The Impact of Recent Policy Changes on the Canadian Mortgage Market

Introduction

Recent policy changes are having a clear impact on the mortgage market. The number of new highly indebted borrowers has fallen, and overall mortgage activity has slowed significantly. Tighter policies around mortgage qualification and higher interest rates are having a direct effect on the quality and quantity of credit. At the same time, provincial and municipal housing measures have weighed on housing activity and price growth in certain markets.

The charts below show how the flow of mortgages has evolved in response to these developments.

Over the past two years, federal authorities have tightened the criteria for mortgage qualification with the aim of improving the quality of new mortgage lending. Mortgage insurance rules for high-ratio mortgages1 (those with loan-to-value ratios greater than 80 per cent) were modified in October 2016 to contain an expanded stress test that checks whether borrowers can still afford payments if interest rates increase.

Borrowers who intend to take on loans that are more than 4.5 times (450 per cent of) their annual income are particularly vulnerable to unfavourable situations, such as higher interest rates or a loss of income. The expanded stress test helped to reduce the proportion of new high-ratio mortgages with loan-to-income (LTI) ratios above 450 per cent—from 20 per cent in the fourth quarter of 2016 to 6 per cent in the second quarter of 2018.

The Office of the Superintendent of Financial Institutions (OSFI) included a similar stress test for low-ratio mortgages in the revised Guideline B-20 that came into effect in January 2018. Subsequently, the share of low-ratio mortgages with a loan-to-income (LTI) ratio over 450 per cent has fallen to 14 per cent as of the second quarter of 2018, down from 20 per cent a year before.

Recent increases in interest rates are also contributing to the reduction in the number of highly indebted households. Since they spend a larger fraction of their income making debt payments, highly indebted borrowers are the most affected by higher interest rates. As average contractual rates for five-year fixed-rate mortgages have increased from 2.7 per cent to 3.3 per cent over the past year, the overall share of new highly indebted borrowers has decreased, from 18 per cent to 13 per cent.

The next several charts, which present data from low-ratio mortgages used to purchase a home, highlight mortgages subject to Guideline B-20.

Some borrowers obtaining a low-ratio mortgage have adjusted their purchasing decisions, as suggested by a shift in the distribution of LTI ratios since the second quarter of 2017. As fewer loans are given to highly indebted borrowers, a larger share of mortgages is now concentrated around LTI ratios between 250 per cent and 450 per cent. Mortgages with an LTI just below 450 per cent are still relatively risky and likely include borrowers obliged to take smaller loans than they would have obtained without a stress test. At the same time, the revised Guideline B-20 has not eliminated high-LTI loans altogether, since lenders can apply other criteria to extend such mortgages, including the applicant’s housing equity and financial assets.2

The number of new low-ratio mortgages for home purchases has declined, beginning in the second half of 2017 and continuing through the first half of 2018. In the first quarter of 2018, old qualifying rules were still being applied to some mortgages that were pre-approved before the updated Guideline B-20 took effect. In the second quarter, when the new rules were fully in effect, the number of new low-ratio mortgages fell 15 per cent from a year earlier.

The most pronounced decline has been in the number of new mortgages extended to highly indebted borrowers, which fell by 39 per cent year-over-year in the second quarter of 2018. Contributing to this decline were the expanded stress test, as well as the effects of higher interest rates on highly indebted borrowers, potentially reinforced by changes in general market sentiment.

However, the significant drop in the number of mortgages with lower LTI ratios indicates that other factors, such as provincial and municipal housing policies, are playing an important part. Most notably, a foreign buyers’ tax for Vancouver was announced in July 2016 and the coverage expanded in February 2018. Ontario’s provincial government implemented a similar tax in April 2017. Other regional policies include an empty house tax, restrictions on short-term rentals and measures to increase affordable housing. These measures helped change market sentiment, which in turn has slowed activity and price growth.

Regional implications

The revisions to Guideline B-20 and the increase in mortgage rates coincided with a decline in the proportion of high-LTI borrowers across most major cities, but to varying extents. Toronto saw its share of highly indebted borrowers fall from a peak of 39 per cent in the fourth quarter of 2017 to 28 per cent in the second quarter of 2018. Hamilton, Toronto’s neighbour, also saw a steep drop in riskier mortgages, although from a lower level. In Vancouver, there has been a shift of the LTI distribution away from mortgages with very high LTIs, but the share above 450 per cent remains substantial.

Other factors have been at play as well. Many cities experienced a slowing in new mortgage activity in the second quarter of 2018. Housing markets in Toronto, Vancouver and their surrounding areas have the highest price-to-income ratios in the country and were also subject to regional housing policy measures over the past two years. These markets slowed before the revised Guideline B-20 was introduced and remained weak through the second quarter of 2018. In Calgary, the real estate market has not recovered since the oil price shock of 2015, with the combination of higher rates, tighter mortgage rules and weak income growth continuing to be a drag on local housing. Ottawa–Gatineau and Halifax are on the opposite end of the spectrum: housing activity in these metropolitan areas has recently picked up while the share of high-LTI borrowers remains modest.

The overall riskiness of new mortgages has therefore decreased because the proportion of risky borrowers has declined across cities. As well, the regional composition has shifted, with a somewhat larger share of new mortgages recently coming from areas outside Toronto and Vancouver.

Guideline B-20 applies only to federally regulated lenders such as banks, although credit unions often use their own mortgage stress tests. Mortgages originated by mortgage finance companies (MFCs) generally meet federal mortgage standards, since the bulk of their funding comes from government-sponsored securitization programs. Private lenders, however, do not need to comply with federal mortgage standards. 3

Areas with high house prices, such as the Greater Toronto Area (GTA), could therefore see more borrowers obtaining mortgages from private lenders because they might not be able to qualify with other lenders. While the volume of new mortgage lending in the GTA fell for all lender types in the second quarter of 2018, private lenders saw smaller declines. This boosted their market share from 6 per cent in the second quarter of 2017 to 9 per cent in the second quarter of 2018, continuing an upward trend observed over the past couple of years. We do not currently have data to verify whether the same trend is evident in markets outside Ontario.

Multiple factors at play in the mortgage market should be followed

Tighter mortgage policies have contributed to a decrease in the creation of highly indebted households. But it isn’t possible to separate the role of these policies from that of other measures in changing mortgage market trends. Higher interest rates are directly increasing the cost of mortgage borrowing and effectively making mortgage underwriting requirements stricter. At the same time, house price dynamics have changed in some large markets, in part because of regional policy measures. Less mortgage borrowing can be both a cause and an effect of changes in house price trends.

With multiple factors at play, we cannot precisely determine the role of individual policies. But over time, more data and the use of models can help us better understand the influence of each factor.

Endnotes