COVID‑19 poses financial challenges for Canadian households
Economic activity came to a sudden halt due to the COVID‑19 pandemic. This has resulted in widespread income losses, creating a challenging situation for many Canadian households, especially those that are highly indebted. From a financial stability perspective, a key concern is whether households can keep up with their debt payments.
The potential for a trigger event to occur when household debt is high has long been cited as the biggest risk facing Canada’s financial system (Slive and Coletti 2018). But the pandemic is different than the triggers and related scenarios that economists typically use to assess the risks to financial stability.
We examine how the COVID‑19 shock affects Canadian households by drawing parallels between pandemics and natural disasters. We then assess the financial health of the household sector when the pandemic began. Finally, we run model simulations to illustrate how payment deferrals and the recovery of the labour market affect mortgage defaults.
Natural disasters versus pandemics
The economic impact of COVID-19 is often compared with past recessions, but this pandemic arguably has more in common with natural disasters. The key feature shared by natural disasters and pandemics is a sudden stop of economic activity caused by a shock that is unrelated to economic factors—in this case, a public health crisis. This contrasts with the 2008 recession, which reflected an underlying fragility in the global financial system that resulted in a lengthy downturn.
We see this when we compare the number of recipients of employment insurance (EI) in Fort McMurray, Alberta, after both the 2008 recession and the 2016 wildfires—the costliest natural disaster in Canadian history (Statistics Canada 2017) (Chart 1). The peak increase in the number of EI recipients was much greater after the wildfires than it was following the recession. But this increase had fully reversed within 10 months of the wildfires. In contrast, the peak occurred near the 10‑month mark following the start of the recession.
Chart 1: Change in the number of employment insurance beneficiaries in Fort McMurray
Sources: Statistics Canada and Bank of Canada calculations
Lessons from Fort McMurray
What lessons about financial stability can we learn from Fort McMurray? To find out, we use anonymous credit files from TransUnion Canada1 and apply what is known as the synthetic control method. This means we compare outcomes in Fort McMurray with those of a control group of residents of Alberta who shared similar demographic and financial characteristics (e.g., age, credit score) before the wildfires. We are interested in studying the rate of mortgage arrears, which is defined as the share of mortgages behind on payments by at least 90 days. Banks use this standard to determine when a payment is in default. The Fort McMurray case study presents insights from research originally conducted to improve the Bank of Canada’s understanding of the relationship between climate change and financial stability (Molico 2019).
The mortgage arrears rate rose sharply after the wildfires, from 0.3 percent to a peak of 1.4 percent (Chart 2). Based on the difference between the arrears rate in Fort McMurray and the control group, we attribute 0.9 percentage points of the increase in arrears to the impact of the wildfires. For perspective, the peak mortgage arrears rate at the national level after the last recession was only 0.45 percent. Importantly, though, the rise in mortgage arrears after the wildfires was short-lived. This likely reflects both the temporary nature of the shock and the effectiveness of disaster relief policies. Indeed, policies implemented to support residents of Fort McMurray were very similar to what is being put in place on a much larger scale during the pandemic (Table 1).
Chart 2: Mortgage arrears rate
Sources: TransUnion Canada and Bank of Canada calculationsLast observation: December 2017
Table 1: Policies to support households
|Direct income support||Debit cards from Alberta government
Emergency funds from Canadian Red Cross
|Canada Emergency Response Benefit (CERB)
Enhancement to Canada Child Benefit (CCB)
Various programs at the federal, provincial and territorial levels
|Mortgage payment deferrals||Up to four months||Up to six months|
|Tax relief||Canada Revenue Agency (CRA) taxpayer relief provisions||CRA taxpayer relief provisions
Increased tax credits
|Expedited employment insurance (EI) claims||Special access code for Fort McMurray residents||Redirection of EI claims to new CERB system|
Note that the sharp but short-lived increase in mortgage arrears shown in Chart 2 is likely overstated because our data include mortgage payment deferrals that we cannot explicitly identify. The mortgage arrears rate does, however, persistently remain about 0.25 percentage points higher than that of the control group. This could mean that Fort McMurray had a greater economic fallout from the 2015–16 decline in oil prices than the parts of Alberta represented in our control group.
This persistent increase in mortgage arrears is also consistent with the broader literature on natural disasters, which uncovers long-term scarring effects of disasters on financial health (Ratcliffe et al. 2019). For example, credit scores usually decline after natural disasters. That initial decline leads to further declines as consumers lose access to or face higher costs for traditional forms of credit. This highlights why ensuring that households receive adequate financial support during this pandemic is important—especially given the high level of debt held by households when the pandemic began.
Economic recovery from the pandemic could be challenging
There are similarities in the way the initial economic impacts of pandemics and natural disasters unfold. But their economic recovery can look quite different.
Disasters tend to be localized and pass relatively quickly. They are also associated with physical destruction of capital that subsequently needs to be rebuilt. This rebuilding process can begin relatively soon after a disaster has occurred, contributing directly to the broader economic recovery.
In contrast, the COVID‑19 pandemic has a global reach, and its aftermath is much more uncertain. Economic activity will undoubtedly rebound as mandated lockdowns are gradually eased. However, this will likely be a sluggish process, meaning some of the macrofinancial effects of the pandemic could linger.
To what extent can households weather the storm? This ultimately depends on:
- the financial health of households when the shock began
- the effectiveness of policy actions aimed at bridging the road to recovery
- the speed at which the labour market recovers
Many mortgage holders have limited financial buffers
One way to gauge how well indebted households can cope with temporary income losses is to calculate the ratio of financial assets to mortgage payments. This ratio measures the number of months households can continue making mortgage payments by drawing only on their liquid assets. This simple metric does not consider other debt or essential expenses. Nevertheless, it provides a useful summary of available financial buffers.
Using the data from the 2016 Survey of Financial Security, we find that one in five households can only make up to two months of mortgage payments using liquid assets and about one-third can make up to four months of payments. We find that households in the occupations most at risk in the near-term—such as sales and trades—also have the weakest financial positions (Chart 3). Almost one-quarter of households in these occupations can only make up to two months of payments.
Chart 3: Months of mortgage payments using liquid assets, by occupation
Chart 3: Months of mortgage payments using liquid assets, by occupation
Source: Statistics Canada, special tabulation based on the 2016 Survey of Financial Security
Home equity lines of credit can provide emergency funds but at the expense of greater borrowing
Although many mortgage holders have limited financial buffers, home equity lines of credit (HELOCs) provide a flexible and relatively low-cost means of accessing cash. Lenders have been increasingly offering HELOCs as part of combined mortgage-HELOC plans, resulting in a large pool of untapped available credit (Al-Mqbali et al. 2019). Using new regulatory data, we find that about 65 percent of the funds authorized through HELOCs have not been used (Chart 4). This amounts to Can$310 billion, or roughly 20 percent, of households’ disposable income.
Chart 4: Distribution of authorized HELOC amounts by utilization rate
Sources: Regulatory filings of Canadian banks (J2 return) and Bank of Canada calculationsLast observation: 2019Q4
From a financial stability perspective, significant increases in HELOC use to cope with COVID-related income losses is not ideal because it could add to financial vulnerabilities in the future. Nevertheless, emergency funds accessed through HELOCs can help prevent the more serious financial stability risk of mortgage default. Government support programs combined with reduced household spending during the containment period should help limit how much HELOCs are relied on in the near term.
Fiscal policy measures are replacing a portion of lost incomes
Fiscal policy measures are playing an important role in supporting households through this difficult period. The most pertinent of these measures is the Canada Emergency Response Benefit (CERB), which provides individuals experiencing income loss due to COVID‑19 with $2,000 per month for up to four months. Other federal measures provide additional support to lower-income households and households with children. Some provinces are also providing direct financial support to households.
To put these policy measures in perspective, we use microdata from the 2017 Survey of Household Spending to document typical monthly expenses for mortgage holders and renters across different income groups (Table 2). We see that the amount of financial support being provided should help support expenditures on core items such as food and shelter, especially for lower-income renters. While some mortgage holders could face greater difficulty meeting their expenses, payment deferrals being offered by most lenders will provide additional financial support. Note that financial stress among renters could have indirect consequences for financial stability. For example, about 700,000 households are also landlords, and 80 percent of them have mortgages.
Table 2: Typical household expenses
|Internet and cell phone ($)||129||157||197||80||161||218|
|All other consumption ($)||1,596||2,416||3,982||1,026||2,121||3,660|
|Proportion of single-person households (%)||48||16||3||66||25||8|
Note: Households in all Canadian provinces are divided into three equal groups by incomes. Figures are the medians for the selected subgroups in each of these terciles.
Sources: Statistics Canada 2017 Survey of Household Spending and Bank of Canada calculations
Model simulations suggest important role for payment deferrals
The link between the COVID‑19 shock and financial stability risks caused by high household debt can be analyzed more formally using the Bank’s Household Risk Assessment Model (HRAM, Peterson and Roberts 2016). Part of the Bank’s stress-testing tool kit, this model has the important advantage of being able to simulate the impact of macroeconomic shocks at the micro level. In other words, HRAM considers the great deal of diversity in household balance sheets, which is key to understanding how financial vulnerabilities interact with a given macroeconomic scenario. The main output of HRAM is the mortgage arrears rate.
We simulate the mortgage arrears rate by using realized data on the impact of the COVID‑19 shock on the labour market and making illustrative assumptions on how the labour market might evolve. In HRAM, the unemployment rate is the most important variable in determining the likely path of the mortgage arrears rate.
Income losses are not fully reflected in the unemployment rate
The unemployment rate is, however, unlikely to fully capture the current scale of employment income losses. This is because many unemployed individuals are being classified as “not in the labour force” due to their inability to actively seek work during the lockdown period. In addition, a large share of the labour force remains employed but is losing all or most of their usual working hours.
Fortunately, Statistics Canada is calculating an alternative labour underutilization rate that we can use in our simulations instead of the traditional unemployment rate. We conduct our simulations with the assumption that April’s underutilization rate of 36 percent (Statistics Canada 2020) represents the peak and a full recovery will occur within four quarters.2 To be clear, this is not a prediction; it is simply an illustration of how much mortgage defaults could rise in a deep but short-lived recession.3
Another important aspect of our simulations is the ability of households to defer mortgage payments. Most lenders in Canada are allowing households affected financially by COVID‑19 to defer payments for up to six months. That means the state of the labour market at the end of the six-month deferral period will ultimately determine mortgage arrears. To put into perspective just how important this factor is, we show our simulations with and without payment deferrals (Chart 5).
Chart 5: Simulated mortgage arrears rate
Source: Bank of Canada calculations
Without deferrals, the mortgage arrears rate reaches a peak of 1.3 percent. This is slightly higher than the historical peak of 1 percent witnessed in the early 1980s. Once we account for payment deferrals, the picture is much more favourable. With deferrals, the arrears rate remains relatively flat over much of 2020, eventually rising to a peak of 0.5 percent in the second quarter of 2021. The increase in arrears in 2021 occurs because, with the assumed four-quarter labour market recovery, the unemployment rate does not fully recover once the six-month deferral period ends.
Ultimately, this exercise illustrates that the effectiveness of deferrals in limiting the rise in arrears depends crucially on the speed of recovery in the labour market.
- 1. To protect the privacy of Canadians, no personal information was provided to the Bank by TransUnion. The TransUnion dataset was “anonymized,” meaning that it does not include information that identifies individual Canadians, such as names, social insurance numbers or addresses. In addition, the dataset has a panel structure, which uses fictitious account and consumer numbers assigned by TransUnion.[←]
- 2. The effective unemployment rate entering the model is a bit lower because we adjust for the historical gap between the unemployment rate and underutilization rate.[←]
- 3. The Financial System Review considers a more severe and persistent macroeconomic scenario, modelled on the lower bound of the range depicted in the April 2020 Monetary Policy Report, to stress test Canada’s six largest banks.[←]
- Al-Mqbali, L., O. Bilyk, S. Caputo and J. Younker. 2019. “Reassessing the Growth of HELOCs in Canada Using New Regulatory Data.” Bank of Canada Staff Analytical Note No. 2019-14.
- Molico, M. 2019. “Researching the Economic Impacts of Climate Change.”
- Peterson, B. and T. Roberts. 2016. “Household Risk Assessment Model.” Bank of Canada Technical Report No. 106.
- Ratcliffe, C., W. J. Congdon, A. Stanczyk, D. Teles, C. Martín and B. Kotapati. 2019. “Insult to Injury: Natural Disasters and Residents’ Financial Health.” Urban Institute Research Report.
- Slive, J. and D. Coletti. 2018. “Keeping the Financial System Healthy.” Bank of Canada The Economy, Plain and Simple, October 16.
- Statistics Canada. 2017. “Infographic: Fort McMurray 2016 Wildfire—Economic Impact.”
- Statistics Canada. 2020. Labour Force Survey, April 2020.
Bank of Canada staff analytical notes are short articles that focus on topical issues relevant to the current economic and financial context, produced independently from the Bank’s Governing Council. This work may support or challenge prevailing policy orthodoxy. Therefore, the views expressed in this note are solely those of the authors and may differ from official Bank of Canada views. No responsibility for them should be attributed to the Bank.