Change theme
Change theme

Rebalancing the economy while managing risks

Available as: PDF


Good afternoon. I am delighted to be here in Regina and would like to thank you for this opportunity to speak with you today.

Before I begin, I would like to take a moment to congratulate a local winner of this year’s Bank of Canada Museum Award for Excellence in Teaching Economics. Andre Boutin Maloney of Bert Fox Community High School in Fort Qu’Appelle developed an interactive trading post simulation that combined the teaching of Indigenous history and economic principles. Congratulations to Andre and the entire class. The 2024 edition of this award launched recently. If you know of a deserving economics teacher, you can submit a nomination on the Museum’s website.

My speech today will review how the events of the past few years have affected households in different ways. Even before the COVID-19 pandemic, every household had different levels of income, wealth and debt. And since the pandemic arrived in Canada in early 2020, economic circumstances have shifted dramatically and repeatedly. Multiple levels of government and the Bank of Canada acted in response.

Policies and economic developments have shaped the ways individual households and businesses experienced the effects of the pandemic. And this, in turn, has influenced our subsequent policy decisions. As we have made decisions in this complex and evolving environment, it has been important to understand how these decisions affect different households and businesses. 

How the pandemic affected individual households

So let me jump into how the pandemic affected individual households. Aggregate economic data do not always tell the full story. This type of data tells a story of averages, but no household is ever truly average. A household’s financial circumstances and spending patterns are as unique as the household itself. Not surprisingly, households experienced the unusual economic events of the past few years in very different ways.

Initially, lockdowns forced some businesses to close entirely and others to limit the number of shoppers they welcomed. Simply, people could spend their money in fewer places. Many businesses saw their sales and revenues fall, and many workers were laid off. Against this backdrop, governments increased support programs and the Bank of Canada eased monetary policy.

An important development during the pandemic was large swings in both demand and supply. Initially, demand spiked for goods that could help people navigate the public health crisis, but, more generally, demand for many other things plummeted. With people staying home and demand in sharp decline, total inflation was actually bouncing around 0% for several months in 2020.

Obviously, that didn’t last. As many workers settled into a new home-centred life, demand for things like home office equipment and sporting goods soared. But new challenges began to emerge. Household spending on goods rebounded quite quickly (especially on items that could be ordered online), but supply was constrained by disruptions to global supply chains. We had seen supply chain disruptions before, but never anything like this. With strong demand and constrained supply, prices for some goods began to rise, and this was an early inflationary force.

Around the same time, an unprecedented boom in real estate activity began to take root and drove house prices to record highs across the country. Similar dynamics were occurring in other countries. Lockdowns had allowed many middle- and upper-income households to accumulate a significant amount of savings. These increased savings along with low interest rates and a desire for more living space fuelled the demand. But the boom affected individual households in starkly different ways, and these differences have continued to influence our policy deliberations. Here’s why.

About two-thirds of Canadian households own their home, and the home equity of these households increased as house prices rose. But some households took on high levels of mortgage debt to finance real estate purchases in a market that strongly favoured sellers. And while the boom eventually slowed, one of its lasting legacies is that some households have higher home equity and others have much higher levels of debt.

The context in which this occurred is important. Even before the pandemic, interest rates had been very low for a long time and some households had accumulated considerable debt. The current elevated debt levels have affected how higher interest rates affect individual households and have changed the relative importance of the channels through which monetary policy is transmitted.

High household debt levels and the channels of monetary policy transmission

Let me delve into a bit more detail about the channels of monetary policy transmission and why one of them has been having an unusually large effect during this tightening cycle.

The simple way of thinking about how monetary policy works is that when we raise the policy interest rate, other interest rates also tend to go up and the Canadian dollar often appreciates. These effects slow both domestic demand and net exports and reduce price pressures. But a lot is going on under the hood. While increases in interest rates tend to slow demand, the impact of these rate increases differs across households, and those differences can influence the ways that monetary policy takes effect.

So far, the largest impacts have been associated with what is known as a cash-flow, or income, channel. This partly reflects the fact that households whose debt-service costs vary with interest rates like prime have felt large direct impacts. To date, about 15% of households are dealing with interest rate increases and have a higher mortgage payment than they did in February 20221. As time goes on, more people will be renewing their mortgages, so more households will face higher payments.

Let me add that a similar dynamic may be driving up rents. Landlords with mortgages may also be paying higher interest costs and passing them along to renters. In addition, higher mortgage rates make it more difficult to qualify for a mortgage, which may push some households into the rental market. The average rent across Canada was up over 6% in August.

Higher interest rates increase the carrying costs of debt that people already have2. Households with variable-rate mortgages and variable payments feel this effect almost immediately when rates rise. So do people with bank loans with variable interest rates, such as home equity lines of credit. Higher shelter costs and debt payments mean that people have less income left over to spend on other things like restaurant meals or vacations.

So far, this cash-flow channel has not only had the largest effects, it has also been more important than in past cycles. I’ll get to why shortly.

Traditionally, the main channel through which monetary policy affects demand has been a change in the timing of people’s spending. To understand this channel, it is important to use the concept of real interest rates. This is the level of interest rates you are used to hearing about, minus expected inflation. The stance of monetary policy, as measured by real interest rates, entered restrictive territory only recently in the wake of a higher policy rate and declining inflation expectations.

Real interest rates influence the choice between buying something immediately or delaying the purchase to a future date. The basic idea is that when real interest rates are high, the return on savings is high compared with the expected increase in prices. So, instead of purchasing something now, you could earn interest on your savings, make the purchase in the future and end up with some money left over. Whether they are viewed as increasing the cost of borrowing or increasing the return to savings, higher real interest rates help reduce demand in the economy and bring inflation down.

A third channel through which policy rate increases can soften demand is a wealth effect. When interest rates increase, the value of many assets—such as houses and stocks—tends to decline. When consumers see their wealth decline, they may pull back on their spending. While asset prices may respond relatively quickly to higher interest rates, the impact of decreasing wealth on consumption tends to be slower.

We take all three of these channels into account when we are setting monetary policy and monitoring how well it is working. We have seen demand slow since we started increasing the policy interest rate. This slowing started in the housing sector but subsequently spread to spending on other interest-rate-sensitive goods. It also showed up as weaker growth of borrowing. On the savings side, we have seen a shift in personal deposits toward term deposits, which tend to have higher interest rates (Chart 1).

And we have seen both consumer price index (CPI) inflation and core inflation decline. But as we have been making decisions, we have been faced with novel challenges in setting monetary policy.

A great balancing act: Accounting for the differences among Canadians when setting monetary policy

The truth is that the past two years have been the biggest test of our inflation-targeting framework. This framework was introduced in 1991. Since then, inflation has sometimes been above the upper end of our inflation control range of 1% to 3%. However, these episodes tended to be short-lived and were frequently associated with temporary surges in only a subset of prices.

This time has been different.

The biggest difference is that inflation rose to a much higher level and has exceeded the control range for longer. As well, outsized increases in prices have been widespread. People at all income levels have been affected, and the effects don’t always link directly to a household’s income level. Higher fuel prices have disproportionate effects for people who live in parts of the country where they heat their homes with oil or where they need to drive long distances. And certain sectors of the economy are impacted much more than others. Agriculture, for example, makes extensive use of gasoline and diesel fuel, and farmers will feel the pinch of higher energy prices much more than most.

The second big difference this time is where things were when all this started. Before the pandemic, we had been in an environment of low interest rates for over a decade (Chart 2). Low interest rates made it easier for households to borrow more money. Average household debt levels were high entering the pandemic. Then, during the pandemic, even as some households used excess savings to pay down debt, low interest rates encouraged others to borrow more.

Because debt was elevated and rates were so low for so long, the cash-flow effect was much more important than in the past.

Low for long, in particular, implies that all five-year fixed-rate mortgages issued before 2022 will likely face significantly higher interest rates when they are renewed (Chart 3). That is what has been happening, and it could continue for an additional three and a half years. The resulting drag on spending could last even longer.

A third difference, as I mentioned earlier, is that during the pandemic, some households gained home equity wealth, some gained stock market wealth and some just consumed a lot less than usual and built up their savings.

In other words, the current context is a complicated one. Because inflation was very high last year, interest rates had to rise a lot. That’s why we acted forcefully and brought them up quickly. Inflation has come down, but it is still too high. And that tends to mean that real interest rates need to remain high.

The current level of inflation is not the only relevant factor, though. Balancing the pandemic paradox of accumulated debt and accumulated wealth has been a defining feature of monetary policy considerations during this cycle of tightening. The combination of elevated savings and pent-up demand can mean that some households have more desire and ability to spend, and their spending is less sensitive to rising interest rates.

The other side of the pandemic paradox is that when households have a lot of debt, their spending is more sensitive to rising interest rates. Gauging the overall impact of these changes on the transmission of monetary policy is key to determining the appropriate degree of tightening. This is just one part of the story behind why we dedicate a lot of time to monitoring detailed data about borrowing, spending and savings. By using a diverse set of sources of information, we are able to understand more about what’s happening at a detailed level and evaluate richer alternative scenarios that put different weights on these different household experiences. 

For instance, from our analysis of anonymized microdata, we know that some households, such as those with variable interest rates and variable mortgage payments, have been really challenged by higher mortgage payments. We have not yet seen a disproportionate increase in mortgage delinquencies for this group. But we do know that the share of indebted households behind on payments for at least 60 days on any credit product has risen sharply since the beginning of last year and is essentially back to the levels observed in the years before the pandemic (Chart 4).

We need to be aware that sometimes special circumstances merit a deeper dive to understand the data. For instance, even though survey data told us that many households are struggling, aggregate consumption was very strong in the first quarter. We assess that this outcome partly reflected a number of special factors, including pent-up demand for services, delayed delivery of some pre-ordered durable goods and unexpectedly strong population growth.

In September, we decided to keep the policy interest rate at 5%. Recent data have provided evidence that our policy rate increases are slowing demand. Household credit growth has eased as the impact of higher rates restrained spending among a wide range of borrowers. In the second quarter, a marked weakening in consumption growth and a decline in housing activity contributed to a sharp slowing of economic growth. And we are mindful that past increases in interest rates will continue to weigh on activity.

On the inflation front, CPI inflation has decreased significantly. It went from a high of 8.1% in June 2022 to 2.8% in June this year, though it rose to 3.3% in July and 4.0% in August. Ups and downs of the size we’ve seen in the past couple of months are not that unusual and are one reason why we look at measures of core inflation—which exclude components with more volatile price movements—to get a sense of what underlying inflation is.

Measures of core inflation have eased, but recent CPI data indicate that inflationary pressures are still broad-based. The number of components of the CPI with price increases greater than 5% is lower than it was, but it is still much higher than it usually is when inflation is stable and close to 2%. More generally, underlying inflation has experienced little recent downward momentum.

The inflation side of things is complex. We examine the details of what is happening to prices within the basket of goods and services in the CPI, not just the overall inflation rate. For instance, we know that mortgage interest costs have risen with increases in our policy interest rate and that these costs have risen by more over the past year than any other component of the CPI. This is partly because mortgage rates were so low for so long. Some have argued that because movements in our policy interest rate can directly and immediately impact some mortgage rates, we should exclude this component when looking at inflation.

Indeed, one advantage of our measures of core inflation is that they exclude extreme price changes—be they increases or decreases. CPI-trim is one of the ways that we measure core inflation. Even though this measure has been excluding mortgage interest costs for more than a year, in recent months it has been at about 3.5% to 4%.

There are some observers who think an upward distortion exists as long as mortgage interest costs are part of the CPI basket. For them, let me provide another calculation. Consider a CPI basket that has all the same goods and services, other than mortgage interest costs, and apply the methodologies to calculate core inflation to this slightly smaller basket. When we do this, we find that the new measures of core inflation are lower, but only by about one-quarter of a percentage point (Chart 5).

That is, underlying inflation is still well above the level that would be consistent with achieving our target of 2% CPI inflation.


When making monetary policy decisions, the Bank of Canada must consider the many ways that shifting economic circumstances affect different households. We don’t set our policy based on what is happening to one subset of households or to the price of any one good or service. But we do our best to understand what is going on at a detailed level. This helps us do a better job of forecasting where the economy is likely to be headed and helps us balance risks.

We monitor information from a wide variety of sources as we balance the risks of under- and over-tightening monetary policy. We know that if we don’t do enough now, we will likely have to do even more later. And that if we tighten too much, we risk unnecessarily hurting the economy.

We don’t make these decisions lightly. We know that higher rates have been very painful for some. But we also know that the burden of persistently high inflation weighs on households of all income levels and in every part of the country. We are seeing signs that monetary policy is working. Both inflation and inflation expectations have come down, and excess demand in the economy is easing. And our past policy actions will continue to have an effect as they work their way through the economy.

However, in our most recent monetary policy decision, we also expressed concern about the persistence of underlying inflation. We will continue to evaluate whether the evolution of excess demand, inflation expectations, wage growth and corporate price-setting behaviour are consistent with achieving the 2% inflation target. We are prepared to raise the policy interest rate further if needed.

Making monetary policy decisions requires both thoughtful consideration and decisive action. We know that the reasons behind the Bank’s decisions are not always obvious to the public, and I hope that I have been able to shed some light on how we have been approaching these decisions. We carefully consider information from a variety of sources, conduct extensive supplementary research and take into account the many different ways monetary policy affects Canadians.

Canada’s experience with inflation targeting has shown that low, stable and predictable inflation is the means by which monetary policy can best promote the economic and financial welfare of Canadians. It is our responsibility to bring inflation back to the 2% inflation target. And the Bank remains resolute in its commitment to restoring price stability for Canadians.

I would like to thank Russell Barnett and Stephen Murchison for their help in preparing this speech.

Related Information

September 19, 2023

Household differences and monetary policy

Speech summary Sharon Kozicki University of Regina Regina, Saskatchewan
Deputy Governor Sharon Kozicki talks about how differences in debt, income and savings across households shaped their experience through the COVID-19 pandemic and how this is affecting monetary policy now.
September 19, 2023

Speech: University of Regina

How household differences have affected monetary policy since the onset of the Covid-19 pandemic — Deputy Governor Sharon Kozicki speaks at the University of Regina. (14:00 (ET) approx.).

  1. 1. This number reflects that about two-thirds of households are homeowners, about 60% of homeowners have mortgages, and about 40% of homeowners with mortgages have faced a higher payment since February 2022.[]
  2. 2. The effect this has on the economy can, in theory, be partially offset by higher interest income that households earn on savings.[]