Reviving Productivity Growth in Canada

2. The productivity challenge

Ali Bargu

Alexander Hijzen

Abstract

This chapter sets the scene for the Review by providing key stylised facts that are relevant to understanding Canada’s productivity challenge. It starts by documenting the evolution of GDP per capita and its main components (e.g. employment, multi-factor productivity and capital deepening) using the OECDs “Going for Growth” methodology. It then proceeds by documenting the evolution of labour productivity using comprehensive linked employer-employee data and the extent to which this is driven by changes in productivity growth within firms related to innovation, technology adoption and learning or changes in the process of efficiency-enhancing job reallocation between firms with an emphasis on job-to-job mobility (the “job ladder”). It concludes with a discussion of the implications of the role of various megatrends in relation to demographic developments (ageing, migration) and structural change (e.g. green transition, AI) for productivity growth in Canada.

In Brief

Key findings

Economic growth in Canada has weakened considerably in recent decades, significantly slowing progress in incomes and living standards. Without effective policy action, weak economic growth will cast a long shadow over future generations.

• Weak economic growth reflects a gradual decline that began in the early 2000s. This decline has been more pronounced than in comparable advanced economies such as the United States and the Euro Area. As a result of the slowdown in economic growth, GDP per capita may be up to one‑third smaller today than its level had economic growth continued at its rate during the late 1990s.

• Reviving economic growth crucially requires boosting labour productivity growth. While there is some scope to raise employment rates, particularly among older workers and women, this will not be sufficient to revive economic growth in the context of an ageing population. Nevertheless, continued efforts to expand labour force participation remain important to stem the adverse effects of an ageing population on economic growth.

• Weak labour productivity growth reflects limited gains in efficiency and weak investment. Multi-factor productivity growth has declined since the start of the 2000s while capital deepening has declined steeply following the sharp fall in commodity prices in 2014. In recent years, weak investment has been the main driver of declining productivity growth.

The chapter provides new evidence on the sources of the slowdown in productivity growth using comprehensive linked employer-employee data with information on firm-level labour productivity for Canada and selected OECD countries. It provides three important stylised facts.

• The slowdown in aggregate productivity growth both reflects slower average productivity growth within firms and slower growth between firms due to a weakening in the pace of efficiency-enhancing job reallocation from less productive firms and sectors to more productive ones.

• The slowdown in productivity growth is concentrated in high productivity firms and manufacturing industries. Its concentration in frontier firms raises important concerns about innovation and the international diffusion of new technologies. The slowdown is broad-based across industries but stronger in manufacturing than in services.

• The job ladder, i.e. job-to-job mobility from less to more productive firms, is the key driver of efficiency-enhancing job reallocation and its decline over time. This highlights the importance of policies that can support voluntary job mobility between firms by creating opportunities for job mobility towards better firms and removing barriers to job mobility.

Population ageing, immigration, the net-zero transition and technological developments in Artificial Intelligence and deglobalisation are reshaping Canada’s productivity landscape with new challenges and opportunities. To respond effectively to the challenges and make the most of new opportunities, a policies environment are needed that actively encourage investment, competition and firm dynamism and provide the necessary enabling conditions, including a flexible and adaptable workforce.

• Increased tariffs on US imports from Canada are weighing on productivity. In 2024, about hree‑quarters of Canada’s goods exports went to the United States. While international trade typically boosts productivity by expanding market size, increasing competitive pressures and enabling technology diffusion, trade disruptions can erode these benefits. Higher tariffs, disputes and supply-chain instability raise uncertainty, curb investment and limit firms’ capacity to innovate and grow. To limit these effects, it is important to strive for stable and predictable trade relations, advance trade diversification through agreements, and reduce interprovincial barriers that fragment domestic markets and impede labour mobility.

• The effects of population ageing for productivity growth have largely materialised in Canada. Between 1980 and 2020, the share of persons aged 55‑64 in the working-age population has increased from 18 to 29% and is expected to broadly constant going forward, reaching 30% by 2060. While ageing may have affected productivity growth through various channels and a full assessment is lacking, population ageing may have weakened efficiency-enhancing job reallocation, accounting for about 10% of the slowdown in productivity growth since the early 2000s in Canada.

• Migration has traditionally been a key driver of employment and productivity growth in Canada but has come under pressure as temporary migration strongly increased in recent years. In 2022, more than one in five people in Canada were foreign-born – one of the highest shares in the OECD, while migration inflows in Canada continued to exceed those of comparable OECD countries. The strong increase in temporary migration reflects a response to acute labour shortages in predominantly low pay, low productivity activities. Yet, it also raises concerns that this has tended to favour low value‑added activities at the expense of activities providing more value added, with adverse effects on aggregate productivity. A re‑balancing of migration policies is currently under way.

• The transition to a net-zero economy raises important challenges for Canada’s energy-extraction industry but also presents new opportunities. Canada’s economy relies more strongly on energy-extraction activities than any other country in the OECD except Norway. They account for 8% of value‑added, twice the OECD average, contribute significantly to national emissions and are characterised by high levels of labour productivity, given the importance of capital in production. Transitioning to net zero thus presents a dual challenge: Canada must reconcile its reliance on emissions-intensive sectors for productivity growth with the need to transition to a net-zero economy. This requires amongst others boosting innovation in clean technology and green job creation in high productivity firms.

Artificial Intelligence (AI) has the potential to significantly raise productivity, particularly in knowledge‑intensive sectors. Yet, estimates of its expected impact on aggregate productivity in Canada vary widely. While OECD estimates suggest AI could raise labour productivity by between 3 to 9% over ten years, these benefits are highly uncertain. To reap the benefits of AI, it needs to be adopted widely. However, adoption remains uneven. To promote AI adoption more needs to be done to prepare infrastructure, firms and workers for the digital age.

2.1. Introduction

Canada’s longstanding productivity challenges have become increasingly urgent amidst significant structural shifts, including population ageing, the green transition, digitalisation, and the resurgence of trade barriers. Over the past two decades, Canada’s productivity growth has fallen to one of the lowest levels observed in the OECD. OECD projections indicate that without appropriate policy action Canada could experience the slowest growth in real GDP per capita among advanced economies from 2020 to 2060 (Guillemette and Turner, 2021[1]).

The objective of this chapter is to provide key stylised facts that are relevant for understanding Canada’s productivity challenge. It starts by examining aggregate trends in economic growth, productivity and employment using the OECD’s Going for Growth methodology (OECD, 2023[2]). Digging deeper using linked employer-employee data from Canada and other OECD countries, it examines whether weak productivity growth is concentrated within firms, and hence reflects the ability of firms to adapt their production processes in response to new opportunities, or between firms and hence reflects the speed with which workers are reallocated across firms according to their most productive use in response to changes in business opportunities across firms. Throughout considerable attention is paid to various mega-trends affecting productivity including demographic transitions, migration patterns, the transition to net-zero emissions, and the prospective impacts of artificial intelligence.

Based on the analysis, the chapter concludes that enhancing labour productivity is essential. The slowdown in productivity growth in Canada has had significant consequences for economic welfare and living standards, and – without effective policy action – will place a burden on future generations. As discussed in the OECD Economic Survey of Canada and the Budget for 2025 (Government of Canada, 2025[3]; OECD, 2025[4]), reviving productivity growth in Canada requires a range of policies related to product, financial, housing and labour markets to foster a competitive environment that rewards good performance and supports innovation, technology adoption and investment. In this context, labour market policies are best seen as enabling conditions that allow productivity gains to materialise once investment, competition, and firm dynamism recover.

2.2. The productivity slowdown in Canada: The macro picture

This section documents aggregate developments in GDP per capita, labour productivity, and its main components using country-level data.

2.2.1. Economic growth has declined strongly

Economic growth has almost come to a halt in Canada. Economic growth, defined here in terms of annual growth in GDP per capita, has declined from over 2.5% in the late nineties to close to zero in the most recent period from 2019-2023 (Figure 2.1). The decline is mainly driven by long-term developments, which may in part be related to the adaptability of firms and workers to structural change.1 Economic growth tended to be particularly weak in periods involving severe economic crises (e.g. global financial crisis and COVID‑19 crisis).

While other advanced economies also have experienced a secular slowdown in economic growth, it has been particularly pronounced in Canada. During the late 1990s, economic growth was similar in Canada, the Euro area and the United States at around 2.5% per year. During the 2000s, it slowed to around 1.5% per year in each of the three blocks. In the period following the global financial crisis until the COVID‑19 crisis, economic growth returned to its level observed before the global financial crisis in the United States and the Euro area, while it declined further in Canada to 0.8% per year. The diverging experience of Canada is in part related to the sharp decline in commodity prices in 2014, which depressed the contribution of the extraction sector to economic growth (Loertscher and Pujolas, 2024[5]). During the COVID‑19 crisis and its recovery, economic growth continued to deteriorate in Canada and weakened markedly in the Euro area, while it remained robust in the United States thanks to its strong performance in the recovery period.

The slowdown in economic growth in Canada has important consequences for economic welfare and living standards. GDP per capita is estimated to have been 32% lower in 2022 than what it would have been had economic growth continue to grow at its rate during the period 1995-2001 (see Box 2.1). This is considerably larger than for the United States and the OECD as a whole where the slowdown in productivity growth was less pronounced (22% and 23% respectively).

Figure 2.1. Economic growth has weakened

Annualised growth rates in real GDP per capita by period and region, percentages

Note: Aggregates are weighted by PPP-based GDP weights across countries.

Source: OECD estimates based on OECD (2026[6]), Annual GDP and consumption per capita, USD, volume, constant PPPs, reference year 2020, https://data-explorer.oecd.org/s/47b.

Box 2.1. The welfare losses of declining economic growth in the long-term

The shortfall in economic growth during the past two decades already has resulted in a significant loss of economic welfare. To get an order of magnitude of the loss in economic welfare due to the slowdown in economic growth since the early 2000s, one can compare actual GDP per capita levels in the most recent year (2023) with a simple counterfactual that assumes that economic growth would have continued at the average annual rate observed during the period 1995-2001. On this yardstick, the shortfall in economic growth experienced by Canada since 2001 is around 35% (Figure 2.2). This is considerably larger than for the United States or the OECD where it was about 22% and 25% respectively. The shortfall in economic growth in the Euro area was similar to that in Canada. Given the very weak economic growth observed in recent years in Canada and projected demographic developments, OECD projections indicate that without appropriate policy action Canada could experience the slowest growth in real GDP per capita among advanced economies from 2020 to 2060 (Guillemette and Turner, 2021[1]).

Figure 2.2. The shortfall in economic growth Canada over the past two decades

Log GDP per capita

Note: Data are measured in constant prices, USD 2020 purchasing power parities. The dotted line is a linear trend through the data points in 1995 and 2001. The trend is normalised to 100 in 2023. The simulations show that seemingly small changes in economic growth can have large consequences in the long run. The estimated shortfall in economic growth can vary significantly depending on the reference period chosen. It will generally be smaller when earlier or later periods are chosen since growth was particularly strong in the late 1990s.

Source: OECD Estimates based on OECD (2026[6]), Annual GDP and consumption per capita, USD, volume, constant PPPs, reference year 2020, https://data-explorer.oecd.org/s/47b.

2.2.2. Reviving economic growth requires boosting productivity growth

The slowdown in economic growth in Canada reflects declining contributions of labour productivity growth and rising employment rates (Figure 2.3)2. More precisely, about two‑thirds of the decline in economic growth is due to weakening labour productivity growth (defined here in terms of GDP per worker) and one‑third due to slower increases in employment rates (defined as the share of employed persons in the working-age population).3 Population ageing (defined here as the share of working age persons in the total population) has not exerted much of an effect on economic growth so far. If anything, demographic developments due to both a combination of rising longevity and declining fertility also contributed to the slowdown in economic growth, as it contributed positively to economic growth during the early 2000s but negatively from the mid‑2010s onwards.

Reviving economic growth crucially requires boosting labour productivity growth. While rising employment rates, as reflected by positive growth rates in the figure, have helped to support economic growth during the past two decades thanks to gradual expansions of the labour force, due to rising female participation, increases in the effective retirement age net and net immigration, its potential to support economic growth going forward is likely to be limited. While there remains some scope for increasing employment rates, future increases may be more difficult to realise, and they will not be sufficient to offset the adverse impact of demographic change, in the form of increasing longevity and declining fertility, on economic growth (see Box 2.2).4

Figure 2.3. The decline in economic growth is driven by slower labour productivity growth and, to a lesser extent, slower employment growth

Annual change in GDP per capita, and the annual change in its components, Canada and OECD, percentages

Source: OECD Economic Outlook 115 Database (June 2024).

Box 2.2. The implications of demographic change for future economic growth in Canada

Like most other OECD countries, Canada faces important demographic changes due to a combination of rising longevity, declining fertility and, more recently, an increase in the number of foreign students. This is expected to reduce employment rates, the share of employed workers in the total population, and drive down economic growth in the absence of policy action (and assuming productivity growth remains constant). OECD estimates suggest that as a result of demographic change in Canada average annual economic growth may decline by about one‑third, from 1.1% during the 2010s to about 0.7% over the period 2024-2060 (OECD, 2025[7]). This Box discusses how policy can alleviate the adverse effects of demographic change for economic growth in Canada by promoting employment rates for different groups of workers based on the simulations exercises presented in Chapter 2 of the OECD Employment Outlook 2025.

• Promoting longer working lives among older workers can significantly alleviate the impact of population ageing on economic growth. By reducing the labour market exit rate to that of the 10th percentile of OECD countries with the lowest rate in each age and gender category above 55, Canada could reduce the projected decline in economic growth by about one half.

• Closing the employment gap between men and women at all ages would reduce the projected decline in economic growth due to population ageing by another quarter. About half of these potential gains would come from closing the gender gap for older workers (aged 55 years or more).

• Migration has helped to offset some of the effects of population ageing in Canada so far, but there is limited scope to expand it further. The priority is to maximise its benefits. Canada’s permanent migration system can support productivity when newcomers are well integrated. By contrast, temporary migration can be useful to stem short-term labour needs but has a limited impact in the long term (see Section 4 below and Chapter 4 for further details).

In short, while promoting employment among older workers and women could go a long way in alleviating the adverse effects of demographic change on economic growth, it would be insufficient to revive productivity growth.

Source: OECD (2025[7]), OECD Employment Outlook 2025, https://doi.org/10.1787/194a947b-en.

2.2.3. Weak labour productivity growth reflects limited gains in efficiency and weak investment

Labour productivity can be decomposed in components related to multi-factor productivity and capital intensity. Multifactor productivity (MFP) captures the efficiency with which inputs like labour and capital are utilised in production.5 It reflects factors such as technological innovation, organisational practices and technology adoption, economies of scale and allocative efficiency (i.e. the allocation of capital and labour across firms that differ in their productivity). Capital intensity measures the quantity of capital, such as machinery, equipment, and infrastructure, employed per unit of labour. An increase in capital intensity means that workers have access to more or superior equipment, potentially boosting their productivity. Increasing capital intensity crucially requires investment.6

Labour productivity growth in Canada declined steadily since the late 1990s due to limited gains in MFP and a slowdown in the pace of capital deepening in the mid‑2010s (Figure 2.4). During the late 1990s, annual labour productivity growth averaged 1.3%, with 0.9% coming from MFP growth and 0.4% from capital deepening. While MFP growth gradually declined during the subsequent two decades, capital deepening remained robust until the collapse in commodity prices in 2014. Indeed, during the period up to 2014, private investment was driven to an important extent by the extraction sector. While this contributed to higher capital intensity and labour productivity growth, it reduced real MFP growth in the oil sector due to amongst others declining economies of scale and the increased cost of resource extraction.7 When the commodity prices collapsed, the weakness of investment in other sectors was laid to bare. Moreover, there are also concerns about the composition of investment, with investment disproportionally concentrated in real estate as opposed to machinery, intellectual property and other intangible assets that are key for introducing and spreading new technologies (OECD, 2025[4]; Allen, Gu and Macdonald, 2025[8]). In the post-COVID period, weak investment has been the main source of declining productivity growth.

The qualitative picture is broadly similar for the OECD as a whole with declining contributions of both MFP and capital deepening. This may suggest that at least to some extent OECD countries face common productivity challenges related to the adaptability of firms and workers to structural change. However, the decline in MFP appears to be more pronounced in Canada than across the OECD on average, while the decline in capital deepening started earlier in most OECD countries because of the lingering effects of the global financial crisis on the supply and demand for credit (e.g. bank and sovereign debt crisis in the Euro area, uncertainty about the macroeconomic outlook). These differences to some extent reflect the reliance of Canada on the resource sector for economic growth and its declining performance over time.

Figure 2.4. The decline in labour productivity growth reflects slower efficiency gains and weaker investment

Annual change in labour productivity and its components, Canada and OECD, percentages

Note: All variables are smoothed except capital stock per worker. The OECD average is obtained by PPP-based GDP weights across countries. For more details, see the description of the OECD potential output estimation methodology (Chalaux and Guillemette, 2019[9]). MFP therein is referred to as “labour efficiency”, given that in the underlying production function technical change is written in labour augmenting form.

Source: OECD Economic Outlook 115 Database (June 2024).

Box 2.3. Productivity growth has not become less broadly shared in Canada

This Box documents the evolution of labour productivity growth in Canada using different measures of productivity (per worker or per hour worked) and compares that with the evolution of average wage growth and wage growth for low-wage workers.

• The evolution of labour productivity growth is qualitatively similar when measuring it in terms of output per hour worked or output per worker. This suggests that the decline in labour productivity growth, based on output per worker is not driven by developments in working time per worker. The main text focusses on output per worker to be consistent with the OECD Going for Growth methodology. However, output per hour is more appropriate when making comparisons with wages. Note that output per worker using the OECD Economic Outlook database, consistent with OECD Going for Growth, is measured in potential terms and hence abstracts from business cycle fluctuations.

• While during the late 1990s productivity gains were associated with falling labour shares and rising wage inequality, productivity and wages have been broadly aligned since. In the period 1998-2001, labour productivity growth far exceeded average wage growth, indicating a declining labour share, while average wage growth exceeded wage growth among median and low-wage workers, indicating rising wage inequality. From 2001-2007 onward, productivity, average wages and the wages of low-wage workers have been much more aligned. For a more detailed discussion of trends in wage inequality and labour shares, see MacGee and Rodrigue (2025[10]).

In sum, unlike in several other OECD countries, there is no indication that productivity growth has become less broadly shared with workers, including those with low wages, in Canada.

Figure 2.5. Productivity, wage and inequality developments in Canada

Annualised growth rates by period, percentages

Source: OECD Economic Outlook 115 Database (June 2024); OECD productivity database; OECD database on annual wages; OECD database on earnings.

Box 2.4. Trade shocks, internal barriers and labour productivity in Canada

Canada is among the more open economies in the OECD, with trade in goods and services (exports plus imports) amounting to around 65% of GDP, compared with an OECD average of about 60%. This degree of openness means that external trade shocks can have disproportionate effects on growth and labour productivity.

The Canadian economy is especially tied to the United States. In 2024, about 76% of Canada’s goods exports were destined for the United States, while 62% of imports originated there (OECD, 2025[4]). Trade with the United States alone represented roughly 16% of Canadian GDP, underscoring the centrality of this relationship for employment and productivity.

International trade generally supports productivity growth by fostering economies of scale, increasing competitive pressures, and facilitating technology and knowledge diffusion. Historical evidence from the 1989 Canada – US Free Trade Agreement suggests that Canadian exporters became more productive, and that manufacturing labour productivity improved in the years following liberalisation (Lileeva and Trefler, 2010[11]; Trefler, 2006[12]).

Trade disruptions can reverse some of these gains. Tariff increases, trade disputes and supply-chain disruptions raise uncertainty, discourage investment and limit firms’ ability to innovate and expand. Recent US tariffs on Canadian exports provide a concrete illustration: Canadian goods exports to the US dropped by over 15% in April 2025, with car exports falling nearly 25% (Bank of Canada, 2025[13]). Employment in sectors highly reliant on US exports declined sharply, while pervasive trade policy uncertainty is holding back hiring and investment. These shocks have been particularly acute in sector’s not covered by the USMCA, as well as Southern Ontario’s manufacturing industries, where exposure to US demand is greatest.

The OECD Economic Survey (2025[4]) highlights priority actions to boost productivity like preserving stable and predictable trade relations with the United States, pursuing further trade diversification through agreements such as CETA and CPTPP and removing interprovincial barriers that fragment internal markets and hinder labour mobility. Complementary measures to boost business investment- particularly in innovation and green technologies – would help offset the drag from trade uncertainty and reinforce productivity growth.

2.3. Beyond aggregate trends: A micro perspective

This sub-section goes beyond aggregate trends by providing a detailed analysis of the sources of weak aggregate productivity growth using administrative linked employer-employee data for Canada.8 It focusses on three key questions. First, to what extent are changes in aggregate productivity growth concentrated within firms, and hence related to changes in the importance of investment, technology adoption and innovation in the workplace, or between firms and hence related to changes in the speed of efficiency-enhancing job reallocation between firms that differ in their productivity? Second, to what extent are changes in productivity growth within firms concentrated in specific groups of firms? Third, to what extent reflect changes in efficiency-enhancing job reallocation changes in the contribution of job-to-job mobility between firms or movements in and out of employment?9

2.3.1. Weak productivity growth reflects both weak growth within firms and weak efficiency-enhancing job reallocation

The trend decline in aggregate labour productivity growth for Canada is also present in the micro data where productivity is measured in terms of real value added per worker (Figure 2.6).10 Employment-weighted average annual productivity growth declined from 1.8% in the period 2002-2007 to 1.3% in 2008-2013 and 0.9% in 2014-2019 (Panel A).11 Employment-weighted average annual productivity growth can be decomposed in a within-firm component, which captures factors related to capital deepening (investment) and MFP growth (technology adoption/diffusion and innovation) in the workplace, and a between-firm component, which captures the role of efficiency-enhancing job reallocation across firms and industries that differ in their productivity (see the Annex for further details on the methodology).12

The decline in productivity growth reflects a slowdown in growth within firms as well as between firms due to weakening efficiency-enhancing job reallocation. In the period 2002-2007, average annual productivity growth within firms amounted to 0.9% (53% of total growth), while growth between firms amounted to 0.8% (47% of total growth). In the period 2008-2013, within-firm growth remained stable at 0.8% (62% of total growth), while between-firm growth weakened to 0.5% (38% of total growth). In the period 2014-2019, within-firm growth slowed to 0.6% (63% of total growth), while between-firm growth weakened further to 0.3% (37% of total growth). In other words, between the first and the last period, about half of the decline in average annual aggregate productivity growth was concentrated within firms (43%) and about half between firms (54%). Patterns by province within Canada are discussed in Box 2.4.

Figure 2.6. Productivity growth within and between firms

Average annual growth rates in real valued per worker, percentages

Notes: The figure provides a worker-level decomposition of aggregate productivity growth on Hahn, Hyatt and Janicki (2021[14]) into components associated with within and between-firm growth. For more details, see Annex 2.A.

Total: average annual aggregate productivity growth. Within firms: average annual aggregate productivity growth among workers staying in the same firm. Between firms: average annual between-firm productivity growth due to the effect of job reallocation on the employment-weighted productivity distribution of firms.

Source: Canadian Employer-Employee Dynamic Database and national linked employer-employee data.

Box 2.5. Productivity developments vary widely across provinces

This Box documents the contribution of provinces to the slowdown in average annual productivity growth in Canada between the period 2002-2007 and the period 2014-2019. The contribution of each province is calculated by i) multiplying the change in average annual productivity growth in each province by its share in aggregate employment and ii) dividing this by the country-wide change in average annual productivity growth between 2002-2007 and 2014-2019. The exercise is conducted for aggregate productivity growth as well as its components within and between firms. The results are visualised in Figure 2.7.

The slowdown in average annual productivity growth between 2002-2007 and 2014-2019 is largely driven by Alberta. It accounts for around 88% of the entire slowdown. This may to some extent be driven by the decline in energy prices in 2015 which generated a collapse in revenue productivity within firms and a shift from energy-generating sectors to other sectors with lower revenue productivity. The disproportionate contribution of Saskatchewan (13%) may also be related to the importance energy sector. British Colombia, Saskatchewan and Manitoba also contributed to the slowdown in productivity growth (18%, 13% and 7% respectively). By contrast, productivity growth increased in Quebec and to a lesser extent Ontario, resulting in negative contributions to the decline in productivity of respectively ‑31% and ‑2%.

The large role of Alberta in the slowdown of productivity growth is partly a result of the definition of the reference periods. Alternative approaches based on year-to-year comparisons might present a different picture but tend to be sensitive to the specific start and end years selected. The choice of longer reference periods here is consistent with the analysis in the main text. As such, the analysis in the box provides useful context for interpreting the patterns documented there.

Figure 2.7. The contribution of provinces to the slowdown in aggregate productivity growth

Contribution of province to change in average annual productivity growth between 2002-2007 and 2014-2019, percentage

Notes: Total: contribution of province to change in average annual aggregate productivity growth. Within firms: contribution of province to change average annual aggregate productivity growth among workers staying in the same firm. Between firms: contribution of province to change average annual between-firm productivity growth due to the effect of job reallocation on the employment-weighted productivity distribution of firms.

Source: Calculations based on Canadian Employer-Employee Dynamic Database

2.3.2. The slowdown in productivity growth is concentrated in high productivity firms and manufacturing industries

The slowdown in productivity growth between 2002-2007 and 2014-2019 in Canada is concentrated in frontier firms, defined as those in the top 10% of the employment-weighted productivity distribution within each 2‑digit industry (Figure 2.8, Panel A). Note that during the early period from 2002 to 2007, growth in frontier firms strongly exceeded that in lagging firms, resulting in productivity divergence. This suggests that strong productivity growth in leading firms was not transmitted to less productivity firms, and that technology diffusion was limited (Gu, 2019[15]). Weak productivity growth among frontier firms during the period 2014-2019 coincides with the decline in aggregate investment and is likely to have depressed job opportunities for job mobility towards more productive firms.

The slowdown in productivity growth was broad-based, but stronger in manufacturing than in services (Figure 2.8, Panel B). This is based on national accounts data by industry which measure labour productivity growth in real terms using industry price deflators.13

Figure 2.8. The decline in within-firm productivity growth is concentrated in frontier and resource‑intensive firms

Average annual growth rates in value added per worker, percentages

Source: Canadian Employer-Employee Dynamic Database (Panel A); Statistics Canada (2026[16]), Table 36‑10‑0 480‑01 Labour productivity and related measures by business sector industry and by non-commercial activity consistent with the industry accounts, https://doi.org/10.25318/3610048001-eng (Panel B).

2.3.3. The decline in efficiency-enhancing job reallocation is driven by a weakening of the job ladder

Efficiency-enhancing job reallocation shifts the structure of employment towards more productive firms, generating positive between-firm aggregate productivity growth. As discussed above, between-firm productivity growth is consistently positive in Canada. It was relatively strong during the period 2002-2007 when it contributed 0.8 percentage points (p.p.) to aggregate productivity.14 The strength of efficiency-enhancing job reallocation has tended to decline over time in Canada, contributing 0.3 p.p. to aggregate productivity on average over the period 2014-2019. At the same time, efficiency-enhancing job reallocation picked up across selected Eurozone economies.

The shift in the structure of employment towards higher productivity firms is entirely driven by direct job transitions between firms (job-to-job mobility), while the contribution of transitions in and out of work (employment mobility) tends to be negative. This insight is based on the decomposition of productivity-enhancing job reallocation in its contributions related to job-to-job mobility, which is likely to be voluntary, and those related to employment mobility, which reflect a range of factors, such as the decision to participate in the labour force, the risk of unemployment and the choice of working in the public sector or being self-employed. The positive role of job-to-job mobility in growth-enhancing job reallocation is sometimes interpreted as a job ladder at work, i.e. the process through which workers climb the rungs of the ladder as they advance in their careers. While the contribution of job-to-job mobility is considerable, it is to a large extent offset by the negative contribution of employment mobility. This mainly reflects cohort effects that arise as young workers start their careers in low productivity firms. Consequently, the large contribution of job-to-job mobility primarily reflects its role in integrating new entrants into the labour market.

The decline in between-firm productivity growth in Canada is largely driven by a decline in the role of job-to-job mobility in efficiency-enhancing job reallocation. Of the 0.5 p.p. decline in average annual between-firm productivity growth between 2002-2007 and 2014-2019 0.4 p.p. is driven by a decline in the contribution of job-to-job mobility (80%) and 0.1 p.p. by a decline in the contribution of employment mobility (20%). The weakening of the job ladder may reflect the growing importance of opportunities for job mobility to better firms, due to for example slower employment growth in frontier firms as documented above or weak investment. The lack of opportunities for job mobility to better firms may be compounded to barriers for doing so. Occupational licensing regulations or the use of non-compete clauses are often mentioned as factors that hinder job mobility and efficiency-enhancing job reallocation (Cf. Chapter 4). These barriers are likely to become more visible when growth in frontier firms picks up.

Figure 2.9. The decline in between-firm productivity growth is driven by a weakening of the job ladder

Average annual between-firm productivity growth and its components due to job-to-job mobility and employment mobility

Notes: The figure provides a worker-level decomposition of between-firm productivity growth in percentage based on Hahn, Hyatt and Janicki (2021[14]) into components associated with net job-to-job mobility and net employment mobility. For more details, see Annex 2.A.

Total between firms: average annual between-firm productivity growth. Job-to-job mobility: average annual between-firm growth due to the effect of job-to-job mobility on the employment-weighted productivity distribution of firms. Employment mobility: average annual between-firm productivity growth due to the effect of employment mobility on the employment-weighted productivity distribution of firms.

Source: National linked employer-employee data.

Box 2.6. Workforce ageing has tended to slow growth-enhancing job reallocation

This Box analyses to what extent workforce ageing has contributed to the slowdown in productivity growth through its impact on efficiency-enhancing job reallocation in Canada. Workforce ageing is likely to slow efficiency-enhancing job reallocation because older workers tend be less mobile and are less likely to leave their firm and be hired by higher productivity firms (Fluchtmann, Hijzen and Puymoyen, 2025[17]). Workforce ageing may also affect productivity growth through other channels, including productivity growth within firms as discussed in Section 4 of this chapter.

To quantify the role of workforce ageing in the slowdown of aggregate productivity growth a shift-share decomposition is implemented. The focus of the decomposition is on the difference in average annual between-firm productivity growth in the first and last sub-period (2002-2007 and 2014-2019) and average annual between-firm productivity growth over the full period (2002-2019) and the extent to which this is driven by changes in efficiency-enhancing job reallocation in each age group (the within-age group component due to changes in job mobility) and changes in the employment share of different age groups (the between-age group component). The between-age group component measures the role of workforce ageing for the change in between-firm productivity growth at constant mobility.

Average annual between-firm productivity growth in Canada declined between 2002-2007 and 2014-2019 by approximately 0.3 p.p. (in the period 2002-2007, it was 0.25 percentage above the average for the full period and in the period 2014-2019, it was 0.05 p.p. below it) (Figure 2.10). This represents 54% of the slowdown in total productivity growth. The between-age group component, which captures the effect of workforce ageing, accounts for a decline in between-firm productivity growth of 0.1 p.p., which corresponds to about one‑third of the decline in between-firm productivity growth and 10% of the decline in total productivity growth. The remainder of the decline in between-firm productivity growth is accounted for by changes in efficiency-enhancing job reallocation within age groups which reflect general changes in labour market dynamism as well as changes in relative mobility across age groups.

Figure 2.10. Workforce ageing contributed to the productivity slowdown

Average annual between-firm productivity growth in each period in deviation from average annual growth over the full period and its components related to average differences in job reallocation within age groups and differences in the structure of employment between age groups (workforce ageing)

Source: Calculations based on Canadian Employer-Employee Dynamic Database

2.4. The role of megatrends

Canada’s long-term productivity trajectory is being reshaped by a variety of structural shifts creating both challenges and opportunities. Demographic changes, notably an ageing population, are constraining employment growth (see Box 2.2), and hence, necessitate a greater reliance on productivity enhancements to sustain economic growth. Migration continues to be a pivotal factor in Canada as a source of skilled labour with potentially important implications for productivity growth. Simultaneously, the nation’s commitment to a net-zero economy and the rapid advancement of digital technologies, including artificial intelligence (AI), are redefining sectoral dynamics and skill requirements. This section delves into these various megatrends and their implications for Canada’s productivity growth.

2.4.1. Population ageing affects productivity growth in various ways and possibly in opposite directions

The effects of population ageing largely have materialised already in Canada (Figure 2.11). Between 1980 and 2020, the share of persons aged 55‑64 in the working-age population has increased from 18 to 29%. According to population projections, it is expected to increase only marginally further, reaching 30% by 2060. Consequently, population ageing may be more relevant for understanding the slowdown in productivity growth in Canada since the 2000s than its trajectory going forward.

Population ageing may affect productivity growth in various ways, possibly in opposite directions, making it difficult to draw strong conclusions on its overall effects on productivity growth. However, ageing makes the productivity challenge more salient by slowing economic growth through its downward impact on employment rates (see Box 2.2). Unfortunately, a detailed assessment of the effects of ageing for productivity growth in Canada and other OECD countries is lacking.15 This sub-section therefore discusses the main channels building on André et al. (2024[18]).

The effects of age on worker and workplace productivity are not clear a priori. While ageing may be associated with declines in physical capacity and certain cognitive skills – particularly when these are not regularly used – these effects tend to manifest relatively late in life and vary by occupation, health status, education, and workplace organisation. Older workers often transition into different types of roles, such as managerial or advisory positions, where accumulated experience, judgement and institutional knowledge can continue to generate high productivity (André, Gal and Schief, 2024[19]; OECD, 2025[7]). The ability of firms to harness complementarities between younger and older workers further plays a key role in determining how ageing influences productivity at the workplace level (OECD, 2020[20]). While skill erosion at older ages in principle can be mitigated through adult learning, participation in training among older workers tends to be low (cf. Chapter 3). Policies that prevent work from impairing one’s health by providing a good quality work environment are also important to sustain productivity among older workers and reap the benefits of an age‑diverse workforce.

The effects of ageing on the structure of employment across firms and sectors are most likely negative. As discussed above in Box 2.2, ageing can reduce labour mobility towards higher wage and more productive firms, slowing down the efficient reallocation of labour across firms, and dampening aggregate productivity growth. Ageing societies may also exhibit lower rates of business entry and exit, with older entrepreneurs and managers potentially crowding out opportunities for younger, more dynamic cohorts – an effect that may hinder innovation and the diffusion of new ideas. In parallel, aggregate demand may shift toward services such as long-term care, which tend to have structurally lower productivity growth, weakening aggregate productivity growth (André, Gal and Schief, 2024[19]). From a policy perspective, this highlights the importance of policies that can promote a flexible workforce at all ages (cf. Chapter 4) as well as measures that can support productivity growth in services sectors with weak productivity performance (OECD, 2025[7]).

Finally, ageing can trigger positive productivity effects by strengthening incentives for investment and innovation. Labour shortages caused by demographic decline may create stronger incentives for firms to invest in automation, labour-saving technologies and organisational innovation. Moreover, increased aggregate saving in older societies – if effectively channelled into productive investment – can support capital deepening and boost labour productivity (André, Gal and Schief, 2024[19]).

Figure 2.11. The share of older individuals of working age in the working-age population has increased significantly

Share of individuals aged 55 to 64 years in the working-age population, various years, percentages

Notes: The working-age population is the population aged 20‑64 years. 2060 is a projection based on the medium scenario. Countries are ranked by their projected p.p. increase between 2023 and 2060, in ascending order. OECD: Weighted average of OECD countries.

Source: OECD (2025[7]), OECD Employment Outlook 2025: Can We Get Through the Demographic Crunch?, https://doi.org/10.1787/194a947b-en.

2.4.2. Balancing labour market pressures and productivity objectives in migration policies

Canada has long relied on skilled labour migration as a key driver of employment and productivity growth. In 2022, more than one in five people in Canada were foreign-born – one of the highest shares in the OECD (Figure 2.12). In recent years, migrant inflows were particularly high by international standards. In 2023, temporary migrant inflows accounted for more than 2% of the overall population, of which many only stayed a short time, while permanent migrant inflows accounted for more than 1%. The high level of temporary migration seen in recent years helped to alleviate acute labour shortages, particularly in low-wage and low-productivity sectors (ESDC, 2021[21]), but also has raised questions about its implications for productivity growth (see discussion below). More recently, from 2024 Canada has started to restrict its Temporary Worker Program (ESDC, 2024[22]) and this will be continued be according to the lates budget (Government of Canada, 2025[3]).16

The Canadian model of skilled labour migration is a points-based system that prioritises immigrants based on factors like education, work experience, language proficiency (English and/or French), and age. The flagship program, Express Entry, manages applications for three federal economic immigration streams. Canada also works with provinces through the Provincial Nominee Program (PNP) to address specific regional labour market needs. The system is designed to attract qualified professionals to support productivity growth and high employment rates in a context of rapid demographic change.

Skilled labour migration can support productivity growth through multiple channels, but its effects do not materialise automatically. It can foster productivity growth by i) supporting a more efficient labour reallocation by alleviating skill shortages in high productivity firms, ii) promoting knowledge diffusion and the dissemination of new business practices (Barabuffi et al., 2025[23]; OECD, 2024[24]) and iii) by improving the allocation of skills to tasks and fostering greater task specialisation within firms (Peri, 2016[25]; Campo, Forte and Portes, 2024[26]).17 Firm-level evidence for Canada suggests that a greater share of immigrant workers is associated with sustained productivity improvements, particularly in knowledge‑intensive sectors (Statistics Canada, 2020[27]). Over time, any possible short-term capital dilution may give way to investment and capital deepening, further supporting labour productivity growth (Bank of Canada, 2023[28]; OECD, 2024[24]; Statisics Canada, 2025[29]).18 Realising these potential gains hinges on policies that support migrant integration and skill matching (C.D. Howe Institute., 2024[30]; OECD, 2024[24]). Yet, immigrants are often employed in low productivity firms and occupations despite post-secondary credentials (cf. Chapter 4). This suggests more effective policies are needed to tackle widespread overqualification and leverage untapped productivity gains from underutilised migrants.19

In addition to permanent skilled migration, Canada has increased the number of temporary foreign workers. After the COVID‑19 pandemic, programmes like the Temporary Foreign Worker Program (TFWP) have expanded to meet employer demands (Box 2.6). While temporary migration programmes aim to address immediate labour shortages, they may inadvertently support the survival or expansion of low-productivity firms by increasing labour supply. Such a dynamic could in principle lead to a misallocation of resources, where less efficient firms persist, potentially at the expense of more innovative and productive enterprises. There is no empirical evidence on the relevance of this possibility for Canada. Whether TFWP are an appropriate response to labour shortages depends on the availability of suitably qualified workers in Canada and the extent to which shortages can be alleviated by offering training and better working conditions.

Figure 2.12. In Canada about one in five persons is born abroad

Foreign-born population as a percentage of the total population in OECD countries, 2013 and 2023

Note: Data refer to 2013 or the closest available year, and to 2023 or the most recent available year. The OECD average is a simple average based on individual rates presented on this chart. For Japan and Korea, the data refer to the foreign population rather than the foreign-born population.

Source: OECD (2024[24]), OECD International Migration Outlook 2024. Figure 1.23, https://stat.link/xbfvqn.

Box 2.7. Recent trends in temporary and permanent migration to Canada

Canada has experienced a marked increase in both temporary and permanent migration in recent years, positioning it among the OECD countries with the highest inflows relative to population. Temporary migrant inflows to Canada increased by 0.6 p.p. from 1.8% of the working-age population in 2019 to 2.1% in 2023 (Figure 2.13). Temporary migrant inflows to Canada in 2023 were much higher than those to the United States (0.2%), the EU‑15 (0.1%) and Australia (0.9%). High temporary inflows to Canada were mainly intended to address sector-specific labour shortages. Inflows of permanent migrants also have increased in recent years from 0.9% in 2019 to 1.2% in 2023 and are high in comparison with other OECD countries (0.3% in the United States; 0.9% in Australia and 0.7% in the EU‑15). This places Canada among the countries with the highest permanent migration rates in the OECD. A significant share of these inflows is linked to the transition of temporary migrants to permanent residence, a feature that is becoming increasingly prominent in countries with hybrid migration systems.

Figure 2.13. Migrant inflows to Canada are high and have increased in recent years 2019-2023

Note: EU 15: Weighted average for Austria, Belgium, Czechia, Denmark, Finland, France, Germany, Italy, Lithuania, Luxembourg, the Netherlands, the Slovak Republic, Slovenia, Spain and Sweden.

Source: Based on OECD (2024[24]), International Migration Outlook 2024, https://doi.org/10.1787/50b0353e-en.

2.4.3. Green jobs are set to grow as oil and gas employment levels off, with significant opportunities and challenges for labour productivity

Canada’s economy features both high-value‑added extractive industries and ambitious greenhouse gas (GHG) reduction targets. Oil and gas production has historically supported robust productivity growth in provinces like Alberta, but it also has contributed to higher national emissions levels. Transitioning to net zero thus presents a dual challenge: Canada must reconcile its reliance on emission-intensive sectors for productivity growth with the need to decarbonise its economy.

The contribution of the energy-supply industries to the economy in Canada is the highest across OECD countries after Norway (Figure 2.14). The share of value‑added in energy-supply industries in GDP in Canada amounted to 8% in 2019, twice the OECD average of 4%. Moreover, its share in employment is considerably lower at 2% of the workforce, but still about twice the OECD average. The difference in the share of value‑added and employment highlights the high level of labour productivity in energy-supply industries. This is a general feature of the energy-supply sector and reflects the importance of capital in production.

To the extent that the transition to a net-zero economy requires a shift away from energy-supply industries, this would tend to depress aggregate productivity growth. This effect can to some extent be mitigated by promoting a shift from fossil-fuel to renewable sources of energy and promoting investments in the use of clean and energy-efficient technologies. Nevertheless, some decline in energy-related employment may be unavoidable, with a modest negative effect on aggregate labour productivity as a result (OECD, 2024[31]). Moreover, the shift in employment out of energy production may have important social consequences if worker transitions between industries are involuntary and associated with sustained periods of unemployment or re‑employment in lower paid jobs (cf. Chapter 4). Regions dependent on fossil fuels are particularly vulnerable to economic disruption, highlighting the need for social protection measures and labour market programmes to support displaced workers (cf. Chapter 4) and facilitate re‑skilling (cf. Chapter 3) as well as economic diversification. Supporting displaced workers not only is important from a social perspective but also helps to maintain broad-based public support for climate‑change mitigation policies and a policy agenda focussed on reviving productivity growth.

The transition to a net zero economy also offers opportunities for productivity growth by stimulating investment in green technologies and the creation of green jobs. Evidence shows that environmental regulations, while sometimes initially reducing productivity, often stimulate cleaner innovation and efficiency gains (OECD, 2023[31]; OECD, 2024[32]; Dechezleprêtre and Sato, 2017[33]). This supports the Porter Hypothesis which argues that well-designed environmental policy can enhance productivity and hence that difficult trade‑offs between environmental objectives and productivity growth can be avoided.

Around one in five Canadian workers are employed in green-driven occupations, i.e. occupations that will likely be positively impacted by the net-zero transition20 (OECD, 2024[34]). These also include jobs that do not directly contribute to emission reductions but produce intermediate goods and services for environmentally sustainable activities. Green-driven occupations are a heterogenous group of jobs: new and emerging occupations are typically high-skill jobs (i.e. managers, professionals and technicians) and employ highly educated workers in urban areas, while the other green-driven occupations employ many more low-educated workers in rural areas. Projections for Canada indicate that green employment will grow substantially between 2020 and 2030, driven by activities such as energy-efficient retrofits, electric-vehicle manufacturing, and renewable‑energy operations (Canadian Chamber of Commerce, 2024[35]).

Thus, the shift to a net-zero economy presents both challenges and opportunities for productivity growth: While the transition demands significant upfront investment in infrastructure, technology, and workforce development, it also may offer long-term gains – particularly when supported by innovation-focussed policies, green investment, and skills development.

Figure 2.14. Canada has the second highest value added from energy supply industries

Share of value added and employment in energy supply industries, 2019

Notes: Energy supply industries: mining and quarrying; manufacture of coke and refined petroleum products and electricity, gas, steam and air conditioning supply.

OECD 25: average for Belgium, Canada, Czechia, Finland, France, Germany, Hungary, Iceland, Ireland, Israel, Italy, Korea, Latvia, Mexico, the Netherlands, New Zealand, Norway, Portugal, Slovenia, Spain, Sweden, Switzerland, Türkiye, the United Kingdom and the United States.

Source: OECD STAN Database for Structural Analysis, 2025 edition, https://data-explorer.oecd.org/s/477.

2.4.4. The productivity gains of Artificial Intelligence are potentially large but require workers with digital skills

Artificial Intelligence holds enormous potential to revive productivity growth.21 Yet macroeconomic assessments of its potential for reviving productivity vary greatly from small positive estimates to large transformative ones (Acemoglu, 2024[36]; Aghion and Bunel, 2024[37]; Briggs and Kodnani, 2023[38]). This not only reflects large uncertainty about its potential but also different views on the time needed before the benefits of AI materialise (Brynjolfsson, Rock and Syverson, 2021[39]). OECD estimates suggest that AI could increase labour productivity in Canada by 3 to 9% over ten years (Filippucci, Gal and Schief, 2024[40]). The estimated effect of AI for Canada is slightly less than for the United States, similar to that for Germany and the United Kingdom and significantly larger than that for France and Italy. The estimates rely on a multi-sector general equilibrium model with input-output linkages in combination with industry-level estimates of the impact of AI on productivity growth that take account of the micro-level efficiency gains of AI, the occupational exposure of economic activities to AI and projected adoption rates of AI.22

AI adoption rates in Canada lag those of several other advanced economies, including the United States, the United Kingdom and Germany (Filippucci, Gal and Schief, 2024[40]). According to a recent survey by (Statistics Canada, 2024[41]), around 6.1% of Canadian businesses incorporate AI-based technologies in their production processes or service delivery. These firms typically report gains in operational efficiency, data-driven decision making, and product innovation. Moreover, sectoral differences in AI adoption are stark (Statistics Canada Survey on Business Conditions, 2024[42]). By far the highest share is observed in information and communication industries, where over 21% of firms reported using AI in their operations. Other adoption leaders include professional, scientific and technical services (14%) and finance and insurance (11%), where analytics, machine learning and chatbots are being leveraged. By contrast, AI use is almost non-existent in some sectors in agriculture and forestry, hospitality and mining and oil extraction. The strong concentration in of AI adoption in information and communication is consistent with cross-evidence presented in Calvino et al. (forthcoming).23

High costs are seen as a major constraint to AI expansion in Canada (Figure 2.15). Key government initiatives have been put forward to reduce the costs of AI adoption in Canada (OECD, 2025[4]). The government is investing in essential infrastructure to support the spread and development of AI technologies. The budget for 2025 proposes to provide CAD 925.6 million over five years, starting in 2025‑2026, to support a large‑scale sovereign public AI infrastructure that will boost AI compute availability and support access to sovereign AI compute capacity for public and private research. Of this amount, USD 800 million will be sourced from funds previously provisioned in the fiscal framework.

The effectiveness of these policies will in part depend on the ability to develop, attract, and retain AI talent. The 2023 OECD AI survey ranks insufficient skills as the most significant non-financial barrier to AI adoption in advanced economies (Figure 2.15). Canadian employers echo this concern, emphasising the shortage of workers with strong digital, machine learning, and data analytics capabilities. To fully leverage AI’s productivity potential, it is essential to develop strategies fostering a workforce that is prepared for the demands of the digital age (see Chapter 3).

Figure 2.15. Cost is the most frequent barrier to AI adoption, government regulation the lowest

Barriers to AI adoption, 2022

Notes: All employers were asked: “I’m going to list a few potential barriers to the adoption of artificial intelligence. In each case, please tell me whether it has ever been a barrier to adopting artificial intelligence in your company: High costs/Lack of skills to adopt artificial intelligence/Government regulation/Not convinced by the technology/Any other barriers not previously mentioned”.

Source: Lane, M., M. Williams and S. Broecke (2023[43]), “The impact of AI on the workplace: Main findings from the OECD AI surveys of employers and workers”, https://doi.org/10.1787/ea0a0fe1-en.

2.5. Concluding remarks

Canada’s economic performance has weakened significantly over the past two decades, placing the country on a path of stagnating real income growth relative to its peers. This is reflected by marked decline in the trend growth rate of GDP per capita, largely due to weaker labour productivity growth.

According to aggregate decompositions based on the OECD’s Going for Growth methodology, the slowdown in productivity growth reflects primarily limited gains in efficiency, and to a lesser extent, weak business investment. Micro-level evidence from linked employer-employee data further indicates that productivity growth has slowed both within firms – particularly among those at the productivity frontier and in manufacturing sectors – and between firms, due to a weakening of efficiency-enhancing job reallocation. This reallocation slowdown is largely the result of declining voluntary job mobility between firms, which has historically played a key role in reallocating labour toward more productive firms and sectors.

While Canada’s productivity challenge is particularly acute, many of its underlying causes – such as diminished business dynamism and weak investment – are not unique to Canada. These issues point to broader structural and demographic factors affecting many OECD countries, and the growing difficulty that firms and workers face in adapting to and capitalising on emerging challenges and opportunities. Given the economic reliance of Canada on energy-supply sectors, the net-zero transition may raise particular productivity and social challenges. This makes it all the more important that effective policies are put in place that support the net zero transition and enable firms and workers to seize upon opportunities related to the development of cleaner or more efficient technologies (e.g. AI). The reliance of Canada on international migration also increases the importance of policies that help ensure that their skills are effectively used.

The central policy challenge – and the primary focus of this review – is how to support the structural transformation and revive productivity growth. While this requires a broad set of policies across financial, product, labour, and housing markets, the remainder of this review focusses specifically on the role of labour markets. In particular, it examines how employment and skills policies can ensure that workers are equipped with the skills required by employers, including those requires for the adoption cleaner and more efficient technologies. It also explores how these skills can be effectively deployed within workplaces and efficiently be allocated across firms, sectors, and regions to support productivity-enhancing structural change. While labour market policies cannot by themselves revive productivity they support an environment that allows productivity gains to materialise once investment, competition, and firm dynamism recover.

References

[46] Abraham, S. et al. (2025), “Potential output in Canada: 2025 assessment”, Staff Analytical Note 14.

[50] Accenture (2024), Canada’s Generative AI Opportunity.

[36] Acemoglu, D. (2024), “The simple macroeconomics of AI”, Economic Policy, Vol. 40/121, pp. 13-58, https://doi.org/10.1093/epolic/eiae042.

[37] Aghion, P. and S. Bunel (2024), AI and Growth: Where Do We Stand?, https://www.frbsf.org/wp content/uploads/AI-and-Growth-Aghion-Bunel.pdf.

[47] Albouy, D. et al. (2019), “Local Labor Markets in Canada and the United States”, Journal of Labor Economics, Vol. 37/S2, pp. S533-S594, https://doi.org/10.1086/703579.

[8] Allen, R., W. Gu and R. Macdonald (2025), “Intangible Capital and Economic Growth in Canada”, Analytical Studies Branch Research Paper Series 482.

[19] André, C., P. Gal and M. Schief (2024), “Enhancing productivity and growth in an ageing society: Key mechanisms and policy options”, OECD Economics Department Working Papers, No. 1807, OECD Publishing, Paris, https://doi.org/10.1787/605b0787-en.

[18] André, C., P. Gal and M. Schief (2024), “Enhancing productivity and growth in an ageing society: Key mechanisms and policy options”, OECD Economics Department Working Papers, No. 1807, OECD Publishing, Paris, https://doi.org/10.1787/605b0787-en.

[45] Baldwin, J. and M. Willox (2016), “The Industry Origins of Canada’s Weaker Labour Productivity Performance and the Role of Structural Adjustment in the Post-2000 Period”, International Productivity Monitor, Vol. 31/Fall, pp. 19-36.

[13] Bank of Canada (2025), The impact of US trade policy on jobs and inflation in Canada, https://www.bankofcanada.ca/2025/06/the-impact-of-us-trade-policy-on-jobs-and-inflation-in-canada/.

[28] Bank of Canada (2023), Assessing the effects of higher immigration on the Canadian economy, https://www.bankofcanada.ca/wp-content/uploads/2023/12/san2023-17.pdf.

[23] Barabuffi, S. et al. (2025), “Knowledge spillovers through high-skilled migration network: evidence from OECD countries”, Industry and Innovation, pp. 1-31, https://doi.org/10.1080/13662716.2025.2451398.

[38] Briggs, J. and D. Kodnani (2023), The Potentially Large Effects of Artificial Intelligence on Economic Growth, Goldman Sachs Economics Research.

[39] Brynjolfsson, E., D. Rock and C. Syverson (2021), “The productivity J-curve: How intangibles complement general purpose technologies”, American Economic Journal: Macroeconomics, Vol. 13/1, pp. 333-372.

[30] C.D. Howe Institute. (2024), Optimizing immigration for economic growth, https://www.cdhowe.org/public-policy-research/optimizing-immigration-economic-growth.

[49] Calver, M. and A. Murray (2016), “Decomposing Multifactor Productivity Growth in Canada by Industry and Province, 1997-2014”, International Productivity Monitor, Vol. vol. 31/Fall, pp. pages 88-112.

[26] Campo, F., G. Forte and J. Portes (2024), The impact of migration on productivity: Evidence from the United Kingdom, https://kclpure.kcl.ac.uk/portal/files/267843415/accepted_manuscript_migration_and_productivity.pdf.

[35] Canadian Chamber of Commerce (2024), Building Canada’s net-zero workforce, https://chamber.ca/building-canadas-net-zero-workforce-report-highlights/.

[9] Chalaux, T. and Y. Guillemette (2019), “The OECD potential output estimation methodology”, OECD Economics Department Working Papers, No. 1563, OECD Publishing, Paris, https://doi.org/10.1787/4357c723-en.

[33] Dechezleprêtre, A. and M. Sato (2017), “The Impacts of Environmental Regulations on Competitiveness”, Review of Environmental Economics and Policy, Vol. 11/2, pp. 183-206, https://doi.org/10.1093/reep/rex013.

[51] Decker, R. et al. (2020), “Changing Business Dynamism and Productivity: Shocks versus Responsiveness”, American Economic Review, Vol. 110/12, pp. 3952-3990, https://doi.org/10.1257/aer.20190680.

[22] ESDC (2024), “Minister Boissonnault announces further Temporary Foreign Worker Program reforms to better protect the Canadian labour market and workers”, News Release, https://www.canada.ca/en/employment-social-development/news/2024/10/minister-boissonnault-announces-further-temporary-foreign-worker-program-reforms-to-better-protect-the-canadian-labour-market-and-workers.html.

[21] ESDC (2021), Evaluation of the Temporary Foreign Worker Program: Report.

[40] Filippucci, F., P. Gal and M. Schief (2024), “Miracle or Myth? Assessing the macroeconomic productivity gains from Artificial Intelligence”, OECD Artificial Intelligence Papers, No. 29, OECD Publishing, Paris, https://doi.org/10.1787/b524a072-en.

[17] Fluchtmann, J., A. Hijzen and A. Puymoyen (2025), Reviving growth in a time of workforce ageing: The role of job mobility, OECD.

[3] Government of Canada (2025), Implementation of Bill C-5: One Canadian economy.

[48] Guénette, J. and L. Shao (2025), “Assessing the Impact of Demographic Composition on Productivity”, Canadian Public Policy, Vol. 51/S1, pp. 22-40, https://doi.org/10.3138/cpp.2024-043.

[1] Guillemette, Y. and D. Turner (2021), “The long game: Fiscal outlooks to 2060 underline need for structural reform”, OECD Economic Policy Papers, No. 29, OECD Publishing, Paris, https://doi.org/10.1787/a112307e-en.

[15] Gu, W. (2019), “Frontier Firms, Productivity Dispersion and Aggregate Productivity Growth in Canada”, International Productivity Monitor, Vol. 37, pp. 96-119, https://ideas.repec.org/a/sls/ipmsls/v37y20194.html (accessed on 2 July 2025).

[14] Hahn, J., H. Hyatt and H. Janicki (2021), “Job ladders and growth in earnings, hours, and wages”, European Economic Review, Vol. 133, p. 103654, https://doi.org/10.1016/j.euroecorev.2021.103654.

[43] Lane, M., M. Williams and S. Broecke (2023), “The impact of AI on the workplace: Main findings from the OECD AI surveys of employers and workers”, OECD Social, Employment and Migration Working Papers, No. 288, OECD Publishing, Paris, https://doi.org/10.1787/ea0a0fe1-en.

[11] Lileeva, A. and D. Trefler (2010), “Improved Access to Foreign Markets Raises Plant-level Productivity…For Some Plants”, The Quarterly Journal of Economics, Vol. 125/3, pp. 1051-1099, https://doi.org/10.1162/QJEC.2010.125.3.1051.

[5] Loertscher, O. and P. Pujolas (2024), “Canadian productivity growth: Stuck in the oil sands”, Canadian Journal of Economics/Revue canadienne d’économique, Vol. 57/2, pp. 478-501, https://doi.org/10.1111/caje.12707.

[10] MacGee, J. and J. Rodrigue (2025), “The Distributional Origins of the Canada-US GDP and Labour Productivity Gaps”, Canadian Public Policy, Vol. 51/S1, pp. 76-103, https://doi.org/10.3138/cpp.2024-049.

[6] OECD (2026), Annual GDP and consumption per capita, US $, volume, constant PPPs, reference year 2020.

[4] OECD (2025), OECD Economic Surveys: Canada 2025, OECD Publishing, Paris, https://doi.org/10.1787/28f9e02c-en.

[7] OECD (2025), OECD Employment Outlook 2025: Can We Get Through the Demographic Crunch?, OECD Publishing, Paris, https://doi.org/10.1787/194a947b-en.

[24] OECD (2024), International Migration Outlook 2024, OECD Publishing, Paris, https://doi.org/10.1787/50b0353e-en.

[34] OECD (2024), OECD Employment Outlook 2024: The Net-Zero Transition and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/ac8b3538-en.

[32] OECD (2024), The Role of Economic Regulators in the Green Transition: Driving Sustainable Change in Network Sectors, OECD Publishing, Paris, https://doi.org/10.1787/7d4704c9-en.

[2] OECD (2023), Economic Policy Reforms 2023: Going for Growth, OECD Publishing, Paris, https://doi.org/10.1787/9953de23-en.

[31] OECD (2023), “Labour and social policies for the green transition”, OECD Social, Employment and Migration Working Papers, No. 295, OECD Publishing, Paris, https://doi.org/10.1787/028ffbeb-en.

[20] OECD (2020), Promoting an Age-Inclusive Workforce: Living, Learning and Earning Longer, OECD Publishing, Paris, https://doi.org/10.1787/59752153-en.

[25] Peri, G. (2016), “Immigrants, Productivity, and Labor Markets”, Journal of Economic Perspectives, Vol. 30/4, pp. 3-30, https://doi.org/10.1257/jep.30.4.3.

[44] Peri, G. (2014), “Do immigrant workers depress the wages of native workers?”, IZA World of Labor, https://doi.org/10.15185/izawol.42.

[29] Statisics Canada (2025), Economic and fiscal performance of immigrant-owned firms in Canada, https://www150.statcan.gc.ca/n1/pub/36-28-0001/2025002/article/00002-eng.htm.

[16] Statistics Canada (2026), Table 36 10 0 480 01 Labour productivity and related measures by business sector industry and by non-commercial activity consistent with the industry accounts, https://doi.org/10.25318/3610048001-eng.

[41] Statistics Canada (2024), Analysis on artificial intelligence use by businesses in Canada, Canadian Survey on Business Conditions, second quarter of 2024.

[27] Statistics Canada (2020), Immigration and firm productivity: Evidence from the Canadian Employer-Employee Dynamics Database, https://www150.statcan.gc.ca/n1/pub/11f0019m/11f0019m2020014-eng.htm.

[42] Statistics Canada Survey on Business Conditions (2024), Canadian Survey on Business Conditions, second quarter of 2024: Analysis on artificial intelligence use by businesses in Canada.

[12] Trefler, D. (2006), The Long and Short of the Canada-U.S. Free Trade Agreement, http://eprints.lse.ac.uk/6721/1/The_Long_and_Short_of_the_Canada-U.S._Free_Trade_Agreement.pdf.

Annex 2.A. Microdata analysis

Decomposing aggregate productivity growth

Productivity growth between t and t‑1 is decomposed into a contribution within and between firms and the between component in turn is decomposed into the contributions of different types of labour mobility using a worker-level decomposition following Hahn, Hyatt and Janicki (2021[14]). In each period t, and for each worker i, dummy variables capture the mobility state for job stayers $\left(r_{it}\right)$, job-to-job movers ($j_{it}$), hires from non-employment (${h_N}_{it}$) and separations to non-employment (${s_N}_{it}$). The total number of each worker type is further defined as $R_t$, $J_t$, ${H_N}_t,$ and ${S_N}_t$, respectively. The number of workers employed in period $t-1$ is then ${{{E_{t-1}= R}_t+J_t+S}_N}_t$ and ${{{E_t= R}_t+J_t+\mathrm{ }H}_N}_t$ in period $t$. With this, $∆y_t$ can be decomposed into the contributions of different labour market mobility types:

(A1) $∆y_t= \underset{Stayers}{\underbrace{\frac{\left(\frac{R_t}{E_t}+\frac{R_t}{E_{t-1}}\right){\sum }_i{r}_{it}∆y_{it} }{2R_t}}} + \underset{Job-to-job movers}{\underbrace{ \frac{\left(\frac{{\mathrm{J}}_{\mathrm{t}}}{E_t}+\frac{{\mathrm{J}}_{\mathrm{t}}}{E_{t-1}}\right){\sum }_i{j}_{it}∆y_{it} }{2{\mathrm{J}}_{\mathrm{t}}}}} +\underset{Entrants from non-employment}{\underbrace{\frac{\frac{{H_N}_t}{E_t}{\sum }_i{h_N}_{it}\left(y_{i,t}-\widetilde{y_t}\right) }{{H_N}_t}}}- \underset{Exits to non-employment}{\underbrace{\frac{\frac{{S_N}_t}{E_{t-1}}{\sum }_i{s_N}_{it}\left(y_{i,t-1}-\widetilde{y_t}\right) }{{S_N}_t}}}$

where $y_{it}$ denotes the productivity of firm j for individual i and year t and $\widetilde{y_t}\mathrm{ }$a reference level for productivity for non-employed workers based on a weighted average for stayers and job-to-job movers:

${\tilde{y}}_t=\frac{R_t}{R_t+ {\mathrm{J}}_{\mathrm{t}}}\left(\frac{{\sum }_i{r}_{it}\left({y_{i,t}+y}_{i,t-1}\right) }{{2R}_t}\right)+\frac{{\mathrm{J}}_{\mathrm{t}}}{R_t+ {\mathrm{J}}_{\mathrm{t}}}\left(\frac{{\sum }_i{j}_{it}\left({y_{i,t}+y}_{i,t-1}\right) }{2{\mathrm{J}}_{\mathrm{t}}}\right)$

Note that job stayers ($R_t$) and job-to-job movers (${\mathrm{J}}_{\mathrm{t}}$) do not change total employment between period t‑1 and t. Any changes in total employment represent hires from non-employment ($H_{N_t}$) and separations to non-employment ($S_{N_t}$).

Annex Table 2.A.1. Data sources

Country	Name	Source	Sample	Period
Canada	Canadian Employer-Employee Dynamics Database	Tax administration	Universe	2001-2019
Denmark	Integrerede Database for Arbejdsmarkedsforskning (IDA) and other data from Statistics Denmark	Tax administration	Universe	2000-2019
Finland	FOLK employment data from Statistics Finland, Employer Payroll Report from Tax Admin.	Tax administration	Universe	2000-2019
Hungary	ADMIN – I – Panel of administrative data (OEP, ONYF, NAV, NMH, OH)	Social security administration	50% random sample of workers	2003-2017
Italy	INPS-INVIND Panel	Social security administration	8.6% random sample of firms	2002-2019
Portugal	Quadros de Pessoal	Mandatory employer survey	Universe	2002-2019
Sweden	Longitudinell integrationsdatabas för sjukförsäkrings- och arbetsmarknadsstudier (LISA), Företagens ekonomi (FEK), Jobbregistret (JOBB)	Social security administration	Universe	2002-2018

Notes

← 1.  The role of different components related to demographics, employment and productivity is discussed below.

← 2.  The figure decomposes the growth rate in GDP per person (Y/Ptot) in its components related to the growth in the working-age population as a share of the total population (working age population, Pwa/Ptot), the growth in the number of employed persons in the working-age population (employment rate, E/Ptot) and the growth in GDP per employed person (labour productivity, Y/E). Formally, this can be represented as follows:

$\mathrm{\%}∆\left(\frac{Y}{P_{tot}}\right)=\mathrm{\%}∆\left(\frac{P_{wa}}{P_{pop}}\right)+\mathrm{\%}∆\left(\frac{E}{P_{wa}}\right)+\mathrm{\%}∆\left(\frac{Y}{E}\right)$

GDP per capita, employment and output are measured in potential terms following the OECD Going for Growth methodology (OECD, 2023[2]). The OECD average is obtained by PPP-based GDP weights across countries.

← 3.  The evolution of labour productivity growth is similar when measured in terms of workers or total hours worked (see Box 2.3). This suggests that the decline in labour productivity growth is not driven by developments in working time.

← 4.  The decomposition of economic growth into the contributions of labour productivity and employment is based the OECDs Going for Growth methodology (OECD, 2023[2]), which uses potential GDP per capita as the basis rather than actual GDP per capita. This explains why actual GDP per capita growth deviates somewhat from what is shown above, particularly during the COVID‑19 crisis and recovery period. Potential GDP per capita refers to the highest level of output per person that an economy can sustain over the long term without generating inflationary pressures. It represents the economy’s productive capacity under normal conditions, assuming that resources such as labour and capital are used efficiently. The use of potential GDP per capita does not have significant implications for the relative importance of changes in the growth rates of labour productivity and employment. See Box 2.3 for descriptive statistics on labour productivity growth using actual growth rates.

← 5.  In OECD data, MFP additionally accounts for improvements in labour composition.

← 6.  See the OECDs Going for Growth methodology for further details (OECD, 2023[2]).

← 7.  There is no consensus on the importance of the resource sector for the slowdown in aggregate productivity growth during the period 2001-2014. While some find that it was main driver behind the slowdown others suggest the slowdown was more broad based (Loertscher and Pujolas, 2024[5]; Baldwin and Willox, 2016[45]). The way relative price effects are taken into account in particular can lead to large variations in the contribution of the resource sector to aggregate productivity growth (Calver and Murray, 2016[49]).

← 8.  The analysis in this section builds on Chapter 5 of the OECD Employment Outlook 2025 “Reviving growth in a time of workforce ageing: The role of job mobility” (Fluchtmann, Hijzen and Puymoyen, 2025[17]). The analysis is based on the Canadian Employer-Employee Dynamic Database and builds on an ongoing collaboration between the OECD and Statistics Canada in the context of LinkEED 2.0 (https://www.oecd.org/en/about/projects/linkeed-200.html). The authors are particularly grateful to Tahsin Mehdi (Statistics Canada) and Jonas Fluchtmann (OECD) for their contributions.

← 9.  The decompositions presented in this section are implemented at the worker-level following the methodology proposed in Hahn, Hyatt and Janicki (2021[14]). An alternative would be to make use of firm-level decompositions as in Decker et al. (2020[51]) and Fluchtmann et al. (2025[17]). The main difference is that the worker-level decomposition used here is not restricted to incumbent firms. The qualitative results are broadly similar.

← 10.  Value added is deflated using national producer price deflators. Consequently, within-country productivity dynamics may partially be driven by relative output price changes between industries and firms.

← 11.  The analysis stops at 2019 to have a consistent end year for all countries and avoid getting into the COVID‑19 period.

← 12.  Matching aggregate trends using microdata is notoriously difficult. The fact that the microdata provide a qualitatively similar trends as the official macrodata is reassuring.

← 13.  The downside of using industry-level data is that productivity developments may reflect both within firm productivity growth as well as the effects of job reallocation across firms within industries. Productivity growth developments within firms by industry based on the microdata tend to be quite volatile and may be driven by output price developments. This is particularly the case for the energy sector (e.g. mining and petroleum). These within-firm developments are therefore not reported here.

← 14.  This was substantially more than on average across selected Eurozone economies (not shown) where the contribution of efficiency-enhancing job reallocation to aggregate productivity growth tended to be negligible (partly driven by Italy where job reallocation tended to reduce overall efficiency).

← 15.  See Guénette and Shao (2025[48]) for a cross-country panel-data analysis of the effects of changes in age composition on productivity (TFP or labour productivity), and a discussion of the implications of their estimates for Canada, China and the United States.

← 16.  Budget 2025 announces that the 2026‑2028 Immigration Levels Plan will stabilise permanent-resident admission targets at 380 000 per year for three years, while increasing the share of economic migrants from 59% to 64%. The plan will also lower the target for new temporary-resident admissions from 673 650 in 2025 to 385 000 in 2026, and to 370 000 in 2027 and 2028.

← 17.  Albouy et al. (2019[47]) further find that migrants in Canada are more responsive to changes in economic conditions, effectively “greasing the wheels” of the national labour market and reducing the local effects of trade shocks.

← 18.  Peri (2014[44]) shows that, although immigration can exert short-term downward pressure on wages, subsequent capital investment and adjustment mean that native workers’ wages are unaffected in the long run.

← 19.  As discussed in more detail in Chapters 3 and 4, overqualification is widespread in Canada not just for immigrants but also for natives.

← 20.  Data refer to 2017‑2019.

← 21.  The potential of AI may be similar to other General Purpose Technologies related to information and communication (e.g. internet, personal computer) and industrial production (e.g. electricity, steam engine).

← 22.  Estimates of AI’s potential impact on productivity vary considerably across studies. (Accenture, 2024[50]) suggests that AI could raise Canadian labour productivity by up to 8% by 2030, primarily via automation of routine tasks, advanced analytics, and more efficient resource allocation. Estimates by the Bank of Canada yield considerably smaller effects on total factor productivity of between 0.3 and 0.5% by 2035 (Abraham et al., 2025[46]).

← 23.  Canadian industries that embraced digital tools experienced higher productivity growth and greater crisis resilience, suggesting a strong association between digitalisation and productivity. The differences in performance are striking: from 2002 to 2019, labour productivity in digitally intensive industries rose 22.1% (cumulatively), over three times the growth in non-digital industries (only 6.3%) (Statistics Canada, 2021 www150.statcan.gc.ca).