Housing Statistics in Canada
Residential real estate sales in 2018: The relationship between house prices and incomes

Introduction

For the first time, data from the Canadian Housing Statistics Program (CHSP) is providing information on the buyers of residential real estate and the properties they purchased from January 1 to December 31, 2018 in Nova Scotia, New Brunswick and British Columbia. The first part of this three-part series provided an overview of the sold properties' characteristics and compared them with the characteristics of all other residential properties, including the geographic distribution of sales, property types, sale prices and living areas.

The second part of the series focused on the characteristics of buyers, including sex, family type, number of buyers as part of a sale, first-time home buyers and immigration status.

As the third and final part of this series, this article examines the relationship between the purchase price of properties and the income of buyers^Note by calculating the price-to-income ratio (PIR) at the property level. The heterogeneity of this ratio across income quintiles is examined and the PIR of properties purchased by immigrants in British Columbia.

In Canada, housing affordability has been historically measured by the shelter-cost-to-income ratio (STIR). The STIR is the average share of before-tax household income spent on shelter costs and has a threshold of 30%, meaning that housing is considered as unaffordable when households spend more than 30% of their income on shelter costs (Statistics Canada, 2019a). To expand on this measure of affordability, the Canada Mortgage and Housing Corporation (CMHC) introduced the housing hardship measure, where a household is in housing hardship when there is not enough disposable income after housing expenditures to afford other basic living expenses (CMHC, 2020). This measure differs from the STIR by taking into account the cost of non-shelter items, different types of households such as households with children, and taxes paid and benefits received.^Note

The PIR presented in this article is a new measure for the Canadian context that builds on the STIR and the housing hardship measure, as it can be used to monitor housing affordability at the time of purchase. By linking the price of the property sold to the income of its buyer, the PIR provides an indication of the financial burden faced by home buyers when purchasing residential real estate. A higher PIR indicates a larger financial burden, requiring buyers to be more indebted or to rely on additional capital for their purchase.^Note This measure can be evaluated over time, with the findings in this article serving as a point of comparison to quantify the effects of the COVID-19 pandemic on housing affordability within Canada.

Internationally, the Organisation for Economic Co-operation and Development (OECD) uses an aggregated price-to-income ratio measure to monitor housing affordability (OECD, 2021). The ratio used by the OECD is calculated as the house price index divided by the disposable income per capita. The income measurement includes all persons in a region, regardless of whether they purchased a property.^Note Comparisons are made between countries to evaluate access to housing markets and the effectiveness of housing policies in addressing affordability.

This article furthers the OECD's approach by looking strictly at the income of property buyers, as opposed to including all persons in a region. Additionally, the PIR in this article uses the income data of the buyers of a property as opposed to national disposable income per capita. A comparison of PIRs by geographical areas, where market dynamics can vary drastically, unveils differences in real estate markets across provinces and cities that are often masked when only looking at the national or provincial levels.

This article includes residential properties sold in a market sale from January 1 to December 31, 2018, purchased by individual (persons) resident buyers. The first section explores the heterogeneity of the PIR of properties purchased in Nova Scotia, New Brunswick and British Columbia. The second section furthers this analysis by examining the ratio across income quintiles. Findings on the PIR of properties purchased by immigrants in British Columbia are also presented. The last section concludes with a summary of the findings in this three-part series.

Section 1: Price-to-income ratios in Nova Scotia, New Brunswick and British Columbia

Vancouver is the least affordable metropolitan area examined

British Columbia had a median price-to-income ratio (PIR) more than double that of Nova Scotia and New Brunswick (Chart 1). This shows that buyers in British Columbia had to commit a larger share of their income to housing, relative to those purchasing properties in Nova Scotia and New Brunswick, and points to housing affordability differences across the provinces. These findings are consistent with previous results, which showed that properties in British Columbia had a significantly larger assessment value-to-income ratio compared with properties in Nova Scotia (Al-Tawil, 2019).^Note

British Columbian buyers faced higher PIRs, despite earning median incomes 1.2 times higher than in Nova Scotia and 1.3 times higher than in New Brunswick. The contrast was greater when looking at median house prices, where the median price of properties purchased by buyers in British Columbia was 3.1 times higher than in Nova Scotia and 3.8 times higher than in New Brunswick. This shows that even though the income was higher in British Columbia, it was not enough to offset the high prices of properties. Therefore, buyers in British Columbia needed significantly more capital to purchase a property, or relied more on debt, than buyers in Nova Scotia and New Brunswick.^Note

In the Vancouver census metropolitan area (CMA), the median PIR was 7.4, considerably higher than the Halifax and Moncton CMAs, where the PIRs were less than 3. Additionally, as shown in Map 1, census subdivisions (CSDs) that are closer to the core of the largest CMAs had a higher median PIR compared with CSDs that were farther from the cores in both British Columbia and Nova Scotia.

In the Vancouver CMA, high property prices were not offset by high incomes, which led to high median PIRs. Of note, the CSDs of West Vancouver (17.2) and Richmond (12.2) had the highest median PIRs of all CSDs observed. Median sale prices in the Richmond CSD ($683,000) were equivalent to those of the broader Vancouver CMA. However, buyers in the Richmond CSD had the lowest median income ($57,900) in the Vancouver CMA. By comparison, buyers in the West Vancouver CSD earned over double the income of buyers in the Richmond CSD, but purchased properties with a median price over three times higher ($2,330,000).^Note The decoupling of property prices and income in the Vancouver CMA may signify that wealth, which is not represented in the income measure, plays an important part in the purchase of properties or that higher levels of indebtedness are required to purchase a home. Additionally, the income measure excludes capital gains and includes only the income reported the year the property was purchased, leaving out all unreported income.

Vacant land properties have the lowest price-to-income ratios

The median PIRs are similar across the different types of properties purchased within each province, with the exception of vacant land, as shown in Table 1. In Nova Scotia and New Brunswick, vacant land properties had a median PIR of 0.2, likely a result of the low price of vacant land.

Buyers may seek to purchase vacant land as an investment, as this property type was more likely to be purchased by at least one multiple-property owner, compared with other property types. This was most pronounced in Nova Scotia, where 83.2% of vacant land properties were purchased by at least one multiple-property owner, compared with 32.8% of single-detached houses. Additionally, buyers of vacant land had a higher median income than those who purchased other property types in New Brunswick and the second-highest median income in British Columbia, following those who purchased properties with multiple units.

Section 2: Analysis of the price-to-income ratio by income quintiles

This section categorizes the buyers into five equal groups based on their income (quintiles, hereafter), calculated independently for each province.^Note This allows for a comparison of property prices and buyer characteristics across different levels of income at the time of purchase and brings new insights by geographical areas.

Buyers in the lowest income quintile purchase properties worth 29 times their income in Vancouver

Buyers in the lowest quintile had the lowest income relative to the price of the properties they purchased, resulting in the highest median PIR of all the quintile groups in the three provinces. Of note, the median PIR of properties purchased by those in the lowest income quintile was over four times higher in British Columbia than in Nova Scotia and New Brunswick, as shown in Chart 2.

The median PIR of the lowest quintile was especially high in the Vancouver CMA, reaching 28.6. By comparison, the lowest quintile had lower median PIRs in the Halifax (7.3) and Moncton CMAs (4.4). The ratios were significantly lower for those in the top income quintile in the Vancouver (3.9), Halifax (1.8) and Moncton (1.4) CMAs.

Buyers in the lowest quintile are older and more likely to be repeat buyers

Different characteristics can be observed across income quintiles. For instance, Al-Tawil (2019) used owner-level data to show that properties of those receiving pension income were more prevalent in the lowest income quintile. The data showed that over half of the properties in the Vancouver, Toronto and Halifax CMAs in the lowest income quintile were owned by individuals receiving pension income, compared with around a quarter of those in the highest income quintile.

Buyers in the lowest income quintile were older than those in the other quintiles. In British Columbia, the median age of the buyers in the lowest quintile was 50, while that of buyers in the third and fourth quintiles was 41. The Survey of Financial Security showed that net worth tends to increase with age, as senior-led families reported the highest median net worth in 2019 (Statistics Canada, 2019b). This suggests that since buyers in the lowest income quintile are older, they may have had additional time to accumulate wealth, which could result in more capital saved for the purchase of their property.

In British Columbia, the lowest income quintile also saw the highest share of properties purchased by repeat buyers,^Note with a proportion over 90%. In Nova Scotia, the lowest income quintile had the second highest share of properties purchased by repeat buyers (80.5%), following those in the highest income quintile (88.9%). This further suggests that buyers who earned the lowest income may have accumulated wealth through the sale of a previous home, especially in British Columbia.

In British Columbia, buyers in the lowest income quintile purchase more expensive properties than those in the second and third quintiles

The high PIRs across all income quintiles in British Columbia were likely driven by the higher property prices, as shown in Table 2. This was especially true in the lowest income quintile, where buyers in British Columbia had the same median income as those in the other two provinces, but purchased properties that were significantly more expensive.

In Nova Scotia and New Brunswick the median sale price of the properties purchased increases with income, with the lowest income quintile acquiring the properties with the lowest prices. Conversely, in British Columbia, the properties purchased by the lowest quintile were priced 10.9% higher than the properties purchased by the second quintile and 1.8% higher than the third quintile.

Properties purchased by immigrants in British Columbia have a higher price-to-income ratio

Taking a closer look at British Columbia reveals that properties purchased by at least one immigrant had a higher median PIR (7.3) compared with properties purchased by non-immigrants (4.7). Differences are observed across all income quintiles, as shown in Table 3. The largest difference was observed in the lowest income quintile, where properties purchased by at least one immigrant buyer had a PIR (26.5) over double that of properties purchased by non-immigrants (13.1).

When the purchase involved at least one immigrant, the median price of properties purchased by the lowest income quintile was 15.5% higher than the second, and 7.8% higher than the third income quintile. Interestingly, when the purchase did not involve an immigrant, buyers in the lowest income quintile purchased properties with the lowest median price.

There were important differences between the ages of the buyers in the lowest income quintile. The median age of buyers, when at least one immigrant buyer was involved in the purchase of the property, was lower (47), compared with the non-immigrant group (54). No significant age differences were observed in the other income quintiles.

This reinforces the notion that income does not always provide a complete measure of an individual's ability to purchase a home. Many demographic or socioeconomic factors such as age, not captured in income measures, may also contribute to the differences observed.^Note

Considerable literature on Canadian immigration (Haan, 2007; Ley, 2007; Rea et al., 2008) highlights the propensity of immigrants to purchase real estate. Using data from the Survey of Financial Security, Gellatly and Morrissette (2019) showed that housing assets comprised a larger share of average wealth among immigrant families compared with Canadian-born families. According to Morissette (2019), the increase in housing wealth was more important for immigrant families relative to non-immigrants. The same study showed that a lower proportion of immigrants invest in registered pension plans compared with Canadian-born individuals. Therefore, immigrants may leverage investments in housing as a retirement asset more than non-immigrants, which can lead to differing home buying preferences.

Conclusion

This three-part series examined the characteristics of home buyers and the properties they purchased from January 1 to December 31, 2018, with a focus on Nova Scotia, New Brunswick and British Columbia.

The first part of this series found that British Columbia had the most active housing market of the provinces examined. Vacant land, a key input in housing supply, was most expensive in British Columbia, where homes were also most expensive. Meanwhile, in Nova Scotia and New Brunswick, properties sold in the reference period had larger living areas relative to other properties.

The second part of this series looked at the characteristics of home buyers, showing that home buyer preferences are shaped by a multitude of socioeconomic factors. In particular, the majority of sales in the three provinces involved more than one buyer, highlighting possible challenges faced by single individuals in becoming homeowners. Additionally, the income gap between first-time home buyers and repeat buyers in British Columbia may point to the difficulties of entering the real estate market in areas where property prices are higher. Finally, immigrants purchased more expensive properties compared with non-immigrants. Properties purchased by immigrants were closer to city centres where there is broader access to social services, community support and job opportunities.

The last part of this series examined housing affordability through the median price-to-income ratio. The divergence between property prices and incomes in some areas demonstrated that sources other than income can play an important role in homeownership, especially in the Vancouver CMA. Areas with high price-to-income ratios pointed to the increased financial burden placed on buyers, and the higher capital requirements to purchase a property. Another important consideration not captured in the income measure was accumulated wealth, acquired through previous homeownership, savings or financial support from individuals not included on the title. Examining these other factors would provide a more complete picture of housing affordability.

Appendix

References

Al-Tawil, J. (2019). Homeownership, income, and residential property values. Housing statistics in Canada. Statistics Canada Catalogue no. 46-29-0001. https://www150.statcan.gc.ca/n1/pub/46-28-0001 /2019001/article/00002-eng.htm

Canada Mortgage and Housing Corporation. (2020). Introducing the Housing Hardship Concept. Research Insight. https://www.cmhc-schl.gc.ca/en/professionals/housing-markets-data-and-research/housing-research/research-reports/housing-needs/research-insight-introducing-housing-hardship-concept

Gellatly, G. & R. Morissette. (2019). Immigrant Ownership of Residential Properties in Toronto and Vancouver. Economic Insights, 087. Statistics Canada Catalogue no. 11-626-X. https://www150.statcan.gc.ca/n1/pub /11-626-x/11-626-x2019001-eng.htm

Haan, M. (2012). The Housing Experiences of New Canadians: Insights from the Longitudinal Survey of Immigrants to Canada. Ref. no.: RR20120301.01. Citizenship and Immigration Canada.https://www.canada.ca/en/immigration-refugees-citizenship/corporate/reports-statistics/research/housing-experiences-new-canadians-insights-longitudinal-survey-immigrants-canada-lsic.html

Haan, M. (2007). Do I buy with a little help from my friends? Homeownership-relevant group characteristics and homeownership disparities among Canadian immigrant group 1971-2001. Housing Studies, 22(6), 921-944. https://doi.org/10.1080/02673030701608142

Ley, D. (2007). Countervailing Immigration and Domestic Migration in Gateway Cities: Australian and Canadian Variations on an American Theme. Economic Geography 83(3), 231-254.https://doi.org/10.1111/j.1944-8287.2007.tb00353.x

Morissette, R. (2019). The Wealth of Immigrant Families in Canada. Analytical Studies Branch Research Paper Series, 422. Statistics Canada Catalogue no. 11F0019M. https://www150.statcan.gc.ca/n1/pub/11f0019m/11f0019m2019010-eng.htm

OECD. (2021). HM1.2 Housing Prices. OECD Affordable Housing Database. OECD Directorate of Employment, Labour and Social Affairs - Social Policy Division. https://www.oecd.org/els/family/HM1-2-Housing-prices.pdf

Rea, W., J. Yeun, J. Engeland & R. Figuera. (2008). The Dynamics of Housing Affordability. Perspectives on Labour and Income, Vol. 9 , no. 1. Statistics Canada Catalogue no. 75-001-X. Ottawa: Statistics Canada. Canada. https://www150.statcan.gc.ca/n1/pub/75-001-x/2008101/article/10511-eng.htm

Statistics Canada. (2019a). Shelter-cost-to-income ratio of private households. Longitudinal Immigration Database (IMDB). https://www23.statcan.gc.ca/imdb/p3Var.pl?Function=DEC&Id=103407

Statistics Canada. (2019b). Survey of Financial Security, 2019. The Daily. https://www150.statcan.gc.ca/n1/daily-quotidien/201222/dq201222b-eng.htm

Notes