Treasury Board of Canada Secretariat
www.tbs-sct.gc.ca
Common menu bar links
Breadcrumb Trail
ARCHIVED - Expenditure Review of Federal Public Sector - Volume One - The Analytical Report and Recommendation
This page has been archived.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.
SECTION THREE
COMPENSATION COMPARABILITY
5. Comparing Federal Total Compensation with Economic Indicators, 1990 to 2003
This chapter compares federal compensation trends to available evidence on trends in the broader Canadian economy.
Federal government employment and salaries
As we noted in Chapter 4, federal employment in the combined core public service and separate employer domains was about:
- 245,000 in the early 1990s, then
- fell below 200,000 in 1997–98 and 1998–99, and
- grew back to about 235,000 by 2002–03.
If we add in the regular members of the Canadian Forces and the regular and civilian members of the Royal Canadian Mounted Police (RCMP) for these years, we would have totals of about:
- 350,000 for 1990–91,
- a bottom of about 275,000 in 1998–99, and
- a total of just under 315,000 for 2002–03.
The difference in pattern from the first set of numbers results from the general downward trend in the Canadian Forces population except in the last few years, and the relatively stable population for the RCMP.
Total salaries for the combined core public service and separate employer domains stood at about $9.1 billion in current dollars in 1990–91, declined to $8.2 billion in 1997–98, and then grew to about $12.5 billion in 2002–03. Adding in the salary mass for the regular armed forces and the RCMP gives us totals of $13.3 billion in 1990–91, $12 billion in 1997–98, and $17.3 billion in 2002–03.
Average salaries have evolved as shown in Table 1017 below.
|
Table 1017 The evolution of average salaries |
|||
|
Domain |
1990–91 |
1997–98 |
2002–03 |
|
Core public service and separate employers |
$37,500 |
$41,500 |
$52,800 |
|
Canadian Forces |
$35,000 |
$42,600 |
$52,700 |
|
RCMP |
$45,400 |
$50,800 |
$59,900 |
|
|
2002–03 $ Constant |
||
|
1990–91 |
1997–98 |
2002–03 |
|
|
Core public service and separate employers |
$47,600 |
$46,300 |
$52,800 |
|
Canadian Forces |
$44,500 |
$47,500 |
$52,700 |
|
RCMP |
$57,600 |
$56,700 |
$59,900 |
Our first perspective on comparability is to summarize the general state of the Canadian economy and of employment and incomes during the period from 1990–91. We then compare rates of growth in employment, salary mass and average salary in the federal public sector with these broad trends in the overall Canadian economy. Appendix J provides the data underlying the figures used in this chapter to illustrate trends in economic indicators.
General economic indicators for Canada
The basic trend lines of the Canadian macro-economy between 1990 and 2003 have been decidedly healthy. Figure 1018 gives details for several key economic indicators. Figure 1019 puts these indicators together in one graph on an index basis (1990–91 = 100). Here we offer a brief description of the evolution of these indicators.
Figure 1018
Key economic indicators in Canada, 1990–91 to 2002–03
Figure 1019
Comparison of rate of change in key economic indicators in Canada, 1990–91 to 2002–03
Nominal Gross Domestic Product (GDP)
This is the total value of goods and services produced during the year in Canada, expressed in current dollars. Between 1990 and 2003, GDP grew from about $680 billion to over $1.22 trillion, an increase of 79%. This growth was fairly steady from year to year, with slowdowns in the early 1990s and in 2001–02.
Population
Although the total Canadian population is not an economic indicator in itself, it is a critical variable for putting the size of government in perspective. Between 1990 and 2003, the Canadian population is estimated to have grown from about 27.74 million to 31.39 million, an expansion of 13%.
Unemployment
This is a key measure of the tightness of the Canadian labour market, and the competition for workers in general. Unemployment was high in the early 1990s, but then fell through the second half of that decade to stabilize in the range of 7% to 7.5% in the period after 1999.
Federal Deficit
This is a measure of the health of federal government finances. The federal deficit remained above $30 billion per year in the first half of the 1990s, reaching over $40 billion in both 1992–93 and 1993–94. In the second half of the 1990s, the deficit turned into a surplus that ranged from a low of nearly $3 billion in 1998–99 to a high of over $9 billion in 2000–01.
Inflation
This indicates how stable prices are in the economy. Starting above 5% in the 1990s, inflation plunged in the next few years to stabilize within the Bank of Canada target range of 1% to 3% per annum from 1995 onwards.
Housing Starts
This is a key leading indicator of the confidence and short-term trend in the economy. Housing starts averaged about 150,000 per year through these years. Activity was soft in 1995, 1996 and 1998, falling as low as 111,000 in 1995. Results were outstanding in 1992, 2001 and especially 2002, with 205,000 starts.
Canadian employment and income
This second set of data still looks at the overall economy, but zeroes in on employment and income indicators. Again, the general picture for the period 1990–2003 is positive with substantial growth in both employment and income per capita. Figure 1020 identifies changes in the levels of employment in the whole economy, and in the private and public sectors respectively, as well as selected indicators of income levels. Figure 1021 puts these indicators into one picture that shows their relative growth since 1990 on an index basis (1990–91 = 100). We offer below a brief comment on each of these indicators.
Figure 1020
Employment and income in Canada, 1990–91 to 2002–03
Figure 1021
Comparison of rate of change in key employment and income indicators in Canada, 1990–91 to 2002–03
Total employment
The number of people with a job in Canada increased from about 13 million in 1990–91 to nearly 15.6 million in 2002–03, a growth of 19%. This compares favourably with the overall population expansion of 13% for the same period. Except for 1991–92 and 1992–93, total employment has grown every year during the period.
Public sector employment
Public sector employment covers public servants for the federal, provincial and territorial governments, as well as employees of municipalities, school boards, health authorities and hospitals, and universities, and publicly owned business enterprises. Despite growth in the Canadian population and overall employment, this indicator has remained quite stable, remaining within a narrow range between a low of 2.65 million in 1997–98 and 1998–99, and a high of 2.94 million in 2002–03.
Private sector employment
This indicator is the difference between the level of total employment and public sector employment. Because public sector employment changed relatively little between 1990 and 2003, private sector employment clearly drove overall employment. Jobs in the private sector expanded from around 10 million in the early 1990s to 12.6 million in 2002–03, an increase of nearly 23%.
Nominal GDP per capita
This is the most general indicator of changes in income per person. This indicator rose from $24,450 in 1990–91 to $38,700 in 2002–03, an increase of nearly 60% in current dollars. The corresponding growth in real terms was about 30%. Most of the expansion occurred in the period from 1996–97 onwards (except for 2001–02 which witnessed a small decrease). Seventy per cent of the real increase took place during that seven-year period.
Average weekly earnings
This is a broad measure of weekly earnings (excluding overtime) for all employees. In 1990–91,[51] the average weekly earnings in Canada were estimated to be $520.89. This grew to $662.47 by 2002–03, an increase of about 27%. As Figure 1020 shows, the growth was fairly steady over the period.
A better measure in principle is average hourly wages. This indicator avoids some pitfalls inherent in average weekly earnings such as defining what a week is. However, the data series is only available from 1996–97, so our chart features the indicator with a more complete data set. Starting with 1996–97, average hourly wages increased from $15.61 to $18.04 in 2002–03, a growth rate of 16%, which is considerably greater than the 11% increase in average weekly earnings in this seven-year period.
Median after-tax family income
This is an important general indicator of the income actually available to families, expressed in constant 2002 dollars. After falling somewhat in the early 1990s, this measure stagnated in the range of $36,000 to $37,000 until 1998–99, after which it has grown steadily. By 2002–03 it reached $41,300, an increase of about 8% compared with 1990–91. From the low point in 1996–97 ($35,900), the increase was 15%.
Comparing employment and salary changes, federal public sector and the broader Canadian economy
We begin by looking at employment in our comparison of the experience of employment and salary change in the Canadian public sector with the performance of the overall economy. Figure 1022 shows the trends in federal employment between 1990–91 and 2002–03. Figure 1023 compares these trends with the broader Canadian economy.
We present two perspectives on federal employment. The first is a broad measure we call "federal public sector employment." This includes employees of the core public service domain, the separate employers, the Canadian Forces, and the regular and civilian members of the Royal Canadian Mounted Police. It excludes federal business enterprises, cultural Crown corporations, reservists, and such special groups as federally appointed judges, parliamentarians, employees of Parliament and ministers and students. The second measure we call "public service and separate employers."This covers the main departments and agencies that form the core public service and the separate employer domains.
On the first measure, as noted at the beginning of this chapter, we see federal public sector employment:
- above 350,000 early in the 1990s,
- falling to about 275,000 in 1998–99, and then
- recovering to about 315,000 in 2002–03.
For the narrower measure covering the core public service and the separate employer domains, we see a similar pattern with the total of:
- about 245,000 in the early 1990s,
- declining below 200,000 in 1997–98 and 1998–99, and
- growing back to about 235,000 in 2002–03.
Figure 1022
Federal and overall Canadian employment trends, 1990–91 to 2002–03
Figure 1023 provides the clearest picture of how changes in federal employment compare with those in the overall economy. From Figure 1023 it is evident that according to both the measures we are reporting, federal employment declined much more steeply than did employment in the rest of the economy. Since 1998–99, the rate of growth in the federal public sector has been greater than for the general economy. Nevertheless, even on the faster growing measure–that for the combined core public service and separate employer domains–federal employment had not reached its 1990–91 level by 2002–03. In contrast, private sector employment in 2002–03 was nearly 23% above 1990–91 levels, and overall public sector employment was 7% higher.
Perhaps the most informative indicator of the place of federal employment within the Canadian economy is federal public sector employment as a percentage of Canadian total employment. This proportion stood at about 2.7% in the early 1990s, declined to 1.9% in 1999–2000, and has since increased only modestly to 2.0%.
Figure 1023
Comparison of trends in federal employment to total Canadian employment,
1990–91 to 2002–03
The next area of comparison is total salary mass, also called the wage bill. Figure 1024 depicts the salary mass for the broader and narrower definitions of the federal public sector used in the previous section, as well as the total Canadian wage bill and the corresponding totals for the private sector, the broad provincial and territorial public sector, and the broad municipal public sector. As in the previous sections, Figure 1025 presents all these indicators in the context of comparative change on an index basis (1990–91 = 100).
Figure 1024
Federal, provincial, municipal and overall Canadian wage bills, 1990–91 to 2002–03
Figure 1025
Comparison of changes in federal, provincial, municipal and
overall Canadian wage bills, 1990–91 to 2002–03
As noted, we distinguish the broader federal public sector, which we define as the core public service and separate employer domains, and the Canadian Forces and the RCMP regular and civilian members, as well as a narrower version limited to just the combined core public service and separate employer domains.
For the broader federal public sector, the salary mass remained for a decade in a fairly tight range extending from a high of $13.3 billion in 1993–94 to a low of $12 billion in 1997–98. Thereafter the total rose substantially, reaching $17.3 billion in 2002–03. From its lowest point in 1997–98 to its highest in 2002–03, the increase was 43%. Looking then at the narrower public service, we see a very similar pattern, with the salary mass fluctuating from a high of $9.8 billion in 1993–94 to a low of $8.2 billion in 1997–98, followed by a steady increase to $12.4 billion in 2002–03. This represents a 36% growth rate from 1990–91 to 2002–03 and a 51% growth rate from 1997–98 to 2002–03.
The total Canadian wage bill rose even more rapidly, with increases each year. The total was $335 billion in 1990–91, and $527 billion in 2002–03, an increase of 57%. The growth in the Canadian private sector was even more pronounced at 69%. By contrast, the overall Canadian public sector increase was about 29%.
We also looked specifically at the broad provincial and territorial public sector.[52] The wage bill in this area grew by 36%, from about $46.6 billion in 1990–91 to around $63.2 billion in 2002–03. Like the federal public sector, the provincial and territorial public sector was very stable for several years, remaining in the range of $49 billion to $51 billion from 1991–92 to 1998–99. As was the case with the federal public sector, most of the increase in the salary mass took place in the last few years.
A very similar pattern applies to the broad municipal public sector.[53] From 1992–1993 to 1998–99, the total municipal salary mass in Canada remained in the range of $34.8 billion to $35.6 billion. Thereafter, this indicator rose to $40.4 billion by 2002–03. Compared with the 1990–91 level of $31.0 billion, this sector increased by 30%.
Looking at Figure 1025, we can observe these various indicators compared on an index basis (1990–91 = 100). The total federal salary mass grew between 1990–91 and 2002–03 by less than total Canadian wages or the private sector wage bill, somewhat faster than the broad municipal public sector, and by about the same proportion as the broader provincial public sector. However, looking only at the period from 1997–98, the rate of increase in both of our indicators of the federal salary mass has been greater[54] than in any of the other measures.
Total salary mass is an important indicator, but it combines the impact of two trends: growth in average earnings and increase in the number of employees.[55] We have already looked at relative changes in employment levels. We now examine indicators of change in average salaries.
Figure 1026 gives the experience from 1990–91 to 2002–03 for average wages in the broader and narrower versions of the federal public sector that we have used throughout this section, as well as average weekly earnings[56] and average hourly wages[57] for the Canadian private sector, and negotiated wage increases for both private and public sector unionized employers with at least 500 employees. Figure 1027 puts all of these indicators together for comparison on an index basis.[58]
Figure 1026
Overview of changes in average salary and earnings in the federal public sector and in the Canadian economy, 1990–91
to 2002–03
Figure 1027
Comparison of changes in average salary in real terms for the federal public sector and for selected indicators
covering the Canadian economy generally, 1990–91 to 2002–03
We define average salary as the total salary mass for a given group divided by the number of employees. For the broader version of the federal public sector, the average salary in current dollars rose from about $37,000 in 1990–91 to nearly $54,900 in 2002–03, an increase of 47%. For the combined core public service and separate employer domains, the change was a bit smaller. Average salaries in this context grew from $37,500 in 1990–91 to about $52,800 in 2002–03, a change of 41%. The corresponding rates of increase in constant dollars were 15.8% for the broader federal public sector, and 11.0% for the combined core public service and separate employer domains.
Looking at the graphs in Figure 1026 of these two federal average salary indicators, we note the fact reported in Volume Two that, particularly in real terms, nearly all of the observed increase occurred from 1998–99 onward.
So how do these increases compare in a macro sense with trends in the overall Canadian economy? Earlier in this chapter, we presented data on average weekly earnings excluding overtime, and average hourly wages. These are reproduced in Figure 1026. Average weekly earnings grew by about 27% between 1991–92 and 2002–03. During the period beginning in 1996–97, when both indicators are available, average hourly earnings grew faster.[59] In real terms, however, both indicators have exhibited little or no growth since 1991–92. (The better rate of increase was in average hourly earnings, which expanded minimally–by about 2% in real terms between 1996–97 and 2002–03.)
Looking now exclusively at the private sector, there is little difference in the pattern observed for average weekly earnings, where the increase was 29.4% in nominal terms and 1.9% in real terms over the years under review. For average hourly wages, we see a similar rate of growth in the private sector compared with the whole economy. In real terms, the private sector saw a small increase of 2.9% in average hourly wages between 1996–97 and 2002–03.
Finally, we look at the year-over-year wage increases in both the public and the private sectors among unionized employers with at least 500 workers on the payroll.[60] For the Canadian overall public sector the cumulative increases in nominal terms from 1990–91 to 2002–03 amounted to 27.2%; for the private sector, the corresponding cumulative increase was 36.5%. In real terms, these compounded increases were 0.2% and 7.5% for the public and private sectors respectively.
We bring these comparisons together on an index basis in Figure 1027. Note that the Figure expresses all the indicators in real terms (i.e. constant dollars). We observe that throughout the period since 1990–91 year-over-year negotiated wage increases for large private sector employers exceeded those for the overall public sector. This was also true specifically for changes in average salary in the federal public sector until 1999–2000 when our broad measure[61] of federal public sector average-salary change surpassed the private sector "agreements in force, year-over-year wage increases" cumulative change. In 2001–2002, the average salary for the combined core public service and separate employer domains also passed the private sector indicator. Average weekly earnings in the private sector grew more rapidly than federal average salaries in the early and mid 1990s. Overall, we can see that federal public sector average salaries increased substantially in real terms after 1997–98, whereas none of the other indicators did.
There are two important caveats to register regarding this analysis of average salaries. First, we know from Chapter 4 that part of the increase in federal public service average salaries arose from changes in the composition of the workforce. This trend toward a greater proportion of specialized knowledge workers within the public service, and more complex work throughout the service has contributed to higher salaries on average. We estimate in Volume Two that about 10 percentage points of the 41% increase in the nominal average salary in the combined core public service and separate employer domains between 1990–91 and 2002–03 was a result of this phenomenon. We do not have comparable information that would permit us to assess the extent to which a similar change in the mix and level of positions with other employers has also changed.
Second, the use of an index approach to compare rates of change in various indicators says nothing about how the starting point for federal average salaries compared with the private sector. We encountered in our research one estimate that bears on this issue. A Treasury Board Secretariat study dated October 1991 claimed that "on a national basis, average salaries in the federal public service were 8.3% behind those of the private sector."[62] Compared with the provincial government sector, federal average wages were found to lag by 2.4% on average. These estimates were based on reports from the Pay Research Bureau (PRB),[63] which carried out job-matching studies for selected occupations in comparison with private and other public employers.
We are not able to determine exactly how these lags were calculated, nor to assess how credible the survey methodology was or how rigorously the surveys were carried out. We do know that employer representatives considered that PRB samples were skewed in favour of large, unionized employers. If we accept this information as the worst-case comparison, Figure 1027 would show a lower starting point (91.7) for the federal public sector average salary line, which would narrow but not eliminate the gap between that line and the index line for year-over-year wage increases in private sector agreements in force. In other words, even assuming such a worst case–an 8.3% federal lag in 1990–91–by 2002–03 the cumulative salary increases benefiting federal public servants would have closed any gap between their average salaries and those in the general Canadian economy. If the original gap were in fact smaller, as seems likely in comparison with the private sector as a whole, this would suggest that a small premium in favour of the federal public sector had emerged by early in the 2000s.
Beyond this comparison of the evolution of average salaries, we also looked at differences in the distribution of salaries within the federal public service in comparison with the broader Canadian economy. With assistance from Statistics Canada, we compared the distribution of earnings as reported in the 2000 Census for four groups of employees in both the federal public sector and the overall Canadian private sector.[64] At the broadest level we found the following distribution of earnings:
|
|
Federal |
Private |
|
Less than $40,000 |
34% |
52% |
|
Between $40,000 and $100,000 |
64% |
40% |
|
Over $100,000 |
2% |
8% |
|
Total |
100% |
100% |
Figure 1028 presents a more detailed comparison using $20,000 income bands, again based on the 2000 Census. Figure 1029 provides the same comparison for the four occupational groups: managers, professionals, secretaries and clerks. In both Figures 1028 and 1029 the data relate to full-time, full-year (at least 48 weeks) workers for occupations with at least 400 Census observations in both the federal government and the private sector.
The general pattern shows a substantially smaller proportion of federal government than private sector workers earning below $40,000, a much greater share from $40,000 to $80,000, a similar share from $80,000 to $100,000, and a progressively smaller relative share for income bands above $100,000. When we look at the four groups, we see a more pronounced version of the general pattern for managers and professionals. For secretaries and clerks we see the private sector dominant below $40,000, and the federal government much more prominent above that income level.
Specifically, we find about 83% of federal government managers between $40,000 and $100,000, whereas in the private sector only about 50% fall in that range. Federal government professionals are even more concentrated, with about 78% between $40,000 and $80,000; for the private sector, professionals are more prominent both below $40,000 and above $80,000.
Figure 1028
Annual earnings in the federal government and the private sector by $20,000 income bands based on the 2000 Census
An important factor influencing the general pattern is the finding from Census data that women in these groups in the federal public sector earn more than women in the private sector. Except for managers and professional women over age 40, this phenomenon was more pronounced in the 2000 Census than in that of 1990. Figure 1030 illustrates these points. For female federal government managers over 40 years old the difference in 1990 was about 12%, but by 2000 it was negligible. For female federal government professionals under 40, the difference grew from about 11% in 1990 to about 13% in 2000. For female secretaries and administrative officers under 40 the gap increased from about 20% in 1990 to 27% in 2000. The corresponding difference for female clerks over 40 expanded from 18% in 1990 to 28% in 2000.
Figure 1029
Annual earnings in the federal government and the private sector by $20,000 income bands for managers,
professionals, secretaries and clerks, based on the 2000 Census
Figure 1030
Percentage wage differences between the federal government and the private sector for men and women above and
below age 40 in the 1990 and 2000 Censuses
Conclusion on comparisons between the federal public sector and the broader Canadian economy
In this section we have presented a variety of indicators regarding change in the general Canadian economy, in employment and in income. Within this context, we have provided a very broad comparison between the federal public sector and the overall Canadian economy in the evolution of employment, salary mass, average salary, and the distribution of earnings by $20,000 pay bands. From all of this description and discussion, we draw five conclusions.
Federal employment has become a smaller proportion of
Canadian employment and population
First, current employment levels in the federal public sector constitute a substantially smaller proportion of total Canadian employment than previously, despite relatively rapid growth in recent years. From a level of 2.7% of total employment in the early 1990s, federal public sector employment declined to 1.9% in 1999–2000, before increasing to 2.0% in 2002–03. To restore the federal public sector employment share of the early 1990s, the federal work force would need to be at least 100,000 more numerous.
Similarly, the federal public sector has shrunk as a proportion of the Canadian population. Whereas in 1990–91, there were 12.65 federal public sector[65] workers per 1,000 Canadians, by 2002–03 that rate had fallen to 9.99. Although the roles of the federal government have shifted over these years (giving up various operational responsibilities such as air traffic control, for example), it appears that the federal public service output per employed Canadian or per Canadian resident has improved over the years under review.
Federal salaries have grown faster than private sector salaries
Second, on a cumulative basis since 1990–91, average federal public sector salaries have increased by a larger proportion (15.8% in constant dollars) than private sector year-over-year wage increases in collective agreements in force (7.5% in real terms).
Federal salaries have also grown much more rapidly than private sector average weekly earnings, which experienced marginal real growth of only 1.9% up to 2002–03.The higher rate of increase in the federal public sector is concentrated in the period following 1997–98.
There is a small wage premium for the federal public sector
Third, at this overall level of comparison between sectors there is at most a small premium in favour of federal public sector salaries versus those paid in the Canadian private sector. Even if we were to accept as a worst case the 1992 Treasury Board Secretariat calculation based on Pay Research Bureau data which indicated that "on a national basis, average salaries in the federal public service were 8.3% behind those of the private sector," by 2002–03 the cumulative salary increases benefiting federal public servants would have closed any gap between their average salaries and those in the general Canadian economy.[66]
Federal wages are relatively egalitarian
Fourth, the distribution of federal public sector wages by level of annual earnings is more concentrated than for the overall Canadian private sector. As Figure 1024 illustrates, about 60% of federal public sector workers were earning between $40,000 and $80,000 as reported in the 2000 Census, versus only about 35% of private sector employees. In this sense, the federal public sector wages distribution could appropriately be described as relatively egalitarian. This pattern has prevailed at least since the time of the Glassco Commission whose report observed the same pattern of paying equal to or better than comparable jobs in private industry in the "lower grades" of the federal public service, with senior administrative (i.e. executive) and professional posts "at a marked monetary disadvantage in competition with private industry."
Women are better paid in the federal public sector than in the
private sector
Fifth, one factor in the pattern stated in the last paragraph is that in general women are relatively better paid in the federal public service than in private industry. This point is well illustrated in Figure 1030, which shows substantial gaps in favour of women in the federal public service versus the private sector, particularly in clerical and secretarial occupations.
This concludes our high-level comparison of indicators of employment and salaries in the federal public service and in the overall Canadian economy.
