Economic and Social Reports
A cross-cohort comparison of economic well-being during retirement years

Abstract

This study analyzes the extent to which pre-retirement lifestyles can be maintained into retirement years by comparing family incomes of five cohorts of individuals as they age from their mid 50s to late 70s. The cohorts considered were from 1984, 1987, 1990, 1993 and 1996. Three main results are uncovered from this study. First, median and average family incomes have generally risen across cohorts. Second, recent cohorts of retirees were able to maintain more of their pre-retirement family incomes compared with previous cohorts, partly driven by an increase in private pensions. Third, family incomes have become more stable across cohorts. The extent of the improvements across cohorts varies across the income distribution.

Author

Tahsin Mehdi is with the Social Analysis and Modelling Division, Analytical Studies and Modelling Branch, at Statistics Canada.

Introduction

The financial security of seniors has long been a concern, especially in recent years, against the backdrop of an aging population (LaRochelle-Côté, Myles and Picot 2008). Working-age individuals usually rely on earnings as a major source of income, but as they age and exit the labour force, they rely to a greater extent on pension benefits coming from the Canada Pension Plan (CPP) and Quebec Pension Plan (QPP), Old Age Security (OAS) and Guaranteed Income Supplement (GIS), registered pension plans (RPPs), and Registered Retirement Savings Plans (RRSPs). This places the issue of financial security of seniors front and centre since it affects not just beneficiaries of pensions, but also workers currently contributing to the CPP, the QPP and RPPs.

Business cycles, in conjunction with the policy landscape, ultimately shape people’s economic experiences. For example, retirees from the 2010s likely had a very different experience than their counterparts from the 1990s. It is important to measure these potential cross-cohort differences in economic outcomes to identify gaps and better understand future policy needs. Using a longitudinal database, this study follows various cohorts of individuals 54 to 56 years of age until they reach ages 78 to 80 and assesses (1) the extent to which pre-retirement living standards are maintained through retirement years and (2) family income stability before and after retirement. Both are important in assessing economic well-being. Income growth alone is not sufficient to conclusively say whether recent cohorts are better off than previous ones. Income instability can also affect well-being in several ways. Unless offset by the tax and transfer system, unexpected fluctuations in family income can cause uncertainty and adversely affect well-being (Morissette and Ostrovsky 2005).

Cross-cohort studies on income replacement and income stability of seniors have been done in the past (LaRochelle-Côté, Myles and Picot 2008, 2010), but the underlying data in those studies are now more than a decade old and may no longer reflect the experiences of more recent cohorts of retirees. Those studies identify individuals at ages 54 to 56 and again at ages 78 to 80, with their incomes compared at those two points in time.^Note  Since Canadians usually retire at age 65, focusing on such a narrowly defined concept of “retirement years” (i.e., 78 to 80 years of age) misses a lot of crucial information leading up to ages 78 to 80 and, more importantly, fails to capture the experiences of retirees who died before reaching that age. Therefore, while this study still maintains 54 to 56 years of age as a suitable proxy for “pre-retirement years,” it opts for a more comprehensive concept of “retirement years” by averaging measures from ages 65 to 80 whenever comparing retirement and pre-retirement indicators, rather than limiting the comparison to just ages 78 to 80.^Note 

While previous Canadian studies on income replacement and stability are not recent, they offer useful information and will partially guide this study’s methodology. LaRochelle-Côté, Myles and Picot (2008) found that lower-income individuals experienced higher levels of income instability than higher-income individuals during their late 50s and early 60s, but the gap closed and incomes became more stable later in life when individuals started receiving the CPP or QPP and OAS and GIS. Morissette and Ostrovsky (2005) also found government transfers to play an important role in reducing income instability. LaRochelle-Côté, Myles and Picot (2010) found that more recent cohorts improved their income positions relative to previous ones, in part because of higher income from private pensions such as RPPs and RRSPs. Schellenberg and Ostrovsky (2009) found a strong correlation between RPP coverage and the likelihood of being retired but did not find any significant difference in income replacement rates between RPP and non-RPP members, although the latter typically retired at older ages.

The topic of income replacement and stability warrants another examination because of recent developments in the labour market, such as the increasing share of women in education, health care, social services and public administration, which could have contributed to higher incomes among more recent cohorts of female retirees resulting from higher RPP coverage. To put this into perspective, the share of women 25 to 54 years of age working full time in public administration increased from 29% in 1980 to 48% in 2020. Similarly, the share increased from 48% to 70% in educational services, while it increased from 71% to 80% in health care (Statistics Canada, 2022a). Moreover, the average job tenure of women in these industries with relatively higher RPP coverage has increased over time (Statistics Canada, 2022b). The purpose of this study is to not only update the results of previous studies with more recent data and more comprehensive indicators, but also analyze the issues within the purview of current policy.

This study highlights three major findings. First, median and average family incomes have generally risen across cohorts. As a result, recent cohorts of retirees have a higher family income at their disposal than previous ones. Second, income replacement rates at ages 65 to 80—family income at ages 65 to 80 as a share of family income at ages 54 to 56—have improved across cohorts, partly driven by an increase in income from RPPs and RRSPs. Third, family income has become more stable for recent cohorts.

Data

This study pulls together data from the Longitudinal Administrative Databank (LAD), which is a 20% sample of all tax filers and their families in Canada. A prominent feature of the LAD is that tax filers are linked across years so their income profiles can be traced over time starting from 1982 (if available). The database, which had annual data up to the year 2020 at the time of this study, contains a rich set of information on income and tax components, as well as basic demographic information. Although the LAD is an individual-level database, it contains useful information on the families of these individuals, such as family size and family income.

The population of interest for this study consists of individuals who were 54 to 56 years of age in 1984, 1987, 1990, 1993 and 1996. Their family incomes were tracked for 24 years, at the end of which they would be 78 to 80 years of age. Four additional cohorts—1999, 2002, 2005 and 2008—were considered, but a complete age–income profile up to ages 78 to 80 could not be constructed since these cohorts are younger.^Note  Individuals who passed away before reaching ages 78 to 80 are included in the sample and hence their incomes still contributed to all the indicator calculations before their death. For this reason, the results of this study are best interpreted as reflecting the economic well-being of all individuals within a cohort, taken collectively.

After-tax family income is the income concept used in this study. Income includes earnings, pensions, investments, net capital gains and other sources. Family income captures potential resource sharing between family members and therefore better measures economic well-being than do individual-based income measures.^Note  In line with the well-being literature, family income is adjusted for family size by dividing it by the square root of the family size to account for economies of scale available to individuals in larger families (OECD 2013). All incomes are converted to 2020 constant dollars using the national all-items Consumer Price Index from Statistics Canada (2022c). To identify specific income sources, certain portions of this study disaggregate family income into its constituent parts. Since not all income sources are taxable, income before taxes is used in those parts of this study.

Methods

Since the income that a family receives in a given year t is partly influenced by positive or negative shocks that do not reflect its earnings capacity, this study uses a measure of “permanent” income that averages family income over three years when considering year t. For example, the permanent income of someone 55 years of age is calculated as the average of their annual adjusted after-tax family incomes when they were 54, 55 and 56 years of age. Further references to “income” in this study should be interpreted as permanent after-tax family income adjusted for family size, unless stated otherwise.

One of the indicators used in this study for measuring economic well-being is the income replacement rate, which is an individual’s income at any age as a share of their income when they were 54 to 56 years of age. Replacement rate in this study is calculated in three steps. First, the rate is calculated at the individual level for each age (e.g., 54 to 56, 55 to 57, …, 77 to 79, 78 to 80). Second, the median or average of the rates at each age is computed across individuals. Third, to calculate the replacement rate at ages 65 to 80, the average of the median rates or the average of the average rates is taken across ages 64 to 66 through 78 to 80.

To capture distributional variations, the initial samples of individuals aged 54 to 56 were divided by sex and into five groups of roughly equal sizes ranked by their income—income quintiles. The income quintile thresholds are calculated separately for women and men to better capture the differential patterns in the indicators.

A common problem with tax data is the inadequate coverage of lower-income individuals, who are less likely to file taxes (although this has improved over time). To overcome this obstacle, the study follows LaRochelle-Côté, Myles and Picot (2008) and excludes individuals who initially had incomes below $10,000 (2005 dollars), which amounts to approximately $13,000 in 2020 constant dollars.^Note  Because of changes in the data (e.g., new income components) and methodology, some of the indicators presented in this study may not be directly comparable with those of LaRochelle-Côté, Myles and Picot (2008, 2010), but, nonetheless, underlying patterns, trends and relative differences can still be compared.

This study also assesses another important aspect of financial security: income instability during retirement. Income instability can have a detrimental effect on retirees’ well-being because it creates uncertainty and may adversely affect consumption levels. To quantify income instability, this study estimates the mean absolute deviation ($MAD$ ) of lifecycle-adjusted family incomes. The $MAD$ is a simple measure of instability and has an intuitive interpretation. It simply measures the deviation from the average, in percentage terms. For example, a $MAD$ of 0.1 would indicate that individuals, on average, deviated from their average income level by 10%. Therefore, a higher $MAD$ value would be associated with greater financial instability.^Note  Annual adjusted after-tax family income is used to analyze income instability since the interest is on examining the degree of year-over-year variation in income levels at different stages of retirement. The $10,000 (2005 dollars) lower-bound restriction from earlier is relaxed in this part of the study. A version of the method described below has been applied by Morissette and Ostrovsky (2005) and LaRochelle-Côté, Myles and Picot (2008).

Income instability is examined separately by sex and for five age groups (55 to 59, 60 to 64, 65 to 69, 70 to 74, and 75 to 79). The method starts out by estimating the following fixed-effects regression model separately by sex and age group:

$$y_{it}={\beta }_0+\beta X_{it}+e_i+u_{it},\text{\hspace{1em}(1)}$$

where $y_{it}$ is the natural logarithm (log) of annual adjusted after-tax family income of individual $i$ at time $t$, $X_{it}$ is a vector of observable personal characteristics (age and age squared in this study), $e_i$ is the individual fixed effect, and $u_{it}$ is the error term. After the model is estimated, lifecycle effects are removed by replacing the actual incomes, $y_{it}{}_{,}$ by the lifecycle-adjusted family incomes, $y_i{t}^{*}$:

$$y_{it}^{*}=y_{it}-{\widehat{\beta }}_0-\widehat{\beta }{X}_{it}={\widehat{e}}_i+{\widehat{u}}_{it}.\text{\hspace{1em}(2)}$$

If $N$ individuals are observed across $T$ years and ${\overline{y}}_i^{*}$ is the average of the lifecycle-adjusted family incomes of individual $i$ over $T$ years, the $MAD$ can be calculated as

$$MAD=\frac{1}{NT}{\displaystyle \sum }_{i=1}^N{\displaystyle \sum }_{t=1}^T\left|y_{it}^{*}-{\overline{y}}_i^{*}\right|.\text{\hspace{1em}(3)}$$

After-tax family incomes of retirees have generally increased across cohorts

Median adjusted after-tax family incomes at various ages for nine cohorts are reported in Table 1. It is difficult to compare the incomes observed across cohorts because of business cycle effects in different periods of time. For instance, someone 60 years of age in 2008 would have had a different experience than someone 60 years of age in 1990. In Table 1, two observations can be made. First, as expected, median incomes tend to fall with age, driven by the fact that earnings decrease as individuals exit the labour market (LaRochelle-Côté, Myles and Picot 2008, 2010). Second, incomes during retirement years have generally increased across cohorts. For example, women 65 to 67 years of age in 1995 (1984 cohort) had a median income of $38,200. At the same stage in life, the 1987 cohort’s median income slightly drops to $36,600. However, this amount gradually increased starting from the 1990 cohort. And by 2019, women 65 to 67 years of age (2008 cohort) had a median income of $48,900. A similar pattern holds for men with the only difference being higher income levels than women.

Average incomes, unlike the median, are more susceptible to outliers. But nonetheless, it is important to ensure that the conclusions drawn from this study are robust to different measures so average incomes still add value to the analysis (Table 2). The averages tell a similar story in terms of cross-cohort income gains, but, within each cohort, the decline in income with age is not as apparent as it is with median incomes. This is because average incomes are influenced by relatively few individuals who might have substantially more income than others.

 

Table 1 and Table 2 illustrate differences in income across cohorts, on average, but they say nothing about the actual distribution of income. To better understand how the income gains across cohorts are distributed, median and average incomes are calculated for the bottom, middle and top income quintiles in Table 3. For ease of presentation, only the incomes at ages 54 to 56 and ages 65 to 80 are reported for the 1984, 1987, 1990, 1993 and 1996 cohorts. Income at ages 65 to 80 refers to median and average incomes averaged across ages 64 to 66 through 78 to 80. Table 3 shows that the income gains at ages 65 to 80 observed across cohorts are, in absolute terms, more pronounced in the middle and top quintiles than in the bottom quintile.

Women 54 to 56 years of age in the top quintile had a median income of $80,500 and average income of $96,500 in 1984. By ages 65 to 80, the average of the median incomes was $59,600, while the average of the average incomes was $80,600. For the 1996 cohort, pre-retirement median income grew by 12%, while average income grew by 16%. The average of the medians and average of the averages at ages 65 to 80 also grew by similar rates, respectively. Similar patterns hold for men except that their income levels were higher.

Median and average incomes of women aged 54 to 56 years in the middle quintile were around $44,000 in 1984. The average of the median incomes at ages 65 to 80 was $35,100, while the average of the average incomes was $38,700. The pre-retirement incomes grew by around 2%, while the retirement incomes grew by 11% for the 1996 cohort. Men had higher incomes and marginally higher growth rates.

Even though women and men from the bottom quintile saw a decrease in pre-retirement median and average incomes across cohorts, the average of median incomes and average of average incomes at ages 65 to 80 remained similar or increased.

Registered pension plans and Registered Retirement Savings Plans represent increasingly important sources of retirement income

So far, the data have shown that incomes have increased across cohorts and most of the gains were made by the middle and top quintiles. But from a policy standpoint, a closer examination of the underlying income sources is needed to gain insights as to which sources are actually driving improvements. Since not all incomes are taxable, decomposing after-tax income is not possible. For this reason, Table 4 decomposes average income before taxes (median income cannot be decomposed) into five sources: (1) earnings (wages and self-employment), (2) pensions (CPP and QPP, OAS and GIS, RPPs, and RRSPs), (3) investments, (4) net capital gains and (5) other sources.

 

As was the case with after-tax income, much of the cross-cohort gains in income before taxes were made by women and men from the middle and top quintiles (Table 4). Pre-retirement incomes at ages 54 to 56 increased significantly for the top quintile across cohorts, and an increase in net capital gains played a major role.^Note 

At ages 65 to 80, women in the top quintile from the 1984 cohort averaged $100,000, while men averaged $111,800. For the 1996 cohort, these amounts increased by $12,700 and $12,600 for women and men, respectively, and much of the gains were because of increases in earnings, RPPs and RRSPs, and net capital gains, with earnings being the dominant source for both women and men. The proportion of women and men with earnings after age 65 increased across cohorts, notably for the top half of the distribution. These increases were mostly because of wages rather than self-employment. The differences in the proportion of individuals with earnings across cohorts start to disappear after age 75.

At ages 65 to 80, women in the middle quintile from the 1984 cohort had an average income before taxes that amounted to $42,900, which steadily increased to $46,900 for the 1996 cohort. More than $3,000 of this increase came from an increase in RPPs and RRSPs. To put this into context, the share of total income from RPPs and RRSPs increased from 29% for the 1984 cohort to 33% for the 1996 cohort. For men, earnings played a more prominent role. Their average income before taxes increased from $46,600 (1984 cohort) to $51,400 (1996 cohort), with $3,300 coming from a cross-cohort increase in earnings and $2,300 coming from an increase in RPPs and RRSPs (investments decreased by $1,100).

Income before taxes at ages 54 to 56 in the bottom quintile decreased by more than $1,000 from the 1984 cohort to the 1996 cohort, but retirement income before taxes at ages 65 to 80 remained similar for women across the different cohorts, while it increased by $2,800 for men from the 1984 cohort to the 1996 cohort. Contrary to what was observed in the middle and top quintiles, RPPs and RRSPs did not contribute to the cross-cohort growth of retirement incomes at ages 65 to 80 in the bottom quintile. OAS and the GIS were the main source of income for retired women in the bottom quintile, making up 40% of total income for the 1996 cohort. OAS and the GIS made up one-third of total income for men in the bottom quintile across cohorts, but a rise in earnings, especially wages, helped increase incomes of men 65 to 80 years of age in the bottom quintile.

Recent cohorts of retirees maintain more of their pre-retirement family income compared with previous cohorts

Median and average incomes are relatively simple yet effective indicators for measuring economic well-being, but they do not fully capture individual welfare. A complementary tool is considered in this section: the income replacement rate, i.e., the adjusted permanent after-tax family income at a given age expressed as a share of income at ages 54 to 56. The average of the median replacement rates and the average of the average replacement rates across ages 64 to 66 through 78 to 80 (referred to as ages 65 to 80) are reported in Table 5. Just like income, the median replacement rates reflect the experience of the typical individual, while average replacement rates are more prone to being influenced by outliers. However, both are valid measures in assessing well-being and serve as complementary measures, especially in ensuring results are robust.

On the basis of the median replacement rate, women at ages 65 to 80 belonging to the 1984 cohort were able to maintain around 85% of their income from when they were aged 54 to 56. Members of the 1996 cohort were able to maintain 90% of their income.^Note  The corresponding percentages for the average replacement rate equal 97% (1984 cohort) and 103% (1996 cohort). Regardless of the metric used, increases in replacement rates were also observed for men. In general, the averages of the average replacement rates were generally higher than the average of the median rates.

Regardless of cohort or sex, income replacement rates were likely to be higher for lower parts of the distribution, and this might appear counterintuitive since previous sections showed that the top parts of the distribution saw the most gains in average and median incomes. Moreover, individuals in the bottom quintile had replacement rates all exceeding 100%, thereby implying that the average person in the bottom quintile ended up with more income during retirement years than they had at ages 54 to 56. These patterns are actually consistent with Table 3, which showed a slight decrease in pre-retirement average and median incomes of the bottom quintile across cohorts but a slight increase in retirement incomes.

The average of the median replacement rates for the 1996 cohort of women in the middle quintile was 86%, which is about a seven percentage point improvement compared with their 1984 counterparts. Similar rates were observed for men. The averages of the average replacement rates tell a similar story but, as mentioned before, are higher than the average of the median rates.

The replacement rates for the top quintile were lower than the other parts of the distribution and did not increase by much across cohorts, and this may seem counterintuitive because the top quintile saw its average and median incomes grow the most. However, this is entirely consistent with Table 3, which showed that individuals in the top quintile saw their incomes at ages 54 to 56 and ages 65 to 80 grow by similar rates across cohorts. The replacement rates and income measures each tell one side of the story, but taken together, they put economic well-being of retirees into a broader context.



Family incomes have become more stable for recent cohorts

Up to this point, the focus has been exclusively on permanent income, which essentially smooths out income over a three-year period in this study. Year-to-year fluctuations in income can cause uncertainty for individuals, so income instability can have a significant impact on financial security. As described earlier, income instability in this study is measured by the $MAD$ of lifecycle-adjusted annual after-tax family income obtained from fixed-effects regression models. It is measured separately by sex and for five age groups: 55 to 59, 60 to 64, 65 to 69, 70 to 74, and 75 to 79.

Table 6 reports the $MAD$ indicator, which measures the absolute deviation from the average income level in percentage terms.^Note  A relatively higher $MAD$ indicates more instability. Regardless of sex, cohort or income position, the $MAD$s indicate that income tends to become more stable with age (lower year-over-year variations) when the CPP and the QPP, as well as OAS and the GIS, kick in. The bottom quintile faces more instability before age 65 than the middle or top quintile, but this instability declines to similar levels as the middle quintile during retirement years when individuals start receiving the CPP and the QPP and OAS and the GIS.



Comparisons across cohorts indicate that incomes have become more stable. For instance, among women 55 to 59 years of age in the bottom quintile, the $MAD$ decreased from 0.293 (1984 cohort) to 0.217 (1996 cohort). A similar result applies to their male counterparts. For the middle and top quintiles, the reductions are not quite as large. In contrast, income instability for individuals aged 70 years or older in the top quintile rose across cohorts, but the increases amount to, at most, a few percentage points and thus are not sizable.

Conclusion

This study assessed economic well-being across five cohorts of retirees or those approaching retirement, by using a longitudinal database and following family incomes of individuals in their mid 50s into their late 70s. The four main cohorts considered in this study consisted of individuals 54 to 56 years of age in 1984, 1987, 1990, 1993 and 1996. Three main findings came out of this study.

First, median and average family incomes have increased across cohorts, albeit not strictly because of business and economic cycle effects. However, living standards varied across the income distribution, with the top half of the distribution experiencing greater income growth. A combination of increases in earnings, RPPs and RRSPs, and net capital gains drove income growth in the top quintile across cohorts of retirees. For retired women in the middle quintile, cross-cohort income growth was mainly driven by increases in RPPs and RRSPs, while for men, it was a combination of RPPs and RRSPs, as well as wages. In the bottom quintile, women and men 54 to 56 years of age saw their family incomes decrease slightly across cohorts but by ages 65 to 80, their incomes had either remained similar across cohorts or increased (in the case of men). OAS and the GIS played a key role in keeping the income of retired women stable across cohorts. OAS and the GIS, combined with an increase in wages, helped the income of retired men in the bottom quintile grow.

The second finding is that recent cohorts of retirees were able to maintain more of their pre-retirement family incomes compared with previous cohorts. The average of the median replacement rates for women across ages 65 to 80 increased from 85% (1984 cohort) to 90% (1996 cohort). For the top quintile, the replacement rate remained at around 72% between the two cohorts. The middle quintile saw its replacement rate increase from 80% to 86%. For the bottom quintile, it increased from 119% to 129%. Similar trends were observed for men.

The third finding is that family incomes have become more stable for recent cohorts. Using the $MAD$ of lifecycle-adjusted family incomes as the metric, this study showed that family incomes have stabilized across cohorts and that family income becomes more stable as individuals age and start to receive government transfers.

Issues surrounding the well-being and financial security of seniors will continue to evolve against the backdrop of an aging population, which will test the resilience of the pension system. The extent to which future cohorts of retirees can maintain their pre-retirement lifestyle remains to be seen.

Appendix

This study followed an unbalanced panel of individuals from ages 54 to 56 until ages 78 to 80. This means that individuals remained in the sample even if they passed away at some point in between those ages. There may be concerns that this might bias some of the estimates because of the correlation between economic and health outcomes, whereby the unbalanced panel might overrepresent higher-income individuals. To address this issue, this study repeated the analyses using a balanced panel and found no material difference in the results. In the 1984 cohort, 62% of women and 50% of men are retained in the balanced panel, while in the 1996 cohort, 66% of women and 57% of men are retained.

The income instability analysis was based on log incomes, and since logs of negative numbers are undefined, only individuals with strictly positive incomes were part of the analysis. Besides excluding individuals with no income, this also effectively excluded self-employed individuals with negative incomes. Therefore, the income instability analysis was repeated using income levels as opposed to logs to capture more individuals, but the results remained the same as the original analysis.

References

LaRochelle-Côté, S., Myles, J., & Picot, G. (2008). Income Security and Stability During Retirement in Canada. Analytical Studies Branch Research Paper Series, Catalogue no. 11F0019M – No. 306. Ottawa: Statistics Canada. https://www150.statcan.gc.ca/n1/en/catalogue/11F0019M2008306

LaRochelle-Côté, S., Myles, J., & Picot, G. (2010). Replacing Family Income During the Retirement Years: How are Canadians Doing? Analytical Studies Branch Research Paper Series, Catalogue no. 11F0019M – No. 328. Ottawa: Statistics Canada. https://www150.statcan.gc.ca/n1/pub/11f0019m/2010328/aftertoc-aprestdm2-eng.htm

Morissette, R., & Ostrovsky, Y. (2005). The Instability of Family Earnings and Family Income in Canada, 1986-1991 and 1996-2001. Canadian Public Policy 31(3): 273-302. https://doi.org/10.2307/3552442

OECD (2013). OECD Framework for Statistics on the Distribution of Household Income, Consumption and Wealth. Paris: OECD Publishing. https://doi.org/10.1787/9789264194830-en

Schellenberg, G., & Ostrovsky, Y. (2009). Pension Coverage, Retirement Status, and Earnings Replacement Rates Among a Cohort of Canadian Seniors. Analytical Studies Branch Research Paper Series, Catalogue no. 11F0019M – No. 321. Ottawa: Statistics Canada. https://www150.statcan.gc.ca/n1/pub/11f0019m/11f0019m2009321-eng.htm

Statistics Canada. (2022a). Table 14-10-0023-01 Labour force characteristics by industry, annual (x 1,000). https://doi.org/10.25318/1410002301-eng

Statistics Canada. (2022b). Table 14-10-0055-01 Job tenure by industry, annual. https://doi.org/10.25318/1410005501-eng

Statistics Canada. (2022c). Table 18-10-0005-01  Consumer Price Index, annual average, not seasonally adjusted. https://doi.org/10.25318/1810000501-eng