Will it be a hard landing or a soft landing? The labour market will decide

Economic insights provided by Alberta Central Chief Economist Charles St-Arnaud.

Key takeaways:

 The Bank of Canada has tightened monetary policy meaningfully in its fight against inflation, raising the policy rate to 5.00%, well into restrictive territory and at its highest since the early 2000s.

• History has shown that it takes between 5 to 7 quarters on average between the moment monetary policy moves into restrictive territory and the start of a downturn, suggesting that the next six months may be the window to observe a more meaningful deterioration in the economy.
• Whether the economy goes through a soft or a hard landing depends on whether we only see a hiring freeze or broad-based layoffs.
• Households are particularly vulnerable to shocks given their high levels of indebtedness and rising insolvencies, suggesting that a rising share of households are already struggling.
• While higher interest rates will continue to have a significant impact on household finances in the near future, we argue that an income shock due to job losses would have a much bigger negative impact on the economic outlook, as borrowers will no longer be able to restructure their borrowing; this would force borrowers to sell assets or default on their obligations.
• Such a scenario could have important negative spillovers to the rest of the economy as housing prices drop and lenders reduce credit availability to protect their balance sheets from further losses and defaults.
• While the labour market looks resilient, it tends to be a lagging indicator of the economic cycle and is usually at its best on the eve of a downturn. Similarly, the intense labour shortages of the past two years may delay layoffs, but it may not prevent them entirely.
• Looking at previous downturns, we find that the Canadian economy typically shed about 600,000 jobs, in today’s equivalent, during a recession, leading to some important income losses. Job losses in previous recessions led to nearly 40,000 bankruptcies in today’s equivalent.
• While such a situation would be manageable, high levels of indebtedness and a decade of very low interest rates could mean that households are more sensitive to shocks than during previous recessions. As a result, the impact of a recession and job losses could be non-linear and bigger than suggested by historical patterns.

In its fight against inflation, the Bank of Canada increased its policy rate from 0.25% in early 2022 to 5.00% currently, its highest level in more than 20 years. The aim of this monetary tightening is to slow the demand side of the economy enough to create some excess supply, thereby easing inflationary pressures. The slack created in the economy will be visible in a rise in the unemployment rate, as low growth will lead to weaker hiring.

However, there is a lot of uncertainty as to what the full impact of the monetary tightening will be on the economy. After rising by 475bp since early 2022, monetary policy is now in restrictive territory and the real policy rate is positive. As we have shown in, “What happened to the recession? The role of the policy stance and demographic“, history shows that it usually takes between 5 to 7 quarters, on average, between the moment monetary policy moves into restrictive territory and the start of a downturn.

Recent economic indicators suggest that economic activity has stalled, with the Canadian economy likely having experienced a mild technical recession in the second and third quarters of this year. Whether the economy will go through a soft or a hard landing depends on whether we only see a hiring freeze or broad-based layoffs.

Households are vulnerable to shocks, especially to income shocks

Households in Canada are heavily indebted, with a total debt of $2.9 trillion, or about $176,000 per household, as of 2023Q2. The debt-to-disposable income remains close to a record high at about 180%. Due to this high level of debt, households are currently spending almost 15% of their disposable income on debt service. Further rises in the debt-service ratio should be expected in the coming years as some of the borrowing gets renewed at higher interest rates.

We have estimated in the past (see The great consumer squeeze) that about 25% of fixed-rate mortgages need to be renewed every year. The interest rate shock on monthly payments following renewals will depend on the difference between interest rates at renewal and the interest rate when the initial mortgage was issued.

Using the change in the 5-year interest rate compared to their level 5 years ago as an estimate, a borrower currently renewing a mortgage taken 5 years ago would see their interest rate increase by about 200bp. It is important to note that, based on this proxy for change in interest rates at renewal, it is the first time in almost four decades that mortgage borrowers will face higher interest rates at renewal. Over the past decade, households have generally renewed their mortgages at lower interest rates.

We estimate that this could translate into an increase in mortgage payments of more than $700 per month, based on the benchmark prices that prevailed at the end of 2018. However, the impact will be much more significant in expensive cities, with the increase closer to $1000 per month in Toronto and $1200 in Vancouver. While these increases are steep, borrowers have options to mitigate the impact of higher interest rates on their mortgage payments and reduce the financial strain; one of these options is to extend the amortization period to spread the rise in monthly payments.

The insolvency data already shows that the number of consumer insolvencies in Canada is back to its pre-pandemic level, signaling that many households are under severe financial strains. However, the details show that the increase in insolvencies is due to a rise in proposals (i.e. a renegotiation of terms) rather than bankruptcies.

The distinction is important. The rise in proposals shows that lenders are being flexible and prefer to modify the loan structure rather than forcing bankruptcy on  borrowers, a win-win situation for both; the lender avoids having to foreclose and take the loss, while the borrower avoids bankruptcy in exchange for a higher lending cost by paying interest over a longer period. However, such arrangements only work if the borrower makes the required payments. It would be a very different story if the borrower could not make his payments because of a decline in income; in this case, the lender would likely foreclose the loan.

Hence, the strength of the labour market matters for the economic outlook. In the case of an underperformance in hiring, the unemployment rate increases, but the negative shock to household income is small. Broad-based job losses would have a more important negative impact on household income. Given the level of indebtedness, it is clear that the average household in Canada requires two income earners to cover its regular expenses and debt repayment. If one source of income disappears, the debt sustainability of that household is likely to evaporate.

The impact of rising layoffs

Job losses and the associated drop in income could cause important negative spillovers to the economy, especially due to its  heavy strain on  household finances.

1. It could lead to forced selling in the housing market, as households who can no longer make their payments sell their homes. The influx of new listings could drastically change the housing market’s equilibrium at a time when demand is also likely to be weaker. Potential buyers will likely stay on the sideline if job security is perceived as low and prices are expected to decline. This would lead to a drop in house prices; also, the associated negative wealth effect and uncertainty will likely reduce household spending.
2. A drop in income would likely lead to a rise in defaults and losses at financial institutions. This would lead to lenders turning risk-averse to protect their balance sheets and reducing the availability of credit to households and businesses, likely leading to higher borrowing costs. The resulting credit crush would have a negative impact on business investment, consumer spending and the housing market as it would be more difficult to obtain a mortgage.

As the economy slows further due to these factors, there is a risk of further job losses and more financial strains on some households, exacerbating the downturn and leading to a deeper recession.

How solid is the labour market? 

The Canadian labour market has been resilient since the BoC’s hiking cycle started. While the unemployment rate has increased by 0.8pp since the beginning of 2022, it remains close to its historical lows. Similarly, the employment rate, the share of the working-age population employed, remains high by historical standards despite solid population growth and an aging population. However, it is important to understand that labour market statistics are usually strongest on the eve of a recession. So, a strong labour market now does not prevent a recession in the near future. Moreover, the labour market tends to be a lagging indicator of the economic cycle, usually deteriorating after the start of a downturn.

One specificity of the current cycle has been the intense labour shortage that occurred as the economy recovered from the economic shutdowns due to the pandemic. As the economy reopened in 2021, many businesses couldn’t hire the workers needed to satisfy the sharp increase in demand, adding to  supply chain disruptions. At the height of the problem at the end of 2021, more than 50% of small businesses reported shortages of skilled labour as a factor limiting business activity. Almost 80% of companies surveyed for the BoC’s Business Outlook Survey said that the labour shortage was becoming more intense. As a result, the vacancy rate, the number of job vacancies relative to employment, reached record high levels, 5.7%, according to official data, with about a million job vacancies in Canada.

With the economy slowing as the impact of higher interest rates is being felt, the intensity of the labour shortage has eased. Employers have responded to a slowing economy by freezing hiring and cancelling job postings. Nevertheless, employment in Canada continues to increase, with about 500k jobs added over the past year. The number of vacancies has gradually declined, and the vacancy rate is now at 3.8%, yet still above its pre-pandemic levels. However, although half of small businesses are still reporting skilled labour shortages, the intensity of the shortages has diminished substantially.

The question is whether the intense labour shortage over the past two years could prevent layoffs as the economy continues to slow. The logic is that businesses had to compete intensely to acquire staff and may be more willing to hold on to them so they do not have to be in the same situation of intense labour shortages when the economy recovers. Such behaviour is likely to delay layoffs, but it is not clear whether it would prevent them entirely; there is a point where a business may have no other choice but to reduce its payrolls to reduce costs.

With history suggesting that a downturn usually occurs 5 to 7 quarters after the policy rate enters into restrictive territory (see What happened to the recession? The role of the policy stance and demographic), the next six months may be the window to observe a more meaningful deterioration in the labour market.

What happens during a recession?

To better understand what would happen to the labour market and income during a downturn, we study how they were affected during previous recessions. First, we look at the decline in employment from peak to through in percentage and also express this change based on today’s size of the labour market (there are currently approximately 20.3M working Canadians).

Second, we look at how income was impacted by this loss in employment and the decline in economic activity. To do so, we use disposable income relative to its pre-recession trend. The use of disposable income is important as it considers the impact of automatic stabilizers on income, such as employment insurance payments, includes other sources of income beyond wages and salaries, and represents the amount available for spending, debt repayment and saving. We also consider the impact on nominal and real income to consider whether the inflationary regime matters. Finally, we examine  income loss up to two years  after the start of the recession as the weakness in the labour market tends to linger even as the economy starts to recover. Hence,  income loss increases over time.

We can make the following observations:

1. On average, employment declines by about 2.5% during a recession. This would equate to a decline of about half a million job losses in today’s equivalent. However, there is considerable volatility. In the early 1980s, employment declined by 5.4% (or almost 1.1 million jobs in today’s equivalent), while it declined by only 0.9% (or 50k) in the early 2000s.
2. The real income losses are slightly bigger on average than the nominal losses. After 1-year, on average nominal disposable income and real disposable income are 2.2% and 2.7% lower, respectively. After 2 years, they are 2.7% and 4.4% lower on average.
3. The nominal income losses during the deflationary recession of the 1970s, 1980s and 1990s have been much bigger than the loss in real terms. This is because of the lower inflation regime after these recessions.

The early 2000s downturn is interesting because it does not qualify as a recession, based on the C.D. Howe Institute Business Cycle Council. However, the BoC had to react meaningfully by providing significant stimulus by cutting the policy rate from 5.75% to 2.00%. Moreover, despite very modest job losses, the declines in income during this time are amongst the biggest of all the periods considered.

What the 2000s experience shows is that it is possible to have a big underperformance in income without significant job losses. However, what matters most is whether the decline in employment and income leads to a rise in the number of struggling households and insolvencies.

Looking at the same recession periods, we find that, on average, bankruptcies increase by almost 30% following a recession. Similarly, the bankruptcy rate, the number of bankruptcies per thousand persons, rises by about 0.7 points if we exclude the 2001 episode. The 2001 downturn is again an outlier with the smallest rise in bankruptcies and bankruptcy rate. This suggests the shortfall in income linked to employment loss generally leads to more financial struggle.

Translating these numbers into today’s context using the change in bankruptcy rate, a rise of 0.6 points in the number of bankruptcies per 1000 workers would mean an increase of about 23,600 bankruptcies. This would bring the number of bankruptcies to about 49,000, or about 90% higher than it is currently. While this seems big, the resulting level of bankruptcies would still be below its pre-pandemic level. Assuming a rise of 0.9 points in the bankruptcy rate, similar to 2008, will mean almost 30,000 bankruptcies.

While these historical patterns and estimates provide a scale of the possible impact of a recession on bankruptcies and suggests that the rise in bankruptcies could be manageable, the current situation is very different. This is because: 1) the share of bankruptcies in total insolvencies is at its lowest level on record at a bit less than 25%, while it was 80% back in 2009. 2) household debt-to-disposable income is much higher than at any point on the eve of a recession, and 3) interest rates have been close to their historical lows for more than a decade and may have led households to be ill-prepared for higher interest rates. As a result, the impact of a recession and job losses could be non-linear and bigger than suggested by historical patterns.

Conclusion

As stated earlier, whether the Canadian economy has a hard or soft landing will depend on whether there is only a hiring freeze or broad-based layoffs. This is especially true since Canadian households, given their indebtedness level and high debt-service ratio, have never been this vulnerable to a potential recession.. While we have seen that they have so far weathered the increase in interest rates rather well, it could be a different story if an income shock, in the form of job losses, occurred. This is especially true if a weaker household sector causes further economic weakness and job losses.

The continued high level of labour shortage and strength in the labour market could reduce the likelihood of significant job losses compared to previous recessions. Nevertheless, looking at a decline in employment and income loss and its impact on bankruptcies during previous recessions, it seems that the Canadian economy could weather a recession.

However, it must be noted that the current level of indebtedness is more significant than at the eve of any past recession and would also follow more than a decade of very low interest rates. This could make households more sensitive to shocks than previous recessions, meaning that looking at past recessions could lead to an underestimation of the impact.