Date Published: August 28, 2017
The number of housing policy changes over the past eighteen months has been unprecedented, at both the federal and provincial level. The first of two major changes (January 2016) at the federal level increased the required down payment on homes worth more than $500,000 for borrowers in need of mortgage insurance. Then, October ushered in a higher bar on stress tests for insured mortgage borrowers, alongside changes to portfolio insurance eligibility. This one-two punch was the broadest and most stringent set of rules imposed over the past decade.
However, the impact on housing demand fell flat relative to expectations. Each successive regulation change at the federal level has left a smaller mark on home buying activity. Our estimates suggest that the most recent federal rule changes may have only shaved 2% off demand nationwide. In contrast, the first regulatory changes implemented in 2008 dampened home sales by roughly 10% (Chart 1). That policy increased the required down payment from 0% to 5% for insured borrowers and lowered the allowable amortization period from 40 years to 35 years.
Why might this be? The answer, in part, lies with the fact that each round of regulatory changes disproportionately impacted borrowers who required mortgage insurance. This incented a shift away from high loan-to-value mortgages into conventional mortgages. New loans that require homebuyer’s insurance now account for less than 20% of all new chartered bank mortgage originations, compared to 40% prior to 2008. So, each round of policy changes has targeted a shrinking share of the overall market. According to Bank of Canada data, insured mortgage originations among Canadian chartered banks took a significant hit in late 2016 and early 2017, falling 43% from the peak in Q4 2015. However, the impact on overall mortgage credit growth and the market was more limited, as buyer’s shifted into conventional mortgages.
As federal policy changes became less effective, those implemented at the provincial level proved to have some bite. Timing was important, as implementation occurred during a rapid erosion in housing affordability. The B.C. government triggered a sharp drop in housing demand when the provincial budget (February 2016) outlined measures to track foreign investment and discourage shadow flipping. Three quarters of the downturn in sales had already occurred by August 2016, when the province introduced a 15% non-resident land transfer tax in Vancouver. Typically, market activity bounces back within six to 12 months following such policy changes. True to form, markets in Vancouver and most of B.C. embarked on a modest recovery earlier this year. But, momentum has since petered out under the weight of higher mortgage rates.
Looking eastward, the Ontario market had a head of steam between January and March this year, despite federal changes to mortgage regulation rules last October. In a prior report, we discussed how speculative behaviour contributed to the rapid rise of demand and prices in the province. Although many commentators had focused on foreign demand, we argued that the Toronto market went beyond those forces, and the influence of domestic speculation could not be ignored.
So in April, Ontario went even further than B.C. with a sweeping sixteen-measure housing market plan. Of these measures, we deem three particularly material for triggering an immediate market response: the 15% non-resident land transfer tax on all purchases in the Greater Golden Horseshoe Area, expanded rent control, and the right to tax vacant land and properties.
The subsequent downturn in Ontario (and Toronto in particular) has been more abrupt than the experience of B.C. (and Vancouver). Since Ontario’s policy measures, existing home sales have tumbled 33%, with Toronto plummeting 44%. Existing home sales in B.C. and Vancouver experienced a similar decline, but over the course of an entire year. Further differentiating the two markets has been a rapid escalation in supply within Toronto (Chart 2). The knee-jerk response supports our view that domestic speculation was playing a larger role in the GTA’s market relative to the forces observed within its western counterpart. Likewise, the movement in prices, while not ringing any alarm bells as of yet, also shows an outsized response relative to B.C. The average home price in the Greater Toronto Area (GTA) has fallen 13% peak-to-trough (Chart 3), with suburban areas disproportionately impacted by the policy changes. On a seasonal and quality adjusted basis, single family home prices in the GTA have edged down by 6.6% so far, and by a greater 11.4% in Oakville-Milton, one of Ontario’s biggest suburbs. By comparison, single-family home prices fell 3.0% peak-to-trough in Vancouver over a six month period (July 2016-January 2017), but have since recovered.
Government policymakers are not done yet with regulatory changes on the mortgage market. The Office of the Superintendent of Financial Services (OSFI) put forward a proposal to subject all borrowers to income test at a rate that is two percentage points higher than the contract rate. Currently, lenders are required to income test at the Bank of Canada’s 5-year posted rate all insured borrowers as well as those uninsured borrowers who take out a variable-rate product or fixed-rate borrowers with terms shorter than five years. Only five-year (and longer) fixed-rate uninsured borrowers are currently exempt from this more stringent requirement. This divergence would no longer be the case if the new measures are put into place, which will cause buyers in the latter group to adjust their behaviour by coming up with a bigger down payment, opting for a lower priced purchase, scaling back other debt, or delaying a home purchase altogether. In the year of implementation, we estimate that this new rule could depress demand by 5% to 10%, and shave 2% to 4% off of our current forecast for the average price level in 2018. This will be yet another force limiting price growth in the future.
The share of foreign buyers in the overall market is now better understood within the Toronto and Vancouver markets:
In Ontario, data collection began after the provincial policy announcements and, therefore, may have underestimated the non-resident purchase pressures that were occuring prior. In addition, although a figure like 5% may not sound material, it can certainly create tension within a market that is already facing tight supply, by making hot markets even hotter. Lastly, non-resident purchases were not uniformally distributed across the geography. Within the GGH, the York region and the City of Toronto reflected foreign purchase shares much higher at 9% and 7%, respectively.
Following the implementation of the non-resident’s tax in Ontario, there has been little evidence of a jump in foreign activity in other regions. Ottawa and Montreal (Chart 4) have heated up in recent months, but they are not yet exhibiting the typical signs of increased foreign investment activity, including a significant jump in the share of sales relative to the size of the population. Moreover, the Quebec Federation of Real Estate has noted that the strength in Montreal has been driven by the $400,000 and under segment of the market, while the the luxury home market has remained amply supplied. In other words, the segment of the market most attractive to first-time homebuyers is driving market activity and price growth.
This question can be unpacked into two components: migration flows and population age structure. In terms of the first, 2016 was a record year for immigration in Canada with over 300,000 immigrants. As a result, population growth has picked up to the highest pace since the mid-1990s. Government policy initiatives in the last year suggest that the level of immigration will hold steady near this pace, or potentially increase with time.
While most provinces have benefited from strong immigration flows, Ontario and B.C. also benefited from a sharp jump in interprovincial migration, with people leaving weak commodity-dependent economies like Alberta, Saskatchewan and Newfoundland. However, migration ebbs and flows with economic conditions. B.C. and Ontario’s ability to attract this group of individuals becomes more of a challenge as economic growth prospects converge with other provinces. In fact, this influence is already playing out. Interprovincial flows peaked in mid-2016 in B.C. and at the end of 2016 in Ontario (Chart 5). The combination of poor housing affordability in these provinces alongside improving economic conditions in Alberta and elsewhere will continue to reduce outflow to those markets.
From a demographic perspective, the age structure of the population has also caused “a clash of the titans” scenario. The two biggest population cohorts are Millennials (aged 21-36) and Boomers (aged 53 to 71). The first group reflects the strongest jump-up in homeownership, and the second group maintains the highest homeownership rate (Chart 6). These two forces are in constant competition for housing. While Millennials desperately look for affordable markets to raise families, Boomers are living longer and healthier in homes that are large relative to the household size. A recent report from the Canadian Centre of Economic Analysis highlighted that one in eight households do not have enough adequate living space. At the same time, half of Ontarians (and three-quarters of those aged 65+) have too many bedrooms and are considered to be over-housed.
Boomers are not yet at a life-stage to compel downsizing in a material way. This is further compounded by the high cost of moving and lack of appropriate new supply. The net effect is a rising homeownership rate among households who are 65 years and older. This places pressure on the mix of supply in the absence of seniors downsizing, particularly within urban centers where both age cohorts maintain strong housing demand due to amenities and proximity to workplaces.
From the Millennial perspective, this has meant that there is still pent-up demand in most markets across Canada, which will help maintain a floor under prices. Census data showed a rise in the share of those aged 20 to 34 living with their parents, from 33.3% in 2011 to 34.7% in 2016. Our estimates suggest that had the share not crept up – or if headship rates for this age group had stayed constant – household formation would have been about 93,000 higher over the past five years, or roughly 18,000 per year. So instead of household formation averaging 150,000 per year between 2011 and 2016, it would have averaged 168,000, which is a non-trivial difference. Should prices start to fall in the absence of an income shock, this data suggests that there are a lot of potential buyers sitting on the sidelines.
The wolf has finally come to the door! After seven years, the Bank of Canada raised the overnight target rate by 25 basis points in July and is expected to further gradually lessen monetary stimulus. Meanwhile, the Federal Reserve in the U.S. will commence a process to unwind its balance sheet and has already hiked rates four times. The European Central Bank is also expected to start reducing the amount of bond purchases it will be conducting in 2018. All these trends point to higher longer term bond yields globally and within Canada, with the 5-year government bond yield already up almost 60 basis points since early June. This has transmitted into roughly a 40 basis point increase in the best 5-year mortgage rate available to homebuyers since early July. And, with the upward path not done, we expect the 5-year government bond yield to rise another 45 basis points by the end of next year. Our soft landing view incorporates a parallel move up in mortgage rates. This is moderate by most standards, but will still crimp demand.
The impact will be felt disproportionately within a few key markets where affordability is the most challenged: Vancouver, Toronto and Montreal. TD Economics uses an affordability metric to measure the market’s sensitivity to higher interest rates. The index measures what share of income an average income earning household would have to devote to mortgage payments if purchasing an average priced home with a conventional mortgage (20% down, a 25 year amortization rate and a five-year fixed mortgage rate). This measure helps us understand how interest rate movements can potentially impact housing demand. Compared to just three years ago, a 40 basis point increase in mortgage rates today has a bigger impact on reducing affordability for the three markets mentioned above. Vancouver is the most sensitive to interest rates, but Toronto is not much better off (Chart 7).
The decline in sales activity in both Vancouver (over the last year) and Toronto (in the last four months) has helped to redistribute the balance of power from a pure seller’s market, back towards buyers, as evidenced by the sales-to-listing ratios (Chart 8). But, first-time homebuyers sitting on the sidelines waiting for higher interest rates to trigger a market crash may be holding their breath for a while. Prices are likely to only reset back to levels that existed prior to a year of exorbitant gains. An important nuance to understand is that a critical feature of a market crash is missing in this cycle. Listings shot up in the GTA following the policy measures, not because homeowners suddenly become incapable of affording their homes, but because speculative activity is being squeezed out. This is evidenced by the lack of a corresponding rise in economic or financial stress. The former is typically evident by rising unemployment rates that places household incomes into distress. However, the opposite has in fact occurred, with income accelerating and unemployment rates pushing to cyclical lows in Canada, and Ontario was no exception (Chart 9 and 10).
In terms of mortgage delinquency rates, these too are near a record low and not showing signs of upward migration. According to the Canadian Housing and Mortgage Corporation and Equifax, just 0.1% of households were more than 90 days delinquent on their loans in the final quarter of 2016 in Toronto. With the unemployment rate holding low, it is unlikely that households will be forced to sell in the near-term. Mortgage defaults follow the unemployment rate more strongly than they do mortgage payments. It’s important to remember that interest rates are rising because the economy is doing better, reflected by job creation and income growth. And, the speed of adjustment in mortgage rates is not anticipated to be rapid. Should it prove otherwise, this could be a recipe for a harder landing.
Under our current rising interest rate forecast, existing mortgage holders, on average, have some room to absorb higher debt-service costs. Canadians have been taking out records amount of debt to fund the purchase of relatively expensive homes. But, what sometimes gets overlooked is the fact that low interest rates have allowed mortgage holders to pay down their principal at an accelerated rate. During the 1990’s, interest costs accounted for well over two-thirds of monthly debt payments. As of the second quarter of 2017, over half of the monthly payments were principal repayment (Chart 11). So as interest rates rise, debt holders can adjust how much principal they are paying off and maintain stable payments by adjusting the type of mortgage (fixed versus variable) and/or changing the amortization period. However, these measures could incur additional costs and reviews to homeowners.
As for prospective buyers, historical comparisons indicate that the pace of mortgage rate increases is consistent with a seven to ten percentage point drop in existing home sales (Chart 12). However, this relationship could be more muted this time around due to past regulatory changes. For instance, increased requirements of income testing should cause demand to be less sensitive to the immediate rise in mortgage rates, because insured mortgages are already income tested at the higher Bank of Canada 5-year posted rate (which is currently 4.84%). This rate has gone up alongside the 5-year government bond yield, but by 20 basis points less. In addition, lenders have also been income testing borrowers who take out a mortgage with a lower term than 5 years at this posted rate. This offers a bit of cushion on how quickly, or to what extent, demand falls following the immediate move in interest rates. Although Toronto, Vancouver and Montreal may benefit less from this cushion due to higher interest rate sensitivity on reduced affordability, a modest rise in interest rates should be well absorbed by markets across Canada.
Yes. In the last few years, most other markets have already gone through a soft landing and are more balanced. In their soft landings, home prices slowed to sub 2% (or below income growth) for a sustained period. Allowing income growth to “catch up” is an argument for why higher mortgage rates will scale back the speed of the market, but not necessarily have the force to derail it.
Vancouver is the poster-child of soft landings. A downturn in housing cycles can be divided into two camps: sharp and short versus sharp and long (Chart 13). In the former camp, the run-up in home prices can last between six months and two years, with home prices rising at an average pace of 15% to 20% per year. These cycles tend to be followed by moderate peak-to-trough home price corrections of 10% to 14%. Vancouver has had five such cycles since 1990, consistent with a soft landing scenario. Existing home sales and prices reset to a level in better alignment with underlying fundamentals. But, the market can remain in a state of high valuations until an income shock comes along.
The second camp captures a sharp and long adjustments, resulting in more than a 20% peak-to-trough decline in home prices. These markets tend to be historically preceeded by a persistence of double-digit home price growth for four to five years. The great Canadian crash of 1989 and the most recent U.S. experience both fall into this camp. The longer the period that home prices rise at a double-digit pace, the more they are likely to divorce from fundamentals and accumulate financial risks. In other words, it’s not just the speed at which home prices rise, but the duration of the cycle that creates financial risks. These have been partially mitigated in the current cycle by regulatory measures, which were less evident in the previous two examples. Lastly, and most importantly, household distress needs to be accompanied by some broader distress in income or the financial system. This is not to say that Canadian households are free and clear of risks. Definitely not. Vulnerability is enhanced in the event that a recession forms. The case we are arguing is that these conditions have not materialized in the current environment as of yet.
So, for the GTA market, the current cycle is likely to be classified as the sharp and short variety. We expect a reset in home prices that would return the average price back to mid-to-late 2016 levels. The average home price has already fallen 13% peak-to-trough, placing it only back to early 2017 levels. Our forecast embeds a 6% contraction in the average home price in Toronto in 2018 and stabilization thereafter. Ultimately, Toronto home prices will remain elevated, similar to the experience of Vancouver. For most other markets, the combination of higher mortgage rates and moderate income growth set against balanced market conditions will help support annual home price growth of between 2-4%.
This report is provided by TD Economics. It is for informational and educational purposes only as of the date of writing, and may not be appropriate for other purposes. The views and opinions expressed may change at any time based on market or other conditions and may not come to pass. This material is not intended to be relied upon as investment advice or recommendations, does not constitute a solicitation to buy or sell securities and should not be considered specific legal, investment or tax advice. The report does not provide material information about the business and affairs of TD Bank Group and the members of TD Economics are not spokespersons for TD Bank Group with respect to its business and affairs. The information contained in this report has been drawn from sources believed to be reliable, but is not guaranteed to be accurate or complete. This report contains economic analysis and views, including about future economic and financial markets performance. These are based on certain assumptions and other factors, and are subject to inherent risks and uncertainties. The actual outcome may be materially different. The Toronto-Dominion Bank and its affiliates and related entities that comprise the TD Bank Group are not liable for any errors or omissions in the information, analysis or views contained in this report, or for any loss or damage suffered.