## wpiea2020042-print-pdf

## Source details

**Canonical URL:** [wpiea2020042-print-pdf](https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020042-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2020/english/wpiea2020042-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2020/english/wpiea2020042-print-pdf.pdf.json)

---

### I. Introduction, motivation, and research questions
- House prices are central to financial stability and systemic risk monitoring; sharp housing-market deterioration can impair household balance sheets, business confidence, banks’ ability to lend, and produce cross-border spillovers.
- The collapse of the U.S. housing market in 2008 precipitated prolonged global financial turbulence and "the largest crisis since the Great Depression."
- Research questions addressed:
  - How do previously identified drivers help measure downside risks to future house price growth?
  - How do downside house price risks vary across countries and cities?
  - What is the role of indicators such as overvaluation, financial conditions, and capital inflows on house price risks?
  - What role does household indebtedness play in influencing house price risks?
  - How did household debt in the U.S. and Canada evolve over time and which types of households are more vulnerable to shocks?

### Data and empirical approach
- Sample: quarterly data from 37 cities in Canada and the U.S. between 1983 to 2018.
- Two-stage methodology:
  - Stage 1: Quantile regressions for each city to model nonlinear relationships between city- and country-level indicators and housing-market vulnerabilities (quantiles q = 0.1, 0.25, 0.5, 0.75, 0.9). Forecast horizons h include 4-quarter and 12-quarter ahead.
    - Dependent variable: future residential real house price growth ΔCityHP_{c,i,t+h}^q
    - Explanatory sets: city-level X_{c,i,t} (autoregressive term, house price-to-income, supply indicator), country-level Z_{c,t} (household debt, FDI, capital flows), OIL_t, FCI_{c,t}.
  - Stage 2: Fit a skewed-t distribution to the predicted quantiles. The lower fifth percentile of the fitted distribution is used as the house prices-at-risk (HaR) measure.
- Definitions and horizons:
  - House prices-at-risk: lower fifth percentile of the fitted skewed-t distribution.
  - Short-term horizon: 4 quarters ahead (one-year ahead).
  - Long-term horizon: 12 quarters ahead (three-year ahead).
- Data sources include S&P/Case & Shiller, FHFA, Statistics Canada, Real Property Solutions, Canadian Real Estate Association, U.S. National Financial Conditions Index, IMF BPM6 dataset, BIS, SCF (U.S.), SFS (Canada), WTI oil price index, and other sources.

### Main contributions
- Granular city-level HaR analysis: each city is estimated univariately to capture distinct sensitivities and heterogeneity that can be obscured at aggregate levels.
- Micro-level household debt analysis for the U.S. and Canada using household survey data to identify debt held by financially weak households and potential amplification effects.

### Main findings — high level
- Finding 1: Divergent trends in downside risks
  - Downside risks to house prices in the U.S. have declined over the past decade.
  - Downside risks to house prices in Canada have increased over the past decade.
  - Compared to historical levels:
    - The U.S. is significantly better off than prior to and during the GFC.
    - Canada is approaching GFC-level risks.
- Finding 2: Horizon-dependent determinants
  - Short-term (4 quarters ahead): tighter financial conditions are detrimental to house price risks.
  - Long-term (12 quarters ahead): the negative effect of tighter financial conditions is dampened.
  - Household leverage:
    - Associated with ameliorating short-term house price risks.
    - Associated with exacerbating long-term house price risks.
  - Capital flows (capital inflows) are significantly associated with future downside risks, with net effects varying by city and horizon; sensitivity correlates with foreign openness (immigration population).
  - City-specific valuation and supply factors are influential determinants.
- Finding 3: Household balance sheets and sectoral vulnerabilities
  - U.S.: overall household debt-to-GDP decreased in recent years; contraction in risky debt suggests lesser over-indebtedness.
  - Canada: household leverage increased, with a rise in the share of risky debt, indicating increased household-sector vulnerabilities.

### Quantitative snapshot and notable statistics (preserved exactly)
- Sample: 37 cities in Canada and the U.S. between 1983 to 2018.
- House prices-at-risk definition: lower fifth percentile of the fitted skewed-t distribution.
- Short-term horizon: 4 quarters ahead.
- Long-term horizon: 12 quarters ahead.
- Canada: about 30 percent overvalued as of 2018Q3 (aggregate valuation).
- Canada household debt: about 98 percent of GDP at end-2018.
- Canada mortgage debt: about 65 percent of total household debt in Canada as of 2019.
- Canada HELOCs: $173 billion in 2018Q4, about 10 percent of household debt.
- IMF (2019b) estimate: household debt service in Canada will increase by 0.4 percentage points to 15.3 percent of disposable income over the following year.
- U.S. household debt: about 80 percent of GDP at end-2018.
- U.S. mortgage debt: about 70 percent of total household debt as of 2019.
- San Francisco house price-to-income ratio: around 220 percent as of 2018 (city-level).
- City short-term HaR (5 percent, 4q-ahead) at end-2018Q3:
  - U.S. aggregate: about -1 percent (annualized).
  - U.S. city extremes (4q): Los Angeles and San Francisco each facing a 4-percentage point decline within the following four quarters.
  - Canada aggregate (4q): about -3 percent at end-2018Q3.
  - Canada city-level extremes at end-2018Q3: Toronto -10.8 percent tail risk; Vancouver -6.5 percent tail risk.

### Drivers of House Prices-at-Risk — empirical highlights
- Valuation (price-to-income)
  - U.S. aggregate: a one standard deviation growth in the price-to-income ratio → 2.9 percentage point deterioration in downside risks.
  - Canada aggregate: a one standard deviation increase in the price-to-income ratio → 1.5 percentage point increase in downside risks.
- Supply-side indicators
  - U.S.: a one standard deviation growth in housing starts → 2.3 percentage point decrease in downside risks (aggregate).
  - Canada: residential investment has heterogeneous effects; mitigates risks in some cities (Edmonton, Quebec City, Ottawa), not in Toronto, Vancouver, Hamilton.
- Financial conditions (FCI)
  - U.S.: ambiguous across cities/quantiles; in some metros (Austin, Cleveland, Seattle, San Francisco) a one standard deviation tightening → 1.7 to 4.8 percentage point increase in house price risks (short term); sensitivity reduced at 12-quarter horizon.
  - Canada: tighter FCI uniformly increases downside risks; most sensitive cities: Toronto, Ottawa, Calgary where a one standard deviation tightening → 1.6 to 3.1 percentage point increase in short-term downside risks; impact moderated at 12-quarter horizon.
- Capital flows
  - U.S.:
    - FDI (short-term): mitigates risks in some tech-friendly cities (Seattle, Portland) by about 1 to 2 percentage points (short term).
    - FDI (long-term): associated with around a 6 to 8 percentage points increase in house price risks in some cities (Tampa, Detroit, Minneapolis).
    - Other capital flows: generally increase downside risks across major U.S. cities; strongest in Sun Belt cities (Las Vegas, Los Angeles, Miami, San Diego).
  - Canada:
    - FDI: generally mitigates downside risks; a one percentage point increase in FDI inflow → reduction of about ½ to 1 percentage points in HaR in Calgary, Toronto, Ottawa, Quebec City (short term).
    - FDI (long-term): associated with 1.5 to 2.5 percentage point decrease in HaR in Quebec City, Toronto, Hamilton.
    - Other capital flows: predict significantly negative effects (increasing risks) in Vancouver, Ottawa, Hamilton in the short term; largely inconsequential long-term except Hamilton.
- Household debt and oil
  - Household debt: generally positive effect on short-term house price risks (credit eases near-term growth), but worsens housing-market risks at 12-quarter horizon.
  - Oil: mostly insignificant for HaR in both countries.

### Cross-country and cross-city patterns
- U.S.: greater short-term risk volatility historically; short-term HaR improved since the GFC; cities show substantial heterogeneity.
- Canada: long-term housing market risks more severe than short-term risks; risks have deteriorated over the past two decades and surpassed GFC-era levels in recent quarters.
- Overvaluation is the largest contributor to short-term downside risks in the U.S.; contributions of drivers vary markedly by city (e.g., Chicago influenced by capital flows; Los Angeles by overvaluation and supply).

### Micro-level evidence: risky debt (definitions and stylized facts)
- Financially weak households defined by:
  - Debt servicing ratio > 40 percent.
  - Debt-to-income ratio > 350 percent.
- Alternative risky-debt thresholds considered: debt-to-income ratio > 450 percent and > 550 percent.
- Risky debt not covered by assets proxied by mortgage debt held by households with loan-to-value (LTV) ratio > 80 percent.
- U.S. trends (2013–2016):
  - Median household debt-to-income declined from 2013 to 2016, converging to 1998 levels.
  - Share of debt held by borrowers with DTI > 350 decreased from about 42 percent in 2013 to about 39 percent in 2016.
  - Share of debt held by borrowers with debt servicing ratio > 40 percent declined by about 5 percentage points since 2004 to about 20 percent of total debt.
  - Minimal change in risky mortgage debt not covered by assets (LTV > 80 percent).
- Canada trends:
  - Greater dispersion across income quantiles relative to the U.S.
  - Low-income households (below 20th percentile) saw DTI decline from 2012 to 2016 but remain vulnerable relative to 1999.
  - Debt of poorer borrowers increased from about 2.5 times annual income in 1999 to about 3.5 times in 2016.
- Policy implications:
  - High household indebtedness raises vulnerability to shocks; debt servicing ratio is an important systemic signal though can be misleading in low-for-long interest rate environments.
  - Macroprudential policies should address externalities from high household leverage and debt-service vulnerabilities.
  - Identifying household types holding risky debt (by DTI, TDS, LTV thresholds) aids systemic risk assessment.

### Policy recommendations and implications
- Consider macroprudential policies targeting highly leveraged households to mitigate externalities and systemic risks.
- Tighten borrower-based policy instruments (loan-to-value, debt-service limits) to better safeguard households against shocks.
- Examine policies regarding capital flows and real estate tax policies as potential tools to address house price risks.
- Design macroprudential frameworks that account for heterogeneity of household indebtedness and housing market imbalances across regions.
- Implement supply-side policies (for example, zoning ordinances) to alleviate overvaluation imbalances where supply is highly inelastic and demand is rising.

### Model robustness and auxiliary results
- Robustness checks included:
  - Alternative lags for capital flows, FCI, and supply measures.
  - Alternative capital inflow types (including portfolio investment).
  - Alternative risky-debt definitions (DTI thresholds 350 and 550 percent; LTV > 90 percent).
  - Baseline model tested at 12-quarter horizon; results largely robust across quantiles and horizons.
- Annex statistics (country-level summary, Obs = 146 for each country) include exact sample moments for key variables (examples preserved in source): United States House Price Growth (YoY) Mean 0.014; Std. Dev 0.052; Min -0.137; Max 0.110. Canada House Price Growth (YoY) Mean -0.024; Std. Dev 0.042; Min -0.134; Max 0.099. (Full sets of summary statistics and city-level examples available in annex tables.)

*Source: wpiea2020042-print-pdf*

### REFERENCES _____________________________________________________________________ 39

### wpiea2020042-print-pdf - REFERENCES _____________________________________________________________________ 39

### I. Introduction and motivation
- House prices play a major role in assessing financial stability and monitoring systemic risk.
- A sharp deterioration of housing markets can have severe adverse effects on:
  - household balance sheets,
  - business confidence,
  - banks’ ability to lend,
  - and can produce cross-border spillovers due to synchronized housing cycles.
- The collapse of the U.S. housing market in 2008 precipitated prolonged global financial turbulence and the largest crisis since the Great Depression.
- Identifying drivers and tail risks to house prices is necessary to prepare for vulnerabilities to household wealth.

### Literature and key mechanisms
- Declines in house prices coincide with reductions in perceived household wealth and are followed by declines in consumption (Campbell and Cocco, 2005; Mian and Sufi, 2016).
- High household debt reduces ability to absorb shocks (interest rate increases, income losses), amplifying consumption declines in downturns.
- House price valuation is linked to macroeconomic and structural factors, including:
  - supply factors,
  - real interest rates (Agnello and Schuknecht, 2011; Grimes and Aitken, 2010),
  - macroprudential and monetary policy measures via credit availability, housing demand, and balance sheet effects,
  - rental yields, capital flows, household income, and leverage (Capozza et al., 2002; Duca, Muellbauer and Murphy, 2011).
- Household debt has economy-wide implications:
  - housing markets in highly leveraged areas are more vulnerable to income shocks (Lamont and Stein, 1999),
  - excessive household debt negatively impacts future GDP growth via debt overhang effects (Mian, Sufi, and Verner, 2017),
  - household debt is associated with higher unemployment and greater probability of banking crises (Mian, Sufi, Verner, 2017; Jorda, Schularick, and Taylor, 2016).

### Research questions addressed
- (1) How do previously identified drivers help measure downside risks to future house price growth?
- (2) How do downside house price risks vary across countries and cities?
- (3) What is the role of indicators such as overvaluation, financial conditions, and capital inflows on house price risks?
- (4) What role does household indebtedness play in influencing house price risks?
- (5) How did household debt in the U.S. and Canada evolve over time and which types of households are more vulnerable to shocks?

### Data and empirical approach
- Sample: quarterly data from 37 cities in Canada and the U.S. between 1983 to 2018.
- Two-stage methodology (building on Adrian, Boyarchenko and Giannone, 2019 and Deghi et al., 2020):
  - Stage 1: Estimate quantile regressions for each city to model nonlinear relationships between city- and country-level indicators and housing-market vulnerabilities, capturing variable contributions at different parts of the forecast distribution.
  - Stage 2: Fit a skewed-t distribution using predicted values for each quantile; the left tail (lower fifth percentile) of the fitted distribution is considered the house prices-at-risk for each city.
- Note: short- and long-term horizons refer to 4 and 12 quarters ahead, respectively.
- Note: “capital flows” refers to capital inflows in this paper.

### Contributions of the paper
- Granular city-level house prices-at-risk analysis:
  - Each city is estimated univariately to measure distinct sensitivities to different drivers of risk.
  - This captures effects that may be obscured at the aggregate (country) level.
- Micro-level analysis of household debt in the U.S. and Canada:
  - Uses household survey data to gauge debt held by financially weak households.
  - Provides insights into potential amplification effects if house prices fall, especially where debt held by financially fragile households is elevated.

### Main findings (threefold)
- Finding 1: Divergent trends in downside risks
  - Downside risks to house prices in the U.S. have declined over the past decade.
  - Downside risks to house prices in Canada have increased over the past decade.
  - Compared to historical levels:
    - The U.S. is significantly better off than prior to and during the GFC.
    - Canada is approaching GFC-level risks, indicative of increasing vulnerabilities.
  - These results are supported by both economy-wide and city-level analyses.
- Finding 2: Horizon-dependent determinants
  - Short-term (4 quarters ahead): tighter financial conditions are detrimental to house price risks.
  - Long-term (12 quarters ahead): the negative effect of tighter financial conditions is dampened.
  - Household leverage:
    - Associated with ameliorating short-term house price risks.
    - Associated with exacerbating long-term house price risks.
  - Capital flows (capital inflows) are significantly associated with future downside risks to key residential housing markets, but the net effect varies by city depending on types of flows and horizon.
  - Sensitivity of future downside risks to capital flows is correlated with a city’s foreign openness (immigration population).
  - City-specific valuation and supply factors are influential determinants of downside risks.
- Finding 3: Household balance sheets and sectoral vulnerabilities
  - U.S. overall share of household debt-to-GDP has decreased in recent years, along with a contraction in risky debt, suggesting a lesser degree of over-indebtedness.
  - Canadian households have become more leveraged, coinciding with an increase in the share of risky debt, indicating increased vulnerabilities of the household sector.

### Notable statistics and descriptors preserved from the source
- Sample size and period: 37 cities in Canada and the U.S. between 1983 to 2018.
- House prices-at-risk definition: lower fifth percentile of the fitted skewed-t distribution.
- Short-term horizon: 4 quarters ahead.
- Long-term horizon: 12 quarters ahead.
- Canada: about 30 percent overvalued as of 2018Q3 (aggregate valuation).
- Reference to the 2008 U.S. housing collapse as leading to the largest crisis since the Great Depression.

*Source: wpiea2020042-print-pdf - REFERENCES _____________________________________________________________________ 39*

### 6. Financial Conditions and Mortgage Debt Service in Canada

### 6. Financial Conditions and Mortgage Debt in Canada

### Conceptual framework — Financial conditions and transmission to housing
- Financial conditions transmit to housing markets through multiple channels: credit supply, leverage, household income, and expectations.
- Empirical evidence cited:
  - Excess corporate bond premia can contract credit supply (Gilchrist and Zakrajšek, 2012).
  - Corporate bond spreads predict fixed investment and economic activity (Philippon, 2009).
  - Tighter financial conditions increase downside risks to GDP growth, especially at short horizons (Adrian, Boyarchenko, and Giannone, 2019).
- Leverage channel: household debt affects house prices and amplifies transmission from financial conditions to housing.

### Conceptual framework — Household debt
- Household debt impacts consumption, net wealth, and GDP growth (Graham, 2018; Caceres, 2019).
- Cross-country and panel findings:
  - Household debt reduces future GDP growth via debt overhang and lower consumption (Mian, Sufi, and Verner, 2017; Alter, Feng, and Valckx, 2018).
  - Debt amplifies the macroeconomic impact of shocks in areas with highly leveraged, poorer households (Mian, Rao and Sufi, 2013).
- Monetary policy transmission can be weakened by high household debt (Sufi, 2015); yet some evidence finds highly indebted households are more sensitive to monetary shocks (Gelos et al., 2019).
- The GFC debate:
  - Subprime lending view: low lending standards increased risky borrowing (Parlour and Plantin, 2008; Dang et al., 2010).
  - Contrasting evidence: investor-driven inflows grew more than first-time buyer/subprime inflows (Bhutta, 2015).
  - Speculative demand and backward-looking expectations linked to house price swings (Case and Shiller, 1988, 1989; Shiller, 1990; Capozza and Seguin, 1996).

### Conceptual framework — Capital flows
- Capital inflows can lower long-term interest rates and contribute to housing booms (Global Savings Glut literature: Bernanke, 2005; Mendoza et al., 2007; Sa and Wieladek, 2015).
- Evidence that foreign demand raises local house prices and distorts affordability (Badarinza and Ramadorai, 2018; Li, Shen, and Zhang, 2019; Ari, Puy, and Shi (forthcoming)).
- Alternative view: international capital flows had minimal role in U.S. house prices; financial liberalization was more important (Favilukis et al., 2012).

### Conceptual framework — Other key indicators
- Housing supply, residential investment, and valuation metrics are important predictors of housing market risk (Capozza et al., 2002; Gattini and Hieber, 2010; Dokko et al., 2014; Fernandez Kranz and Hon, 2006).
- Demographics (urbanization, population growth) positively associated with house prices (Girouard et al., 2006; Lin et al., 2018).

### Stylized facts — Debt, prices, and composition (U.S. vs Canada)
- Historical co-movement: household debt and house prices generally moved together; correlation weakened in the U.S. post-GFC.
- Canada:
  - Real house price index has continuously increased alongside household debt-to-GDP for decades.
  - Household debt in Canada was about 98 percent of GDP at end-2018.
  - Mortgage debt makes up about 65 percent of total household debt in Canada as of 2019.
  - Home equity loans (HELOCs) in Canada stood at $173 billion in 2018Q4, about 10 percent of household debt.
  - IMF (2019b) estimate: household debt service in Canada will increase by 0.4 percentage points to 15.3 percent of disposable income over the following year.
  - Long-term trend: about 30 percent overvaluation in aggregate Canadian house prices; Toronto about one and a half times overvalued and Vancouver about 40 percent overvalued relative to long-term averages (2018).
- United States:
  - Household debt about 80 percent of GDP at end-2018.
  - Mortgage debt about 70 percent of total household debt as of 2019.
  - Aggregate U.S. house price valuation has remained stable since 2013.
  - Some cities that were overvalued pre-GFC experienced large declines during the bust (Miami, Las Vegas, Los Angeles, Tampa).
  - San Francisco house price-to-income ratio around 220 percent as of 2018 (city-level).

### Policy responses and macroprudential measures in Canada
- Bank of Canada:
  - Lowered interest rates in 2008-09, kept them close to the zero-lower bound, then tightened in 2017-18.
  - Did not deploy unconventional policy tools; focused on inflation targeting and spillover management.
- Canadian macroprudential tightening since 2008:
  - Six rounds of mortgage insurance rule tightenings targeting loan-to-value limits, credit score requirements, and amortization limits.
  - Provincial demand-side measures: buyer stamp duties, speculation and vacancy taxes (British Columbia and Ontario, 2016-18).
- Evidence on effectiveness:
  - Mixed findings: early rounds curbed mortgage credit growth but high household debt and elevated house prices persisted (Krznar and Morsink, 2014; Kuncl, 2016).
  - Loan-to-value limits more effective on first-time buyers’ demand than debt-service ratio limits (Allen et al., 2017).

### Data and empirical approach
- Data sources:
  - U.S. house price indices: S&P/Case & Shiller Home Price Indices; FHFA for city-level.
  - Canada house price indices: Statistics Canada; Real Property Solutions; Canadian Real Estate Association.
  - Financial conditions: U.S. National Financial Conditions Index (Federal Reserve Bank of Chicago); Canada: price-based FCI based on IMF (2017) methodology.
  - Capital inflows: IMF BPM6 dataset (net changes in liabilities, scaled by GDP).
  - Household debt: first difference in household debt-to-GDP from the BIS.
  - Other variables: residential investment (Canada), housing starts (U.S.), WTI oil price index (quarter-on-quarter change).
- Household-level analysis:
  - U.S.: Survey of Consumer Finances (SCF).
  - Canada: Survey of Financial Security (SFS).
- Empirical methodology:
  - Two-stage city-level HaR (House prices-at-Risk) following Adrian, Boyarchenko and Giannone (2019).
  - Stage 1: Quantile regression for each city:
    - Dependent variable: future residential real house price growth ΔCityHP_{c,i,t+h}^q
    - Explanatory sets: city-level X_{c,i,t} (autoregressive term, house price-to-income, supply indicator), country-level Z_{c,t} (household debt, FDI, capital flows), OIL_t, FCI_{c,t}.
    - Forecast horizons h include 4-quarter and 12-quarter ahead; quantiles q = 0.1, 0.25, 0.5, 0.75, 0.9.
  - Stage 2: Fit a skewed-t distribution to the predicted quantiles to estimate the lower tail (5 percent house prices-at-risk).
    - Parameters (location, scale, skew) chosen by minimizing squared distance between estimated quantiles and skewed-t quantiles.
  - Short-term downside risks defined as the 5 percent value-at-risk for a one-year ahead horizon; long-term downside risks refer to the 5 percent value-at-risk for a three-year ahead horizon.

### Empirical findings — Through the cycle and cross-section
- Cross-country summary:
  - U.S. housing markets: generally riskier in short term, with larger magnitude and volatility of downside risks; more susceptible to abrupt booms and busts.
  - Canada: long-term housing market risks are more severe than short-term risks; risks have deteriorated over the past two decades, surpassing GFC-era levels in recent quarters.
- Short-term downside risks (one-year ahead, 5 percent HaR) at end-2018Q3:
  - U.S. aggregate: about -1 percent (annualized). Interpretation: there is a 5 percent likelihood that house prices will experience a 1-percentage point decline over the following four quarters.
  - City extremes in U.S.: Los Angeles and San Francisco each facing a 4-percentage point decline within the following four quarters.
  - Canada aggregate (4-quarter horizon): the 5 percent house price-at-risk is about -3 percent at end-2018Q3.
  - Canada city-level extremes at end-2018Q3: Toronto -10.8 percent tail risk; Vancouver -6.5 percent tail risk.
- Cross-time summary:
  - U.S. short-term downside risks improved since the GFC (2007 vs 2018), decreasing across cities.
  - Canada: house price risks high by historical standards and worsening since 2007 in several cities (Toronto, Quebec, Hamilton).
  - Structural contrast: U.S. experienced rapid rise and large fall in prices around the GFC and have not returned to concerning levels; Canada experienced minimal decline at GFC and prices have generally continued to rise thereafter.
- Valuation and risk link:
  - Cities with high pre-GFC overvaluation (Chicago, Phoenix, Atlanta) had 5 percent HaR between -10 and -25 percent before the GFC and have seen valuation and risk decline since.
  - Some metro areas remained significantly overvalued in 2018 (e.g., San Francisco ~220 percent house price-to-income).
  - Canadian metros (Toronto, Vancouver, Hamilton) have valuations that surpassed levels supported by borrowing capacity and net present value of rental income (Andrle and Plasil, 2019).
- Heterogeneity:
  - Considerable dispersion in downside risks across cities in both countries, reflected in volatility of lower-bound percentiles of the cross-sectional distribution.
  - Drivers of house price risks differ across cities; identifying these drivers is necessary for targeted policy responses.

*Source: wpiea2020042-print-pdf - 6. Financial Conditions and Mortgage Debt in Canada (IMF staff compilation, figures and citations as provided in the chapter).*

### 1. Downside Risks across American cities

### 1. Downside Risks across American cities

### Summary of cross-city downside risk measures
- Panels reported for: 5 percent, 4q-ahead; 5 percent, 12q-ahead; and valuation comparisons (2007Q3 and 2018Q3).
- Cross-sectional series: green = 90th percentile; blue = 10th percentile; black = cross-sectional average; yellow dotted = aggregate country level.
- Each series depicts the average of the previous four quarters.

### Key findings on geographic heterogeneity
- City-level valuation is the main contributor to house price risk across cities.
- Tail risks to the U.S. housing market have generally improved in recent years; tail risks to Canadian house prices have generally increased.

---

### B. Drivers of House Prices-at-Risk

### Valuation (price-to-income ratio)
- U.S.: A one standard deviation growth in the price-to-income ratio results in a 2.9 percentage point deterioration in downside risks to the U.S. housing market (aggregate level).
- Canada: A one standard deviation increase in the price-to-income ratio results in a 1.5 percentage point increase in downside risks to the overall housing market (aggregate level).
- The association between house price valuation and downside risks is statistically significant across quantiles in both countries.

### Supply-side indicators
- U.S.: A one standard deviation growth in housing starts is associated with a 2.3 percentage point decrease in downside risks to housing markets (aggregate level); housing starts alleviate risks in most U.S. cities.
- Canada: Residential investment shows heterogeneous effects:
  - At the country level, increased housing supply mitigates risks across quantiles.
  - Positive (mitigating) effects observed in cities including Edmonton, Quebec City, and Ottawa.
  - Negative association (i.e., increases in supply do not mitigate risks) observed in Toronto, Vancouver, and Hamilton—interpreted as supply increases being insufficient to satisfy demand due to natural barriers and supply restrictions.

### Financial conditions (FCI)
- U.S.: Effect ambiguous across cities and quantiles.
  - In metro areas like Austin, Cleveland, Seattle, and San Francisco, a one standard deviation tightening (increase) of financial conditions leads to a 1.7 to 4.8 percentage point increase in house price risks across these cities (short-term/lower quantile).
  - For some cities, tighter financial conditions appear to mitigate downside risks at a one-year horizon.
  - San Francisco and Seattle are most sensitive to tightening in financial conditions.
  - Impact is less severe at the 12-quarter horizon.
- Canada: Tighter financial conditions are associated with increases in downside risks uniformly across Canadian cities and statistically significant for most.
  - Most sensitive cities: Toronto, Ottawa, and Calgary, where a one standard deviation tightening (increase) in financial conditions leads to a 1.6 to 3.1 percentage point increase in short-term downside risks.
  - Impact moderated in the long run (12-quarter horizon).

### Capital flows
- Types: foreign direct investment (FDI) and other capital flows (bank transactions from foreigners; not portfolio or FDI).
- U.S.:
  - FDI short-term: alleviates risks in some tech-friendly cities (Seattle, Portland) — associated with a reduction of about 1 to 2 percentage points in house prices-at-risk (short term).
  - FDI long-term (12-quarter): associated with around a 6 to 8 percentage points increase in house price risks in cities like Tampa, Detroit, and Minneapolis.
  - Other capital flows: generally associated with increased downside risks across almost all major U.S. cities; strongest effects in Sun Belt cities (Las Vegas, Los Angeles, Miami, San Diego).
  - Sensitivity to other capital flows correlates with foreign openness (share of immigrants to total population).
- Canada:
  - FDI: generally mitigates downside risks; a one percentage point increase in FDI inflow to Canada is significantly associated with a reduction of about ½ to 1 percentage points in house prices-at-risk in Calgary, Toronto, Ottawa, and Quebec City (short term).
  - FDI long-term (12-quarter): associated with a 1.5 to 2.5 percentage point decrease in house price risks in Quebec City, Toronto, and Hamilton.
  - Other capital flows: predict significantly negative effects (increasing risks) in Vancouver, Ottawa, and Hamilton in the short term; largely inconsequential to long-term house price risks across Canadian cities aside from Hamilton.

### Household debt and oil
- Household debt:
  - Generally positive effect on house price risks in the short term (ease of credit stimulates near-term growth).
  - Worsens housing market risks at a 12-quarter horizon (long-term negative effects of high household indebtedness).
- Oil:
  - Mostly insignificant effect on house prices-at-risk in both Canada and the U.S.

### Aggregate and city-level contributions
- Overvaluation is the largest contributor to short-term downside risks to housing markets in the U.S.
- Other factors (oil, household debt) have mostly mitigating, albeit minimal, effects over time until around the GFC.
- Contributions vary by city: e.g., Chicago's short-term tail risks significantly influenced by capital flows; Los Angeles dominated by overvaluation and supply-side factors.
- Overall: city-level valuation is the dominant driver of house price risk across cities.

---

### VI. Micro-level evidence: Risky Debt

### Definitions and measurement
- Financially weak households defined by:
  - Debt servicing ratio > 40 percent.
  - Debt-to-income ratio > 350 percent.
- Alternative risky-debt thresholds considered: debt-to-income ratio > 450 percent and > 550 percent.
- Risky debt not covered by assets proxied by mortgage debt held by households with loan-to-value (LTV) ratio > 80 percent.
- Debt servicing ratio = ratio of monthly payments (mortgage installments and credit card payments) to disposable income (includes government transfers).
- Debt-to-income ratio = total household debt divided by disposable income.
- Loan-to-value ratio = final value of mortgage loan divided by value of principal residence.

### Trends and distributional patterns
- U.S.:
  - Median household debt-to-income ratio across all income quantiles declined from 2013 to 2016, converging back to 1998 levels.
  - Share of debt held by borrowers with debt-to-income ratio > 350 decreased from about 42 percent in 2013 to about 39 percent in 2016.
  - Share of debt held by borrowers with debt servicing ratio > 40 percent has declined by about 5 percentage points since 2004 to about 20 percent of total debt.
  - Minimal change in the amount of risky debt not covered by assets (mortgage debt with LTV > 80 percent).
- Canada:
  - Greater dispersion across income quantiles compared to the U.S.
  - Low income households (below 20th percentile) saw debt-to-income ratio decline from 2012 to 2016, but remain particularly vulnerable relative to 1999 levels.
  - Debt of poorer borrowers increased from about 2.5 times annual income in 1999 to about 3.5 times in 2016, indicating increased financial distress.

### Policy-relevant implications
- High household indebtedness raises household vulnerability to shocks; debt servicing ratio is an important signal of systemic risk though can be misleading in low-for-long interest rate environments.
- Macroprudential policies should address externalities from high household leverage and debt-service vulnerabilities.
- Identifying types of households holding risky debt (based on the defined thresholds) aids assessment of potential systemic risk from a housing market bust, especially where high LTV coincides with high leverage.

*Source: CREA; Statistics Canada; Haver; IMF staff calculations.*

### 1. Debt-to-Income (DTI) in the US

### 1. Debt-to-Income (DTI) in the US

### Key findings on household indebtedness and risky debt
- DTI ratio is calculated as total amount of debt divided by household disposable income, including government transfers. TDS is the ratio of mortgage payments plus credit card payments divided by disposable income, including government transfers. LTV is the ratio of mortgage loan (final value) to the value of the principal residence.
- The level of household debt in the U.S. has decreased in recent years, along with the share of debt held by risky borrowers.
- In contrast, the level of household debt in Canada has increased in recent years, coinciding with the share of debt held by risky borrowers.
- From 2012 to 2016, Canadian households experienced a small increase in the share of debt belonging to borrowers with a debt-to-income ratio above 350, which stood at about 41 percent of total debt as of 2016.
- Since 2005 the share of risky debt in Canada has increased significantly when considering both the debt-to-income ratio and the debt servicing ratio, signifying increasing financial vulnerabilities.
- The share of risky debt not covered by assets in Canada has remained stable since 2005 and is comparable to U.S. levels.

### Robustness of empirical results
- Model selection used a variety of country- and city-level variables with the Pseudo R2 as guidance for variable inclusion.
- The baseline model was tested at a 12-quarter horizon; results are available in annex tables and notable differences are mentioned in the main text.
- Influence of drivers is horizon-dependent:
  - FDI is significantly impactful to housing markets in the long run, reflecting sizable foreign investment into real estate.
  - Financial condition indices (FCI) capture current states of financial markets; FCI is not expected to have a sizable influence on long-term house price risks.
  - Household debt shows a negative association with long-term house price risks; short-run effects of household debt are mostly positive across quantiles, with some outliers.
  - Supply indicators, overvaluation, and oil show no significant difference in coefficients across horizons, indicating robustness.
- Across quantiles:
  - Influence of valuation is harmonious and robust across economic growth stages (10th percentile = busts, 90th percentile = booms, median = average growth).
  - Impact of capital flows on house price risks remains broadly consistent across quantiles.
  - Financial conditions are more strongly associated with future house prices at the left tail of the distribution than at the median (consistent with “Growth-at-Risk” literature).
  - Influence of supply factors and oil varies across quantiles.
- Validity checks included:
  - Various lags (including 1 and 4 quarter lags) on capital flow measures, financial conditions, and supply measures.
  - Alternative capital inflow types, including portfolio investment.
  - Alternative definitions of risky debt (e.g., DTI above 350 and 550 percent; LTV above 90 percent).
  - Results are generally robust to alternative specifications.

### Conclusion and interpretation
- The health of housing markets and household indebtedness is an important pillar for financial stability due to financial sector exposure to housing (mortgages, asset-backed securities) and the nexus between household balance sheets, house prices, and consumption.
- Using a sample of 37 cities with quarterly data from 1983-2018, the paper identifies housing overvaluation, supply-side drivers, financial conditions, household debt and capital flows as significant associates of house price risks, with effects varying by city and horizon.
- Sensitivity of future house price risks to capital inflows appears to correlate with foreign openness.
- Overall assessment:
  - U.S. housing market: less risky with improving credit conditions and fundamentals, especially relative to the GFC, but with substantial heterogeneity across cities and horizons.
  - Canada housing market: gradually deteriorated over the past few years, with household leverage and housing market risks rising to levels last seen around the GFC.

### Policy recommendations
- Consider macroprudential policies targeting highly leveraged households to mitigate externalities and lessen systemic risks.
- Tighten borrower-based policy instruments to better safeguard the household sector against unexpected shocks.
- Examine policies regarding capital flows and real estate tax policies as potential tools to address house price risks.
- Design macroprudential policies that account for heterogeneity of household indebtedness and housing market imbalances across regions.
- Implement supply-side policies (for example, zoning ordinances) to alleviate overvaluation imbalances; note that highly inelastic housing supply coinciding with rising demand restricts the effectiveness of macroprudential and tax-based demand-side measures (IMF 2018).

### Key statistics and summary figures (select from country and city tables)
- Sample: 37 cities, quarterly data from 1983-2018.
- Country-level summary statistics (United States, Obs = 146):
  - House Price Growth (YoY): Mean 0.014; Std. Dev 0.052; Min -0.137; Max 0.110
  - House Price-to-Income: Mean 1.131; Std. Dev 0.154; Min 0.883; Max 1.469
  - Housing Starts (YoY): Mean 0.007; Std. Dev 0.190; Min -0.704; Max 0.616
  - FCI: Mean -0.287; Std. Dev 0.597; Min -0.900; Max 2.940
  - Δ Household Debt (% of GDP): Mean 0.201; Std. Dev 0.758; Min -1.900; Max 2.800
  - FDI (% of GDP): Mean 0.014; Std. Dev 0.008; Min 0.002; Max 0.037
  - Other Capital Flows (% of GDP): Mean 0.016; Std. Dev 0.015; Min -0.036; Max 0.057
  - Oil (QoQ): Mean 0.005; Std. Dev 0.148; Min -0.704; Max 0.393
- Country-level summary statistics (Canada, Obs = 146):
  - House Price Growth (YoY): Mean -0.024; Std. Dev 0.042; Min -0.134; Max 0.099
  - House Price-to-Income: Mean 0.938; Std. Dev 0.079; Min 0.740; Max 1.091
  - Residential Investment (% of GDP): Mean 0.013; Std. Dev 0.004; Min 0.006; Max 0.022
  - FCI: Mean -0.141; Std. Dev 0.282; Min -0.704; Max 1.065
  - Δ Household Debt (% of GDP): Mean 0.405; Std. Dev 0.731; Min -1.200; Max 3.700
  - FDI (% of GDP): Mean 0.024; Std. Dev 0.018; Min -0.002; Max 0.085
  - Other Capital Flows (% of GDP): Mean 0.012; Std. Dev 0.014; Min -0.016; Max 0.065
  - Oil (QoQ): Mean 0.005; Std. Dev 0.148; Min -0.704; Max 0.393
- Selected city-level statistics (examples, Obs = 146 for each city):
  - Los Angeles: Real House Price Growth Mean 0.028; Std. Dev 0.098; Min -0.262; Max 0.295; House Price-to-Income Mean 1.451; Std. Dev 0.263; Min 1.074; Max 2.164
  - Toronto: Real House Price Growth Mean 0.035; Std. Dev 0.076; Min -0.181; Max 0.325; House Price-to-Income Mean 0.870; Std. Dev 0.242; Min 0.509; Max 1.575
  - Vancouver: Real House Price Growth Mean 0.009; Std. Dev 0.065; Min -0.254; Max 0.181; House Price-to-Income Mean 0.818; Std. Dev 0.253; Min 0.418; Max 1.362

*Sources: Statistics Canada’s Survey of Financial Security, US Survey of Consumer Finances, IFS, FRED, Haver, CREA, BIS, IMF staff calculations.*

### Annex Table 1. United States: Short-term City-level Coefficients

### Annex Table 1. United States: Short-term City-level Coefficients

### Dependent variable and variables included
- Dependent variable: city-level residential house price year-on-year growth, 4 quarters ahead.
- Explanatory variables (as used in the table): House Price-to-Income; Housing Starts (log growth of state-level housing starts); FCI (price-based financial conditions index); Household Debt (scaled by GDP); FDI (scaled by GDP); Other Capital Flows (scaled by GDP); Oil (quarter-on-quarter growth of the oil price index); Intercept; AR.

### Selected United States short-term coefficients (point estimate and standard error)
- House Price-to-Income: -0.190*** (0.018)
- Housing Starts: 0.120*** (0.030)
- FCI: -0.033** (0.016)
- Household Debt: 0.031*** (0.006)

(Note: The table reports these and many additional city-level coefficients and standard errors for Austin, Boston, Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Houston, Las Vegas, Los Angeles, Miami, Minneapolis, New York, Philadelphia, Phoenix, Portland, San Diego, San Francisco, Seattle, Tampa, Washington DC, and an aggregate United States column. Coefficients are presented across different quantiles/tau values in the Annex table.)

### Key table features and interpretation cues
- Coefficients are reported with standard errors in parentheses.
- Asterisks denote statistical significance (as in the table).
- Variables measuring financial flows (Household Debt, FDI, Other Capital Flows) are scaled by GDP.
- Oil is measured as quarter-on-quarter growth of the oil price index (as stated in the table note).
- Housing starts are measured as log growth of state-level housing starts.
- The table presents city-level heterogeneity in the short-term relationship between explanatory variables and 4-quarter-ahead residential house price growth.

*Source: Annex Table 1. United States: Short-term City-level Coefficients, wpiea2020042-print-pdf*

### Annex Table 3. United States: Long-term City-level Coefficients

### Annex Table 3. United States: Long-term City-level Coefficients

### Note
- The dependent variable is city-level residential house price year-on-year growth, 12 quarters ahead.
- Household debt, FDI, and other capital flows are scaled by GDP.
- FCI is the price-based financial conditions index.
- House price to income is the city-level house price over state-level income.
- Housing starts is the log growth of state-level housing starts.
- Oil is the quarter-on-quarter growth of the oil price index.

### Variables included in the table (as labeled)
- House Price-to-Income
- Housing Starts
- FCI
- Household Debt
- FDI
- Other Capital Flows
- Oil
- Intercept
- AR

### Representative long-term coefficients (city-specific estimates shown exactly as in the table; significance stars and standard errors in parentheses are preserved where shown)
- Atlanta
  - Example coefficients across tau rows (selected): 0.138*** (0.035); -0.447 (0.005); 0.007 (0.019); -0.014*** (0.003); 0.450 (0.015); 0.028 (0.031); -0.038 (0.060); 0.178** (0.082); 0.124*** (0.024); 0.093 (0.002); 0.002 (0.018); -0.013*** (0.003); 0.594 (0.007); 0.046 (0.033); -0.006 (0.051); 0.000 (—)
- Boston
  - Example coefficients across tau rows (selected): -0.016 (0.021); 0.656** (0.003); 0.008 (0.010); 0.001 (0.003); -0.594 (0.006); 0.030 (0.030); 0.407*** (0.046); -0.262*** (0.027); 0.299*** (0.051); -0.008 (0.014); 0.483** (0.002); 0.012 (0.012); 0.004 (0.003); -0.815 (0.007); 0.051** (0.023); 0.466*** (0.062); -0.285*** (0.036); 0.243*** (0.056); 0.082*** (0.015); 0.607** (0.003); 0.020** (0.009); 0.004** (0.002); -0.717 (0.004); 0.058* (0.032); 0.482*** (0.029); -0.282*** (0.015); 0.290*** (0.067)
- Chicago
  - Example coefficients across tau rows (selected): 0.024 (0.030); -0.355 (0.004); 0.003 (0.019); -0.008*** (0.003); -4.709* (0.025); 0.033 (0.034); 0.181* (0.106); -0.103 (0.063); 0.224* (0.114); 0.078*** (0.024); -0.441*** (0.001); -0.017* (0.010); -0.005*** (0.002); -0.203 (0.005); 0.010 (0.014); 0.060* (0.032); -0.040* (0.022); 0.183* (0.099)
- Los Angeles
  - Example coefficients across tau rows (selected): -0.001 (0.023); -0.983*** (0.003); -0.030*** (0.011); -0.001 (0.002); -0.418 (0.008); -0.035 (0.029); 0.141** (0.056); -0.098*** (0.032); 0.025 (0.175); 0.014 (0.021); -0.514*** (0.002); -0.018** (0.008); -0.001 (0.002); 0.099 (0.005); -0.009 (0.021); 0.156*** (0.036); -0.103*** (0.021); 0.376*** (0.188)
- Miami
  - Example coefficients across tau rows (selected): 0.094** (0.043); 1.512* (0.009); -0.004 (0.041); -0.002 (0.007); -14.484*** (0.037); 0.064 (0.071); 0.591*** (0.112); -0.497*** (0.105); 0.028 (0.099); 0.042 (0.034); 1.278** (0.006); 0.026** (0.010); 0.002 (0.005); -8.956** (0.039); -0.031 (0.050); 0.450*** (0.089); -0.373*** (0.057); 0.588*** (0.148)
- New York
  - Example coefficients across tau rows (selected): 0.056 (0.056); 0.297 (0.005); 0.019 (0.019); -0.005 (0.004); 0.624 (0.017); 0.090** (0.043); 0.301*** (0.057); -0.239*** (0.032); 0.000 (0.063); 0.118*** (0.020); -0.378 (0.004); 0.006 (0.016); 0.002 (0.005); 1.456 (0.011); 0.029 (0.055); 0.262*** (0.060); -0.197*** (0.032); -0.048 (0.055)
- San Francisco
  - Example coefficients across tau rows (selected): 0.085*** (0.026); -0.993*** (0.004); -0.046*** (0.016); 0.010*** (0.002); -1.106 (0.010); 0.030 (0.025); 0.389*** (0.031); -0.318*** (0.021); -0.256*** (0.053); 0.163*** (0.029); -0.4720.022 (0.003) [note: value grouping is as in table]; -0.043*** (0.015); 0.007** (0.003); 1.837** (0.008); 0.043 (0.028); 0.369*** (0.040); -0.287*** (0.023); -0.275*** (0.078)
- Seattle
  - Example coefficients across tau rows (selected): 0.172*** (0.026); -0.993*** (0.004); -0.022** (0.016); -0.005 (0.004); 2.438*** (0.007); -0.049 (0.026); 0.120** (0.060); -0.137*** (0.042); -0.118 (0.053); 0.025 (0.022); -0.145 (0.003); -0.034** (0.015); -0.006** (0.003); 2.968*** (0.008); -0.032 (0.028); 0.176*** (0.042); -0.139*** (0.027); -0.231*** (0.078)

(Note: the table contains many more city-specific coefficients and associated standard errors and significance indicators for multiple tau values. The examples above reproduce coefficients and parentheses exactly as they appear in the source for select cities and selected tau rows.)

### Table structure and additional elements
- The table presents coefficients for multiple tau quantiles (visible tau labels: 0.9, 0.5, 0.1) for each variable across cities.
- Standard errors are reported in parentheses under the coefficient estimates.
- Significance is indicated with stars (***, **, *) next to coefficient estimates as shown.
- The page contains continuation of Annex Tables for Canada and additional Annex Tables (Annex Table 4, Annex Table 5, Annex Table 6, Annex Table 7) and dataset descriptions following the United States table.

*Annex Table 3. United States: Long-term City-level Coefficients (page content reproduced exactly as provided).*

### REFERENCES

### REFERENCES

### Macro-financial linkages and growth
- Adrian, T., N. Boyarchenko, and D. Giannone, 2019. “Vulnerable Growth”, American Economic Review, 109(4): 1263–1289.  
- Adrian, T., F. Grinberg, N. Liang, and S. Malik, 2018. “The Term Structure of Growth at Risk” International Monetary Fund WP/18/180.  
- Eggertsson, G.B. and P. Krugman, 2012. “Debt, Deleveraging, and the Liquidity Trap: A Fisher-Minsky-Koo Approach.” Quarterly Journal of Economics 127 (3): 1469–513.  
- Gilchrist, S., and E. Zakrajšek, 2012. “Credit Spreads and Business Cycle Fluctations.” American Economic Review, 102(4), 1692-1720.  
- Jordà, Ò., M. Schularick and A.M. Taylor, 2016. “The Great Mortgaging: Housing Finance, Crises and Business Cycles.” Economic Policy 31 (85): 107–52.  
- Mian, A., and A. Sufi. 2016. “Who Bears the Cost of Recessions? The Role of House Prices and Household Debt.” Chapter 5 in Handbook of Macroeconomics, edited by John B. Taylor and Harald Uhlig. Amsterdam: Elsevier  
- Mian, A., A. Sufi and E. Verner, 2017. ‘Household Debt and Business Cycles Worldwide.’ Quarterly Journal of Economics (132)4: 1–63.  
- Mian, A., Sufi, A. and K. Rao, 2013. “Household Balance Sheets, Consumption, and the Economic Slump.” Quarterly Journal of Economics, 128(4), 1687-1726  
- Jones, C., V. Midrigan, and T. Philippon, 2018. “Household Leverage and the Recession” NBER Working Paper 16965, National Bureau of Economic Research

### House prices, consumption, and household debt
- Campbell, J. and J. Cocco, 2007. "How do house prices affect consumption? Evidence from micro data," Journal of Monetary Economics, Elsevier, Volume 54(3), 591-621  
- Capozza, D., P. Hendershott, C. Mack, and C. Mayer, 2002. “Determinants of Real House Price Dynamics.” NBER Working Paper 9262, National Bureau of Economic Research  
- Case, K. and R. Shiller, 1988. “The Efficiency of the Market for Single Family Homes” NBER working paper 2506, National Bureau of Economic Research  
- Case, K. and R. Shiller, 1989. “Survey evidence on diffusion of interest and information among investors” Journal of Economic Behavior and Organization, volume 12(1), pages 47-66  
- Bhutta, N. 2015. “The Ins and Outs of Mortgage Debt During the Housing Boom and Bust” Journal of Monetary Economics, volume 76, pages 284-298  
- Duca, J., J. Muellbauer and A. Murphy, 2011. “House Prices and Credit Constraints: Making Sense of the U.S. Experience.” Economic Journal, Volume 121 (552): 533–51.  
- Campbell, J. and J. Cocco, 2007. "How do house prices affect consumption? Evidence from micro data," Journal of Monetary Economics, Elsevier, Volume 54(3), 591-621  
- Capozza, D. and P. Seguin, 1996. “Expectations, Efficiency, and Euphoria in the Housing Market” Regional Science and Urban Economics, volume 26(3-4), pages 369-386  
- Ortalo-Magne, F., and R. Rady, 2006. “House Market Dynamics: On the Contribution of Income Shocks and Credit Constraints” The Review of Economics Studies, 73(2), 459-485  
- Zabai, A., 2017. “Household Debt: Recent Developments and Challenges” BIS Quarterly Review.  
- Bank of Canada, 2017. “Home Equity Lines of Credit: Market Trends and Consumer Issues”, Public Research Report, June 2017.  
- Al-Mqbali, L., O. Bilyk, S. Caputo, and J. Younker, 2019. “Reassessing the Growth of HELOCs in Canada Using New Regulatory Data,” Bank of Canada, Staff Analytical Note 2019-14.  
- Bilyk, O. and M. teNyenhuis, 2018. “The Impact of Recent Policy Changes on the Canadian Mortgage Market”, Bank of Canada – Staff Analytical Note 2018-35.  
- Peterson, B. and T. Roberts, 2016. “Household Risk Assessment Matrix”, Bank of Canada, Technical Report No. 106.  
- Baker, S. 2015. “Debt and the Consumption Response to Household Income Shocks”.  
- Campbell, J. and J. Cocco, 2007. "How do house prices affect consumption? Evidence from micro data," Journal of Monetary Economics, Elsevier, Volume 54(3), 591-621

(Note: the source lists multiple contributions on house prices, consumption, mortgage markets, and HELOCs across academic journals, NBER and central bank staff notes.)

### Macroprudential policy, housing finance, and financial stability
- Alam, Z., A. Alter, J. Eiseman, R. G. Gelos, H. Kang, M. Narita, E. Nier, and N. Wang, 2019. “Digging Deeper--Evidence on the Effects of Macroprudential Policies from a New Database” IMF working paper 19/66.  
- Allen, J., T. Grieder, B. Peterson, and T. Roberts, 2017. “The Impact of Macroprudential Housing Finance Tools in Canada”, Journal of Financial Intermediation.  
- Krznar, J., and J. Morsink, 2014. “With Great Power Comes Great Responsibility: Macroprudential Tools at Work in Canada.” IMF working paper 14/83.  
- Kuncl, M., 2016. "Assessment of the Effects of Macroprudential Tightening in Canada," Staff Analytical Notes 16-12, Bank of Canada.  
- Alter, A., J. Dokko, and D. Seneviratne, 2018. “House Price Synchronicity, Banking Integration, and Global Financial Conditions”, International Monetary Fund, WP/18/250.  
- Alter, A., A. X. Feng, and N. Valckx, 2018. “Understanding the Macro-Financial Effects of Household Debt: A Global Perspective.” International Monetary Fund, WP 18/76.  
- Drehmann, M. and M. Juselius, 2012. “Do Debt Service Costs Affec Macroeconomic and Financial Stability?” BIS Quarterly Review.  
- Deghi, A., M. Katagiri, S. Shahid, and N. Valckx, 2020. “Predicting Downside Risks to House Prices and Macro-Financial Stability” International Monetary Fund.  
- International Monetary Fund, 2017. “Household Debt and Financial Stability”, Global Financial Stability Report, Chapter 2, IMF, October 2017.  
- International Monetary Fund, 2017. “Financial Conditions and Growth at Risk”, Global Financial Stability Report, Chapter 3, IMF, October 2017.  
- International Monetary Fund, 2019a. “Downside Risks to House Prices”, Global Financial Stability Report, Chapter 2, IMF, April 2019.

### Capital flows, foreign demand, and global factors
- Caballero, R., E. Farhi, and P. Gourinchas, 2008. “An Equilibrium Model of "Global Imbalances" and Low Interest Rates” American Economic Review, Volume 98(1), pages 358-393  
- Favilukis, K., S. Ludvigson, and S. Van Nieuwerburgh, 2012. “International Capital Flows and House Prices: Theory and Evidence” NBER Working Paper 17751, National Bureau of Economic Research  
- Sá, F., P. Towbin, and T. Wieladek, 2013. “Capital Inflows, Financial Structure and Housing Booms” Journal of the European Economic Association, Volume 12(2), pages 522-546  
- Sá, F. and T. Wieladek, 2015. “Capital Inflows and the US Housing Boom”, Journal of Money, Credit and Banking, Volume 47, Issue S1, pages 221-256.  
- Aizenman, J., and Y. Jinjarak, 2009 “Current Account Patterns and National Real Estate Markets” Journal of Urban Economics, 66(2), 75-89.  
- Borensztein, E., J.D. Gregorio, and J.W. Lee, 1998. “How Does Foreign Direct Investment Affect Economic Growth?” Journal of International Economics, 45(1), 115-135  
- Iamsiraroj, S. and M.A. Ulubasoglu, 2015. “Foreign direct investment and economic growth: A real relationship or wishful thinking?” Economic Modelling, 51: 200-213.  
- Globerman, S., 2019. “Canadian Foreign Direct Investment: Recent Patterns and Interpretation” Fraser Research Bulletin.  
- Ari, A., D. Puy, and Y. Shi. “Are Foreigners Driving Local House Prices? Evidence from the US” IMF Working Paper, forthcoming.  
- Badarinza. C and T. Ramadorai, 2018. “Home away from Home? Foreign Demand and London House Prices”, Journal of Financial Economics, Volume 130(3), pages 532-555  
- Li, Z., L.S. Shen, and C. Zhang, 2019. “Capital Flows, House Prices, and the Real Economy: A “China Shock” in the US Real Estate Market.”

### Empirical methods, forecasting, and models of housing dynamics
- Gattini, L., and P. Hiebert, 2010. “Forecasting and Assessing Euro Area House Prices through the Lens of Key Fundamentals.” Working Paper 1249, European Central Bank, Frankfurt, September.  
- Gelos, R. G., F. Grinberg, T. Mancini-Griffoli, S. Khan, M. Narita, and U. Rawat, 2019. “Has Higher Household Indebtedness Weakened Monetary Policy Transmission?” IMF Working Paper 19/11.  
- Deghi, A., M. Katagiri, S. Shahid, and N. Valckx, 2020. “Predicting Downside Risks to House Prices and Macro-Financial Stability” International Monetary Fund.  
- Capozza, D. and P. Seguin, 1996. “Expectations, Efficiency, and Euphoria in the Housing Market” Regional Science and Urban Economics, volume 26(3-4), pages 369-386  
- Lamont, O., and J. Stein, 1999. “Leverage and House-Price Dynamics in U.S. Cities.” RAND Journal of Economics 30 (3): 498-514.  
- Laurent, R. D. 1988. “An Interest Rate-Based Indicator of Monetary Policy,” Federal Reserve Bank of Chicago Economic Perspectives, 3–14.  
- Shiller, R., 1990. “Speculative Prices and Popular Models” Real Estate Economics, volume 18(3) pages 253-273.  
- Phillipon, T., 2009. “The Bond Market’s q*” The Quarterly Journal of Economics, 124(3), 1101-1056.  
- Dang, T.V., G. Gorton, and B. Holmström, 2010. “Opacity and the Optimality of Debt in Liquidity Provision” working paper, MIT.  
- Peterson, B. and T. Roberts, 2016. “Household Risk Assessment Matrix”, Bank of Canada, Technical Report No. 106.

*References listed in the source PDF: wpiea2020042-print-pdf - REFERENCES*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2020/english/wpiea2020042-print-pdf.pdf_
