2001 to 2005.

At the broadest level, the public service as a whole has shifted from 59% male and 41% female in 1981, to majority female (48% male/52% female) by 2001, to almost female-dominant in 2005 according to the Equal Wages Guidelines, 1986 criteria, at 46% male/54% female. Notable shifts in particular occupational groups include the examples shown in Figure 1056, below.

Figure 1056
Changes in gender balance in selected classification groups, 1981-2005

Display full size graphic

* This figure is for 1986, since the Executive Group did not exist in 1981 in its present form. The predecessor "Senior Manager" group was 96% male in 1981.

Almost without exception, the 16 groups listed above are relatively well-paid professional groups. At this high end of the public service, pay equity appears to be emerging through employment equity, as well as women's increasingly greater participation in the academic and professional training needed to fill such positions. A deeper analysis of these groups by level would likely show the proportion of men increasing at higher classification levels. The broader historical trend, however, can be expected to move towards strengthened gender balance by level over time.

In looking at the shift in gender balance within occupational groups over time, it would also be important to look for any evidence that feminizing groups are sliding in salary levels relative to similar male-dominated groups. Such changes would need to be assessed carefully to determine whether they resulted from gender-based discrimination.

There remain some groups, normally fairly small, that are heavily weighted towards male participation. Examples include Architecture and Town Planning (91% male in 2005), Electronics (96% male), Engineering and Land Survey (85%), Forestry (82%), Firefighters (98% male), General Labour and Trades (95% male), Heating Power and Stationary Plant Operation (99% male), Ships Crew (93% male), Ships Repair (97% male), and Technical Inspection (89% male). Many of these groups are at most in the middle ranks on pay levels.

On the female-dominant side, there are several groups that have remained essentially unchanged since 1981. The largest occupational group, Clerical and Regulatory, is 82% female. Hospital Services is 58% female, Library Science 74% female, Nursing 85% female, Occupational and Physical Therapy 73% female, and Secretarial, Stenographic and Typing is 98% female. The last group is declining rapidly in numbers and can be expected to virtually disappear in the next few years.

The point of this brief exposition on the evolution of gender proportions by occupational groups over the past 25 years is not to make any precise finding relating to equal pay for work of equal value. What is evident, however, is that the federal public service workforce has been transforming in many ways in the place accorded to men and women. This can only alter profoundly the relationship between the genders on matters of compensation. 

This concludes our commentary on issues relating to the application of the human rights principle of equal pay for work of equal value in the federal public service. Perhaps the biggest problem with the past generation's struggle to interpret and apply this concept has been an unresolved ambiguity at the heart of pay equity as to its real purpose. Formally, International Labour Organization Convention 100 and the Canadian legislation giving effect to this international undertaking, address individual employers and require them to avoid gender-based discrimination on pay. But the proponents of pay equity are understandably driven by a much larger agenda of bringing society as close as possible to equal incomes for men and women. Thus pay differences arising from less education, or shorter labour force attachment, or shorter working hours, while logically non-discriminatory at the level of an individual employer, can still be seen as the results of larger social forces that themselves discriminate against women. This broader social outlook leads pay equity advocates to resist accepting non-discriminatory factors as mitigating an employer's obligation to close fully a given gender pay gap.

Pay equity is thus aimed squarely at correcting wage discrimination in the overall labour market. Some egregious historical examples of undervaluing women's work in the economy generally can be cited. Few would now argue, for example, that nurses were treated fairly in the health professions until recently. But taking on the labour market as an individual employer, even such a large and prominent employer as the Government of Canada, is a perilous enterprise. Internal wage differentials that do not correspond with the external labour market will make it too easy to recruit and retain some kinds of employees, giving those successful in landing such jobs a salary premium not enjoyed by other Canadians in similar circumstances. On the other hand, these differentials risk making the public service less attractive to other kinds of workers, who can generally expect to be more highly valued by private employers less influenced by the pay equity system.

It is especially unfortunate that these imbalances in the weight attached to work characteristics tend to downplay the importance of knowledge, which is more and more the key to a competitive workplace. It is doubly regrettable if the processes used to arrive at this situation are themselves expensive, little understood or trusted, and time-consuming. Organizations that succeed in serving their clients with excellence cannot afford to divert scarce talent over many years into the search for an all-purpose formula that will push the organization toward greater gender equality in compensation.

The main danger for the federal public service in the legislative reforms proposed by the Pay Equity Task Force is that once again a set of ill-defined concepts will be enshrined in law, with little guidance on how to understand them, apply them, or reconcile them with other important principles like free collective bargaining. This in turn risks spawning a new industry of complex, arcane seeming science. After inevitable procrastination, litigation and bad feelings on all sides, substantial expenditures will be required to advance income equality between men and women, but with little regard for the draining of energy from the substantive work of the public service, or its capacity to attract and retain critical talent. Now is not the time to launch further decades of trial and error in the name of a powerful human rights principle. Rather, it is long overdue that we focus on identifying breaches of the equal pay for work of equal value requirement in a pragmatic manner, and that we get on with eliminating directly those gaps that cannot reasonably be explained by reasonable factors other than discrimination.

In summary, then, our recommendations on pay equity in the federal public service are these:

With this chapter we have brought to an end our series of substantive proposals for improving the management of compensation in the federal public service. Before concluding this Recommendations Section, we provide some thoughts and recommendations in the next chapter on how best to proceed with following up on our proposals.

 

• Previous Page
• Table of Contents
• Next Page