Insights on Canadian Society
Homeownership, mortgage debt and types of mortgage among Canadian families

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

by Sharanjit Uppal

Start of text box

End of text box

Introduction

In recent years, a great deal of research has focused on the levels of debt that families carry and the increase in debt.^Note  Most of the increase in total debt is from mortgage debt, and the increase in mortgage debt can be attributed to two factors. The first is the rapid increase in housing prices. According to the monthly Home Price Index published by the Canadian Real Estate Association (CREA), housing prices rose by 109% (in nominal terms) from January 2005 to December 2016.^Note  In comparison, during the same period, the Consumer Price Index (CPI) showed an overall price increase of 22%.^Note  The second factor is mortgage interest rates, which have been at historically low levels for most of the last decade, making borrowing costs low and acting as an incentive for families to take on new or bigger mortgages.^Note 

There is a possibility that these historically low mortgage rates would increase as the Bank of Canada normalizes monetary policy. Because mortgage debt is high, there are concerns that certain households may not be able to withstand interest rate hikes. The short-term effect of an interest rate hike depends, in part, on the type of mortgage rate the debt carries. If the mortgage rate is fixed, then a rate hike will not have any immediate effect on mortgage payments or on the outstanding amount to be paid until the end of the term. However, if the rate is variable, an increase in the rate will have either of the following consequences, depending on the type of agreement: (1) it will lead to an immediate increase in payment or an increase in payment after the interest rate reaches a certain threshold; or (2) it will not change the payment, but decrease the repayment of principal and increase the accrued interest. Even for families with a fixed mortgage rate, a higher interest rate may have an impact for those who are not prepared for the possibility of facing higher interest rates at the end of their term. In view of the above, a study on homeownership, current outstanding mortgage debt and types of mortgage rates is important.

While examples of Canadian studies focusing on homeownership and mortgage debt can be found in the literature,^Note  little is available on the types of mortgage rates across regions or family types. For the first time, the 2016 Survey of Financial Security (SFS) asked about the type of mortgage rate that a family took on their principal residence mortgage debt (see the Data sources, methods and definitions section).

The first part of the article will look at homeownership rates across family characteristics. It will then look at the characteristics of families who had paid off their mortgage, and the amount of mortgage outstanding among families who had not paid off their mortgage debt. In the second part, family characteristics associated with various types of mortgage rates will be examined. Where possible, comparisons will be drawn over time.

In 2016, more than 6 in 10 families owned their principal residence

In 2016, 63% of Canadian families owned their principal residence, up slightly from 60% in 1999 (Table 1).^Note Note  Almost all of this increase is attributable to a demographic shift—the Canadian population became older, and older age groups are more likely to own their homes.^Note 

Recent immigrants (those who immigrated during the five years preceding the survey) had the largest increase in homeownership.^Note  The homeownership rate for this group increased from 17% in 1999 to 39% in 2016. Established immigrants (those who immigrated more than five years before the survey) also had a notable increase (6 percentage points).

The homeownership rate for families whose major income earner had a non-university postsecondary certificate increased by 6 percentage points. In contrast, the rate for the lowest educational category (i.e., those who did not complete high school) decreased by 5 percentage points. Families in the top two income quintiles also experienced a significant increase in homeownership (4 percentage points and 7 percentage points, respectively), as well as unattached individuals^Note  (6 percentage points) and those who maintained a family budget (4 percentage points). Regionally, there were no statistically significant changes in homeownership.

There is a clear association between homeownership and age, largely because homeownership is a life cycle phenomenon. Younger individuals aged less than 35 are less likely to own homes, because many are still in school and others have recently graduated and are in the process of saving money for a down payment. The homeownership rate is higher among middle aged individuals (i.e., those aged 45 to 64); in 2016, homeownership was 71% among families whose major income earner was aged 45 to 54 and 78% among those aged 55 to 64.

Homeownership rates were higher among families whose major income earner had a level of education above a high school diploma (largely because higher educated people have higher incomes). Couple families were more likely to own homes, compared with lone parents and unattached individuals. Families whose major income earner was self-employed were more likely to own their homes than families of paid employees and people who were not in the labour force. Despite the large increase in homeownership rates of recent immigrants, they remained less likely to own their homes than established immigrants and Canadian-born individuals. As expected, families in the top income quintile were more likely to own their homes than those in the bottom quintile.

Certain financial behavioural factors can also be associated with homeownership. In the 2016 SFS, questions related to financial behaviour included whether or not the family had a household budget, had borrowed money through payday loans in the past three years, or had anyone in the family who had ever declared bankruptcy or who made a formal proposal or informal financial arrangements to creditors. Information on payments on credit card balances was also collected.^Note  Families who paid off their monthly credit card balances in full were more likely to own their homes. However, financial instability (whether measured as having declared bankruptcy or made formal or informal arrangements, or as having borrowed through a payday loan) was negatively associated with homeownership. Declaring bankruptcy or making formal proposals to creditors would make it more difficult to obtain credit from a financial institution, while having borrowed from a payday loan reflects a low level of financial resources and therefore a reduced ability to make a down payment.

There were some regional variations in homeownership. Families in Montréal (51%) were the least likely to be homeowners, whereas families in Newfoundland and Labrador (74%) and areas in Manitoba other than Winnipeg (78%) were the most likely.

Because various personal characteristics are expected to be correlated with each other, a logistic regression was estimated to calculate predicted probabilities of homeownership. Controlling for various characteristics changed some of the conclusions based on descriptive statistics. The predicted probability of owning a home was now higher among those aged 65 and over (0.75), compared with 0.63 for those aged 45 to 54 and 0.46 for those aged 25 to 34 (Table 2). Seniors are more likely to have lower levels of education, be unattached and belong to the lower income quintiles—characteristics that are associated with lower homeownership rates.

When financial behavioural characteristics are included in the model, the result for university education becomes statistically insignificant. This shows that university education has a positive association with homeownership through its effect on financial behaviour. In other words, university education has a positive association with positive or prudent financial behaviours, which in turn have a positive association with homeownership.

The probabilities of owning a home were relatively higher among families living in the Atlantic provinces (from 0.66 to 0.73). Additionally, areas outside large census metropolitan areas (CMAs) in the rest of Canada had higher ownership rates. Areas outside CMAs usually have higher homeownership for two reasons: relatively less expensive house prices and lower availability of rental units. Finally, the probability of owning a home in Montréal (0.54) was the lowest of all regions. The low homeownership in Montréal (and in Quebec in general) can be attributed to the favourable shelter costs of renting compared with owning.^Note 

In 2016, more than 4 in 10 families had paid off their mortgage

In 2016, 43% of homeowners did not have any mortgage debt on their principal residence; in other words, they had paid off their house. However, this proportion varied across characteristics, particularly age. This is expected, since mortgage debt typically involves large amounts that are contracted at a relatively young age to pay for a house and then slowly repaid over time.

Therefore, it is not surprising that older people are more likely to have paid off their mortgage. In 2016, over 8 in 10 families whose major income earner was aged 65 or over had paid off their mortgage, compared with less than 1 in 10 of those whose major income earner was aged 25 to 34. Other characteristics, however, are also associated with having no mortgage debt. These characteristics can be examined in a regression model controlling for age, with results expressed in predicted probabilities (Table 3).

Couples with children were less likely to have paid off their mortgage (0.35), compared with unattached individuals (0.44).^Note  This is likely because they have more financial responsibilities given the presence of children. Couples with children also tend to have larger houses, which are more costly. Conversely, individuals who were not in the labour force (0.54) were more likely to have paid off their mortgage than those who were in the labour force. This result is not necessarily a surprise, since the group of people outside the labour force includes retirees, and people often continue to work until their debts are paid in full.^Note 

Recent and established immigrants had a lower probability of having paid off their mortgage (0.34 and 0.39, respectively) than Canadian-born individuals (0.44). The results associated with financial behaviour variables show that the probability of having paid off the mortgage was higher among those who paid off credit card balances in full and those who had never declared bankruptcy.

The results did not show much variation across regions, since most probabilities were not statistically different from the reference region (Toronto). However, homeowners in Newfoundland and Labrador and in Saskatchewan excluding Regina and Saskatoon were more likely to have paid off their mortgage, compared with homeowners in Toronto. In contrast, homeowners in Montréal and British Columbia excluding Vancouver were less likely to have paid off their mortgage.

It is also important to examine changes over time. From 1999 to 2016, the proportion of families who had paid off their mortgage decreased from 46% to 43%. This decline occurred against a backdrop of population aging, which would be associated with a higher rate because older people are more likely to have paid off their mortgage. Population aging slowed down the decline in the proportion of homeowners who had paid off their mortgage. If the age structure had remained the same as in 1999, the decline in the proportion of families who had paid off their mortgage debt would have been significantly larger—falling from 46% in 1999 to 36% in 2016.

From 1999 to 2016, almost all groups experienced a decline in the proportion of homeowners who had paid off their mortgage. The magnitude, however, varied across family characteristics. The decline was relatively large for families whose major income earner was aged 35 to 64—an 11 to 12 percentage point decrease. The younger age group, those aged 25 to 34, saw a 4 percentage point decrease (Table 4).

The decline in the proportion of homeowners who had paid off the mortgage on their principal residence was around 5 to 6 percentage points for the bottom three income quintiles. This proportion did not change for the upper two income quintiles. Other groups that saw a significant decrease in this proportion included couples with children (12 percentage points) and without children (5 percentage points), other family types (15 percentage points), established immigrants (8 percentage points), homeowners not in the labour force (6 percentage points), and homeowners who did not maintain a family budget (7 percentage points). Regionally, significant declines were seen in Newfoundland and Labrador (14 percentage points), Winnipeg (11 percentage points), Edmonton (10 percentage points) and areas of Saskatchewan other than Regina and Saskatoon (13 percentage points).

From 2012 to 2016, the increase in total debt was entirely due to mortgage debt

Mortgage debt constitutes the majority of a family’s total debt. In 2016, it accounted for 81% of the total debt of Canadian families, up from 77% in 1999.^Note  From 1999 to 2016, the average total family debt more than doubled, from $75,300 to $162,400 in 2016 dollars.^Note  During the same period, average mortgage debt increased from $58,200 to $131,100, an increase of 125%. Consumer debt also rose, but at a slower pace (from $17,000 to $31,300).^Note  Therefore, most of the increase in total debt resulted from the rise in mortgage debt.

In fact, mortgage debt contributed to 84% of the increase in total debt from 1999 to 2016. Interestingly, the contribution of mortgage debt to the increase in total debt has been more significant in recent years. From 1999 to 2005, 68% of the increase in total debt resulted from the increase in mortgage debt (Chart 1), rising to 82% from 2005 to 2012, then to 100% from 2012 to 2016. The average consumer debt remained unchanged between 2012 and 2016 (at $31,300).

Mortgage debt on the principal residence doubled from 1999 to 2016

Among homeowners who had a mortgage, the median mortgage debt was $180,000 in 2016 (Table 5).^Note  As expected, the level of mortgage debt decreases with age. The median mortgage debt of families whose major income earner was aged 25 to 44 was twice as much as those whose major income earner was aged 65 and over. Higher mortgage debt in the middle age group is consistent with the life-cycle theory of asset accumulation.

Families whose major income earner held a university degree had more than twice as much mortgage debt than those whose major income earner did not complete high school. University-educated individuals typically have higher incomes, which allows for better access to credit.

Mortgage debt also varied across the income distribution. The median level of mortgage debt among families in the top income quintile was $228,000, compared with $145,000 among families in the bottom quintile. Mortgage debt was higher for recent and established immigrants ($293,000 and $230,000, respectively) than for Canadian-born individuals ($158,600). Among family types, couples with children ($205,000) had the most mortgage debt, and unattached individuals ($135,000) had the least.

Homeowners who had not borrowed through a payday loan in the past three years and those who had never declared bankruptcy or made a formal proposal or informal financial arrangements to creditors had higher levels of mortgage debt, reflecting a better ability to obtain credit.

Regionally, families in Toronto ($280,000), Vancouver ($266,000), Calgary ($260,000) and Edmonton ($250,000) had the highest median levels of mortgage debt. These markets are among the most expensive in the country. Homeowners residing in New Brunswick ($89,000) and Prince Edward Island ($80,700), where housing prices are lower, had the lowest median mortgage debt.

From 1999 to 2016, mortgage debt doubled from $91,900 to $180,000 in 2016 dollars. This is not a surprise, given that housing prices almost doubled in Canada from 2005 to 2016. Mortgage debt increased significantly among all groups, but there were differences across groups. In percentage terms, among all age groups, families whose major income earner was aged 45 to 54 experienced the largest increase in mortgage debt (121%, or $95,200), followed by those whose major income earner was aged 35 to 44 (107%, or $103,200).

The median mortgage debt more than doubled among families in the bottom two income quintiles and among families in the top quintile. Relatively large increases in mortgage debt were also seen among couples with children (112%, or $108,200) and lone-parent families (109%, or $86,200). In percentage terms, homeowners living in the Prairies experienced the largest increases, and homeowners living in Prince Edward Island were the only ones not to record a significant increase in median mortgage debt over the period.

Almost three-quarters of families had a fixed-rate mortgage in 2016

Given the increase in the relative importance of mortgage debt among Canadian families, it is also important to examine how Canadians finance this debt. For the first time, the 2016 SFS collected information on the type of mortgage rate selected by Canadians on their principal residence: a fixed mortgage rate, a variable mortgage rate or some combination of both.

Homeowners can be motivated to choose a variable rate over a fixed rate in order to benefit from a lower interest rate (and hence lower mortgage payments), even if this means being exposed to a higher risk of rate increase in the near future. Conversely, homeowners who are more risk averse may prefer a fixed rate. These choices depend on the spread between the fixed and variable rate, and the type of mortgage products offered by the financial institutions for different amortization periods. The SFS did not ask respondents as to why they chose a variable rate, fixed rate, or combination of both. As a result, the section below is limited to a discussion of mortgage rate types across family characteristics.

In Canada, the majority of homeowners with a mortgage have a fixed mortgage rate.^Note  About 74% had a fixed mortgage rate, while 21% had a variable rate and 5% had a combination of both (Table 6). ^Note  Families in the youngest age group, aged 19 to 24, were the least likely to have a fixed mortgage rate (63%).^Note  Those aged 25 to 34 (80%) and 65 and over (79%) were the most likely to have a fixed mortgage rate. Among the remaining age groups, about 72% had a fixed mortgage rate.

Homeowners who had a relatively small number of years remaining to pay off their mortgage were less likely to have a fixed rate, compared with those whose remaining amortization period was longer. Among those who reported that they would take at least 25 years to pay the full balance of their mortgage, 83% had a fixed rate, compared with 68% among those who had 6 to 10 years left to pay and 74% among those who had 0 to 5 years left.

A model was estimated to study the characteristics of families who were more (or less) likely to choose a fixed mortgage rate over a variable mortgage rate.^Note  This analysis is not meant to estimate a model of mortgage choice, but to see which types of families are likely to have a particular type of mortgage rate.

The results by age indicate that the youngest families (aged 19 to 24) were less likely than families with a major income earner aged 25 to 34 to have a fixed mortgage rate, but other age differences were statistically insignificant (Table 7). Couples with children (with a predicted probability of 0.72) and other family types (0.71) were less likely to have a fixed rate than unattached individuals (0.80).

Families in the fourth and fifth income quintiles were less likely to have a fixed-rate mortgage, compared with those in the bottom quintile. In addition, families whose major income earner was self-employed were less likely to have a fixed rate than families whose major income earner was a paid employee.

While the amount of outstanding mortgage debt did not have any association with the type of mortgage rate, the amortization period did.^Note  Compared with those who had 6 to 10 years left (0.69) to pay the full balance of their mortgage, homeowners who had 0 to 5 years (0.75), 21 to 25 years (0.78) and more than 25 years (0.83) left were more likely to have a fixed-rate mortgage.^Note  The results for homeowners who had 11 to 20 years left to pay their mortgage were statistically insignificant. As more of the mortgage payment goes to principal, families with just a few years remaining on their mortgage may be tempted to secure their payments over last few years of their amortization period; they may also benefit from advantageous fixed rates offered by lenders. At the other end of the spectrum, people with long amortization periods may want to opt for more stability, given the long time left before their mortgage is paid off.  

Finally, when financial behaviour variables are added in the model, the results showed that families who maintained a household budget were more likely to pick a fixed rate, perhaps reflecting the fact that those who budget prefer the certainty of fixed payments over time.

Conclusion

Although homeownership rates have remained relatively constant in Canada, debt has been increasing in recent years. From 2012 to 2016, mortgage debt accounted for the entire increase in debt among Canadian families with debt. In 2016, mortgage debt accounted for 81% of total debt among Canadian families. The increase in mortgage debt has several implications for the housing market in Canada.

First, the increase in mortgage debt implies that Canadians could take longer to pay off their houses. From 1999 to 2016, the proportion of Canadians who had paid their houses in full declined from 46% to 43%. If not for the aging of the population, the decline would have been even more pronounced. This is because older individuals are typically more likely to have paid off their house in full than younger people.

Second, the results suggest that even though mortgage debt increased in almost every demographic group, the magnitude of the increase was not the same for every group. Specifically, mortgage debt was higher among couples with children, homeowners aged 25 to 44, homeowners with higher education and higher-income families. These groups also recorded the largest increase in mortgage debt from 1999 and 2016. However, debt also increased notably among some vulnerable groups, including lone parents and recent immigrants.

Third, the increase in mortgage debt also suggests that the financial choices of Canadians regarding the type of mortgage rate are important. In 2016, three-quarters of Canadians financed their mortgages via fixed-rate loans, with some types of families being particularly more likely to choose a fixed rate, including older Canadians, homeowners in low income quintiles, and homeowners with a longer amortization period.

Finally, the results suggest that there are some important regional variations in homeownership, mortgage debt and the type of mortgage rate. These differences may be related to variations in prices, regional economic conditions, differences in housing conditions, and differences in local preferences and choices. More research will be needed to better understand regional differences in mortgage debt, homeownership and type of mortgage rate.

Sharanjit Uppal is a senior researcher with Statistics Canada’s Insights on Canadian Society.

Start of text box 1

End of text box 1

Start of text box 2

End of text box 2