The Long-Term Evolution of House Prices: An International Perspective Canadian Association for Business Economics Kingston, Ontario 25 August 2015

Lawrence Schembri Deputy Governor

Bank of Canada

Outline

Introductory remarks:

Motivation: Housing is important; useful to provide context Approach: Examine house prices across time & countries (Advanced SOEs)

Long-run determinants of house prices:

Demand: Macroeconomic, demographic, and credit conditions, Supply: Regulation and geography

Policy implications:

Macroprudential policy: effectiveness & complementary role

Key messages

2

50

100

150

200

250

1975Q1 1980Q1 1985Q1 1990Q1 1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

Index

Spain United Kingdom Ireland Italy United States

b. Real house price index; post-crisis correction

Index: average = 100

50

100

150

200

250

1975Q1 1980Q1 1985Q1 1990Q1 1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

Index

Australia Canada Norway New Zealand Sweden

a. Real house price index; no post-crisis correction Index: average = 100

Real house prices have increased globally since 1995

3

Conceptual issues – multiple dimensions

Housing is both a consumption good (housing services) and an asset (store of wealth)

Owner-occupied housing versus renting

Single-family versus multiple unit

Composite good: structure and land

Existing versus new house

Focus: owner-occupied; existing singles & multiples; composite price

4

Demand Factors

5

Demand factors

Macroeconomic Rising disposable incomes; lower long-term real interest rates

Demographic Population growth; migration

Credit conditions Improvements due to financial liberalization and innovation

Other possible factors Macro-policy framework improvements, international investment,

preference shifts and regulatory/tax changes

6

80

100

120

140

160

180

200

1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

Index

Australia Canada Norway New Zealand Sweden

b. Real personal disposable income

Index: average = 100

80

100

120

140

160

180

1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

Index

Australia Canada Norway New Zealand Sweden

a. Nominal house price to income ratio

Index: average = 100

House prices have increased relative to incomes

7

0.0

0.2

0.4

0.6

0.8

%

Australia Canada Norway New Zealand Sweden

b. Net migration to total population Average annual as a per cent

0.0

0.5

1.0

1.5

%

Australia Canada Norway New Zealand Sweden

a. Total population growth Average annual growth rate

Population growth has contributed to house price increases, aided by net migration: 1975-94 and 1995-2014

8

0.0

0.1

0.2

0.3

0.4

0.5

%

Australia Canada Norway New Zealand Sweden

Urbanization increase: 1975-94 and 1995-2014 Average annual growth rate

Increasing urbanization is important

9

Credit conditions for housing finance have eased

10

Financial liberalization

Financial innovation

Lower long-term global interest rates

Lower mortgage rates

Broader and more efficient access to housing finance

A

F

F

O

R

D

A

B

I

L

I

T

Y

-2

0

2

4

6

8

10

1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

%

Australia Canada Norway New Zealand Sweden

Real 10-year government bond yields: 1995-2015

Real long-term interest rates have trended downward since 2000

11

60

65

70

75

80

85

90

1995 2000 2005 2010

%

Australia Canada Norway Sweden

b. Home-ownership rates

Canada

United States

Australia Sweden

Norway

United Kingdom

Netherlands

Denmark

Italy France

Spain

0.0

0.2

0.4

0.6

0.8

1.0

0 20 40 60 80 100

Index

Residential mortgage-debt-to-GDP ratio (2001-06 average, %)

a. IMF mortgage market index and ratio of

residential mortgage-debt-to-GDP

Housing finance index, mortgage market depth and home ownership rates

12

0

15

30

45

60

1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015

Securitization share

(%)

Private Securitization - United States

Public/Agency Securitization - United States

Private Securitization - Canada

Public /Agency Securitization - Canada

Total Securitization - Australia

b. Mortgage securitization outstanding as

a share of total mortgage credit

0

50

100

150

200

0

2

4

6

8

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

% %

Canada (left scale) New Zealand (left scale)

Australia (left scale) Norway (right scale)

Sweden (right scale) Denmark (right scale)

a. Covered mortgage bonds outstanding

as a share of total household credit

Mortgage funding by covered bonds and securitization has risen

13

Other factors supporting demand for houses

Demand for

houses

Macro & financial

policy frameworks

International investment

Shifting preferences

Other government

policies

14

Supply Factors

15

Regulatory and physical constraints limit the price responsiveness of housing supply

16

Regulation

Land use/zoning regulations

Greenbelts

Development fees

Geography

Bodies of water

Inhospitable terrain

0 3 6 9 12 15 18

Detroit

Houston

Chicago

Ottawa

Winnipeg

Las Vegas

Edmonton

Saskatoon

Calgary

Montreal

Tokyo

New York

Toronto

Los Angeles

London (GLA)

Melbourne

San Francisco

Sydney

Vancouver

Hong Kong

Median multiple

Median price-to-income ratio and land-use regulation: 2014

More-restrictive regulation

(International)

More-restrictive regulation

(Canada)

Less-restrictive regulation

(International)

Less-restrictive regulation

(Canada)

Land-use regulation and geographical constraints increase price-to-income ratio

17

House prices rise with increases in urban density: 1995-2014

18

Vancouver

Toronto

Montréal

Edmonton

Calgary

Winnipeg

Ottawa

Quebec City

100

125

150

175

200

225

250

-20 -10 0 10 20 30 40 50 60

Price change %

Change in population per square kilometre (%)

With population density increasing and binding supply constraints , multi-unit starts are dominating in Canada’s biggest cities

19

1600

6600

11600

16600

21600

1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

Single family house starts Multi-unit starts

a. Vancouver

2000

9000

16000

23000

30000

1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

Single family house starts Multi-unit starts

b. Montréal

4200

13200

22200

31200

40200

1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

Single family house starts Multi-unit starts

c. Toronto

9600

22600

35600

48600

61600

1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

Single family house starts Multi-unit starts

d. Rest of Canada

000s of units 000s of units

000s of units 000s of units

Policy Implications

20

Post-crisis monetary policy in advanced SOEs was effective

Countercyclical MP response to

decline in demand

Effective MP transmission:

credible IT framework &

resilient financial system

Lower real interest rates

boosted domestic demand

Stronger demand for houses and

rising house prices

21

Implications of higher house prices for financial stability

22

Two related vulnerabilities:

Associated with household indebtedness and leverage

Potential source of house price misalignment

0.2

0.4

0.6

0.8

1.0

1.2

1975Q1 1980Q1 1985Q1 1990Q1 1995Q1 2000Q1 2005Q1 2010Q1 2015Q1

Ratio

Australia Canada Norway New Zealand Sweden

Ratio of household-debt-to-GDP

Household leverage has grown, especially since 1995

23

Complementary role of macroprudential policy

Countercyclical monetary policy

increases demand for houses

Macroprudential policy restricts access

to credit to more qualified borrowers

Increased borrowing by the most

financially capable households

24

Macroprudential policy: Effectiveness

Evidence on the effectiveness of loan-to-value and debt-to-income constraints and other policies

Reduce procyclicality of household credit and bank leverage

Moderate credit growth

Improve quality of borrowers

Lower rate of house price appreciation

25

Macroprudential tightening in Canada: 2008-12

Maximum amortization (insured mortgages)

40 years 25 years

LTV limit for new mortgages

100% 95%

LTV limit for mortgage refinancing

95% 80%

LTV limit for investment properties

95% 80%

Debt-service criteria TDS capped at 45% GDS capped at 39% and TDS ratio at 44%

26

0

10

20

30

40

50

60

70

80

90

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

%

Under 600 600-659 660-699 700 and over

a. Distribution of credit scores at origination

(CMHC high-ratio mortgages)

0

5

10

15

20

2007 2008 2009 2010 2011 2012 2013 2014 2015

%

b. Mortgage credit

Annualized 3-month growth rates

Canadian post-crisis macroprudential policies contributed to higher borrower quality and lower household credit growth

27

Key messages

Real house prices have been rising relative to income in Canada and other comparable countries for about 20 years due to:

Demand: Demographic forces (population growth, migration), improving credit conditions, other factors

Supply: geography and regulation reduced supply elasticity, especially in urban areas

Macroprudential policy experience:

complementary to monetary policy and better suited to address financial vulnerabilities

28