RBC Global Asset Management: Surprisingly Sustainable Canadian Housing

Economic Compass
Global Perspectives for Investors
OUTLOOK
Near term Medium term Weight Confidence
Household
debt Neutral Slight
negative
15% Medium
Housing
affordability Neutral Major
negative
25% Medium
Construction
sustainability Neutral Slight
negative
20% High
Condo
appetite
Slight
negative
Slight
negative
15% Low
Foreign buyers
and investors Neutral Neutral 10% Low
Distribution
of debt Neutral Negative 15% Medium
Economic
implications Neutral Negative Medium
Note: "Near term" defined as over the next year, "Medium term" as 1–5 years.
"Confidence" refers to confidence in forecast. Source: RBC GAM
CANADIAN HOUSING IN SIX QUESTIONS
Among the many variables vying for influence over the Canadian
economic outlook – prominently including a weaker currency,
lower oil prices and a strengthening U.S. economy – the
Canadian housing market has tended to capture the imagination
of the public, the press and investors more than the rest.
There are three reasons for this fascination. First, the majority
of Canadians own their home, making developments in the
housing market of obvious relevance. Second, home prices have
increased at a spectacular rate over the past decade, inducing
glee in those already in the housing market and despair among
those who are not. Third, there has long been a feeling of
uneasiness about housing’s future prospects.
This paper identifies six key questions whose answers together
determine the extent of the challenges awaiting the Canadian
housing market:
1) Is household debt unsustainable?
2) Is housing affordability precarious?
3) Is residential construction exceeding demand?
4) Is the condo market especially overbuilt?
5) Are foreign buyers and investors a source of vulnerability?
6) Does the distribution of household debt reveal additional
problems?
In summary (Exhibit 1), we find that many of the concerns are
overblown. In the near term, household-debt levels are perfectly
sustainable, housing affordability is surprisingly normal and
construction is merely keeping pace with demand. The condo
market is perhaps more vulnerable than is the market for
single-family homes – but less so than it appears – and support
from foreign buyers and investors is not likely to dry up soon.
HIGHLIGHTS
ERIC LASCELLES
Chief Economist
RBC Global Asset Management Inc.
ƒƒ The Canadian housing market has long defied expectations of collapse, though fears linger.
ƒƒ The near-term outlook remains quite benign. There are no particular signs of household
distress, affordability is fine given low mortgage rates and construction is running precisely
as it should. Worries about excessive condo activities and the influence of investors are
overblown.
ƒƒNaturally, the medium-term outlook is somewhat more negative, dominated by deteriorating
affordability due to rising mortgage rates. Still, the potential construction downside is
surprisingly tame, limiting the likely economic damage to no more than a quarter percentage
point of GDP per year.
ƒƒWhile more bearish scenarios are conceivable, they remain unlikely. As such, the Canadian
housing market arguably takes a back seat to more pressing Canadian economic impulses,
such as a lower loonie (good), lower oil prices (bad) and a stronger U.S. economy (good).
ISSUE 33 • NOVEMBER 2014
Exhibit 1: Canadian housing scorecard
The distribution of debt naturally reveals a few additional
vulnerabilities, but nothing shocking.
On the other hand, the medium-term outlook is still somewhat
negative. Looking further out, we expect household debt to
become more burdensome, housing affordability to deteriorate
significantly and construction to gradually ebb, reflecting
slowing population growth. Still, disaster is unlikely.
Thus, while Canada’s housing market certainly merits a watchful
eye, it arguably attracts too much attention relative to other
more relevant economic impulses, such as those involving the
loonie, oil prices and the U.S. economy.

2 ECONOMIC COMPASS Issue 33 • November 2014
Note: Based on latest data available. Debt for households and non-profit
institutions serving households. Figures differ slightly from made-in-Canada
calculations. Source: Haver Analytics, RBC GAM
Note: Debt-service ratio defined as cost of interest payments on debt only.
Source: Statistics Canada, RBC GAM
1) Is household debt unsustainable?
Canadian household debt is now at a historically large 164%
of personal disposable income – near a record high and well
above the U.S. and U.K. These figures have naturally led to fears
that Canadians could eventually be forced to undertake a brutal
deleveraging akin to the U.S. following the 2008 financial crisis.
Near-term calm
Fortunately, there is little immediate evidence of distress. While
Canadian household-debt levels are high, they are nowhere
near the highest in the world. Countries including Norway,
Denmark, the Netherlands and Australia survive, and in some
cases, thrive, with materially more debt (Exhibit 2).
In Canada’s case, the cost of servicing the interest on all of this
debt is quite low, and in fact commands the smallest share of
income in decades (Exhibit 3). When principal payments are
included, the Bank of Canada calculates that the ratio is still no
higher than normal.
Furthermore, the era of excessive household-credit growth
seems to be over. The debt-to-income ratio has stabilized as
household-credit growth has been pared to a tame 4% per year.
Part of the reason for this deceleration lies in self-regulating
households that are wary of taking on additional debt.
Another part may be explained by macroprudential rule changes
that have served to limit access to credit via stricter eligibility
rules. Given international evidence that each macroprudential
rule change seems only capable of undercutting demand for
a quarter or two, Canada Mortgage and Housing Corporation
(CMHC) has delivered a steady stream of rule changes in recent
years. The latest tweaks have cut access to CMHC mortgage
insurance for homes costing more than $1 million and for
investors seeking to finance second homes. We expect further
rule-tightening to continue as necessary to keep credit growth in
check.1
Medium-term decline
The medium-term outlook for household debt is somewhat
worse, as higher borrowing costs will eventually push the
debt-service ratio – at least the version that includes principal
payments – into worse than usual terrain. Fortunately, a
variety of mitigating forces should serve to limit the damage,
and leave the overall medium-term outlook no worse than a
slight negative.
Any parallels to the U.S. housing crash and household-deleveraging
experience are limited, as the U.S. experience was
caused only in part by high home prices and elevated levels
of household debt (and not at all by rising rates). The more
important contributors were sub-prime mortgages that lured far
too many into the housing market, a securitization process that
concealed the underlying risk, a credit market unprepared for
adverse conditions and a spike in unemployment that brought
the whole thing tumbling down. None of these conditions exist
or appear likely in Canada.
History demonstrates that credit crashes have little to do with
how much debt a country is carrying, but instead key off of how
quickly and how recently the debt has been accumulated. The
Bank of International Settlements has identified a technique
for quantifying this risk via the departure of credit growth from
its long-term trend. We implement this for Canadian household
debt and find that the downside risk has faded over the past
few years, from an extremely elevated risk to entirely normal
readings today (Exhibit 4).
The type of debt that households have been accumulating also
has a bearing on the risk (Exhibit 5). All types of borrowing boost
the economy in the short run. The variation is in what happens
Denmark
Netherlands
Norway
Switzerland
Australia
Sweden
Canada
U.K.
Spain
Japan
France
Greece
Eurozone
U.S.
Germany
Italy
0
50
100
150
200
250
300
Household Debt to Disposable
Income (%)
Exhibit 2: Canadian household debt nowhere near highest in the
world
6
7
8
9
10
11
12
1990 1994 1998 2002 2006 2010 2014
Debt-Service Ratio (%)
Historical
average
Exhibit 3: Cost of servicing Canadian household debt at
historic low

Economic Compass
3
Note: Real home price is % change from 1980 level; price-to-income and price-to-rent
versus average since 1975; carrying cost versus average since 1980.
Source: The Economist, Haver Analytics, RBC GAM
Note: Trend calculated using HP filter on quarterly data with lambda of 500,000.
Source: Haver Analytics, BIS, Bank of Canada, RBC GAM
-8
-6
-4
-2
0
2
4
6
8
10
1980 1987 1994 2001 2008 2015
Deviation from Credit-to-GDP
Growth Trend (Percentage Points)
This metric now
argues
household credit
vulnerability
gone
Household
credit growth
pointed to
extreme
vulnerability
Credit overshoot
Credit
undershoot
Exhibit 4: Household credit vulnerability has faded
High
High
High
High
High
No
Medium
Yes
Medium
Yes
Low
Low
PURPOSE OF CREDIT
CAPITAL HOUSING SPENDING
Benefit to
short-term GDP:
Benefit to
long-term GDP:
Asset
accumulation:
Bubble risk:
Exhibit 5: Not all credit is created equal
+130%
+32%
+74%
+1% -4%
-10
10
30
50
70
90
110
130
Real Home
Price
Home Price-to-
Income
Home Price-to-
Rent
Carrying
Cost-to-
Income
(Fixed Rate
Mortgage)
Carrying
Cost-to-
Income
(Variable
Rate
Mortgage)
Misvaluation of Canadian
House Prices (%)
Exhibit 6: Housing affordability depends on the measure
over the longer term. The most economically useful borrowing
is deployed into capital investment – the stuff of machinery
and bridges – as this increases the productive capacity of the
economy without overheating it or blowing bubbles.
Of course, households aren’t usually in a position to do
much of this. Instead, they borrow to buy a house or finance
discretionary spending. Fortunately, Canadians have been
doing the more prudent of the two – buying homes that at least
increase the asset side of the balance sheet, leaving their net
financial position unaltered. In fact, for all of the accumulated
household debt, household assets now outweigh liabilities by a
remarkable 5.4 times.
Moreover, everyone has to live somewhere. A mortgage
payment and the debt associated with it helpfully eliminates the
cost of paying rent.
2) Is housing affordability precarious?
As Canadian home prices have risen, so has chatter about
deteriorating affordability. Confusingly, various affordability
metrics yield wildly different readings, with home valuation
estimates ranging from 130% too high to 4% too low (Exhibit
6). We have strong opinions on which of these readings can be
trusted, and which cannot.
Real home price
The real home price metric assumes that home prices should not
rise any faster than inflation. It is tempting to agree – after all, a
home is just a pile of bricks, copper and other commodities
placed on a fixed plot of land. This measure notes that home
prices have outpaced inflation by a whopping 130% since 1980.
In practice, however, real home prices are a poor measure of
affordability. A key reason is that household incomes rise over
time. Buyers can afford to pay more whether or not a house
actually costs more to build.
Second, land is a scarce resource, especially in the context of
a rising and urbanizing population. The price of land can and
should expect to outpace inflation over time.
Third, the quality and size of homes have increased steadily
over the years.
Home price-to-income
The home price-to-income ratio suggests that home prices
should increase at the same pace as personal incomes.
This addresses one of the key flaws of the prior measure.
Nevertheless, this measure estimates that home prices are still
32% too high.
However, it neglects a further crucial consideration: few
Canadians pay cash for their homes. The vast majority must
borrow to do so, making interest rates a relevant but overlooked
Source: RBC GAM

Note: Current carrying cost of a home versus the historical norm.
Source: CREA, Statistics Canada, Haver Analytics, RBC GAM
Note: As of October 213. Source: CMHC, RBC GAM
-60
-50
-40
-30
-20
-10
0
10
20
30
40
50
1985 1990 1995 2000 2005 2010 2015
% Deviation From Fair Value
Fixed
Variable
Good home
affordability
Poor home
affordability
Fixed-rate
mortgage is
around
average
Variable-rate
mortgage is
cheap
Exhibit 8: Canadian housing affordability OK for now
$1,526
$1,743
$1,287
$1,580
$1,028
$1,200
$1,005
$1,281
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
1-Bedroom 2-Bedroom 1-Bedroom 2-Bedroom
Average Rents ($)
Condo Apartments
Purpose-Built Apartments
Toronto Vancouver
Almost 50%
premium
Exhibit 7: Condo rentals are more expensive
influence. Interest rates have declined for an unprecedented
30 years, leaving the effective cost of owning a home far lower
than a simple ratio of a home’s price to its owner’s income
would suggest.
Home price-to-rent
The home price-to-rent ratio takes a totally different approach.
It ignores incomes, inflation and interest rates, and instead
focuses on the relative allure of renting versus buying to meet
one’s shelter needs. The theory, then, is that the price of a home
should be equal to a fixed (and presumably rather large) number
of months of rental payments. The home price-to-rent ratio
claims that Canadian home prices are a startling 74% too high.
However, there are four problems with this affordability metric.
First, and crucially, the methodology underlying the Canadian
home price-to-rent ratio is flawed via its exclusive focus on
purpose-built rentals (and exclusion of condo rentals). Purpose-built
rentals are increasingly dated, as very few have been built
in recent decades. Meanwhile, condo rentals are excluded
despite representing practically the entirety of the new rental
stock, with average condo rents running 25% to 50% higher
than purpose-built rentals (Exhibit 7). Thus, the true home price-to-
rent ratio is not nearly as extreme as official reports claim.
Second, and as with the prior affordability measures, the home
price-to-rent ratio ignores the structural decline in the cost of
borrowing, thus erring in making home-buying look relatively
more expensive than it really is.
Third, buying and renting are not true substitutes. The selection
of single-family homes for rent is relatively slim in Canada, and
they are sprinkled unevenly across neighbourhoods. Renting
also introduces an element of geographic risk, as the tenant
does not have complete control over the duration of their stay.
Furthermore, the transactional costs and effort required to
sell a home, find another and physically move are quite high,
rendering arbitraging cost gaps quite difficult in practice.
Fourth, rent controls artificially repress rents in some parts of
the country. Similarly, Canada has a longstanding culture of
home-buying, meaning Canadians are willing to pay at least a
small premium for the privilege.
Carrying cost
The serious flaws in each of these metrics prod us toward the
least flawed of the bunch: carrying-cost measures.2 These come
closest to approximating how Canadians themselves evaluate a
prospective home purchase: by how much they earn versus how
much their mortgage will cost on a monthly basis.
By this metric, home prices have actually behaved quite
reasonably. Yes, prices have soared, but mortgage rates have
plummeted and incomes have edged higher. The interplay
between the three variables has left the carrying-cost measures
almost precisely at fair value. A home financed with a fixed-rate
mortgage is a mere 2% too pricey, while a home financed with
a variable-rate mortgage is actually 4% cheaper than it should
be (Exhibit 8). In other words, Canadians have quite responsibly
calibrated their purchases to what they can afford.
And with home prices puttering upwards at 2% to 6% per
year (Exhibit 9), affordability doesn’t appear to be deteriorating
significantly.
Bigger issue later
Of course, once mortgage rates climb to normal levels,3 the
carrying-cost affordability calculations suddenly become much
less friendly, lurching to the conclusion that home prices are
15% too high (Exhibit 10).
There are two ways this mismatch can be resolved. The first
is painful and involves home prices falling by an abrupt 10%

Economic Compass
5
Note: Based on data since 1990 where data is available. Box represents the range
of 25th and 75th percentile. CREA national residential average price; Teranet/
National Bank of Canada Composite 11 Home Price Index; Statistics Canada New
Housing Price Index. Source: Haver Analytics, RBC GAM
Note: Fixed Floor imposes a minimum "normal" mortgage rate on the affordability
calculations, and so in the current context reveals how affordability would look at
normal mortgage rates. Source: CREA, Statistics Canada, Haver Analytics,
RBC GAM
4.3%
6.3%
2.0%
5.5%
5.4%
1.6%
0
1
2
3
4
5
6
7
8
9
10
Teranet/National
Bank
New Housing
Price Index
YoY % Change
Historical Average
Latest
CREA Existing
Home Prices
-50
-40
-30
-20
-10
0
10
20
30
40
1985 1990 1995 2000 2005 2010 2015
% Deviation From Fair Value
Fixed Floor
Fixed Actual
Good Home
Affordability
Poor Home
Affordability
Affordability fine at
current rates,
somewhat expensive
at normal rates
Exhibit 9: Canadian home prices rising at normal clip
Exhibit 10: Canadian affordability will fall when rates normalize
over the span of a few years, with rising incomes eroding the
remainder of the affordability gap. The second possibility is
much more muted: home prices simply flatline for four or five
years, leaving the entire gap to be closed via rising incomes
(increasing at a rate of perhaps 3–4% per year).
The latter scenario is the more likely, as the pain of higher
mortgage rates won’t bite quickly enough to result in a more
extreme scenario. There are several reasons to anticipate a
“slow burn”:
ƒƒ Central banks, including the Bank of Canada, are unlikely
to tighten rates particularly quickly given their concern over
the resilience of the economic recovery.
ƒƒ The increase in bond yields (and thus term mortgage rates)
should be restrained by investors substituting away from
even lower yields in Japan and the Eurozone.
ƒƒ A rising majority of Canadian mortgages4 are locked into
fixed rates, usually for a term of five years. This means it
will take several years for the effects of higher mortgage
rates to saturate the housing market.5
It is nevertheless the case that, in all scenarios, home prices
eventually slip 15% behind their prior upward trajectory. This
brings real consequences that we quantify later.
3) Is construction exceeding demand?
The conventional wisdom has long been that Canadian
construction is stubbornly exceeding demographically
sustainable demand. Whereas 200,000 to 250,000 annual new
units have been the norm of the past decade, the rule of thumb
has long been that only around 175,000 new housing units are
actually needed each year.
However, this assumption is slightly off. The necessary housing
stock varies quite substantially according to population growth
and underlying demographic shifts. Although population growth
is no longer strong, the sustainable household formation rate is
nevertheless running along at a significantly faster clip
(Exhibit 11). Sluggish population growth among young people
(who live mainly with their parents anyway) is being trumped by
faster growth among seniors (who are much more likely to live
alone).6 Every household, of course, needs somewhere to live.
Housing flow
We calculate that the Canadian economy currently needs
200,000 housing starts per year to keep up with demographic
demand. This is well above the aforementioned rule of thumb,
and even a bit above the recent trend.7
The approximate appropriateness of current construction rates
can be independently corroborated by looking at the share
of GDP dedicated to residential investment (Exhibit 12). The Source: Statistics Canada, United Nations, Haver Analytics, RBC GAM
1.0%
0.3%
2.5%
1.3%
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Population Aged <25 Aged 75+ Households
Annualized % Change, 2005-2020
Population
growth is
just so-so...
...but growth by
age is skewed
toward seniors,
who form far more
households than
children do...
...resulting in
unusually
strong
expected
household
formation
Exhibit 11: Surprisingly good household formation in Canada

Note: Sales-to-New Listings ratio is calculated by dividing existing home sales by
existing home new listings. Shaded area indicates balanced market as defined
by a sales-to-new listings ratio in the range of the 34th to 67th percentile between
1988 and the present. Source: CREA, Haver Analytics, RBC GAM
Source: Statistics Canada, Haver Analytics, RBC GAM
70
80
90
100
110
120
130
1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014
Residential Investment as % of GDP
(1981 = 100)
Nominal Investment
Real Investment
Exhibit 12: Nominal residential investment is high, but not real
investment
-15
-10
-5
0
5
10
15
20
25
30
0.3
0.4
0.5
0.6
0.7
0.8
0.9
2001 2003 2005 2007 2009 2011 2013
Average Price (YoY % Change)
Sales-to-New Listings Ratio
Sales-to-New Listings Ratio
Existing Home Price
Buyer's
Market
Seller's
Market
Exhibit 13: Resale housing market seems balanced
nominal figure looks worryingly high, but this proves to be an
illusion once inflation has been accounted for.8 The volume of
residential construction in Canada actually constitutes a normal
share of output.
Market dynamics in the resale home market also continue
to confirm a view that the market is healthy and reasonably
balanced (Exhibit 13).
Housing stock over the medium term
When examined over the medium term, however, the
construction outlook becomes slightly negative. This is due to
the dimensions of the pre-existing housing stock, the average
Canadian home size and shifting demographics.
It is one thing for the flow of construction to be proceeding
in line with the growth in households, as it currently is. It is
something very different for the aggregate housing stock to be
aligned with the overall number of households.
Our estimates9 suggest that Canada started its recent housing
boom with an inadequate housing stock. Subsequent strength
has pushed the stock upwards out of this hole, and ultimately
into a moderately overbuilt position (Exhibit 14). Fortunately,
the excess is not enormous. We figure there are around 73,000
too many homes, representing just 0.5% of the overall housing
stock. Some underbuilding will be necessary to restore balance.
A further cause for caution is that Canadian homes appear to
have more rooms per person than other countries – including,
surprisingly, more than the U.S. (Exhibit 15). Naturally, this
would seem to bolster the view that the Canadian housing
market is stretched. However, there are several tempering
interpretations:
ƒƒ First, current Canadian construction is skewed toward
condos (which have fewer rooms), so regardless of the
Note: Natural demand calculated using UN population forecasts and historically
normal age-based population-to-household ratios. Actual housing stock
calculated using reported figures through 2000, then estimated via the rate
at which new homes are completed (with a 1.12 multiplier from housing
completions to increases in the housing stock due presumably to the conversion
of some properties into multi-unit apartments), a 1.055 multiplier between the
number of households and number of dwellings to reflect seasonal properties
and a 0.015% annual teardown rate on the existing housing stock.
Source: CMHC, Statistics Canada, UN, Haver Analytics, RBC GAM
-150
-100
-50
0
50
100
150
1986 1989 1992 1995 1998 2001 2004 2007 2010 2013
Gap Between Actual Residential
Dwellings and Natural Demand
(Thousand Units)
Overbuilt
market
Underbuilt market
Currently too many
homes, but the gap
is only equivalent to
five months of
excess
construction
Housing boom was
initially about restoring
the housing stock to a
normal level, but then
continued into
overbuilt territory
Exhibit 14: Canadian housing stock slightly too large
origin of this excess, it is unlikely that the surplus of rooms
is new to the latest housing boom or getting worse.
ƒƒ Second, if Canadian homes have more rooms, it helps to
validate high home valuations.
ƒƒ Third, Canada has among the lowest population densities
in the world, meaning more land per household. It can be
no coincidence that Australia is the other country at the
head of the class, with the U.S. not far behind.
ƒƒ Fourth, Canadian homes may have more rooms than
the U.S. because of the near-universality of basements
(which are less the norm in the U.S. West and South). It

Economic Compass
7
Note: Bars calculated using weighted average of major Canada cities. Circle
calculations based on change from 2011 to 2013. Source: Conference Board of
Canada/Genworth Canada, RBC GAM
Note: Housing starts-equivalent measure of demand calculated using UN
population forecasts, historically normal age-based population-to-household
ratios, a 1.03 multiplier from housing completions to housing starts, a 1.12
multiplier from housing completions to increases in the housing stock (due
presumably to the conversion of some properties into multi-unit apartments after
completion), a 1.055 multiplier to reflect the existence of seasonal properties and
a 0.015% annual teardown rate on the existing housing stock. Source: CMHC,
Statistics Canada, UN, Haver Analytics, RBC GAM
Source: OECD Better Life Index 2014, RBC GAM
0.5
1.0
1.5
2.0
2.5
Canada
Australia
New Zealand
U.S.
Ireland
Denmark
Netherlands
Norway
Spain
Switzerland
U.K.
France
Germany
Japan
Sweden
Portugal
Italy
Korea
Brazil
Turkey
Mexico
Russia
Number of Rooms Per Person
Exhibit 15: Canadian homes have more rooms
is debatable whether these basements – even renovated
ones – deserve equal billing as functional rooms.
Lastly, and perhaps most importantly, are the deteriorating
demographic considerations. Slowing population growth
reduces the demand for new homes. In five years, Canada's
needs will fade from 200,000 units annually to 190,000 units.
This means a mild 1% construction decline per year. In 25 years,
the figure falls all the way to just 125,000 units (Exhibit 16).
4) Is the condo market especially overbuilt?
The large number of cranes darkening the skies over Toronto
and Vancouver are viewed by many as evidence of reckless
condo overbuilding.
100
120
140
160
180
200
220
240
260
1987 1997 2007 2017 2027 2037
Thousand Units
Housing Demand
Housing Starts
Exhibit 16: Construction matches current demand, but must
gradually fall
Note: Multi-units defined as row houses and apartments, excluding semi-detached.
Source: CMHC, Haver Analytics, RBC GAM
22
26
30
34
38
42
46
50
54
58
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Multi-units as % of Total Housing
Starts
Share of
existing
housing stock
(2)... but not
unprecedented...
(3)... and prior share
was unnaturally low
(1) Multi-units now
over half of
construction...
Exhibit 17: Multi-unit starts shift higher
There is no denying a basic uptick in condo construction.
Multi-unit construction10 now accounts for the majority (52%) of
housing starts, up from just 29% in 1997 (Exhibit 17).
Equally, it is fair to acknowledge that the condo market currently
looks slightly less robust than the single-family market, both
from the perspective of inventories and prices.
With regard to inventories, condo units appear to be selling less
quickly than before. The supply of resale condos for sale has
increased moderately over the past few years in most cities. The
trend is more mixed for new condos (Exhibit 18).
0
1
2
3
4
5
6
7
8
9
10
Resale Condo Market New Condo Market
Number of Month's Supply For Sale
2011
2012
2013
Rising in 75%
of major cities
Rising in 63%
of major cities
Worsening
month's
supply
Slight
improvement in
month's supply
Exhibit 18: Canadian condo market internals soften modestly

Source: Canadian Housing Observer 2013, CMHC, RBC GAM
Note: Apartments and row houses per 100,000 people over 25 years of age.
Historical average since 1976. Source: CMHC, Statistics Canada, Haver Analytics,
RBC GAM
-10
-5
0
5
10
15
20
2006 2007 2008 2009 2010 2011 2012 2013 2014
MLS Home Price Index
(YoY % Change)
Single-Family Home
Apartment Condo prices persistently
underperformed since
financial crisis
0
2
4
6
8
10
12
14
15-24
25-29
30-34
35-39
40-44
45-49
50-54
55-59
60-64
65-69
70-74
75+
Canadian Condominium Owners
as % of All Households
Age Group
2011 2006
2001 1996
Young and old
have shown
greatest increase
Source: Canadian Real Estate Association, RBC GAM
Exhibit 19: Condo prices lag in Canada
Exhibit 20: Condo appetite rising over time
Similarly, condo prices are rising more slowly than in the single-family
market, a gap that has been in place since the onset of
the financial crisis (Exhibit 19).
Exaggerated concerns
However, we suspect some of the more extreme concerns about
the condo market are unrealistic.
Given that the overall rate of dwelling construction in Canada is
about right, any excess condo construction must simultaneously
mean that there are too few single-family homes being
constructed. Thus, this is a debate about composition rather
than absolute excesses.
The overall housing stock11 is 40% multi-unit properties. This
means that the 29% multi-unit construction share of the late
1990s clearly constituted underbuilding. Therefore, part of the
recent shift toward condo construction merely represents a
counterbalance to this earlier era. The current construction share
remains shy of the all-time high of 58% set in the late 1960s.
Shift in preferences
Changing demographics and tastes go a long way toward
justifying the remainder of the recent shift toward condos.
The demographic aspect of this shift has already been laid out:
the rising number of one-person households and childless
couples naturally tilts demand toward condos over single-family
homes. Furthermore, an aging population increasingly seeks to
avoid the hassles of home maintenance and drive less.
Supplementing this is an apparent shift in housing preferences
among the young. Whereas the classic aspiration was once to
settle down in the suburbs, many young people increasingly
prefer to remain downtown (necessarily, in multi-unit dwellings).
There are several plausible reasons why.
Growing cities compounded by geographic impediments (such
as Vancouver’s ocean, waterways and mountains, and Toronto’s
lake and greenbelt) render each new iteration of suburban
homes – the classic stomping ground for new families – ever
more distant from the urban core and thus less practical from a
commuting perspective.
Moreover, young people seem less interested in owning a car
or driving than previous generations. From 1998 to 2008, there
was a 28% drop in the fraction of Americans aged 16–19 with a
driver's license.12 Increasingly, young people want to live where
they work and play.
All of this has led to an increased appetite for condo living,
regardless of age, though especially among the young and old
(Exhibit 20).
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
1980 1987 1994 2001 2008 2015
# of Standard Deviations from
Normal
Multi-Unit Under Construction
Multi-Unit Starts
Multi-Unit Completions
Loads of
construction...
...versus
merely warm
starts and
completions
Exhibit 21: Worries of a condo construction glut…
Source: Canadian Real Estate Association, RBC GAM

Economic Compass
9
The condo pipeline
A remaining concern about the condo market is that the pipeline
of condos under construction seems to be far bigger than usual,
and completely out of line with the recent rate of condo starts
and completions (Exhibit 21). The fear is that, once completed,
all of this construction could completely overwhelm demand.
However, we are decreasingly worried by this scenario.
The key to understanding this is that the modern residential
building appears to take around twice as long to build as those
of decades past (20 months rather than 10 months). There are
several plausible reasons why. The desire to “pre-sell” as many
condos as possible may lead to some foot-dragging in the early
stages of construction. The fact that multi-unit dwellings are
increasingly built in downtown cores means projects are more
complicated and require more coordination. Furthermore, the
fact that multi-unit dwellings appear to be growing ever taller
also increases the complexity of the projects (and physically
reduces the ratio of usable to overall constructed space13).
If condos take twice as long to build, then the market requires
twice as much construction at any time to ensure a normal
supply of completed condos (Exhibit 22). As a result, we suspect
the supposed condo-construction bulge does not actually exist.
Remaining condo worries
The risk to condo builders actually seems fairly tame.
CMHC estimates that 89% of condos under construction
are presold. At a minimum, banks usually require at least
75% of units sold before extending financing. This means
that, barring an extreme home-price correction that
completely wipes out the value of the 15%–20% deposit
that most buyers make on their condos, builders are
reasonably insulated from housing-market gyrations.
Another worry is that condos might be too expensive relative
to single-family homes. Toronto condos regularly cost more on
a square-foot basis than a house, despite the absence of land.
However, this is less troubling than it first looks:
ƒƒ First, location is everything. Condos tend to be in extremely
attractive locations relative to single-family homes.
ƒƒ Second, the condo stock is much newer and of a higher
quality than the existing single-family housing stock.
The fact that condos rent out for 25% to 50% more than
purpose-built rentals confirms this.
ƒƒ Third, condos offer amenities such as gyms, pools and
recreation rooms that do not figure into their square
footage.
ƒƒ Fourth, condo fees do not vanish into a sinkhole – they
mostly cover the maintenance costs that homeowners of all
stripes incur.14
ƒƒ Fifth, the cost per square foot of a dwelling usually declines
as its size increases. This makes sense: even the smallest
homes have kitchens and bathrooms, which are expensive
to build. Most of the extra square footage in larger homes
is due to relatively inexpensive family rooms, additional
bedrooms and renovated basements. Therefore, while
condos are smaller than the average stand-alone property,
they are not necessarily much cheaper to construct.
5) Are foreign buyers and investors a
source of vulnerability?
Some pundits worry that Canada’s housing market is unduly
exposed to a large number of foreign buyers and/or investors
(the two groups substantially overlap) who might suddenly flee
en masse, leaving a gaping hole. To the contrary, we actually
view these groups as a fairly stable source of demand.
Foreign buyers
There is little reliable data about the extent of foreign buying of
Canadian homes. Anecdotes tend to make the foreign fraction
seem quite high, but these likely exaggerate reality. A key
source of confusion is that it can be difficult on the surface
to distinguish “foreign” buyers – those who own Canadian
property, but do not live in the country – from immigrant buyers.
Immigrants now represent 40% of Vancouver’s population and
46% of Toronto's.
More credible estimates of foreign buyers put them at around
5% of the total demand. Of course, this varies by region and
sector. Foreigners may represent as much as 40% of the
Vancouver luxury property market, for instance.
Foreigners appear unlikely to retreat from the Canadian housing
market. One reason is that even “foreign” buyers usually have
fairly deep Canadian connections, either via visas, citizenship or
Note: Dynamic Sum of Apartment Starts scales the average construction time
steadily upward from 10 months to 20 months between 2000 and 2009, then
holds constant at 20 months thereafter. Source: Haver Analytics, RBC GAM
0
20
40
60
80
100
120
140
1974 1978 1982 1986 1990 1994 1998 2002 2006 2010 2014
Number of Apartments (Thousands)
Apartments Under Construction
10m Sum of Apartment Starts
Dynamic Sum of Apartment Starts
Gap
explained by
lengthier
construction time
Exhibit 22: …Condo construction glut vanishes upon close
examination

family – the latter frequently in the form of children or spouses
living in the very homes they have purchased (even though they
themselves do not spend most of the year there). Frequently,
they aspire to move to Canada themselves or plan to retire
there. Moreover, in selecting Canada, they seek not so much an
appreciating asset as the freedom, education, health care, clean
air, natural beauty and cultural mosaic associated with
the country.
To the extent that a large fraction of foreign buyers come from
emerging markets, especially China, there are four other
considerations. First, residents of emerging market countries
continue to build wealth at a rapid rate, increasing the supply
of potentially well-to-do home buyers. Astonishingly, a recent
survey in the South China Morning Post found that half of
China’s millionaires plan to leave the country within five years.15
Second, these foreign buyers appear relatively inelastic in their
demand. They have shown little inclination to shift away from
the expensive Vancouver market and toward less expensive
destinations. In fact, many view the Vancouver housing market
as relatively cheap, given that other popular destinations are
the even pricier Hong Kong, Singapore, London and Sydney.
As further evidence of this inelasticity, swings in the Canadian
dollar have had little obvious effect on the appetite for Canadian
homes over the past several years. To the extent that they buy
homes with cash, the spectre of rising rates is irrelevant.
Third, Chinese investors tend to be skeptical about traditional
portfolio investments, such as stocks and bonds. This is not
entirely unreasonable given that the Chinese stock market is
dominated by state-owned enterprises with poor governance
track records, and the bond market continues to be weighed
down by repressed interest rates. Consequently, they view
housing as their primary investment vehicle. This attitude may
fade over time, but it is unlikely to vanish overnight.
Fourth, the recent crackdown on Chinese corruption could in
the short run push more Chinese money out of the country and
into Canada.
Of course, the outlook is not completely risk-free. One important
consideration is the regulatory landscape. An important point
of access for foreign buyers – Canada's Immigrant Investor
Program – was cancelled earlier this year due to perceptions
that it did not attract and encourage the sort of entrepreneurial
mindset that was intended.
This curb could reduce foreign demand in the short run, but
it is unlikely to evaporate altogether. Quebec has retained
a scaled-back investor program, which continues to provide
backdoor access to the rest of the country. Many foreigners gain
access via other immigration programs. And a new improved
immigrant investor pilot program is expected shortly. More
generally, Canada is in the process of substantially revising its
immigration rules, with the goal of attracting the same number
of immigrants, but better aligned with the economy’s needs. It is
hard to fathom this being a net negative for the housing market
over the long run.16
Condo investors
The bulk of investment purchases involves condos. There is no
single definitive figure for the share of condos held by investors,
but it appears to be fairly high. CMHC figures that investors
account for 17% of condo purchases. Builders estimate that the
figure is closer to 50% or 60%, though they are usually referring
to Toronto, where even CMHC's conservative figures find that
43% of newly completed Toronto condos are investor-bought.17
The question, of course, is whether the big contribution to sales
from investors is problematic. We believe it to be a fairly benign
trend, and historically quite normal. Whereas the single-home
market has always been dominated by homeowners (currently
91% are owner-occupied), only 35% of existing multi-unit
dwellings are owner-occupied. Thus, the introduction of new
condos that are 40% to 50% owner-occupied is actually nudging
the overall multi-unit share higher.
Healthy investor attitude
A CMHC survey of condo investors yields results that are
inconsistent with a sudden retreat (Exhibit 23):
ƒƒ Forty-two percent of condo investors are mortgage-free,
and just one-fifth put less than 20% down. Few are highly
levered investors, meaning that the pain of rising mortgage
rates shouldn’t be unusually problematic for them.
ƒƒ Only 8% of condo investors plan to sell their property
within two years, suggesting the vast majority are not
flippers looking for a quick buck. In fact, over half have
purchased their condo for rental income rather than the
prospect of capital gains.
Note: Based on survey conducted in 2013. Source: CMHC, RBC GAM
42%
8%
53%
0
10
20
30
40
50
60
70
80
Mortgage-Free Plan to Sell Within
Two Years
Own for Rental
Income
Share of Condo Investors (%)
Most condo
investors not
highly
leveraged...
... few looking
to flip their
property...
... most hold for
rental income rather
than capital gains
Exhibit 23: Condo investors are a level-headed bunch

Economic Compass
11
ƒƒ Investors do not harbour unrealistic expectations of
riches. Just 48% of investors expect Toronto condo prices
to rise, and 37% expect higher prices in Vancouver over the
next year.
Investor returns
The available condo rental yield of between 1% and 5% (on
levered and unlevered investments alike) is hardly compelling.18
However, in the current environment of ultra-low bond yields,
such returns are frankly not a bad substitute for the bond
market. Rising rents and home prices should increase the
returns to existing unlevered investors over the long run.
Levered investors also enjoy these benefits, but these pluses
must be weighed against rising mortgage rates.
Additionally, many so-called investors (in that their condo is
not their primary residence) are not truly investors in the classic
sense. They do not care about rental yields or capital gains,
instead using the condo as a second home perhaps for holidays,
on weekends or for business trips.
Rental absorption
Despite rumours of high condo vacancy rates, the official figures
seem quite tame, averaging a 2% or lower condo rental vacancy
rate across major Canadian markets. This is comparable to
purpose-built rental buildings (Exhibit 24).
An alternate estimate from a CMHC survey finds that condos
held by investors have a 6.9% vacancy rate. Even this higher
number is miles below anecdotes of half-vacant buildings. The
gap between the two sets of figures may possibly be reconciled
by the “investors” who treat their condo as a second home and
leave it empty most of the time.
Investors are irrelevant
In the spring of 2014, CMHC stopped insuring mortgages on
second homes. This may dampen investor demand, but arguably
not by much since most investors make large enough down-payments
that CMHC was never part of the calculation.
But never mind that – fundamentally, investors just don’t
matter as much as they first seem to. Keep in mind that overall
residential construction is running approximately in line with
demographically supported demand. It doesn’t matter who buys
these new homes – people need them to live in. If individual
investors were to retreat, builders would likely shift construction
away from condos and toward rental buildings to meet the
underlying tenant demand.
6) Does the distribution of household debt
reveal additional problems?
Averages can conceal important information about where
the greatest household debt pressures and risks lie. More
granular data is thus useful for identifying the areas of greatest
vulnerability.
To provide an example of why the underlying distribution of
debt matters, it is not at all unusual for a young family in a major
Canadian city to have – and thrive – with a debt-to-income
ratio of 400%, far in excess of the 164% national average. They
manage this because they are at a point in their careers when
salaries often rise briskly, they likely enjoy the stability of two
incomes and they have a long period in the workforce ahead of
them. In contrast, it is concerning if a household on the cusp of
retirement has one-quarter that amount of debt. The extremes
matter, as does the context around them.
The underlying distribution of household debt is for the most
part reassuring. Canadian lenders and borrowers seem to be
fairly adept at gauging what constitutes a reasonable amount of
debt given a household’s specific circumstances and prospects.
This makes sense: banks wish to avoid lending to people who
are unlikely to repay the debt, and borrowers don’t wish to go
through the pain of a forced home sale, mortgage foreclosure
or bankruptcy.
Illustrating this, higher-income Canadians carry far more debt
on average than low-income households, and people with
higher credit scores have undertaken the bulk of household
borrowing in recent years (Exhibit 25).
Most Canadians have little debt
It is worth stepping back for a moment and recognizing just
how few Canadian households are heavily indebted (Exhibit
26). Almost a third are debt-free, another 42% have less than
$100,000 of debt and a mere 26% owe more than $100,000.
Framed differently, only 34% of Canadian households have
a mortgage, and the average balance among these is a tame
$155,000. Thus, most Canadians will be only minimally affected
by rising interest rates over the next several years.
Note: As of October 2013. Source: CMHC, RBC GAM
1.1
1.8
1.0
1.7 1.7
1.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Vancouver Toronto Calgary
Rental Vacancy Rate (%)
Condo
Purpose-built
Condo rental vacancy rates
are comparable to
purpose-built rentals
Exhibit 24: Rental vacancy rates are low

Note: Based on latest data available. Shaded area represents the range of
historical median home price-to-median income ratios of 228 metropolitan areas
in the U.S. sampled. Source: NAHB, Haver Analytics, RBC GAM
Note: As at 2014. Source: Ipsos Reid, RBC GAM
Source: Equifax Canada Inc., Bank of Canada, RBC GAM
-10
0
10
20
30
40
50
60
70
< 611 611 - 706 707 - 758 759 - 807 > 807
Share of Consumer Credit Growth,
Q3 2010 to Q3 2013 (%)
Credit Score
More loans to those
with high credit
scores
Fewer loans
to those with
low credit
scores
Exhibit 25: Rational borrower and lender behaviour limits risk
32%
25%
17%
14% 12%
0
5
10
15
20
25
30
35
Debt
Free
$5K to
<$25K
$25K to
<$100K
$100K to
<200K
>=$200K Distribution of Households
by Amount of Debt (%)
Only 26%
have $100K or
more of debt
Exhibit 26: Most Canadian households not too heavily indebted
0
2
4
6
8
10
Median Home Price-to-Median
Income Ratio by Metropolitan Area
Metropolitan area with
lowest ratio
Metropolitan area
with highest ratio
Exhibit 27: "Normal" housing affordability varies enormously
Geographic distribution
Some cities are certainly more vulnerable than others, though
it is not as simple as identifying where home prices are highest
on an absolute basis or relative to incomes. What matters is
how far from its localized “normal” each market is. Exhibit
27 demonstrates that the definition of normal across the U.S.
varies between home prices that cost less than two times annual
income in some regions to more than eight times in others. As
the distribution around each data point shows, home prices
encounter resistance when these costs depart significantly from
the local norm, whether or not the absolute level is high.
In this context, what is relevant is not that Vancouverites normally
spend 59% of their income servicing their mortgage, versus 40%
in Toronto. What matters is that the latest Vancouver reading is
slightly more elevated relative to its norm than Toronto.
Age-based distribution
The age-based distribution of debt is also important, as people
in their 30s and 40s are much better positioned to carry and
eventually pay down debt. Fortunately, the debt profile broadly
aligns with this ideal. The heaviest debt loads in Canada are
held by those who are early to mid-career – carrying a mortgage
and a car loan, and paying for childcare. In contrast, those just
starting out usually have somewhat less debt, and those in
retirement usually have significantly less debt – around one-third
the level of the peak age group.
Seniors nevertheless constitute a potential risk point for
Canadian household debt. Whereas in recent years the fraction
of households in debt has declined nicely for younger age
groups, there has been an increase among households headed
by a person 65 or older. The fraction of indebted seniors has
lately risen above 50%. And there is reason to think that when
the current middle-aged cohort reaches retirement age, it could
have even more debt to grapple with given the high prices
they have paid for their homes and the likely burden of rising
mortgage rates.
Evaluating the most vulnerable
Lastly, we cut to the chase by focusing on those households that
are most at risk.
At present, there is no sign of serious distress: mortgage
arrears and credit card delinquency rates are low and declining,
signalling that even households at the most vulnerable end of
the spectrum are avoiding trouble (Exhibit 28).
However, the fraction of Canadians exposed to a dangerously
high debt-service ratio (defined in this instance as interest plus
principal) of at least 40% has been inching upwards over the
past several years, from 5.6% in 2007 to 5.9% more recently.
This isn’t an especially problematic level or increase by itself, but
it has happened at the same time that the fraction of households

Economic Compass
13
Note: Residential mortgage in arrears for 3+ months. Credit card delinquency rate
of 90+ days for VISA and MasterCard. Source: Canadian Bankers Association, RBC
GAM
Note: As at 2014. Debt-service ratio excludes credit card debt, includes principal
payments. Source: Ipsos Reid, RBC GAM
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
1990 1994 1998 2002 2006 2010 2014
Credit Card Delinquency Rate (%)
% of Mortgages in Arrears
Mortgages in Arrears (LHS)
Credit Card Delinquency Rate (RHS)
Improving
Exhibit 28: Little evidence of borrower stress
98%
27%
5.6%
5.9%
4
5
6
7
8
0
20
40
60
80
100
120
% of Households with
Mortgages with 4%
or Higher Rate (LHS)
% of Indebted Households
Spending 40% or More of
Disposable Income Servicing
Debt (RHS)
%
%
2007 2014
Collapse in
mortgage
rates...
… yet fraction of
households in precarious
debt-servicing position has
edged higher
Exhibit 29: Some households vulnerable to higher rates
Residential construction/
Renovations/Transactions
Housing
wealth effect
BASE CASE
BEARISH
SCENARIO
-0.5ppt -0.75ppt
-1.75ppt -2.25ppt
Overall
-1.25ppt
-4.0ppt
Exhibit 30: Economic implications
Source: RBC GAM
paying a 4.00% mortgage rate or higher has plummeted from
98% to just 27% (Exhibit 29). As mortgage rates rise in the
future, the share of households that are extremely vulnerable will
rise to higher-than-normal levels. Higher mortgage and credit-card
delinquency rates should follow.
Conclusion
The six key questions posed in this report return quite a mixed
interpretation of Canada’s housing market (summarized in the
Exhibit 1 scorecard). Broadly, the near-term outlook appears
benign, tilting only slightly in a negative direction.
On the other hand, the medium-term outlook is distinctly
negative. The coming drags may not be quite as large as many
imagine: household debt levels are less extreme than they look,
housing affordability will be poor but not atrocious, construction
should slow but not collapse, investors and foreign buyers are
unlikely to flee en masse, and only a small fraction of Canadian
households are quite vulnerable to rising rates. Nevertheless,
the housing market is set to soften over the coming years,
mainly as affordability deteriorates.
Base case forecast
The economic implications of a housing slowdown over the next
five years can be divided into construction and affordability
components (Exhibit 30).
Our base case forecast is that construction activity will impose
a mere 0.5 percentage point drag on the economy, spread over
the next five years because building activity need only decline
slightly over this period to remain aligned with demand (we also
assume a moderate decline in renovations). However, because
higher rates are set to significantly worsen affordability, a
diminished home-price trajectory should curtail spending (via
the so-called “wealth effect”) by a cumulative 0.75 percentage
point off GDP growth.19
Combined, then, we imagine that the economy could expand
by a cumulative 1.25 percentage points less than otherwise over
the next five years. This is a noticeable but not crippling burden,
amounting to 0.25 percentage point less growth per year (refer
to Appendix A for a more bearish scenario). This drag should be
distributed fairly broadly, affecting builders (via diminished
new construction and fewer renovations), slowing consumption
and dimming government revenues (via diminished land-transfer
fees, lower builder fees and slower growth in the
property tax base).
Overall, though, other factors look set to play a more central role
in the Canadian economic outlook, among them lower oil prices
(bad), a weaker Canadian dollar (good) and a stronger U.S.
economy (good).

A significantly worse scenario is conceivable, but
improbable. A key reason is that potent housing corrections
usually require two active ingredients (Exhibit 31). Rising
rates and/or diminishing credit availability usually provide
the first ingredient. What is needed is the addition of a
second ingredient, usually a spike in unemployment.20
Rising rates are quite likely in the coming years, but the
probability of a large increase in unemployment seems
fairly low given our economic forecasts. Furthermore, their
collective probability – the odds of both triggering at the
same time – are even lower, especially since it is hard to
fathom the Bank of Canada pushing interest rates higher if
the labour market goes into a tailspin.
Despite the low odds, it is nevertheless worth
contemplating an adverse scenario given the high stakes.
We do this with the help of four analytic approaches.
A) Scorecard stress testing
The first approach mechanically revisits the scorecard
of Exhibit 1 and forces a more negative conclusion by
downgrading the outlook on a sliding scale according to our
relative confidence in each of the key conclusions reached
in this report.21
This transformation results in a slight negative near-term
outlook and a major negative medium-term forecast. Thus,
a nastier housing correction is theoretically conceivable
given the uncertainty that exists around our base case
forecast.
APPENDIX A: BEARISH SCENARIO
Individual probability
Collective probability of
severe housing downturn
Fairly high Fairly low
Low
Materially
higher
rates
Materially
higher
unemployment
Exhibit 31: Housing crisis requires two ingredients
Source: RBC GAM
5%
23%
72%
< 10%
10%–24.9%
≥ 25%
Home Equity Level of
Mortgage Holders
Exhibit 32: Decent home equity levels in Canada
Note: Percentage of mortgage holders with different amounts of home
equity. May not add to 100% due to rounding. Source: Looking for a "New
Normal" in the Residential Mortgage Market, May 2014. CAAMP. RBC GAM
B) GDP stress testing
The second approach is depicted as the “bearish scenario”
in Exhibit 30. It ventures beyond the base case forecast by
imagining that construction plummets to 125,000 units
annualized and that home prices fall by 25% over the next
several years (instead of flat-lining in the base case – which
itself represents a 15 percentage point undershoot of the
normal upward trend). These adjustments cause the net
economic drag to explode from just 1.25 percentage points to a
hefty 4.0 percentage points. This would eliminate almost half of
Canada’s economic growth over the next five years.
C) Venn diagram stress testing
Approach C differs from Approach B in that it evaluates the
specific burdens that could befall the credit market, as opposed
to merely calculating an aggregate economic impact.
In the event of an across-the-board 25% home-price correction,
28% of mortgage holders would find themselves without any
equity in their home (Exhibit 32). This is an important group,
because the recent U.S. experience demonstrates that they are
about 30% more likely to default on their mortgage than other
mortgage holders.
Of course, Canada has far fewer “no recourse” mortgages than
the U.S., meaning it is more difficult for borrowers to abandon a
home and its associated mortgage at the first sign of trouble.
Furthermore, the vast majority of Canadians would continue
to be gainfully employed in even the most adverse economic
scenario. Where the risk lies is in the resulting Venn diagram

Economic Compass
15
overlap of households with negative equity and households
suffering a job loss.
Canada’s mortgage delinquency rate would spike from its
current reading of just 0.3% to around 1.3% in the event of a
three-percentage-point unemployment rate increase.22 This
represents more than a quadrupling, and would be about twice
as bad as the worst reading since the data series began in
1990. On the other hand, it would be a far cry from the worst
of the U.S. experience, where mortgage delinquencies peaked
at an unfathomable 11% for single-family homes. Replicating
the U.S. experience would require several additional adverse
triggers, including bank capital shortfalls and an unwillingness
by lenders to roll mortgages.
D) Third-party stress testing
Lastly, the Bank of Canada stress tests Canadian household
debt as part of its semi-annual Financial System Review.
The central bank estimates that the combination of a
220-basis-point increase in mortgage rates (which seems
plausible if a bit aggressive) and the aforementioned three-percentage-
point increase in the unemployment rate would
result in past-due mortgages rising by around 0.8 percentage
point – very close to our own conclusion in Approach C. The
fraction of households paying 40% or more of their income for
debt servicing would rise from a moderate 6% to a high 8%,
signalling materially increased distress among households.
Overall, a bear case scenario seems quite unlikely, but
would create serious problems for the economy and credit
market were it to transpire, if nowhere near the scale of the U.S.
housing bust.

Notes:
1 We believe the best further steps would be to cap the debt-to-income ratio at a lower level and to test it against historically normal mortgage rates.
2 Carrying cost affordability measures are also not perfect. For instance, our measures compare the average income to the average monthly mortgage payment. It might
be preferable to examine the median measures, since high incomes at the absolute top of the income spectrum likely distort the average income. It is true that the
average home price may also be distorted higher, providing some amount of offset, but higher income households generally spend a smaller share of their money on
housing.
3 Our forecast for “normal” mortgage rates assumes a normal government 10-year yield of around 3.75% (as articulated in our November 2013 Economic Compass
entitled “Estimating a Normal Yield”).
4 70% of Canadian mortgages are currently fixed rate, though fewer HELOCs are. When the two are combined, perhaps 60% of home loans are for a fixed rate.
5 Additionally, even as mortgage rates rise, some homeowners renewing their mortgage will find that their new rate is nevertheless lower than it was five years before
when they previously locked in.
6 Looking forward, although anecdotally there is an increased inclination for young people to return to the family nest after school or to live with roommates (thus
reducing household formation rates), a large chunk of this is cyclical, not permanent. Moreover, CMHC projects that the number of single-person households and couples
without children will grow markedly in the coming years, versus a decline in couples with children.
7 Could construction therefore be running too low, rather than too high? We wouldn’t want to push that notion too aggressively (as the housing stock numbers will soon
explain).
8 One plausible reason for this is that homes prices have increased too much (as discussed in the affordability section), sending the dollar value of residential investment
higher. To consider it a further point of vulnerability here would be double counting.
9 Estimating demand requires first converting age-based population figures into an approximation of the number of households per age cohort, and then summing the
total across age groups. The link between the two is only formalized via census data once every five years, requiring interpolation and extrapolation for the rest. This
is then stretched to reflect the fact that each household in Canada normally possesses an average of 1.06 homes (seasonal properties and vacancies reflecting the
excess). Estimating the net level of the housing stock requires taking official estimates that were discontinued after the year 2000 and mapping them forward via housing
completions, minus the usual rate of teardowns (0.15%), plus an extra 12% assumed increase in the housing stock above and beyond completions that appears to
originate from a fraction of new homes being converted into multi-unit dwellings (such as renting out the basement) after construction is complete.
10 Multi-unit defined for this purpose as apartments plus row houses, but excluding semi-detached homes.
11 According to the 2011 census.
12 64% of eligible Americans 19 years and under had a driver’s license in 1998, versus just 46% in 2008.
13 After all, no one lives in the empty elevator shaft soaring into the sky.
14 Statistical agencies figure that the upkeep of a home requires reinvesting 1.5% to 2.0% of the home’s value each year.
15 Though it is hard to fathom that many actually doing so.
16 It is theoretically possible that Canada could impose targeted stamp taxes on foreign buyers, much as Hong Kong and Singapore have done. But it would arguably
be out of keeping with Canadian values to exclude one particular group, and any concern about excessive demand is more usefully addressed by curtailing investment
activity more generally, regardless of the origin of the investor.
17 CMHC finds that, overall, 26% of the Toronto condo stock is rented out.
18 There are additional subtle drivers that may render the estimated yield better than it looks. One is that rented condos tend to be disproportionately single-bedroom
units and located on lower (less valuable) floors. Thus, in comparing the average condo’s price to the average condo rental rate, there is a mismatch in the relative quality
of the average property under consideration for each.
19 A recent Bank of Canada paper ("Household Borrowing and Spending in Canada", 2012) found that home-equity extraction drove as much as 2% of Canadian
consumption in 2010. However, this equity extraction was quite stable despite fluctuating home prices. In turn, one should not assume that home equity extraction
would collapse altogether if home prices were to soften in the future. On a related note, a recent Canadian Association of Accredited Mortgage Professionals (CAAMP)
estimate figures that among homeowners who took equity out of their home over the past year, 32% went to consolidate their debt, 25% went to home maintenance and
renovations, and 24% went to investments. Only 19% went directly toward consumption. Thus, any wealth effect hit would be distributed in part into renovations and
other investment asset classes.
20 Recall, for instance, that the Toronto housing bubble of the late 1980s/early 1990s was popped by rising borrowing costs followed by a local unemployment rate that
came close to tripling.
21 To illustrate how this works, we leave a forecast unchanged if our confidence in it is “high,” we downgrade it by one notch if our confidence is “medium,” and by two
notches if our confidence is “low.” To illustrate, a one notch downgrade would entail a “neutral” outlook becoming “slight negative.” A two notch downgrade would take a
“slight negative” outlook to “major negative.”
22 On the assumption that half of the households with negative equity and a lost job would default on their mortgage.

Economic Compass
This report has been provided by RBC Global Asset Management Inc. (RBC GAM Inc.) for informational purposes only and may not
be reproduced, distributed or published without the written consent of RBC GAM Inc. In the United States, this report is provided
by RBC Global Asset Management (U.S.) Inc., a federally registered investment adviser founded in 1983. In Europe and the Middle
East, this report is provided by RBC Global Asset Management (UK) Limited, which is authorised and regulated by the Financial
Conduct Authority. RBC Global Asset Management (RBC GAM) is the asset management division of Royal Bank of Canada (RBC)
which includes RBC Global Asset Management Inc., RBC Global Asset Management (U.S.) Inc., RBC Global Asset Management
(UK) Limited, RBC Alternative Asset Management Inc., and BlueBay Asset Management LLP, which are separate, but affiliated
corporate entities.
This report is not intended to provide legal, accounting, tax, investment, financial or other advice and such information should
not be relied upon for providing such advice. RBC GAM takes reasonable steps to provide up-to-date, accurate and reliable
information, and believes the information to be so when printed. Due to the possibility of human and mechanical error as well as
other factors, including but not limited to technical or other inaccuracies or typographical errors or omissions, RBC GAM is not
responsible for any errors or omissions contained herein. RBC GAM reserves the right at any time and without notice to change,
amend or cease publication of the information.
Any investment and economic outlook information contained in this report has been compiled by RBC GAM from various sources.
Information obtained from third parties is believed to be reliable, but no representation or warranty, express or implied, is made
by RBC GAM, its affiliates or any other person as to its accuracy, completeness or correctness. RBC GAM and its affiliates assume
no responsibility for any errors or omissions.
All opinions and estimates contained in this report constitute our judgment as of the indicated date of the information, are
subject to change without notice and are provided in good faith but without legal responsibility. To the full extent permitted by
law, neither RBC GAM nor any of its affiliates nor any other person accepts any liability whatsoever for any direct or consequential
loss arising from any use of the outlook information contained herein. Interest rates and market conditions are subject to change.
A Note on Forward-Looking Statements
This report may contain forward-looking statements about future performance, strategies or prospects, and possible future
action. The words “may,” “could,” “should,” “would,” “suspect,” “outlook,” “believe,” “plan,” “anticipate,” “estimate,”
“expect,” “intend,” “forecast,” “objective” and similar expressions are intended to identify forward-looking statements.
Forward-looking statements are not guarantees of future performance. Forward-looking statements involve inherent risks and
uncertainties about general economic factors, so it is possible that predictions, forecasts, projections and other forward-looking
statements will not be achieved. We caution you not to place undue reliance on these statements as a number of important
factors could cause actual events or results to differ materially from those expressed or implied in any forward-looking statement
made. These factors include, but are not limited to, general economic, political and market factors in Canada, the United States
and internationally, interest and foreign exchange rates, global equity and capital markets, business competition, technological
changes, changes in laws and regulations, judicial or regulatory judgments, legal proceedings and catastrophic events. The
above list of important factors that may affect future results is not exhaustive. Before making any investment decisions, we
encourage you to consider these and other factors carefully. All opinions contained in forward-looking statements are subject to
change without notice and are provided in good faith but without legal responsibility.
® / TM Trademark(s) of Royal Bank of Canada. Used under licence.
© RBC Global Asset Management Inc. 2014
EC (33/2014)/E

RBC Global Asset Management: Surprisingly Sustainable Canadian Housing

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to RBC Global Asset Management: Surprisingly Sustainable Canadian Housing

Similar to RBC Global Asset Management: Surprisingly Sustainable Canadian Housing (20)

Recently uploaded

Recently uploaded (20)

RBC Global Asset Management: Surprisingly Sustainable Canadian Housing