Treasury Board of Canada Secretariat
Symbol of the Government of Canada

ARCHIVED - Expenditure Review of Federal Public Sector - Overview

Warning This page has been archived.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.

Some observations on the 2002–03 snapshot

Differences and commonalities among domains

Comparisons among the five domains must be handled with care. Exploring differences is always interesting, but the unique nature and business purposes of each domain mean that such differences are unlikely to be meaningful in policy terms. The core public service is diffuse and highly varied in its business missions and employee characteristics. The various separate employers each tend to have fairly focused mandates, but which are as different as food inspection and tax collection. The Canadian Forces must be prepared to apply lethal force to defend Canada and its citizens and interests; the RCMP of course has policing as its vocation. These varied purposes drive different labour force requirements, which in turn imply distinctive salary patterns. There is no reason to expect average salaries, or salary distributions in a given domain, for example, to have a particular relation with those in another domain.

On the other hand, we can observe clear evidence of structural commonalities across the various domains studied. First, the components of total compensation are similar throughout the federal public sector. It is rare to find anomalies such as the absence of a Supplementary Death Benefit in the case of the RCMP, or the presence of a type of location-based cost-of-living allowance such as the Canadian Forces' Post Living Differential. Second, many of the insurance and benefit plans are unified across several domains. Finally, there has evidently been an intention to hard-wire the relationships among salaries for senior officials across all the domains.[6]

The influence of external labour markets

Broadly speaking, external labour markets affected federal public sector compensation in 2002–03 in three ways:

  • as one factor influencing collective bargaining;
  • as a standard for setting salaries, within the context of comparable total compensation, for the first level of the Executive group in the core public service; and
  • as an influence on determining salaries for regular federally appointed judges.

Generally speaking, attrition in the federal public sector in 2002–03 was low, except perhaps in the Canadian Forces. In the core public service, only about 1% of employees resigned. Departures from the armed forces constituted about 7% of serving members. Detailed and consistently recorded statistics on the number of qualified candidates applying in external competitions are not available. However, it appears that external demand for advertised positions in the federal public sector was high in 2002–03, except for highly specialized positions for which the external labour market is active.

Changes in federal public sector compensation: 1990‑91 to 2002‑03

Having documented the components and associated expenditures comprising federal public sector compensation in 2002–03, we then tell the story of how total compensation has evolved since 1990–91. We focus particularly on changes since the end of Program Review downsizing in 1997–98. By examining both changes in expenditure levels and the underlying forces driving those changes, we in effect explain how we arrived at the levels and pattern of compensation spending in 2002–03.

Looking at change very broadly, the 1990s saw the federal public sector[7] shrink by about 75,000 employees (or 21%) by 1998–99. Thereafter, employment increased by at least 40,000, to reach 90% of the 1990–91 level. The overall level of both salary spending and total compensation declined in the federal public sector in the early 1990s as the level of employment fell.

After 1997–98, however:

  • salary spending increased by 44% in current dollars and
  • total compensation costs grew by 49%.

In constant 2002–03 dollars, the growth in salary costs between 1997–98 and 2002–03 was 29%; for total compensation, the constant dollar increase was 34%. Total salary expenditures in 2002–03 were 131% of the 1990–91 level; total compensation spending equalled 139% of 1990–91 expenditures.

The largest drivers of change in total compensation were salaries and employer pension contributions. Increases in total salaries paid accounted for two thirds of the growth in total compensation from 1997–98 to 2002–03. Larger employer pension contributions in regard to current service represented 15%, or nearly half of the remaining increase in total compensation costs. Other areas of substantial increase were allowances and premiums, and health and dental plan employer costs, each contributing nearly 5% to the growth.[8] The appendix shows the employer costs by principal component in the combined core public service and separate employer domains, the Canadian Forces, and the RCMP, as well as the federal public sector as a whole, for 1990–91, 1997–98 and 2002–03.

Changes in employment

Our principal focus was on the core public service and the separate employers.[9] In these combined domains, total employment fell from as high as 245,000 in the early 1990s to about 195,000 in 1998–99, and then grew back to around 235,000 by 2002–03. However, employment levels over those years are not strictly comparable. We conclude that increases in expenditures on service contracts from 1998–99 onwards, as well as on transfers of functions to the private sector or other governments where the federal government effectively still pays for the activity, provided an equivalent of at least 10,000 person years of work. By 2002–03 then, the core federal government's effective size was at least as great as in the early 1990s.

Figure 2
Employment in the combined core public service and separate employer domains, the Canadian Forces and the regular and civilian members of the RCMP, 1990–91 to 2002–03

Display full size graphic

Figure 2

Employment changes in the uniformed federal public sector followed different patterns. For the Canadian Forces, the number of regular members fell through almost the whole period under review, from nearly 89,000 in 1990–91 to a low of 59,000 in 1999–2000 and 2000–01, rising slightly to 62,000 in 2002–03.[10] For the RCMP, the total number of regular and civilian members was quite stable. From a maximum of 18,000 in the early 1990s, the total diminished to just below 17,000 in 1998–99 and 1999–2000, returning to about 18,000 in 2002–03.

Changes in the salary mass

For the combined core public service and separate employer domains, total salaries:

  • fell from $9.1 billion in 1990–91 to
  • $8.2 billion in 1997–98, and then rose by 50% to
  • around $12.4 billion in 2002–03.

In constant 2003 dollars, there was a decline of 22% from the peak level in the early 1990s to 1997–98, followed by a 35% increase by 2002–03.

For the Canadian Forces, total military payroll fell from a high of:

  • $3.4 billion in 1993–94 to a low of
  • $2.9 billion in 1996–97 and 1997–98, thereafter rebounding to
  • about $3.7 billion in 2002–03.

In constant 2003 dollars, there was a decline of 23% between 1990–91 and 1997–98, with a subsequent increase of 14% by 2002–03.

For the regular and civilian members of the RCMP, total salaries rose from:

  • $803 million in 1990–91 to
  • $1.08 billion in 2002–03.

While there was some fluctuation in the intervening years, there was no distinct down-and-then-up pattern for total salaries as in the other domains. In constant 2003 dollars, however, the RCMP salary mass declined by 13% between 1993–94 and 1997–98, and then grew by 12% as of 2002–03.

Principal drivers of growth in the salary mass after 1997–98

We focused mainly on determining the principal drivers in the 50% increase in total salaries in the combined core public service and separate employer domains between 1997–98 and 2002–03. We estimate that growth in the number of public servants accounted for over 40% of this increase, and growth in average salaries explained the remaining portion.

Figure 3
Components of change in the total salary mass for the combined corepublic service and separate employer domains, 1997–98 to 2002–03

Display full size graphic

Figure 3

In looking at population increases, we attempted to reconstruct how these were financed. Although the historical data is somewhat ambiguous and awkward to work with, it appears that between 70% and 90% of the growth in employment resulted from specific approvals by the Treasury Board to increase salary budgets as part of decisions to implement new or revised programs, or to recognize workload increases. The remaining 10% to 30% was funded through transfers initiated by departmental managers from approved non-salary budgets into salary budgets.[11]

By 2002–03, average salaries in the combined core public service and separate employer domains reached $52,800, a cumulative increase since 1998–99 of 27.3% in current dollars, which was equivalent to 14.1% in constant 2003 dollars. By contrast, between 1990–91 and 1998–99, the average salary in constant 2003 dollars fluctuated in a narrow band between $46,300 in 1997–98 and $48,100 in 1994–95. In fact, this pattern had prevailed since 1982–83. Thus, after 15 years when the average federal public service salary remained essentially unchanged in terms of real income through periods of both collective bargaining and salary controls or freezes, from 1997–98 to 2002–03 average real salaries grew by about one seventh.[12]

Unravelling the factors driving change in the average salary in the combined core public service and separate employer domains is complex. Figure 4 summarizes those factors. Cumulative inflation between 1997–98 and 2002–03 was 11.6%. Collective bargaining outcomes covered that inflation, and also contributed 7.9 percentage points (well over half) of the 14.1% real salary increase in those years. General economic increases accounted for 3.7 of these percentage points. Changes in the structure of salaries such as adding or eliminating salary increments, or combining rates for different groups or regions resulted in 4.1 percentage points of the real increase.

Figure 4
Components of change in average salary in the combined core public service and separate employer domains, 1997–98 to 2002–03

Display full size graphic

Figure 4

Change in the composition of the federal public service workforce accounted for over one third (5.3 percentage points) of the real increase. In essence, higher paid classification groups associated with analytical and professional work grew, while traditional administrative and operational groups shrank. For example, the Clerical and Regulatory group declined between 1991 and 2003 by over 16,600 members, or 28%. The Secretarial, Stenographic and Typing group fell by 82%. Conversely, the greatest growth occurred in the Computer Systems group (with an increase of 174%), the Law group (+132%), and the Economics, Sociology and Statistics group (+115%). Substantial expansion in the Administrative Services classification group (+64%) was at least partly explained by a shift of former clerks and secretaries into broader administrative work.

In addition, salary changes resulting from pay equity settlements during the period contributed a real increase of about 2% overall to the average salary in the combined core public service and separate employer domains. Pay equity settlements covering more than a dozen female-dominated classification groups between 1980 and 2003 resulted in significant payouts to the affected employees. Lump-sum payments totalled about $3.4 billion, including about $900 million in interest. Cumulative ongoing costs from the time of settlements up to March 2003 were estimated to exceed $1.8 billion.

Executive employment in the core public service followed in a more pronounced way the pattern of the broader public service. The population fell 32% between 1990–91 and 1997–98, and then rose by 37% by 2002–03 to reach about 4,400, or 93% of the 1990–91 level. Average salaries for these executives grew by 13% in constant dollars between 1997–98 and 2002–03. Including improved performance pay introduced in the late 1990s, core public service executives saw their average combined salary and performance pay increase by about 18% in real terms over the five year period.

As with the cost of employment growth, the cost of increased average salaries in the combined core public service and separate employer domains between 1997–98 and 2002–03 was financed partly by transfers from the Treasury Board (between 60% and 75%), and partly by departmental transfers from approved non-salary budgets to salaries (between 25% and 40%). The total value of departmentally initiated transfers from non-salary to salary budgets was $1.2 billion higher in 2002–03 than in 1997–98, contributing nearly 30% of the total increase in salary mass over that five year period.

Average salary growth for the Canadian Forces

The pattern of change in real average salaries in the Canadian Forces was more variable than for the combined core public service and separate employer domains (for simplicity, we will call this the "regular public service"). A clear upward trend in the early 1990s (from $44,500 in 1990–91 to $49,900 in 1994–95 in constant 2003 dollars), was followed by a dip to $47,000 in 1996–97, then rapid growth to $54,200 in 2000–01, and finally a decline to $52,700 in 2002–03. The endpoint is virtually identical to the regular public service average of $52,800 in that year. Between 1997–98 and 2002–03, the Canadian Forces constant dollar average salary increase was about 10.9%, or 3.1 percentage points lower than for the regular public service.

The incidence of structural change as an influence on average salaries was much less prominent in the Canadian Forces than in the regular public service. An analysis of the proportion of regular Canadian Forces members serving at the various ranks shows an upward shift in average salary of about 2.7% between 1990–91 and 2002–03 as a result of structural change. For the period 1997–98 to 2002–03, this factor was actually a negative 1.5%, compared with a contribution to average salary of 5.3% in the regular public service in those years from compositional change.

Average salary growth for the RCMP

For the regular and civilian members of the RCMP, the average salary was higher than for either the regular public service or the Canadian Forces, but its growth has been more modest. In constant 2003 dollars, the average salary increased modestly from $57,600 in 1990–91 to $62,000 in 1994–95, then declined to $56,700 in 1997–98, grew to $60,900 in 2001–02, and finally dipped to $59,900 in 2002–03. The increase, then, from 1997–98 to 2002–03 in real terms was around 5.6%, less than half the increase in the regular public service. For the RCMP, changes in average salary arising from realignment in the distribution of members by rank would if anything be negative.

Average salary growth for other groups

For federally appointed judges, salary changes approved in the Judges Act covering the period 1997–98 to 2002–03 resulted in a 21% constant 2003 dollar increase in judicial salaries. For Members of Parliament and Senators, the salary increase between 1997–98 and 2002–03 was about 19.6% in constant 2003 dollars.[13]

Pension contributions

Pensions are by far the largest non-salary benefit for employees, and the largest non-salary cost for the employer. During the period from 1990–91 to 2002–03, there has been a substantial increase in the proportion of contributions for current service assumed by the public service employer. The employer/employee contribution ratio for the main Public Service Pension Plan was 51/49 in 1990–91, and employees still paid over 40% as recently as 1995–96. By 1999–2000 the ratio was 72/28; it was 74/26 in 2002–03. The total contribution paid by employees in 1991–92 was $540 million. This declined to $453 million in 1997–98 with a reduced population, and then grew to $644 million in 2002–03. The employer contribution increased in all but one year during this period, rising from $561 million in 1990–91 to $1.8 billion in 2002–03.[14]

There were two main reasons for this shift. First, the combined employee contribution to the Public Service Pension Plan and to the Canada and Quebec Pension Plans was capped at 7.5% of salary until 2000. As CPP/QPP contributions increased, employee contributions to the Public Service Pension Plan declined from 5.7% to 4.0%.[15] Second, actuarial assumptions became more unfavourable in later years, as expectations of high real interest rates and low real salary increases were reversed. With defined benefit pension plans such as the Public Service Pension Plan, the employer is obliged to cover the expected costs of the plan, less whatever is contributed by the employees.

A similar pattern of an increasing employer share of current service pension costs was evident in the other principal federal public sector pension plans. For the Canadian Forces Pension Plan, the employer/member cost-sharing ratio shifted from 68/32 in 1991–92 to 78/22 in 2002–03.

For the regular and civilian members of the RCMP the evolution was from 64/36 in 1991–92 to 76/24 in 2002–03.

For federally appointed judges' annuities the change was from 75/25 in 1990–91 to 84/16 in 2002–03.

For the Members of Parliament retiring allowances, the cost-sharing ratio between the Government and members fluctuated with changes in plan terms, from as even as 50/50 in 1990–91 to 84/16 in 2002–03.

The reason for a higher employer share in the current service pension costs of these latter groups is that most groups contribute the same percentage of salary as in the regular public service, but fewer years of service are required to qualify for a pension, and in some cases the retirement age is also lower.

Other benefits

Compensation costs for elements other than salaries and employer pension contributions grew from about $3.1 billion in the federal public sector in 1990–91 to almost $4.4 billion in 2002–03. Most of the growth occurred after 1997–98, when the total was around $2.8 billion.[16] Total expansion in this area between 1997–98 and 2002–03, then, was about 57% in current dollars, or 35% in constant 2003 dollars.

In several cases, increased costs resulted from new or substantially revised policies, as described below.

Recruitment and retention (terminable) allowances

For certain unionized groups determined to be in high demand in the external labour market, new recruitment and retention allowances were introduced. This policy began in 1997 with the Computer Systems group. Spending in this area grew to $77.4 million in 2002–03 in the core public service. Similar allowances in the separate employer domain totalled at least $20 million in that year.

Performance pay

On the recommendation of the independent Advisory Committee on Senior Level Retention and Compensation, a new performance pay system for Executives and Deputy Ministers was phased in between 1998 and 2000. The total of lump sum payments in the core public service relating to performance increased from about $11 million in 1997–98 to around $40 million in 2002–03. Payments for this purpose in other domains amounted to at least $20 million in that year.

Maternity and parental leave

As a result of changes in the Employment Insurance Act in 2001, the duration of supplementary allowances for those taking maternity or parental leave was doubled to a maximum of a year. In the core public service the value of this allowance grew from $22 million in 2000–01 to $62 million in 2002–03. Expenditures in that year for the Canadian Forces and the RCMP combined reached about $25 million.

Canadian Forces Post Living Differential

The Canadian Forces introduced a Post Living Differential in 2000–01 to reduce variations in standard of living as members are posted around the country. By 2002–03 this allowance cost $65 million, compared with about $4 million for the Accommodation Assistance Allowance which was its closest predecessor. Military operational allowances increased with the tempo and danger of foreign operations undertaken by the Canadian Forces, increasing from about $36 million in 1997–98 to $60 million in 2002–03.

Pensioners' dental plan

In 2001, a voluntary Pensioners' Dental Services Plan was introduced to cover public sector pensioners and their dependants. By 2002–03, the employer share was about $23 million.

Leave entitlements

Employee leave entitlements grew over the period. Average total days of leave per capita (including vacation, sick leave, family and other leave, and statutory holidays) grew from 40 days in 1991–92 to 41.2 days in 2002–03. We estimate that the proportion of paid time not worked increased from 15.7% in 1990–91 to 18.3% in 2002–03. This difference has a combined value of about $225 million in current dollars.

CPP/QPP contributions

Changes in contribution rates for the Canada and Quebec Pension Plans in the late 1990s led to a major increase in employer payments for this purpose. Contributions covering the regular public service, the Canadian Forces and the RCMP grew from about $250 million per year in the early 1990s to about $590 million in 2002–03.

Other cost drivers

In some other areas little has changed in policy terms, but costs have risen substantially. In the area of health and dental programs, the main employer cost drivers were increases in the share of costs borne by the employer and increases in supplier prices. The employer share of health and dental plans increased substantially just before and in the first few years of the 1990s. For employees, the employer share of Public Service Health Care Program (PSHCP) costs increased from 50% to 75% in 1989, to 90% in 1991, and to 100% in 1992.[17] For pensioners, the employer share rose from 50% to 75% in 1991, and to 80% in 1992. For the Public Service Dental Care Plan (PSDCP) the 50% employer share set at the program's inception in 1987 was increased to 100% shortly afterwards as part of an out-of-court settlement with the unions.

Employer costs for the Public Service Health Care Plan are estimated to have been about $84 million in 1990–91, rising to around $175 million in 1997–98, and to $422 million in 2002–03. Claims paid per member rose from $300 in 1990, to $560 in 1997, to $841 in 2002. Between 1997 and 2002, prescription drug costs rose from $129 million to $267 million, accounting for most of the increase in program expenditure.[18]

Employer costs for the Public Service Dental Care Plan more than doubled from $77 million in 1990 to $156 million in 2002. Claims paid per member rose from $250 in 1990 to $472 in 2002. Most of the increase related to increasing charges for routine preventative care.

Disability claims for the federal public sector's two main insurance plans more than doubled between 1990 and 2002. Claims paid by the larger plan, that for unionized employees, rose from $85 million in 1990 to $172 million in 2002; in the plan for unrepresented employees claims rose from $12 million in 1990 to $29 million in 2002. The number on disability claim increased as a proportion of members from 2.88% in 1990 in the main plan (1.68% in the unrepresented plan) to 4.28% in 2002 (2.84% in the other plan). Much of the increase resulted from growth in cases relating to depression and anxiety.

For the RCMP there was little change in the nature of major allowances. However, their cost rose through rate increases. For example, the shift differential/stand by category rose from $3.7 million in 1990–91 to $14.5 million in 2002–03. The Senior Constable Allowance grew from $3.3 million in the earlier year to $13.4 in 2002–03. The plainclothes/kit upkeep category expanded from $4.8 million in 1990–91 to $16.1 million in the later year.

In a few areas employer expenditures declined in relative terms. A key example is overtime. In the combined core public service and separate employer domains, this item declined from 3.2% of payroll in 1991–92 to 2.8% in 2002–03. The amount paid (about $350 million) was nearly the same in constant 2003 dollars in the two years. With the understandable exception of the Program Review years in the mid-1990s, severance pay and termination benefits have declined somewhat, both absolutely and in constant dollars.

Some observations on changes since 1990–91

Collective bargaining and salary controls

As Figure 5 illustrates, over the 36 years between the introduction of collective bargaining in the federal public service in 1967 and 2003, the collective bargaining process was suspended in whole or in part for at least 11 years in three distinct periods. In addition, in 1999 the Government acted to force settlements on two particular groups.

The relative instability of the collective bargaining system raises questions about its effectiveness. Apparently, the system is only capable of yielding salary and benefit increases, typically above the rate inflation or the growth in the economy, whatever else is happening in the external labour market or general economic conditions.

Since 1997–98 and the end of the last period of controls, salaries have increased in real terms in a way not seen since at least the early 1980s. If this has not created pressure for new controls, two explanations suggest themselves. First, similar real after-tax family income increases in the general Canadian economy over the same years imply that some real income increase was not extraordinary. Second, the string of federal budget surpluses since 1998 has meant that pressure to scrutinize federal operational costs has not been intense. If overall economic conditions or the state of federal finances weaken, it remains to be seen how the collective bargaining system in the federal public service will respond. With employees and unions habituated to continued increases in real salaries, it might prove difficult to accept restraint voluntarily.

Total compensation and compensation management

Another key point that emerges is the fragmentation of decisions affecting compensation. Collective bargaining in the federal public service is usually limited to dealing with salaries, allowances and leave. Classification and pensions are excluded by statute from bargaining, although they are often discussed informally with the unions. Other benefits such as insurance, health and dental plans are negotiated separately, and then included in the collective agreements. Such a system discourages transparency, and provides little scope for trade-offs among compensation elements. Since the cost of pensions and other benefits mainly falls on the employer, dealing with compensation components in different forums at different times is likely disadvantageous from the perspective of controlling costs. This practice may also make it more difficult to achieve a balance in compensation elements that best suits employees.

Figure 5
Periods of collective bargaining and compensation controls, 1967 to 2003

Display full size graphic

Figure 5

Developing and pursuing a holistic view of compensation, across both the domains of the federal public sector and the various components of compensation, has been hampered by a fragmentation of responsibility on the employer side. As is often the case in large organizations, distinct groups have managed different aspects of compensation such as collective bargaining, salary determination for unrepresented groups, executive compensation, classification, pensions, and insurance and health and dental plans, generally with light coordination at most. In the absence of greater integration in compensation management, both coherence and control are difficult to attain.

During the 1980s, the Treasury Board Secretariat attempted to design analytical means to track total compensation costs and to use this information to influence collective bargaining. However, this was easily turned aside by the unions. First, several elements of compensation such as pensions were off limits, so they were not on the negotiating table. Second, the calculations tended to be abstract estimates of value to the employee, as opposed to costs to the employer. The concept of total compensation as the framework for determining salaries and other benefits persisted only with groups, such as the Canadian Forces, the RCMP or Executives, for whom the Treasury Board determined compensation directly, albeit taking account of advisory groups in most cases.

Comparability of federal compensation with the Canadian labour market

Historical context

Over the past century, successive federal governments have acknowledged the importance of maintaining comparability with the external labour market when determining federal public service compensation. The general idea has been that in the absence of a bottom line, there is no inherent discipline in the level of public sector compensation. The private sector, however, can be relied upon to pay only what is necessary to secure needed talent, because failure to do so would threaten a private business's ability to compete and survive in the marketplace. By maintaining comparability with the external labour market, the public sector can both attract and retain the employees it needs, and assure taxpayers that what it spends is reasonable.

In practice, however, other factors have tended to displace comparability as a determinant of salaries and benefits. These have varied in importance over time, depending on changing circumstances. Among these factors we noted pressure to maintain internal relativities, pragmatic considerations such as finding a saleable deal, the Human Rights Act requirement of equal pay for work of equal value, and the desire to set an example as an enlightened employer, as well as such national imperatives as controlling inflation or reducing government spending.      

So where was federal public sector compensation in 2002–03 in comparison with the broader Canadian labour market?

Salary comparability

Available data suggest that as of 2003, there was likely a small premium on average in favour of federal public sector salaries versus those paid in the Canadian private sector. However, in the early 2000s, the rate of increase in average salaries in the federal public sector was greater than for broad indicators of salary increase in the private sector such as year-over-year wage increases for private sector collective agreements in force, average weekly earnings, or average hourly earnings. If this pattern of faster growth persists, federal salaries on average will surpass those in the private sector by an increasingly significant margin. Figure 6 illustrates the relative rate of growth of salaries in the federal public sector and the overall Canadian private sector between 1990–91 and 2002–03.

Some studies put forward in recent years suggest that there is already a substantial wage premium in the federal public sector versus the Canadian private sector. These studies generally rely on broad sources such as the Census or the Labour Force Survey.

Figure 6
Comparison of the rate of change in key average income indicators in the federal public sector and the private sector in Canada, 1990–91 to 2002–03

Display full size graphic

Figure 6

However, we found that as these comparisons were tailored more and more rigorously to track the actual characteristics of particular jobs and occupations in the federal public service, the purported salary premiums shrank in significance. The one survey which matches specific jobs, that conducted by the Institut de la statistique du Qubec, indicated a federal public service premium of only about 1% versus the Quebec private sector in its 2003 report. Nevertheless, all these general studies found, as we did, a trend toward relatively faster increases in average salaries in the federal public service than in the broader economy in the early 2000s.

We also looked at several studies comparing salaries in the federal public service and the private sector for particular occupational groups. All these studies found that public service salaries lagged their private sector comparators. In our view, these studies were questionable in that they tended to be undertaken in the lead up to collective bargaining, and they appeared to be unbalanced in the points highlighted. In general, there is an absence of established approaches or standards to discipline such studies.

The distribution of federal public sector wages by level of annual earnings is much more concentrated than for the overall Canadian private sector. According to data from the 2001 Census, over half of private sector workers earned less than $40,000 per year, compared with only about one third of federal public sector employees. At the top end, about 13% earned more than $80,000 in the private sector, versus 8% in the federal public sector. In the range between $40,000 and $80,000, then, we found 59% of the federal public sector, but only 35% in the private sector.

Consistent with this distribution, the prevailing pattern in the federal public service over several decades has been to pay equal or better salaries than for comparable jobs in the private sector in the lower grades except for the trades, while paying less for those in executive positions and some professional posts.

Pay equity settlements in the federal public service appear to have accentuated this pattern. The ongoing share of the wage bill for pay equity in the combined core public service and separate employer domains is estimated to be in the range of 2% to 3% overall. Federal public service investments in pay equity appear to have been as large as or larger than those implemented in other jurisdictions. Settlements in Ontario, for example, for organizations with 500 or more employees were reported to equal 0.6% of payroll in the private sector, and 2.2% in the public sector.

Pension comparability

The Federal Public Service Pension Plan can reasonably be considered to be in the top 10% of all registered pension plans in Canada in terms of its value to employees and cost to the employer. While employee contributions are generally higher than for employees enrolled in major private sector plans, the federal plan's benefits are also better than all but a small group of private sector plans. The federal plan is particularly favourable in regard to ancillary benefits such as early retirement, normally with at least 30 years' service at age 55, and full indexation to compensate for inflation. Other federal pension plans, namely those for the Canadian Forces, the Royal Canadian Mounted Police, Members of Parliament and Senators, and federally appointed judges, provide even better value in that pensions may be earned in fewer years of service (mainly for judges and parliamentarians), or commenced earlier than age 55 in certain cases for members of the military or the RCMP.

Employees pay relatively high pension contribution rates: 4% of earnings for that part of salary below the amount covered by the Canada and Quebec Pension Plans ($39,900 in 2003), and 7.5% above that income level. However, these contributions are lower than what most provincial public service pension plans require for similar benefits. Overall, in the early 2000s, federal public servants paid no more than 28% of the current service costs of their pensions, significantly below the 40% to 50% shares set for most other public sector plans in Canada.[19]

Comparability of other benefits

Although some aspects of other benefits are less favourable than for employees of large private sector enterprises, overall the other benefits provided for federal public sector employees compare very well with what is offered in the Canadian private sector.

Public Service Health Care Plan

In general, the Public Service Health Care Plan (PSHCP) provided for active employees and their dependants was comparable with middle-of-the-road private sector plans. Areas that were somewhat less competitive included the absence of a drug card, reimbursement and deductible levels, hospital room coverage, limits on many paramedical specialists, vision care limits, as well as private duty nursing and out-of-country coverage.

Few employers provide the same level of coverage for retirees as for active employees. Thus, PSHCP coverage for public service retirees was very competitive, especially in the scope of coverage and in the area of cost-sharing for hospital coverage.

Public Service Dental Care Plan

The Public Service Dental Care Plan (PSDCP) was less competitive in areas such as the waiting period, deductible, and the combined maximum for basic and major dental services. The PSDCP was more competitive, however, on the lifetime maximum for orthodontics.

Again, employer support for a retirees' dental plan is not common. Thus, the federal public sector Pensioners' Dental Services Plan was advantageous.

Long-term disability

The long-term disability plans for the federal public sector were quite competitive compared with those offered by other employers, especially in regard to the benefit level, inflation protection and maximum coverage.

Severance pay

Severance pay at the time of leaving the federal public service was relatively generous in that only about two thirds of private sector collective agreements include such a benefit, and it is rare in the broader Canadian labour market.

Paid leave

The federal public sector was generally more generous than most private sector employers in its policies on paid leave. It was particularly generous in the duration (up to a full year) of the top-up of Employment Insurance benefits (to 93% of regular income) for the combination of maternity and parental leave.