It is a pleasure to be back in Vancouver, a place often rated as the world’s most liveable city. Less frequently is it viewed as the most affordable. In the past three years, the average Vancouver house price is up about 30 per cent, making this city an extreme example of general developments in Canadian housing.
In my remarks, I will discuss the Canadian residential real estate market, consider some of the lessons from recent international experience and, finally, draw the policy implications.
The single biggest investment most Canadian households will ever make is in their home. Housing represents almost 40 per cent of the average family’s total assets, roughly equivalent to their investments in the stock market, insurance and pension plans combined. In recent years, housing has proved a very good investment indeed. The value of residential real estate holdings in Canada has climbed more than 250 per cent in the past 20 years, vastly outpacing increases in consumer prices and disposable income over that period (Chart 1).
However, Canada is arguably no better off because of it. That’s because while homeowners may feel wealthier because of this rise in prices, housing is not net national wealth. Some Canadians are long housing; others are short. Housing developments can have important implications for equality both across and between generations. Though some people in this room may have been enriched, their children and neighbours may have been relatively impoverished.
Housing not only meets a fundamental human need, it also has a wider economic significance. The Bank of Canada cares about the housing market because it can affect both price and financial stability. Let me address these in turn.
The objective of Canadian monetary policy is low, stable and predictable inflation, defined as a 2 per cent annual rate of increase in the consumer price index (CPI). Housing can influence inflation in three principal ways.
First, it is a major element of the CPI, accounting for more than a fifth of the basket of goods and services that makes up the index. The prominence of housing in the Bank’s target index provides an important advantage in that it ensures that the most important asset price in the economy is directly relevant to monetary policy.
Second, the real estate sector makes a significant, if volatile, call on resources and labour in our economy, thereby influencing wages and price pressures more generally. While residential investment has accounted for only 6 per cent of GDP on average over the past 30 years, activity in the sector has been four times more volatile than the overall economy over that period. Thus developments in housing substantially affect the business cycle and by extension inflation.
Third, while changes in housing values may not lead to changes in net wealth, they do influence consumption by affecting households’ access to credit. Through this “financial-accelerator” effect, homeowners can borrow more against increases in home equity to finance home renovations, the purchase of a second house, or other goods and services. Such expenditures can accelerate the increase in house prices, reinforcing the growth in collateral values and access to borrowing, leading to a further rise in household spending. Of course, this financial accelerator can also work in reverse: a decrease in house prices tends to reduce household borrowing capacity, and amplify the decline in spending.1,2
A home purchase triggers the biggest liability most families will ever take on. The value of housing-related debt in Canada has nearly tripled over the past decade to $1.3 trillion. This debt is also the single largest exposure for Canadian financial institutions, with real estate loans making up more than 40 per cent of the assets of Canadian banks, up from about 30 per cent a decade ago (Chart 2).
This unprecedented exposure exists in the context of a Canadian mortgage market that is subject to more stringent checks and balances than in the United States. For instance, almost all Canadian mortgages are full recourse, mortgage interest is not tax-deductible, and high-ratio lending standards are generally prudent. These factors help instil responsibility and discipline on both homeowners and lenders.
Nonetheless, the central position of housing assets and liabilities on the balance sheets of both households and financial institutions means that any housing excesses could generate important vulnerabilities in the financial system.
A review of the recent history is instructive.
A benign global macroeconomic environment and rapid financial innovation contributed to a global housing market boom through much of the past decade. In some countries, it went too far. Most prominently, in the United States, a substantial deterioration in underwriting standards turned boom to bubble and, ultimately, to bust. U.S. housing starts are now down three-quarters from their peak and prices have fallen by one-third. There have been over nine million foreclosures initiated since the cycle turned, and almost one in four mortgages is now in a negative equity position.
While some details differ, most notably in the scale of defaults, recent trends have been similar in the United Kingdom, Spain and Ireland, where housing markets remain acutely challenged.
Elsewhere, however, the global financial crisis proved to be only a brief setback, with the growth of house prices resuming, or even exceeding, its former pace.3
Canada falls into this latter group. Nationally, our house prices have risen 31 per cent from their trough in early 2009, to stand 13 per cent above their pre-crisis peak (Chart 3).4 Here in Vancouver, the recovery has been even stronger, with prices up 55 per cent from their trough to a level 29 per cent above the prior peak. The rebound in housing market transactions and new construction, both locally and nationally, has been similarly robust.
The performance of Canadian housing during the recent cycle has been unusual. For example, it took nearly 12 years for real residential investment to regain its level on the eve of the 1990s recession; this time it took only a year and a half (Chart 4). This rapid recovery importantly reflects the evolution of monetary policy during the recent recession. In response to the sharp, synchronous global recession, the Bank lowered its policy rate rapidly to its lowest possible level, doubled our balance sheet, and provided exceptional guidance on the likely path of our target rate.
With the initial rapid narrowing of the output gap, the return of employment to a level above its pre-crisis peak, the highly effective transmission of monetary policy in Canada, and the sustained momentum in household borrowing, the need for such emergency policies passed (Chart 5). As a consequence, the conditional commitment was removed, our balance sheet was normalized, and rates were successively tightened until they reached 1 per cent last autumn, where they have remained.
These policies provided considerable stimulus to the Canadian economy during a period of very weak global economic conditions and major downside risks. The housing market, highly sensitive to interest rates, responded. The rapid bounce-back in housing activity was also supported by federal government initiatives, notably the Insured Mortgage Purchase Plan and the Home Renovation Tax Credit. As a result, the recovery in the housing market played an important role in ensuring that the recession in Canada, while sharp, was also short.
With this renewed vigour building on the decade-long boom that preceded the crisis, the average level of house prices nationally now stands at nearly four-and-a-half times average household disposable income. This compares with an average ratio of three-and-a-half over the past quarter-century (Chart 6). Simple house price-to-rent comparisons also suggest elevated valuations (Chart 7). While neither of these metrics reflect the impact of low interest rates, even after adjusting for these effects, valuations look very firm. For example, the ratio between the all-in monthly costs of owning a home and renting a home, as measured in the CPI, is close to its highest level since these series were first kept in 1949 (Chart 8).
Financial vulnerabilities have increased as a result. Canadians are now as indebted (relative to their income) as the Americans and the British (Chart 9). The Bank estimates that the proportion of Canadian households that would be highly vulnerable to an adverse economic shock has risen to its highest level in nine years, despite improving economic conditions and the ongoing low level of interest rates.5 This partly reflects the fact that the increase in aggregate household debt over the past decade has been driven by households with the highest debt levels (Chart 10).
There are some offsets. Debt is largely fixed rate and household net worth is at an all-time high. However, borrowers should remember that a fixed-rate mortgage will reprice a number of times over the life of the mortgage and, while asset prices can rise and fall, debt endures.
The fact that the “official” personal savings rate in Canada has remained consistently positive is of limited comfort.6 The personal savings rate has fallen to historically low levels, despite the fact that the baby-boom generation is entering its highest saving years.7 Adjusting for housing expenditures, Canadian households have now collectively run a net financial deficit for 40 consecutive quarters, in effect, demanding funds from the rest of the economy, rather than providing them, as had been the case through the 1960s, 1970s, 1980s and 1990s (Chart 11).
How concerned should we be? To answer requires a deeper examination of the fundamental forces of supply and demand in the Canadian housing market.
Canada’s housing supply is relatively flexible, compared with other countries (Chart 12), and it appears to have grown at rates broadly consistent with underlying demand forces, the most important of which is the rate of household formation (Chart 13).8
Some excesses may exist in certain areas and market segments. In particular, the elevated level of “multiples” inventories (Chart 14), the ample pipeline of developments under way, and heavy investor demand (much of it foreign) reinforces the possibility of an overshoot in the condo market in some major cities.
Moreover, looking at the amount being spent on Canadian housing, rather than simply the number of houses being built, suggests that overall activity has been quite robust. Residential investment as a whole (including new home construction, renovations and ownership transfer costs) has consistently exceeded its long-term average share of the overall economic activity for more than seven years. Residential investment is now at levels that have previously proved to be peaks in Canada and, on a relative basis, in the United States (Chart 15). This partly reflects the generally strong performance of Canada’s labour market over the past decade, which has driven solid gains in employment and income. However, it also reflects historically favourable borrowing conditions, and potentially overly optimistic assumptions about future developments.
Lower interest rates reduce the cost of financing the purchase or construction of houses and increase the value of collateral, enabling more borrowing than would otherwise be possible. While monetary policy rates influence mortgage rates across the yield curve, mortgage rates are ultimately set in the market, where a broad set of domestic and global forces play a role.
Among these forces, the so-called “global savings glut” has been particularly important in underpinning the trend decline in long-term borrowing rates over the past decade. Flows of excess savings from emerging markets into the U.S. Treasury market have restrained the long-term U.S. interest rates that provide the benchmark for yield curves globally. Rough calculations suggest that the lower real rates from this phenomenon could justify a substantial proportion of the increase in house valuations across mature markets.9 The eventual rebalancing of the global economy from deficit countries, like the United States, to surplus countries, like China, should dampen this effect.
As in many other countries, cheap credit has been used to bid up the price of Canadian houses, a non-tradable good, rather than invest in expanding the productive capacity and export competitiveness of our businesses. For example, over the past decade, housing debt grew by more than 150 per cent, while business borrowing rose by only 40 per cent. As a result, the stock of housing-related debt went from less than business debt to almost two-thirds more.
Domestic demand factors are not the only forces at work. Some Asian wealth is being invested in selected international housing markets as those investors seek out diversification and hard assets. This has become a familiar phenomenon in this city. Partly as a consequence, the average selling price of a home in Vancouver is now nearly 11 times the average Vancouver family’s household income, a multiple similar to those seen in Hong Kong and Sydney—cities that have also become part of a more globalized real estate market. Such valuations are extreme in both Canada and globally (Chart 16).
Given such developments, one cannot totally discount the possibility that some pockets of the Canadian housing market are taking on characteristics of financial asset markets, where expectations can dominate underlying forces of supply and demand. The risk is that expectations become extrapolative, prompting the classic market emotions of greed and fear—greed among speculators and investors—and fear among households that getting a foot on the property ladder is a now-or-never proposition.
The Bank manages policy for the economy as a whole, rather than any specific region or sector. In this context, what does the Bank of Canada expect for housing? In a word: moderation.
In the Bank’s view, Canadian housing market developments in recent years have largely reflected the evolution of supply and demand. While supply of new homes should remain relatively flexible, many of the supportive demand forces are now increasingly played out.
For example, while measures of housing affordability remain favourable, this is largely because interest rates are unusually low. Rates will not remain at their current levels forever. The impact of eventual increases is likely to be greater than in previous cycles, given the higher stock of debt owed by Canadian households. At a 4 per cent real mortgage interest rate—equivalent to the average rate since 1995—affordability falls to its worst level in 16 years (Chart 17).10 As I have observed, some markets are already severely unaffordable even at current rates.
Since 2008, the federal government has taken a series of prudent and timely measures to tighten mortgage insurance requirements in order to support the long-term stability of the Canadian housing market. These will reduce the possibility that prices are further driven up simply through higher leverage.
The Bank has been expecting moderation in the housing sector as part of a broader rebalancing of demand in Canada as the expansion progresses. Overall economic growth is expected to rely less on household and government spending, and more on business investment and net exports. Household expenditures are expected to converge toward their historic share of overall demand in Canada (Chart 18), with expenditures growing more in line with household income. In this context, the Bank anticipates a slowing in both the rate of household credit growth and the upward trajectory of household debt-to-income ratios.
There are conflicting signals regarding the extent to which this moderation is proceeding (Chart 19). While growth in consumer spending slowed markedly in the first quarter, housing investment re-accelerated, as did household borrowing, with mortgage credit growing at a double-digit annual rate (Chart 20). It is likely that some of this resurgence in borrowing is transitory, reflecting the lagged effects of the surge in existing home sales in the fourth quarter of last year, as well as recent changes in mortgage insurance regulations that may have resulted in some activity being pulled forward into the first quarter. Mortgage credit growth slowed in April, reinforcing the view that the particularly strong increase in borrowing in the first quarter was temporary. Nonetheless, at a 5 per cent annual rate, growth in mortgage credit in April was slower but not slow, particularly given the sustained above-trend increases of recent years.
As the Bank emphasized in its recent rate announcement, the possibility of greater momentum in household borrowing and spending in Canada represents an upside risk to inflation.
In conclusion, historically low policy rates, even if appropriate to achieve the inflation target, create their own risks.
Canadian authorities will need to remain as vigilant as they have been in the past to the possibility of financial imbalances developing in an environment of still-low interest rates and relative price stability. We continue to co-operate closely and monitor the financial situation of the household sector.
In the short term, economic growth in Canada is expected to slow to modest rates, due to a number of temporary factors. These include supply chain disruptions that will dampen automotive production and the drag from adjusting to higher energy prices on consumer spending in Canada and the United States. Overall, the Bank expects a re-acceleration of growth in the second half of the year, consistent with a renewed narrowing of the output gap.
While global uncertainties persist, to the extent that the Canadian economic expansion continues and the current material excess supply in the economy is gradually absorbed, some of the considerable monetary policy stimulus currently in place will be eventually withdrawn, consistent with achieving the 2 per cent inflation target. Such reduction would need to be carefully considered.
With monetary policy continuing to be set to achieve the inflation target, our institutions should not be lulled into a false sense of security by current low rates. Similarly, households will need to be prudent in their borrowing, recognising that over the life of a mortgage, interest rates will often be much higher.