Has wage setting changed in Canada? Evidence from the pre-pandemic 2020 Wage-Setting Survey

Key messages

This note presents results from the Bank of Canada’s 2020 Wage-Setting Survey. Although its original target was 150 firms, the Bank was able to survey only 42 before health restrictions were put in place in Canada to control the spread of the COVID‑19 virus.1

One of the survey’s main goals was to assess possible factors behind an unusual trend in Canada: wage growth from early 2018 to mid-2019 was subdued despite labour market tightening.2 By the time of the survey, however, this wage puzzle was already showing signs of resolving. Indeed, the findings suggest that labour markets were functioning appropriately, pushing wage growth up as they tightened. Specifically, firms noted:

• ad hoc wage increases to compete with wage offers by other employers
• increased labour market churn as a growing number of people quit their job
• wage growth driven by a rise in the market wage rate and firms’ need to attract and retain staff

At the same time, the survey results highlight changes in other factors that may have important impacts on wage dynamics:

• Increased use of digitalization and automation. Firms reported:
  • having better data to track workers’ performance
  • substituting capital for labour
  • seeing improved matching of jobs to workers through online recruiting resources
  • increasing their use of websites and social media to gather data on market wages
  These innovations allow businesses to set wages more efficiently. But they also provide potential channels for labour market outcomes to become more polarized—by creating winners and losers.
• Greater importance of non-wage compensation. Compared with results from the Bank’s 2007–08 Wage-Setting Survey, a greater proportion of firms reported that they include non-wage elements (e.g., flexible work arrangements, job enrichment) in their compensation packages to respond to shifts in workers’ preferences.3 Also, firms commonly reported bargaining over non-wage conditions when recruiting or trying to retain workers.
• Expanded labour supply. Firms noted they had better access to new or previously untapped labour, such as recent immigrants (including foreign students) and gig workers,4 to alleviate labour shortages and cost pressures.

The survey findings also reveal that some fundamental wage-setting practices of Canadian firms have not changed since 2007–08. This suggests that firms’ wage-setting behaviour in a low inflation environment is slow to change.

• More than half of firms that consider inflation when setting wages look at inflation over the past 12 months. In other words, they have adaptive or backward-looking expectations.
• Most businesses report no link between the timing of changes in their output prices and changes in their wages.
• Businesses largely adjust wages on an annual basis and mostly in the first half of the year. One possible implication of this could be that monetary policy changes that occur later in the year may have larger real effects on output.

1. Motivation and sample

Throughout 2018 and most of 2019, wage growth in Canada was below levels generally believed to be consistent with the late stages of economic cycles and tightening labour markets. Initially, one of the main motivations for the 2020 Bank of Canada Wage-Setting Survey (WSS20) was to explore factors behind this trend. However, by the time the survey was conducted, this wage puzzle was already showing signs of resolving, with some wage measures strengthening.

The survey explored whether factors in addition to tightening labour markets were having an impact on wage dynamics, such as increased use of digitalization and the importance of non-wage compensation. It also updated the Bank of Canada’s understanding of fundamental wage-setting behaviours among Canadian firms operating in a relatively low inflation environment. The results provide an interesting look at how firms were setting wages and thinking about labour market issues in the period immediately before the COVID‑19 pandemic. The survey results help fill in data gaps where little or no timely firm-level data exist. It will also be a useful baseline for future work.

Bank staff collected extensive qualitative information through in-person meetings with senior business leaders of private sector firms. Questions focused on the firms’ non-unionized labour force.5 The Bank originally targeted a sample size of 150 firms, but the interview period was cut short in early March 2020 by public health directives aimed at stopping the spread of the COVID‑19 virus in Canada. As a result, the sample for analysis includes responses from 42 firms. While the sample is small, it has enough diversity to allow for some high-level observations (see Appendix A for details). However, results presented in this note focus exclusively on overall findings rather than regional, sectoral or firm-size differences.

Where possible, we compare survey results from the WSS20 with those from:

• the 2007–08 Bank of Canada Wage-Setting Survey (WSS08), which was conducted to assess how firms made wage-setting decisions in a period of relatively tight labour markets (the WSS08 took place after years of slightly elevated inflation and before the global financial crisis)
• a subset of wage-setting questions added to the quarterly Business Outlook Survey in the first quarter of 2019 (BOS19)
• the same subset of wage-setting questions from the BOS19 added to an experimental online business survey conducted in the first quarter of 2019 (e-BOS19)6 to reach very small firms that were not targeted in either the WSS20 or WSS08

Appendix A presents information about the sample distributions of all four surveys by sector, region and firm size.

The remaining sections of this note present the survey results. Section 2 provides evidence of positive pressures on labour markets and wage growth before the pandemic. Section 3 describes how digitalization is impacting wage dynamics. Section 4 discusses the role of non-pay elements in firms’ total compensation packages. Section 5 shows that fundamental wage-setting practices are generally similar to earlier surveys. And Section 6 offers some concluding messages and suggests areas for future research.

2. Tightening labour markets were driving up wage growth before the pandemic

2.1 Firms’ labour market churn increased

One of the key drivers of wage growth is job change. Most firms in the WSS20 said they had seen an increased rate of turnover in their labour force in the three to five years before the survey (Chart 1). This is consistent with findings from Kostyshyna and Luu (2019a). Those authors show that in 2017 and 2018, labour market churn in Canada improved from the low levels reached during the 2015 oil price shock but was still below levels experienced before the global financial crisis.

Many firms in the WSS20 reported that employee quit rates had increased in the past three to five years because of:

• better external employment options
• greater competition for labour in a tightening labour market

Turnover at firms was also heavily influenced by more retirements, a result in line with Canada’s aging labour force.

2.2 Labour market pressures were pushing wages up before the pandemic

When asked about the factors that influence their wage-setting decisions, firms ranked market wage rates as one of the most important considerations (Chart 2). This is similar to results from the WSS08. Two other highly relevant factors were the firm’s:

• ability to attract and retain workers
• profitability

To a slightly lesser extent, a worker’s productivity, fairness considerations and inflation were also reported as key considerations. And less than half of respondents viewed other wages (e.g., public sector, union and minimum wages) as important when adjusting their firm’s wages. These results suggest that the factors considered when setting wages in 2019 and 2020—that ultimately influence inflation—were not materially different than those in 2007 and 2008.

As for the direction of the pressure these factors were placing on wage adjustments, many businesses reported that market wage rates and the ability to attract or retain workers—factors that are often related—had been pushing their wage adjustments up in the two years before the survey (Chart 3). These findings suggest that competition for workers was leading to wage and cost pressures that could prompt firms to raise prices and therefore contribute to higher overall inflation over time, which is consistent with a well-functioning labour market.

Firms generally also said that fairness considerations, inflation and worker productivity were pushing up workers’ pay. To a lesser extent, businesses reported that other wages, such as public sector, union and minimum wages, were also a source of positive wage pressure.

As well, the number of businesses that said profitability was contributing to recent wage changes was roughly the same as the number that said it was having a negative effect. This suggests that firms’ profitability was having a near-neutral overall impact on wage-setting behaviour.

3. Firms see digitalization affecting wage dynamics

3.1 Improved data about work performance are helping managers set wages

Firms in the WSS20 were asked to comment on the statement, “Compared with three to five years ago, we now have better data to track workers’ performance.” The results are summarized as follows:

• Almost three-quarters of firms confirmed that their data have improved (Chart 4, panel a). This idea of better data was often linked to the digitalization of work processes, which allows managers to more easily differentiate the output of individual workers or to compare work teams (e.g., making store-to-store comparisons).
• Still, firms noted that improved data have helped them track performance for some workers—such as shop floor staff, truck drivers and sales agents—but not for staff in occupations where worker production is more difficult to track.
• Several firms that rejected the statement said they had been doing this for more than three to five years—longer than the period in the statement. Together with the firms that said their data have improved, this suggests that performance tracking is widespread.
• Most firms that agreed with the statement said they use this improved data to make better decisions about pay (Chart 4, panel b). This reflects the possibility that labour markets are becoming more efficient and also potentially more polarized, as individual workers’ pay more fully reflects their marginal productivity.

Chart 4: Firms have access to better data, which they use to set wages

Chart 4: Firms have access to better data, which they use to set wages

3.2 Firms are substituting capital for labour, which may lead to upward pressure on wages

Two-thirds of firms acknowledged that they plan to or had already substituted capital for labour over the past three to five years (Chart 5, panel a). Such capital substitutions were through either investments in hard assets, such as machinery and equipment, or investments in softer assets like software and process reorganizations. Many firms said they were making the type of investments that generate the improved performance tracking data mentioned above. Those that rejected the statement about the substitutability of capital for labour were roughly split between those choosing not to adopt current available technologies (for various reasons) and those believing such substitution was impossible in their sector. A similar question in the WSS08 found that only 39% of firms had made or intended to make this type of substitution. This suggests a more rapid adoption rate of technology in recent years.

While most firms in both the WSS20 and WSS08 said automation had no impact on wages, about one-third in 2020 said this substitution was putting upward pressure on wages. In other words, capital investment in automation had boosted worker productivity and was allowing firms to sustainably pay certain workers a higher wage.

Chart 5: Automation is more widespread than in the previous survey

Chart 5: Automation is more widespread than in the previous survey

Note: WSS08 is the Bank of Canada’s 2007–08 Wage-Setting Survey; WSS20 is the Bank of Canada’s 2020 Wage-Setting Survey.

3.3 Online resources help firms reach new pools of labour and reduce cost pressures

More than half of respondents agreed that online human resources (HR) tools (e.g., Workopolis, Indeed, LinkedIn) enable them to more easily and efficiently match their job postings to available workers. Some firms stated that these HR resources had significantly shortened their recruiting cycle. Other firms said these platforms allow them to reach better candidates or find new pools of labour.

In response to a related question, slightly more than half of firms reported that they had recently gained access to a new or previously untapped source of workers and that this new access had allowed them to deal with labour shortages and associated cost pressures. The most widely cited sources of new labour were (in order of importance):

• recent immigrants to Canada
• students, including foreign students
• independent contract workers
• domestic firms contracted to do work

Together, these results suggest that digitalization is helping Canadian firms reach new sources of labour, which has helped control labour costs (Chart 6).7

3.4 Firms are shifting to online sources to gather market wage information

Almost all firms reported that market wage rates are somewhat or very important for setting their wages, and roughly half said they were obtaining this relevant information by word of mouth or other informal networks (e.g., contacting other local employers) (Chart 7). The second most common source of market wage information was national, industry or regional surveys. However, gathering market wage data through word of mouth and surveys was cited less often in the WSS20 than in the WSS08. The results suggest that these sources are being partly replaced by either job-matching websites or social media sources.8 Internet-based sources allow for easier comparison of wages across employers and may therefore make wage setting more efficient, with more firms offering wages closer to the prevailing market wage. The increased use of online sources for information on market wages may have important impacts on wage dynamics.

4. Many firms prioritize non-pay compensation

4.1 More firms now include non-pay elements in their compensation packages

The ranking of factors that firms view as essential components of labour compensation is unchanged from the WSS08 (Chart 8). For example, base pay and benefits remain at the top of the list and were reported by nearly all firms. However, some non–base pay elements have become more important. Roughly 8 out of 10 firms said they include the following in their compensation packages:

• incentive pay
• job enrichment opportunities (e.g., training, conferences)
• work arrangements (e.g., working remotely, flexible work schedules, extra vacation days)

Many firms linked the importance of flexible work arrangements to younger workers’ preferences. If this shift in preferences persists or becomes more prominent because of the pandemic, then, all else being equal, wage growth may be more subdued in the future. Indeed, the shift toward non-pay elements (i.e., more flexible work arrangements and job enrichment opportunities) may help explain weaker-than-expected wage growth during 2018 and 2019 when labour markets were generally tightening. Still, more firms than in WSS08 also recognized that to better align individual and corporate goals, it’s important to motivate workers to do their best work by providing incentive or bonus pay measures.

Pension plans or other forms of shared retirement savings programs were included in almost three-quarters of firms’ compensation packages. Some businesses mentioned that they have transitioned from a defined-benefit to a defined-contribution plan. Variable pay (e.g., commissions) pertains mostly to sales staff, and half of firms surveyed said they use it. For both elements—pensions and variable pay—reports of each being included in overall compensation increased slightly from WSS08 levels.

4.2 To recruit and retain staff, many businesses negotiated non-pay conditions

At the time of the survey, almost three-quarters of firms said they had greater difficulty recruiting staff than they did three to five years ago, and just under half reported that it was harder to retain workers (Chart 9). Roughly half of businesses said that when recruiting, they bargained with potential hires over pay or a combination of pay and non-pay conditions (Chart 10). Similarly, half of firms stated that they had negotiated with staff who had recently quit or threatened to quit—with roughly equal proportions saying discussions were about pay, non-pay or some combination of the two. Non-pay elements were usually related to work arrangements (e.g., flexibility, vacation time) and opportunities for progression. These survey results suggest that non-pay elements play a significant role in negotiations and that wage data alone may not fully reflect improvements in overall compensation.

5. Fundamental wage-setting practices remain unchanged

5.1 Most firms adjust wages once a year, typically in the first six months

Around 90% of firms surveyed in the WSS20, the WSS08 and the BOS19 reported adjusting wages on a regular cycle (Chart 11, panel a).9 In contrast, just under two-thirds of businesses in the e-BOS19 said they change wages regularly. This is likely explained by the fact that small and micro-sized firms, which made up most of the e-BOS19 sample, typically do not have dedicated HR teams and are less likely to have a formal wage structure or compensation packages.10

Chart 11: Most firms adjust wages annually in the first half of the year

Chart 11: Most firms adjust wages annually in the first half of the year

In all four surveys, almost all businesses that adjust wages on a regular cycle reported doing so once a year (Chart 11, panel b). When firms were asked about when they make their wage adjustments, most reported they do so in the first quarter, while some said they adjust in the second quarter.11 The wage changes were often linked to budgeting and planning cycles as part of projections for the upcoming fiscal year. This suggests that wage adjustments are not evenly distributed across the year and that wage stickiness may have a seasonal cycle. One implication, as suggested by Olivei and Tenreyro (2007, 2010), is that monetary policy changes that occur in the second half of the year will have larger real effects on output and employment than those that take place in the first half of the year.

Among firms in the WSS20 that adjust wages on a regular cycle, more than half said they also adjust wages outside of the scheduled dates and almost always upward, which is similar to findings in the WSS08 (Chart 11, panel c). However, in most cases, these adjustments are given to a small portion of employees—for example, on a case-by-case basis to individuals or to very small groups. These ad hoc increases are often related to changes in the minimum wage or to changes in wages offered by other employers—another potential channel for wage inflation.

5.2 Most firms continue to account for inflation when setting wages

Nearly 80% of firms in the WSS20 reported that inflation is an important factor when setting wages. This is similar to results from the WSS08 (Chart 12).

Results from both surveys are similar for firms that consider inflation when setting wages (Chart 13):

• Most of these firms do not follow a formal rule. In the WSS20, about one-third of these businesses reported automatically linking wages to inflation.
• When inflation is used, either formally or informally, more than half of firms look at past inflation exclusively.
• Almost all businesses taking past inflation into account use a 12-month window.

Generally, these findings suggest little change in firms’ wage-setting behaviour in the low inflation environments of 2007–08 and 2018–19. Insights from the WSS08 suggest that managers view wage setting based on past inflation as safer than wage setting based on expected future inflation. In particular, using past inflation avoids the potential costs of forecast errors—that is, if wages overshoot inflation but firms do not have the ability to restrain wages in future years, or if wages undershoot inflation and staff are disgruntled until the next wage adjustment date. In the WSS20, firms’ widespread reliance on past inflation for wage setting in a low inflation environment is consistent with theories of subdued wage growth and findings from other jurisdictions (Cormier et al. 2019). These survey results suggest that lower-than-expected wage growth during the period of labour market tightening in 2018 and early 2019 could be linked to low past inflation.

5.3 Most firms report no link between the timing of price and wage adjustments

When asked to describe the timing between changes to their wages and to their prices, roughly two-thirds of businesses interviewed in the WSS20 reported there was no link (Table 1). However, a small share of firms did say that their price changes follow wage changes. A few businesses reported either that their price- and wage-change decisions are made simultaneously or that there is a link between wage and price changes but no clear pattern. Overall, these results are remarkably similar to the findings in the WSS08 survey and suggest that the link between the timing of changes in output prices and wages continues to be weak.

6. Conclusions and areas for future work

Findings from the Bank’s 2020 Wage-Setting Survey show that pre-pandemic labour markets were tight and that this was leading to cyclical wage pressures, which is consistent with a well-functioning labour market. The results also suggest that some fundamental wage-setting practices of Canadian firms—related to the timing of wage adjustments and the linkages between wages, inflation and firms’ selling prices—have not changed since the Bank’s 2007–08 Wage-Setting Survey. In addition, the 2020 survey identifies changes in slow-moving factors that may have important impacts on wage dynamics:

• Firms reported increased use of digitalization related to their labour force. This may be making labour markets more efficient and concentrating wage gains among top employees.
• Workers’ preferences may have shifted slightly more toward non-wage compensation, such as flexible work arrangements.

Results from this survey can be a useful baseline for research on wage setting and labour market behaviour in a post-pandemic context. Further research is warranted, particularly in the areas of:

• Questions related to structural changes resulting from more employees working remotely
  • How will the improved work-life flexibility experienced during the pandemic play a role in compensation packages in the future?
  • How has remote work affected staff productivity? Do firms think productivity can be maintained if a higher share of their workforce continues to work remotely?
  • How do businesses see increased remote work impacting their investment decisions (labour versus capital)?
• Questions related to a different cyclical environment
  • How will the factors driving firms’ wage decisions change after the pandemic?
  • In a period when inflation is elevated, how does the link to wage growth change?
  • To what degree did firms accelerate their adoption of digital and other technologies during the pandemic? What impact did this have on firms’ labour decisions and how they set wages?

Appendix A: The 2020 Wage-Setting Survey sample

The sample for our analysis consists of responses from 42 firms. Efforts to reach the full intended sample of 150 firms were halted in early March 2020 because of public health directives aimed at stopping the spread of COVID‑19 in Canada.

Given the small sample, two caveats about these results are worth noting:

• We do not present any analysis by sector, region or firm size. Only overall results are discussed.
• This final sample does not perfectly match the sector, region or firm-size composition originally planned for the full sample (Table A-1). It underrepresents firms that are medium-sized, firms in the services sector and firms in the Prairies. The context of the underrepresentation of firms from the Prairies is particularly important to note—the energy sector, which dominates this regional economy, was facing challenges at the time of the survey. Therefore, the results may suggest slightly more labour market tightness than was actually present at the time of the survey.

Appendix B: Theories about changes in firms’ wage-setting behaviour

The 2020 Wage-Setting Survey asked respondents to evaluate the relevance of seven plain language statements. The main results are presented in Chart B-1. The exact wording of these statements, taken directly from the questionnaire, is shown in the subsequent text.

The first four statements in Chart B-1 were widely accepted. This generally suggests that firms are increasingly using hard and soft forms of assets, including those related to digitalization, that may impact wage dynamics (those results are summarized in Section 3). The three remaining statements received less agreement from respondents:

• Only one-third of firms agreed that the “next best” employment options for their staff were less attractive compared with three to five years ago. That is, most firms believed that their workers had either better or similar employment options; some referred to the recent strength of national or regional labour markets or to new sectors they now increasingly compete with for talent.
• Firms largely rejected the statement that consolidation had impacted (either increased or decreased) their bargaining power. This is at odds with research that suggests workers have lost some of their bargaining power (Krueger 2018).
• Only one in five businesses agreed with the statement that they were paying bonuses in good years rather than raising base pay, and they were not doing this three to five years ago.

Statements from questionnaire

1.1. In the past three to five years, improvements in our access to new or previously untapped sources of labour have allowed us to alleviate shortages and/or labour cost pressures.

• Not at all true
• Somewhat true
• Very true
• Don’t know/not sure
• Not applicable

1.2. Which sources did you tap into to alleviate shortages and/or labour cost pressures? (check all that apply)

1.3. Compared with three to five years ago, we now pay bonuses in good years rather than raising general wages so that our cost structure is more flexible.

• Not at all true
• Somewhat true
• Very true
• Don’t know/not sure
• Not applicable

1.4. The “next best” employment options for our workers are less attractive than they were three to five years ago.

• Not at all true
• Somewhat true
• Very true
• Don’t know/not sure
• Not applicable

1.5. Consolidation (fewer, often bigger competitors for workers/talent) in our industry has changed our bargaining power in setting wages (either increased or decreased).

• Yes, increased our bargaining power
• Yes, decreased our bargaining power
• No change in our bargaining power
• Don’t know/not sure
• Not applicable/no consolidation/more competitors for labour

1.6. Over the past three to five years, we have substituted capital for labour, or we intend to do so in the coming year (investment in machinery and equipment, software, process reorganizations).

• Yes, we have recently
• Yes, we intend to do so in the coming year
• Yes, both
• No, we could but have chosen not to
• No, this type of substitution is impossible to do
• Don’t know/not sure

1.7. If yes, did it have (or do you expect it to have) an impact on wages at your firm?

• Yes, upward pressure on wages
• Yes, downward pressure on wages
• No impact
• Don’t know/not sure
• Not applicable

1.8. Compared with three to five years ago, we now have better data to track workers’ performance.

• Not at all true
• Somewhat true
• Very true
• Don’t know/not sure
• Not applicable

1.9. If true, does this better data help you make better decisions about which workers merit a bigger (or smaller) wage gain?

• No
• Yes, sometimes
• Yes, most or all of the time
• Don’t know/not sure
• Not applicable

1.10. Online resources (such as Workopolis, Indeed, LinkedIn) allow us to match available workers to our job postings more easily and efficiently than three to five years ago.

• Not at all true
• Somewhat true
• Very true
• Don’t know/not sure
• Not applicable

Endnotes

References

1. Amirault, D., P. Fenton and T. Laflèche. 2013. “Asking About Wages: Results from the Bank of Canada’s Wage Setting Survey of Canadian Companies.” Bank of Canada Staff Discussion Paper No. 2013-1.
2. Cormier, A.-K., M. Francis, K. Hess and G. Poulin-Bellisle. 2019. “Drivers of Weak Wage Growth in Advanced Economies.” Bank of Canada Staff Analytical Note No. 2019-3.
3. Cunningham, R., V. Rai and K. Hess. 2019. “Exploring Wage Phillips Curves in Advanced Economies.” Bank of Canada Staff Discussion Paper No. 2019-8.
4. de Walque, G., J. Jimeno, M. Krause, H. Le Bihan, S. Millard and F. Smets. 2010. “Some Macroeconomic and Monetary Policy Implications of New Micro Evidence on Wage Dynamics.” Journal of the European Economic Association 8 (2–3): 506–513.
5. Druant, M., S. Fabiani, G. Kezdi, A. Lamo, F. Martins and R. Sabbatini. 2009. “How Are Firms’ Wages and Prices Linked: Survey Evidence in Europe.” European Central Bank Working Paper Series No. 1084 (August).
6. D’Souza, C., J. Fudurich and F. Suvankulov. 2021. “Small and Smaller: How the Economic Outlook of Small Firms Relates to Size.” Bank of Canada Staff Analytical Note No. 2021-14.
7. Kostyshyna, O. and C. Luu. 2019a. “The State of Labour Market Churn in Canada.” Bank of Canada Staff Analytical Note No. 2019-4.
8. Kostyshyna, O. and C. Luu. 2019b. “The Size and Characteristics of Informal (“Gig”) Work in Canada.” Bank of Canada Staff Analytical Note No. 2019-6.
9. Krueger, A. B. 2018. “Reflections on Dwindling Worker Bargaining Power and Monetary Policy.” Luncheon address at the Jackson Hole Economic Symposium, Jackson Hole, Wyoming, August 24.
10. Olivei, G. and S. Tenreyro. 2007. “The Timing of Monetary Policy Shocks.” American Economic Review 97 (3): 636–663.
11. Olivei, G. and S. Tenreyro. 2010. “Wage Setting Patterns and Monetary Policy: International Evidence.” Federal Reserve Bank of Boston, Research Department Working Papers 2010 Series, No. 10-8.

Acknowledgements

We thank Olena Kostyshyna and Corinne Luu for their contributions to the design of the questionnaire. We also greatly appreciate the valuable comments and suggestions from Joshua Slive, Yang Zhang and Brigitte Desroches. We thank our team members in the Regional Analysis Division for their hard work setting up interviews, providing feedback on the questionnaire and collecting results. We also thank Meredith Fraser-Ohman and Carole Hubbard for editorial assistance and Janki Shah and Bethany Madsen for research assistance.