## _wp09235

## Source details

**Canonical URL:** [_wp09235](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09235.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09235.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09235.pdf.json)

---

### Overview and scope
- Focus: Fundamental determinants of house price developments in five large Canadian provinces (Alberta, British Columbia, Ontario, Quebec, and Saskatchewan).
- Data and sample period: Quarterly regional data for the period 1993Q1-2009Q2, including disposable income, demographic developments (accounting for inter-provincial and international migration trends), and mortgage credit.
- Methodology: Econometric long-run (cointegration) relationship between house prices and determinants, embedded in an error-correction mechanism to compute equilibrium prices as determined by demand and supply factors.

### Key empirical findings
- Aggregate price movements:
  - House prices rose by around 60 percent in nominal terms (45 percent in real terms) from 2003 to their end-2007 peak.
  - House prices fell by around 10 percent (both in real and nominal terms) by early 2009.
- Regional valuation assessments (model estimates, various dates):
  - At the peak of the housing boom, house prices were significantly overvalued in the west.
  - As of end-2009Q2:
    - Alberta and British Columbia remain slightly overvalued at around 8 percent according to the model estimation.
    - Ontario, Quebec, and Saskatchewan are now close to equilibrium.
  - Peak overvaluation in Alberta and British Columbia was estimated at double-digit levels during the housing boom.
- Section VI — peak and later deviations (using CREA existing-home prices deflated by each province’s CPI, 1993Q1-2009Q2):
  - At the peak of house prices (by 2007), prices were around 25 percent above equilibrium in Alberta.
  - At peak, prices were around 17 percent above equilibrium in Saskatchewan and British Columbia.
  - By late 2007/early 2008 declines reduced overvaluation to around 8 percent for British Columbia and Alberta, and 5 percent for Saskatchewan.
  - Ontario experienced some price overvaluation in 2007, but prices now appear to be in line with equilibrium.
  - Quebec appeared to have weathered the housing boom without any overvaluation.
- Historical context:
  - The large run-up in house prices in the west in the 2000s largely reflects a catch-up from stubbornly undervalued levels following the housing collapse in the early 1990s.
- Overall conclusion:
  - While western provinces showed some overvaluation, aggregate Canadian house prices are close to equilibrium—confirming results from Tsounta (2009) and IMF (2004, 2008, 2009).

### Determinants and mechanisms analyzed
- Demand-side variables: Disposable income, demographic developments (including inter-provincial and international migration), mortgage credit.
- Supply-side variables: Factors influencing the available housing stock (as captured in the long-run model specification).
- Modeling approach: Long-run cointegration relationship between house prices and determinants, with short-run dynamics handled via an error-correction mechanism.

### Empirical model results (select coefficients and statistics, 1993-2009Q2)
- Dependent Variable: Real House Price (growth)
- Lagged dependent variable — Lagged real house price (growth) by province:
  - Alberta: 0.47 [3.47]
  - British Columbia: 0.02 [0.15]
  - Ontario: -0.37 [-3.88]
  - Quebec: -0.60 [-5.90]
  - Saskatchewan: 0.08 [0.58]
- Reversion — Error correction coefficient by province:
  - Alberta: -0.04 [-0.94]
  - British Columbia: 0.04 [0.78]
  - Ontario: -0.23 [-6.70]
  - Quebec: 0.06 [5.09]
  - Saskatchewan: -0.08 [-0.94]
- Fundamentals — Selected elasticities and t-statistics:
  - Real earnings (per capita, growth):
    - Alberta: 0.57 [2.06]
    - British Columbia: 1.84 [3.89]
    - Ontario: 1.50 [4.14]
    - Quebec: -0.20 [-0.66]
    - Saskatchewan: 1.17 [2.67]
  - Real credit (growth): 0.10 [2.02]; -0.01 [-0.12]; -0.13 [-2.14] (coefficients and corresponding T-statistics listed in table)
  - Population (growth): -3.44 [-0.93]; -19.03 [-3.62]; 4.64 [1.74] (coefficients and corresponding T-statistics listed in table)

### Key statistics from tables and figures
- Table 1 — Real Prices of New Houses and Land, 1995-2009Q2 (percentage change, year over year)
  - Houses and Land:
    - 1995-2005: 0.3 1.3 0.6 2.0 -2.0 1.2
    - 2006Q4: 9.7 3.2 2.4 40.7 5.9 10.7
    - 2007Q4: 3.7 2.3 0.5 7.7 4.4 33.2
    - 2008Q4: -1.0 3.3 0.5 -7.3 -3.5 7.9
    - 2009Q1: -2.9 3.4 -0.6 -10.8 -6.9 0.7
    - 2009Q2: -3.2 3.1 -0.9 -9.2 -8.5 -5.2
  - New Houses Only:
    - 1995-2005: 0.9 1.5 1.5 2.6 -2.5 1.4
    - 2006Q4: 10.0 3.0 3.2 38.5 9.6 10.2
    - 2007Q4: 3.4 2.4 1.0 5.0 4.3 37.8
    - 2008Q4: -2.0 2.7 1.0 -11.5 -5.3 5.7
    - 2009Q1: -4.0 2.8 -0.5 -14.0 -10.9 -2.6
    - 2009Q2: -4.4 2.7 -1.2 -10.9 -14.5 -8.3
  - Land Only:
    - 1995-2005: -0.9 0.9 -1.2 0.7 -1.3 0.4
    - 2006Q4: 9.0 3.9 0.8 44.9 -0.2 12.5
    - 2007Q4: 4.4 2.0 -0.6 13.7 4.8 17.8
    - 2008Q4: 1.2 5.3 -0.4 2.0 -1.1 15.5
    - 2009Q1: -0.6 5.6 -0.9 -4.1 -0.9 13.0
    - 2009Q2: -1.1 4.7 -0.3 -5.9 0.1 6.2
- Figures (selected captions/notes):
  - Figure 1: Cummulative Increase in Resale House Prices (End 2002-Peak, in percent) — Sources: Canadian Real Estate Association, and author's calculations.
  - Figure 2: G-7: Price-to-income and price-to-rent ratios, 1970-2009 — Source: OECD; and author's calculations.
  - Figure 3 and Figure 4: Over/Undervaluation series, 1993Q1-2009Q2, for select eastern and western provinces (in percent).

### Policy actions taken by authorities (measures and microprudential steps)
- Monetary and liquidity measures:
  - Purchase of mortgage-backed securities.
  - Expansion of the Canada Mortgage Bond program to 10-year (from five-year) maturity.
  - Enhanced liquidity measures and lowered mortgage rates.
  - Announcement of a temporary home-renovation tax credit.
- CMHC microprudential and market-structure measures:
  - Lowered the maximum amortization period for new government-backed mortgages to 35 years.
  - Required a minimum down-payment of five per cent for new government-backed mortgages.
  - Established a consistent minimum credit score requirement.
  - Introduced new loan documentation standards.
- Effect: These measures limited the risk of relaxation of lending standards and constrained Canada’s exposure to the subprime market.

### Implications and outlook
- Household and financial-sector implications:
  - House is the largest single asset for most households; valuation changes have important consequences for household balance sheets and spending behavior.
  - Mortgages and other real-estate related assets represent an important component of financial institutions’ balance sheet—almost half of chartered banks’ loans—implying housing market developments could affect financial system health, profitability, and soundness.
- Macro and policy implications:
  - House prices affect the consumer price index, thereby influencing inflationary trends and expectations and having implications for monetary policy in its role to preserve price and financial stability.
  - Revenues from real estate transactions (including construction-related income and excise taxes) affect fiscal position.
- Reasons why a large adverse spillover to the rest of Canada is not expected:
  - (i) The degree of misalignment is small.
  - (ii) The housing market was already showing some encouraging signs of revival.
  - (iii) Commodity prices, closely linked with earnings performance in the west, had risen considerably from their lows in late 2008.
- Given these factors, an abrupt correction in Alberta and British Columbia—which together account for almost a third of Canada’s GDP and a fourth of Canada’s employment—is not expected to produce large negative nationwide effects.

*Source: _wp09235 - References (PDF).*

### References..............................................................................................................

### _wp09235 - References

### Overview and scope
- Focus: Fundamental determinants of house price developments in five large Canadian provinces (Alberta, British Columbia, Ontario, Quebec, and Saskatchewan).
- Data and sample period: Quarterly regional data for the period 1993Q1-2009Q2, including disposable income, demographic developments (accounting for inter-provincial and international migration trends), and mortgage credit.
- Methodology: Econometric long-run (cointegration) relationship between house prices and determinants, embedded in an error-correction mechanism to compute equilibrium prices as determined by demand and supply factors.

### Key empirical findings
- Aggregate price movements:
  - House prices rose by around 60 percent in nominal terms (45 percent in real terms) from 2003 to their end-2007 peak.
  - House prices fell by around 10 percent (both in real and nominal terms) by early 2009.
- Regional valuation assessments:
  - At the peak of the housing boom, house prices were significantly overvalued in the west.
  - As of end-2009Q2:
    - Alberta and British Columbia remain slightly overvalued at around 8 percent according to the model estimation.
    - Ontario, Quebec, and Saskatchewan are now close to equilibrium.
  - Peak overvaluation in Alberta and British Columbia was estimated at double-digit levels during the housing boom.
- Historical context:
  - The large run-up in house prices in the west in the 2000s largely reflects a catch-up from stubbornly undervalued levels following the housing collapse in the early 1990s.

### Determinants and mechanisms analyzed
- Demand-side variables: Disposable income, demographic developments (including inter-provincial and international migration), mortgage credit.
- Supply-side variables: Factors influencing the available housing stock (as captured in the long-run model specification).
- Modeling approach: Long-run cointegration relationship between house prices and determinants, with short-run dynamics handled via an error-correction mechanism.

### Implications highlighted
- Households’ behavior:
  - House is the largest single asset for most households; valuation changes have important consequences for household balance sheets and spending behavior.
- Financial soundness:
  - Mortgages and other real-estate related assets represent an important component of financial institutions’ balance sheet—almost half of chartered banks’ loans—implying housing market developments could affect financial system health, profitability, and soundness.
- Policy implications:
  - House prices affect the consumer price index, thereby influencing inflationary trends and expectations and having implications for monetary policy in its role to preserve price and financial stability.
  - Revenues from real estate transactions (including construction-related income and excise taxes) affect fiscal position.

### Structure of the paper (as indicated)
- Section II: Recent housing market developments in Canada.
- Section III: Significance of housing developments from macroeconomic and financial sector perspectives.
- Section IV: Related literature review.
- Section V: Econometric model description and results.

*Source: _wp09235 - References*

### Section VI concludes this paper.

### Section VI concludes this paper.

### Summary of main findings
- Home prices rose significantly in Canada in the late 2000s, especially in the western provinces, and then declined markedly.
- Using an error-correction model with CREA existing-home prices (deflated by each province’s CPI, 1993Q1-2009Q2), the paper finds:
  - At the peak of house prices (by 2007), prices were around 25 percent above equilibrium in Alberta.
  - At peak, prices were around 17 percent above equilibrium in Saskatchewan and British Columbia.
  - By late 2007/early 2008 declines reduced overvaluation to around 8 percent for British Columbia and Alberta, and 5 percent for Saskatchewan.
- For eastern provinces:
  - Ontario experienced some price overvaluation in 2007, but prices now appear to be in line with equilibrium.
  - Quebec appeared to have weathered the housing boom without any overvaluation.
- Overall conclusion: while western provinces showed some overvaluation, aggregate Canadian house prices are close to equilibrium—confirming results from Tsounta (2009) and IMF (2004, 2008, 2009).

### Interpretation and drivers
- Part of the late-2000s increase in real house prices in the west reflects a return to equilibrium following an “underreaction” of house prices during the 1990s housing collapse.
- More recently (pre-2007 peak), marked increases led real house prices to overshoot equilibrium levels, but these overshoots were transient deviations rather than sustained speculative bubbles.
- Given the subdued degree of overvaluation in the west, large negative spillovers to the rest of Canada are not expected.

### Policy actions taken by authorities
- Measures implemented to address housing market concerns included:
  - Purchase of mortgage-backed securities.
  - Expansion of the Canada Mortgage Bond program to 10-year (from five-year) maturity.
  - Enhanced liquidity measures and lowered mortgage rates.
  - Announcement of a temporary home-renovation tax credit.
- Microprudential and market-structure measures by CMHC:
  - Lowered the maximum amortization period for new government-backed mortgages to 35 years.
  - Required a minimum down-payment of five per cent for new government-backed mortgages.
  - Established a consistent minimum credit score requirement.
  - Introduced new loan documentation standards.
- These measures limited the risk of relaxation of lending standards and constrained Canada’s exposure to the subprime market.

### Implications and outlook
- Reasons why a large adverse spillover to the rest of Canada is not expected:
  - (i) The degree of misalignment is small.
  - (ii) The housing market was already showing some encouraging signs of revival.
  - (iii) Commodity prices, closely linked with earnings performance in the west, had risen considerably from their lows in late 2008.
- Given these factors, an abrupt correction in Alberta and British Columbia—which together account for almost a third of Canada’s GDP and a fourth of Canada’s employment—is not expected to produce large negative nationwide effects.

*Source: Section VI concludes this paper.*

### References

### _wp09235 - References

### References
- Allen, J., R. Amano, D. R. Byrne, and A.W. Gregory (2006), “Canadian City Housing Prices and Urban Market Segmentation,” Working Paper 2006-49, Bank of Canada.
- Cunningham, R. and I. Kolet (2007), “Housing Market Cycles and Duration Dependence in the United States and Canada,” Working Paper 2007-2, Bank of Canada.
- Gallin, J. (2003), “The Long-run Relationship Between House Prices and Incomes: Evidence from Local Housing Markets,” Finance and Economics Discussion Series, No. 2003-17, Board of Governors of the Federal Reserve System.
- Girouard, N., M. Kennedy, P. van den Noord, and C. André (2006), “Recent House Price Developments: The Role of Fundamentals,” Economics Department Working Papers, No. 475, OECD Publishing.
- Goldman Sachs (2008a), “Houses in the Industrial Countries: Built on Rock, Or on Sand?” Global Economics Weekly No. 08/18, May.
- Goldman Sachs (2008b), Global Economics Weekly, April.
- Gurkaynak, R. (2005), “Econometric Tests of Asset Price Bubbles: Taking Stock,” Finance and Economics Discussion Series, No. 2005-04, Board of Governors of the Federal Reserve System.
- IMF (2009), World Economic Outlook: Crisis and Recovery, April, Washington, DC: International Monetary Fund.
- IMF (2008), World Economic Outlook: Housing and the Business Cycle, April, Washington, DC: International Monetary Fund.
- IMF (2004), World Economic Outlook: The Global Demographic Transition, September, International Monetary Fund.
- Kiff J. (2009), “Canadian Residential Mortgage Markets: Boring But Effective?” Working Paper No. 09/130, Washington, DC: International Monetary Fund.
- Klyuev, V. (2008), “What Goes Up Must Come Down? House Price Dynamics in the United States,” Working Paper No. 08/187, Washington, DC: International Monetary Fund.
- Malpezzi, S. (1999), “A Simple Error Correction Model of House Prices,” Journal of Housing Economics, Vol. 8, pp. 27–62.
- Merrill Lynch (2008), “The Tipping Point?” Canadian Economic Commentary, September.
- Somerville, S. and K. Swann (2008), “Are Canadian Housing Markets Over-priced?” Working Paper 2008-01, Center for Urban Economics and Real Estate, University of British Columbia.
- Sutton, G. (2002), “Explaining Changes in House Prices,” Quarterly Review, September, Bank for International Settlement.
- Tomura, H. (2008), “A Model of Housing Boom and Bust in a Small Open Economy,” Working Paper, 2008-9, Bank of Canada.
- Tsounta, E. (2009), ”Is the Canadian Housing Market Overvalued? A Tale of Two Regions,” Canada―Selected Issues, Country Report No. 09/163, Washington, DC: International Monetary Fund.

### Figures (captions and notes)
- Figure 1. Canada: Cummulative Increase in Resale House Prices (End 2002-Peak, in percent)
  - Sources: Canadian Real Estate Association, and author's calculations.
  - Labels appearing in chart text: 020406080100120; Nominal Prices; Real Prices.
  - Provinces listed in chart text: Saskatchewan; Alberta; Manitoba; British Columbia; Newfoundland; Canada; Quebec; New Brunswick; Nova Scotia; Ontario; Prince Edward Island; Newfoundland and Labrador.
- Figure 2. G-7: Price-to-income and price-to-rent ratios, 1970-2009
  - Source: OECD; and author's calculations.
  - Chart labels: 40 60 80 100 120 140 160 180 19701973197619791982198519881991199419972000200320062009
  - Country series labels: USA; JPN; DEU; FRA; ITA; GBR; CAN.
  - Notes: "Price to income ratio (Sample average=100)"; "Price to rent ratio (Sample average=100)".
- Figure 3. Select Eastern Canadian Provinces: House Price Over/Undervaluation, 1993Q1-2009Q2 (In percent)
  - Source: Author's calculations.
  - Chart axis labels and series examples: -5 -4 -3 -2 -1 0 1 2 3 4 5 1993Q11996Q11999Q12002Q12005Q12008Q1 (Quebec); -10 -8 -6 -4 -2 0 2 4 6 8 10 12 1993Q11996Q11999Q12002Q12005Q12008Q1 (Ontario).
- Figure 4. Select Western Canadian Provinces: House Price Over/Undervaluation, 1993Q1-2009Q2 (In percent)
  - Sources: Author's calculations.
  - Chart axis labels and series examples: -30 -25 -20 -15 -10 -5 0 5 10 15 20 1993Q11996Q11999Q12002Q12005Q12008Q1 (Saskatchewan); -15 -10 -5 0 5 10 15 20 25 30 1993Q11996Q11999Q12002Q12005Q12008Q1 (Alberta); -15 -10 -5 0 5 10 15 20 25 1993Q11996Q11999Q12002Q12005Q12008Q1 (British Columbia).

### Table 1: Real Prices of New Houses and Land, 1995-2009Q2 (percentage change, year over year)
- Source: Haver Analytics; Fund staff calculations.
- Panel label: Canada Quebec Ontario Alberta British Columbia Saskatchewan
- Houses and Land
  - 1995-2005: 0.3 1.3 0.6 2.0 -2.0 1.2
  - 2006Q4: 9.7 3.2 2.4 40.7 5.9 10.7
  - 2007Q4: 3.7 2.3 0.5 7.7 4.4 33.2
  - 2008Q4: -1.0 3.3 0.5 -7.3 -3.5 7.9
  - 2009Q1: -2.9 3.4 -0.6 -10.8 -6.9 0.7
  - 2009Q2: -3.2 3.1 -0.9 -9.2 -8.5 -5.2
- New Houses Only
  - 1995-2005: 0.9 1.5 1.5 2.6 -2.5 1.4
  - 2006Q4: 10.0 3.0 3.2 38.5 9.6 10.2
  - 2007Q4: 3.4 2.4 1.0 5.0 4.3 37.8
  - 2008Q4: -2.0 2.7 1.0 -11.5 -5.3 5.7
  - 2009Q1: -4.0 2.8 -0.5 -14.0 -10.9 -2.6
  - 2009Q2: -4.4 2.7 -1.2 -10.9 -14.5 -8.3
- Land Only
  - 1995-2005: -0.9 0.9 -1.2 0.7 -1.3 0.4
  - 2006Q4: 9.0 3.9 0.8 44.9 -0.2 12.5
  - 2007Q4: 4.4 2.0 -0.6 13.7 4.8 17.8
  - 2008Q4: 1.2 5.3 -0.4 2.0 -1.1 15.5
  - 2009Q1: -0.6 5.6 -0.9 -4.1 -0.9 13.0
  - 2009Q2: -1.1 4.7 -0.3 -5.9 0.1 6.2

### Table 2. Determinants of House Prices in Select Canadian Provinces (Summary of Empirical Results, 1993-2009Q2)
- Source: Staff estimates.
- Note: T-statistics are listed within brackets.
- Dependent Variable: Real House Price (growth)
- Explanatory Variables and estimates by province (Alberta, British Columbia, Ontario, Quebec, Saskatchewan):
  - Lagged dependent variable
    - Lagged real house price (growth): 0.47 0.02 -0.37 -0.60 0.08
    - T-statistics: [3.47] [0.15] [-3.88] [-5.90] [0.58]
  - Reversion
    - Error correction coefficient: -0.04 0.04 -0.23 0.06 -0.08
    - T-statistics: [-0.94] [0.78] [-6.70] [5.09] [-0.94]
  - Fundamentals
    - Real earnings (per capita, growth): 0.57 1.84 1.50 -0.20 1.17
    - T-statistics: [2.06] [3.89] [4.14] [-0.66] [2.67]
    - Real credit (growth): 0.10 -0.01 -0.13
    - T-statistics: [2.02] [-0.12] [-2.14]
    - Population (growth): -3.44 -19.03 4.64
    - T-statistics: [-0.93] [-3.62] [1.74]

*Source: _wp09235 - References (PDF).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09235.pdf_