Jul 13, 2023

That 70s show: The evolution of income inequality in Canada

A protestor holding a placard stating the inequality of capitalism during the Occupy Toronto Movement on October 17, 2011 in Toronto, Canada. © arindambanerjee via Shutterstock


This brief looks at the evolution of inequality going back to 1976. Drawing on Statistics Canada’s Canadian Income Survey, it reviews changes in the distribution of income by decile (groupings of ten percent of households ranked by income from lowest to highest income), and asks a hypothetical question: what would today’s incomes look like if they followed the same distribution pattern as in the late 1970s?

Key findings:

The distribution of market income (wages and salaries, self-employment and investment income) has become  much more unequal than in the 1970s: 

  • By the late 2010s, individuals in the top 10% received over $23,600 more per person than they would have under a 1970s-like income distribution.
  • The next 10% also saw an improvement, securing nearly $3,900 more per person.
  • However, the bottom 70% of the income distribution were worse off, with a pronounced impact on the lowest deciles.
  • As of 2021, the top 10% of households commanded 30% of all market income, an increase of 5 percentage points relative to the 1970s. 

Income taxes and transfer payments offset some of the much greater market inequality but not fully:

  • In the late 2010s, individuals in the top 10% were better off by over $8,200 per person in after-tax income when compared to a 1970s distribution.
  • Those in the bottom 10% saw a modest improvement after taxes and transfers, with an increase of $473.
  • Middle-income earners (those in the third to seventh deciles) were worse off compared to a 1970s distribution, facing reductions between $1,000-2,000 per person, or 2-5% of income.

Inequality based on income after taxes and transfers was at its worst in the early 2000s, the height of neoliberal policy implementation in Canada. Some progress has been made since then to improve incomes at the bottom but much remains to be done. The COVID-19 pandemic years of 2020 and 2021 demonstrated the potential for robust income supports to alleviate poverty and reduce inequality.

Inequality and the neoliberal era 

Beyond the rise of Disco and Star Wars, the late 1970s are widely regarded by economists as the end of the post-war “Golden Age” of economic growth and broad-based prosperity. Economic life was far from perfect in the 1970s, of course, and such depictions by economists overlook the more severe discrimination faced by women, people of colour and LGBTQ+ people. In broad macroeconomic terms, however, it was an era when labour power was strong, unemployment was low and real incomes were growing.

The early 1980s launched the neoliberal era—a shift to market-oriented policies like tax cuts and smaller government, free trade and deregulation—with the elections of Margaret Thatcher in the United Kingdom and Ronald Reagan in the United States. By 1984 and the election of the Mulroney government, Canada made its own shift to neoliberal policies, including the Canada-US Free Trade Agreement, privatization, preferential tax treatment of capital gains, the Goods and Services Tax (GST) and reduced public spending. In the 1990s and early 2000s federal and provincial tax and spending cuts reduced the size of the public sector and favoured high-income earners.

Also important in this era was a shift to higher interest rate policies by central banks to fight inflation, a development that weakened economic performance, including a deep recession in the early 1990s, worsened unemployment and wiped out income at the bottom of the distribution. Globalization, technological change, declines in union membership all intersected with neoliberal policies over this time period to deliver a more unequal income distribution. As a result, the push towards market-oriented policies was accompanied by rising inequality—a growing gap between the richest and the poorest. 

Neoliberal ideas hit their peak after the turn of the millennium. Erosion of faith in the marketplace came in the aftermath of the 9/11 terror attacks in the United States and the crash of the tech bubble. A full-blown financial crisis in 2008-09 further battered neoliberalism and revived beliefs in the public sector as a force for good. Indeed, in the face of challenges like climate change, ensuring affordable housing, fighting the COVID-19 pandemic and reducing inequality, the public sector is indispensable. 

Economist Lars Osberg’s sweeping review of inequality trends spanning 75 years in Canada notes that it is unclear where we head next in the wake of the COVID-19 pandemic: 

The longer-term issue, therefore, is one of political economy—whether the “new normal” role for government of the post-COVID-19 world pivots back to Zombie Neo-Liberalism or is replaced by a new vision. It is unlikely that the political economy of the post-COVID-19 world can be shaped by the same faith in low taxes and market-based individualism that fuelled Neo-Liberalism.

Inequality trends by decile

Rolling back the clock to 1976 (the first year in the Statistics Canada income series used below) we can measure income in several ways: market income (taxable income in the form of wages and salaries, self-employment and investment income); total income (market income plus income transfers received from government); and after-tax income (total income less income taxes paid, typically called disposable income).1[1]

Data used below are adjusted for inflation (amounts stated are in 2021 dollars) and for family size. The family size adjustment accounts for economies of scale within the household so average incomes reported below can be thought of as per person amounts for those living in multi-person households.2[2] Note, however, that these are not necessarily the same people in the top, bottom or the middle deciles in each year.

 Figure 1 shows the average income by decile for each of the three ways of measuring income in 2021, the most-recent data year. The highest decile has a much larger share of income than all others, however defined. Government transfers—whose effects are shown in the shift between the blue and orange bars—raise the incomes of all deciles but have a bigger impact in both dollar and percentage terms for those closer to the bottom. Likewise, income taxes—whose effects are shown in the shift from orange to grey bars—reduce total income for all deciles but have a bigger impact on the incomes of higher deciles.

An important caveat is that in 2021, Canada was emerging economically from the COVID-19 pandemic. In the first half of the year, in particular, total income and after-tax income were larger than pre-pandemic levels due to temporary increases in transfer income (for example, the Canada Emergency Response Benefit or CERB). COVID-19 income supports substantially reduced poverty on a scale not seen before and reverting to pre-COVID-19 policies will have an adverse impact on income distribution in 2022 and 2023. 

The top and the bottom

Market incomes for the top decile are much larger than other deciles by definition but the degree to which this is the case is surprising. In 2021 the top 10% commanded 30% of all market income.The average market income of the top decile is 79% higher than the next decile. Taxes and transfers compress the income distribution but even after tax, the top decile still takes home 56% more on average than the next decile. 

Figure 2 focuses on the top 10% share of market income (in orange) and after-tax income (in blue) going back to 1976. An additional 5% of the market income pie went to the top 10% in 2021 compared to the 1970s. The critical period for inequality growth is from the early 1980s to the early 2000s. After taxes and transfers, the top income share falls to 23%, or 6-7 percentage points lower than its market income share. This gap between market and after-tax income shares was smaller in the 1970s. For after-tax income, there has been some moderation since the early 2000s, although the after-tax income share is still slightly higher than in the 1970s. 

Canada is now taxing income at the top of the distribution slightly more than it did in the 1970s. After taxes and transfers, the top 10% receive a highly disproportionate share of the income pie but that share has declined from its peak in the 2002 to 2004 period when tax cuts were a top policy priority in Canada. How the top decile share changes in 2022 and 2023 is of central importance, and with the end of COVID-19-related income supports that largely benefit those lower in the distribution, the top income share is likely to rise.

Figure 3 shows the same data for the bottom decile. In the 1990s recession and slow economic recovery characterized by high unemployment—abetted by high interest rate policies from the Bank of Canada and austerity from federal and provincial governments —the bottom decile share of market income was close to zero. The bottom decile is largely pensioners, people who are unemployed and on social assistance, and single parents. They made almost no money in the paid labour market when times were bad, but generally received less than 0.5% of total income even in good times. 

This lesson is in need of re-learning as the Bank of Canada has embarked on a new inflation fighting crusade with interest rate hikes since March 2022. For all of the pejorative talk about hard work being the reason for people’s success in life, the volatility of market incomes over time shows the vulnerability of low-income people to the state of the economy.

Income from government transfers both increases and stabilizes average income in the bottom decile. Market income share (in blue) is well under 1% in all years but after tax income share (in grey) is close to 3%. For the second decile (next 10%) market income share (not shown) also grows substantially after accounting for income transfers.

The share of after-tax and transfer income for the bottom two deciles fell from the mid-1990s to the mid-2010s, with some moderate recovery over the 2015-2019 pre-COVID-19 period. Outcomes in the main COVID-19 year, 2020 (and 2021 to a lesser extent), are extraordinary with the highest income shares over the entire time frame analyzed going to the bottom decile (the same is true for the second decile, not shown). 

This demonstrates the inequality-reducing impacts that can be realized by robust income transfers, including the 1998 implementation and subsequent expansions of child tax benefits. More-recent large income transfer programs during COVID-19 show that poverty and inequality can be reduced substantially and quickly. Alas, these were viewed only as temporary measures during the pandemic. However, there is nothing inevitable about rising inequality and public policy could step in with stronger income supports that reduce poverty and inequality.

Back to the 1970s

Table 1 shows a hypothetical situation of what incomes would have looked like had the income distribution stayed the same as the late 1970s. Because the numbers jump around a bit, the average of the 1976 to 1980 period is compared with the average of 2015 to 2019, the period before COVID-19. The late-2010s period should be a better indicator of current inequality than the COVID-19 years, which substantially affected both market incomes and income transfers.

The top decile market income in the late 2010s was more than $23,600 per person higher than it would have been under a 1970s income distribution. The next 10% also came out ahead relative to the 1970s, by almost $3,900 and the third decile by $760 per person. All other deciles, representing the bottom 70% of the distribution, were worse off in terms of market income in the late 2010s than they would have been under a 1970s distribution. If we use 2021 as the comparator (not in the table), top-decile income surged even higher at $29,000 per person higher than under a 1970s distribution.

Taxes and transfers offset the much-greater market inequality but not fully. The top decile is still better off by more than $8,200 per person in after-tax income in the late 2010s when compared to a 1970s distribution. The bottom decile was modestly better off than under a 1970s distribution by $473. In contrast, the second to eighth deciles were all worse off when compared to a 1970s distribution. For the middle deciles this amounts to between $1,000 to 2,000 per person (2% to 5% of income). 

In broader terms, the federal system of income transfers (including the GST Credit, Child Tax Benefits, Old Age Security and Guaranteed Income Supplement) improved in the late 1980s and early 1990s, and again in the mid-2010s. In the opposite direction was a period of tax cuts in the late 1990s and early 2000s. The 1995 elimination of cost-shared transfers to the provinces under the Community Assistance Plan is also notable and was accompanied by the atrophying of provincial social assistance programs for the working age population and people with disabilities. Provincial income supports remain a weak spot in Canada’s social safety net. 

While these data look at income across all households, we know that various groups will be disproportionately located in the lower income deciles, including seniors, women, single parents, Indigenous people, racialized and LGBTQ+ households, and people with disabilities. Policies that lead to increased income for the bottom of the distribution thus have a disproportionate benefit in combating racial, gender and other injustices.

In addition, the evidence suggests that both the very poor and the very rich are underrepresented in the income survey data. At the very bottom are households that are missed or undercounted by the survey used to generate the data, including people who are homeless and other marginalized people, those living on First Nations reserves, as well as those living in remote areas.

Within the top decile, affluent households are mixed in with the super rich. Breaking out the top of the distribution, similar patterns will be present where a disproportionate share of the gains of the top 10% went to the top 1%, and the top 0.1% got a large share of income in the top 1%, and so on. Lars Osberg looks at this for tax filers, finding that even as the top 1% doubled their income between 1982 and 2018, this was skewed by a top 0.01% income jump of 189%.

A more equal society is possible

Left to its own devices, the market will deliver a very unequal distribution of income. But public policy can and must intervene through taxation and the provision of both income transfers and public services. The recent COVID-19 experience shows that income inequality can be effectively combatted without the sky falling. However, it is not clear whether policy makers will revert back to neoliberal policies that lead to greater inequality.

To reduce inequality, public policy should aim first and foremost to reduce the inequality of market income. Reforms to compress market income inequality include measures like sectoral bargaining that would lead to higher unionization rates, higher minimum wages and labour market regulations to boost incomes for precarious workers. At the higher end, more limits are needed on excessive CEO pay and other compensation while cracking down on tax evasion and avoidance. 

That said, market measures do little to help the situation of seniors who have little or no market income. Improving income transfers would also reduce after-tax inequality, as would additional public investments in non-market housing and a range of public services and infrastructure are needed to further reduce inequality. Indeed, the high cost of housing today is a major difference with the 1970s when housing prices were much lower relative to income.

A more progressive income tax system could include higher top marginal rates, full taxation of capital gains, reductions in deductions and credits that disproportionately favour top earners (such as RRSP deductions and mineral exploration credits) and reducing basic exemptions in favour of income transfers to low-income households. 

The analysis above does not account for the distributional impacts of other taxes besides income taxes, many of which are regressive in nature. Although this discussion is focused on income, there’s also a strong case to be made for a wealth tax.

Transfer income also plays a strong role in outcomes at all levels, a fact reinforced by the recent COVID-19 experience, but especially for low-income households. The federal system of income transfers could be greatly augmented as seen with the temporary income supports provided during COVID-19 and more recently the federal “grocery rebate.” The CCPA’s Alternative Federal Budget outlines reforms and improvements to child, seniors and disability benefits, along with a “livable income” benefit for working-age adults.

The federal role in income support could re-examine the terms of the Community Assistance Plan (CAP) transfers to the provinces, mentioned above. From 1966 to 1991, CAP was open-ended meaning the federal government would match any eligible provincial expenditures and thereby share the risk associated with higher income transfers expenditures when there’s an economic downturn. Alternatively, major reform could include uploading responsibility for all income support to the federal government (as is already the case for child and seniors benefits and employment insurance).

Overall, income support in Canada is woefully inadequate and needs to be significantly improved as part of reducing inequality. At the same time, the expansion of public services and infrastructure is also a key element of the redistribution of income and can provide cash-free access to necessities for the poorest and everyone else, as Armine Yalnizyan points out

At a time when society has abject homelessness at one extreme and lavish wealth at the other, it’s important to recognize that outcomes are not the result of economic forces beyond our control but choices we make about how the pie is sliced.

Thanks to the following for comments on an earlier draft of this post: Shannon Daub, Alex Hemingway, Iglika Ivanova, David Macdonald, Lars Osberg and Armine Yalnizyan. Any errors are the responsibility of the author.


[1] It is also possible to look at more comprehensive or broad income measures that include all other sources of income that give people command over resources (inheritances and gifts, employer-provided benefits, corporate profits) and all other tax measures (sales taxes, corporate taxes, payroll taxes) but for this paper we stick to the conventional definitions.

[2] The convention for the family size adjustment is to divide income by the square root of family size. For example, four can live more cheaply per person than one person alone, so a family of four with $200,000 income is adjusted to be equivalent to a single individual with $100,000 income. In the absence of this adjustment, the bottom of the distribution would be dominated by single earners and the top by two or more earner households. This adjustment is also useful when considering longer periods of time, such as the 1970s when household size was larger on average.


Topics: , ,