Did the COVID-19 pandemic zombify the economy? A look at zombie firms

Zombie firms are businesses that persistently perform poorly over time without exiting, and their prevalence has been rising over time across many advanced economies. They negatively affect economic growth as they tend to be unproductive and compete with other healthy firms for scarce resources. In Canada, while the share of zombie firms was falling leading up to the COVID-19 pandemic, they were becoming less productive over time, were negatively affecting healthy firms and were increasingly lowering aggregate productivity. In 2019, aggregate productivity would have been up to 5% higher had zombie firms exited (Amundsen et al., 2023).

From 2020 to 2022, aggregate labour productivity declined in the business sector in Canada. Of interest is the potential long-term effects of the economic downturn that accompanied the pandemic. In addition, a large amount of support was offered to sustain Canadian businesses during the pandemic, including the Canada Emergency Wage Subsidy (CEWS), the Canada Emergency Commercial Rent Assistance (CECRA), the Canada Emergency Rent Subsidy (CERS) and the Canada Emergency Business Account (CEBA). Some have noted that the scale and swift implementation of pandemic-related measures globally had the potential to accelerate zombification and constrain the post-pandemic recovery (for example, OECD, 2020).

This article summarizes recent research by Amundsen et al. (2025) that examined zombie firm prevalence and the dynamics of these firms over the 2020-to-2022 pandemic years, the role of business supports in helping to prop up zombie firms or encourage their survival, and their contribution to the decline in Canada’s productivity. Zombie firms are defined as having earnings before interest and taxes less than their interest payments for 3 consecutive years and as being at least 10 years old.

The zombie firm share declined over the pandemic

From 2002 to 2019, the share of zombie firms in Canada hovered around 5% to 7% (Chart?1). It peaked in 2011 and declined after that until 2019. Over the 2020-to-2022 pandemic years, the share continued to decline. From 2011 to 2019, the share of zombie firms fell from 7.1% to 5.6%. From 2019 to 2022, that share dropped to 4.4%. This decrease was accompanied by corresponding declines in the zombie employment share and zombie capital share, although the zombie capital share generally increased up to 2020. From 2019 to 2022, the zombie employment share fell from 5.4% to 4.5%, while the zombie capital share fell from 8.4% to 7.5%. The capital share, however, remained elevated compared with historical levels. The declining prevalence of zombie firms highlights that the pandemic did not lead to a mass proliferation of these firms, which could have inhibited the post-pandemic recovery.

Data table for Chart 1

Zombie firms were more likely to exit over the pandemic

To begin evaluating the factors driving the lower zombie firm share, Table?1 presents the probability of transitioning between zombie and non-zombie status and exiting the economy. For the average of all three-year transition rates from 2002 to 2019, a zombie firm had a 24.5% chance of remaining a zombie firm three years later, a 46.7% chance of becoming a non-zombie firm and a 28.8% chance of exiting. By comparison, for the three-year transition period from 2019 to 2022, zombie firms were less likely to remain zombie firms, at 19.3%. This decline in persistence is entirely attributable to a greater exit rate, 34.4%, because the probability of a zombie firm recovering to non-zombie status declined only marginally from 46.7% to 46.4%. This shows that, compared with historical trends, zombie firms had a higher probability of exiting the economy over the 2020-to-2022 pandemic years, but their chance of recovering was only slightly diminished.

Table 1 Three-year transition matrix

Non-zombie firms also had a higher probability of exiting from 2019 to 2022—a period that includes the onset of the pandemic—compared with before the pandemic (+2.7 percentage points). However, unlike for zombie firms, where the increase was associated almost entirely with a decline in zombie persistence, the higher exit rate for non-zombie firms was associated more strongly with a lower probability of becoming a zombie firm (-1.7 percentage points) and less strongly with a decline in non-zombie persistence (-1.0 percentage point). This finding suggests that non-zombie firms that in the past would have become zombie firms instead exited during the pandemic.

The economic conditions led more firms to exit, while business supports helped firms survive and operate as non-zombie firms

What changes in the transitions are responsible for the decline in the prevalence of zombies? To answer this question, the contribution of each component of the transition matrix to the change in the zombie share is examined.1 Findings show that most of the decline in the zombie firm share was attributable to fewer non-zombie firms transitioning to zombie status and exiting instead. The contribution of the higher exit of zombies is smaller, because the zombie firm population is much smaller than the non-zombie firm population.

What factors are responsible for the changes in the transition rates, in particular the lower transition of non-zombie firms to zombie firms and their higher propensity to exit? Here, a multivariate regression model is used to assess the impact of firm characteristics, industry shocks, macroeconomic conditions and business supports (CEWS, CECRA, CERS and CEBA) on the probability of transitioning into a zombie firm, transitioning into a non-zombie firm and exiting.2 The negative impacts of the pandemic captured by aggregate and industry shocks in the model were found to cause zombie and non-zombie firms to exit in greater numbers than before the pandemic, reduce the probability that zombie firms would recover, and reduce the probability that non-zombie firms would become zombie firms. The presence of business supports mitigated the increase in exits caused by the pandemic. However, it did so mainly by helping non-zombie firms retain their non-zombie status and helping zombie firms recover to become non-zombies. The presence of business supports was associated with higher rates of non-zombie firms becoming zombie firms and greater zombie persistence, but the impacts were relatively small. The supports are associated with a 0.3 percentage point increase in the transition rate of non-zombie firms to zombie status, compared with the 1.7 percentage point decline that occurred. They are also related to a 0.9 percentage point increase in zombie persistence, compared with the actual 5.2 percentage point decline. Overall, the evidence points to the pandemic increasing the probability of zombie firm exit, while business supports helped zombie firms to recover into non-zombie ones. Both are consistent with zombification declining over the pandemic.

Zombie firms were not responsible for the decline in aggregate labour productivity over the pandemic

The study also examines whether zombie firms had an impact on aggregate labour productivity. It uses the decomposition proposed by Griliches and Regev (1995), which breaks down changes in aggregate productivity into changes in productivity within firms, reallocation of employment across firms, entry and exit. It is modified by separating some components into non-zombie and zombie firm elements to highlight certain dynamics.

Table?2 presents the results of the decomposition for the pandemic period. Aggregate labour productivity in the business sector (real total revenue per employee) decreased by $9,299 from 2019 to 2022. Negative contributions came from the decline in the productivity of continuing firms (-$8,015), which was made up of continuing non-zombie firms (-$6,668) and continuing zombie firms (-$1,348). Continuing firms that switched from zombie to non-zombie status made a positive contribution (+$1,104) to productivity, while continuing firms that switched from non-zombie to zombie status made a negative contribution (-$1,167).

Table 2 Decomposition of the change in aggregate labour productivity

The reallocation between continuing firms contributed positively to productivity (+$1,009). This finding shows that over the 2020-to-2022 pandemic years, employment in firms that had above-average productivity grew relative to that in firms with below-average productivity.

The combined entry and exit dynamics made a negative contribution (-$2,293). The entry of firms made a negative contribution to the change in aggregate productivity (-$10,411), highlighting that entrants are on average less productive than incumbents. The exit of firms made a positive contribution to productivity (+$8,118), which is consistent with exiting firms being less productive than the average. This contribution is composed of exiting non-zombie firms (+$6,553) and exiting zombie firms (+$1,565). The exit of zombie firms contributed disproportionately compared with their share in the economy, which is reflective of their lower productivity than exiting non-zombie firms.

Overall, it is the within-firm contribution of non-zombie firms that is the source of the decline in aggregate labour productivity during the pandemic, with zombie firms playing a limited role.3

Authors

Alexander Amundsen is with the International Monetary Fund. Amélie Lafrance-Cooke is with the Economic Analysis Division, Analytical Studies and Modelling Branch, at Statistics Canada. Danny Leung is with the Department of Finance Canada.

Disclaimer

The views expressed in this paper are those of the author(s) and do not necessarily represent the views of the International Monetary Fund (IMF), its Executive Board or IMF management. The views expressed herein do not reflect, in any way, the views of the Department of Finance Canada.

References

Amundsen, A., Lafrance, A., & Leung, D. (2023). Winter is coming? Zombie firms and ownership type in Canada.

Amundsen, A., Lafrance, A., & Leung, D. (2025). Firm Performance, Business Supports and Zombification over the Pandemic. IMF Working Paper (forthcoming). International Monetary Fund, Washington, DC.

Griliches, Z., & Regev, R. (1995). Firm productivity in Israeli industry 1979–1988. Journal of Econometrics. 65(1): 175–203.

OECD (Organisation for Economic Co-operation and Development). (2020). Government support and the COVID-19 pandemic. OECD Policy Responses to Coronavirus (COVID-19).

Notes

Note 1

Counterfactual zombie shares are calculated by changing one component of the transition matrix from its pre-pandemic value to its value during the pandemic, and then adjusting the other components proportionally so that the transition probabilities continue to sum to 1.

Return to note?referrer

Note 2

Multinomial logit models are used to model the transition probabilities of non-zombie and zombie firms before and during the pandemic. The estimated coefficients are then used to produce counterfactual scenarios on the transition rates.

Return to note?referrer

Note 3

A more detailed analysis by Amundsen et al. (2025) shows that the within-firm contribution of zombie firms has also not changed materially from historical levels.