## CHAPTER 3 COMMERCIAL REAL ESTATE: FINANCIAL STABILITY RISkS duRING ThE COVId-19 CRISIS ANd BEYONd

## Source details

**Canonical URL:** [CHAPTER 3 COMMERCIAL REAL ESTATE: FINANCIAL STABILITY RISkS duRING ThE COVId-19 CRISIS ANd BEYONd](https://www.imf.org/-/media/files/publications/gfsr/2021/april/english/ch3.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/gfsr/2021/april/english/ch3.pdf.md)
- [Structured JSON version](/-/media/files/publications/gfsr/2021/april/english/ch3.pdf.json)

---

### Chapter 3 at a Glance
- Key near-term developments and segmental impacts:
  - The COVID-19 crisis has hit the commercial real estate sector hard and increased uncertainty about the outlook for some of its segments due to possible structural shifts in demand, warranting enhanced supervisory attention.
  - Commercial property transaction volumes and prices plummeted globally in the second quarter of 2020 as containment measures eroded economic activity; the sector has recovered somewhat since then, especially in Asia, but generally remains depressed.
  - Among major segments, retail, hotels, and offices have been the most affected; the industrial segment has fared relatively better.
  - Indicators:
    - Prices of publicly listed mortgage and equity real estate investment trusts showed signs of recovery in line with prices of other listed securities, reflecting unprecedented policy support since March 2020.
    - Net operating income dropped sharply in the hotel and retail segments.
    - CMBS delinquency rates increased sharply during the crisis, with rates for the retail and hotel segments reaching all-time highs in 2020:Q2.
- Valuation dynamics and misalignments:
  - While there was little evidence of large price misalignments at the onset of the pandemic, signs of overvaluation have emerged in some economies as actual prices have not fallen as much as prices implied by fundamentals.
  - In the run-up to the pandemic, the median commercial real estate price across many economies steadily increased; in Sweden and the United States, real commercial real estate prices almost doubled between 2009 and 2019.
  - Some segments (notably retail) were already facing structural headwinds from a shift toward e-commerce, putting downward pressure on revenues and prices (capital growth).
- Financial stability relevance and transmission channels:
  - Analysis covers 30 economies over 2000:Q1–2020:Q2: Australia, Austria, Belgium, Canada, China, the Czech Republic, Denmark, France, Germany, Hong Kong SAR, Hungary, Indonesia, Ireland, Italy, Japan, Korea, Malaysia, The Netherlands, New Zealand, Norway, Poland, Portugal, Singapore, South Africa, Spain, Sweden, Switzerland, Thailand, the United Kingdom, and the United States.
  - Three key channels:
    - Bank solvency channel: banks are exposed via CRE loans and CMBS holdings; borrower defaults and CMBS price drops weaken bank capital and may reduce credit supply.
    - Collateral channel: CRE properties pledged as collateral; lower CRE values raise loan-to-value ratios and risk-weighted assets, reducing regulatory capital ratios and curtailing corporate borrowing and investment.
    - Nonbank financial institution (NBFI) channel: insurers, pension funds, and investment funds hold CRE debt and equity; asset value declines, redemptions, and fire sales can amplify price declines and affect funding availability.
- Size and concentration of exposures (selected figures):
  - The commercial real estate sector had total assets of about 20 percent of GDP as of the end of 2019, on average, across major advanced and emerging market economies, up from 17 percent a decade ago.
  - Market shares by economy can be much higher: as high as 50 percent or more in Singapore, Sweden, and Switzerland.
  - In the United States and some European economies (for example, Estonia and Poland), direct lending related to commercial real estate constituted more than 50 percent of total bank lending to nonfinancial corporations in 2019.
  - In the United States, commercial real estate lending is highly concentrated among smaller banks (total assets < $100 billion): over 165 percent of their regulatory capital was committed to commercial real estate and construction lending in 2019, compared with 50 percent for large banks.
  - Listed firm leverage: the median debt-to-total-assets ratio is 35 percent for listed real estate firms versus 20 percent for other firms as of the end of 2019:Q4.
  - Loan-to-value ratios on new commercial real estate loans averaged about 60 percent in 2019 compared with 82 percent in 2007 in the United States and the European Union (per market contacts).
  - U.S. CMBS market context: annual issuance in the run-up to the global financial crisis reached about $230 billion; issuance fell to a few billion dollars in 2008–09 and dropped again during the COVID-19 crisis.
- Historical experience and potential losses:
  - Past crises show substantial CRE-related losses (examples cited include Sweden in the early 1990s, the U.S. savings and loan crisis, Ireland 2008–11, and the U.S. financial crisis of 2007–09).
  - During the U.S. 2007–09 crisis, the cumulative loss rate for CMBS was about 14 percent and for commercial real estate loans about 8 percent; banks with high CRE exposures had a much higher likelihood of failure.
- Policy implications and recommendations:
  - Near-term: Policy support to maintain the flow of credit to the nonfinancial corporate sector and to stimulate aggregate demand will help facilitate the recovery in the commercial real estate sector.
  - Medium-term / targeted actions:
    - To the extent that large price misalignments persist, policymakers should swiftly deploy targeted macroprudential measures to contain vulnerabilities in the sector as warranted.
    - Capital flow management measures could be considered under specific circumstances to limit potential risks from excessive cross-border inflows.
  - Supervisory priority: Enhanced supervisory attention is warranted given the increased uncertainty and potential for structural shifts in demand across segments (for example, retail, offices, hotels).

### Market size, returns, and valuation dynamics
- Nominal annual capital appreciation averaged about 3 percent globally.
- The capitalization rate (ratio of net operating income to commercial real estate prices) declined steadily since the global financial crisis to its lowest level since the global financial crisis.
- The decline in the capitalization rate has been broadly in line with the reduction in real long-term government bond yields; the spread between the two series has remained within a narrow range over the past 15 years.
- Real commercial property prices in most economies have risen above their levels before the global financial crisis.
- Retail performance was weak even before the pandemic.
- Pre-pandemic fair-value model results:
  - For the core sample, the average deviation of commercial real estate prices from fair values before the pandemic is estimated at about minus 2 percent.
  - Before the global financial crisis the estimated overvaluation was 8 percent.
  - For individual segments (offices, retail) price misalignments appear to have been limited before the pandemic for most economies in the sample.
  - In the United States, the sharp decline in aggregate demand and net operating income during 2020 put downward pressure on fair values, implying an overvaluation that large monetary easing did not fully offset.
- Pandemic-era misalignments and scenario analysis:
  - Commercial real estate price misalignments increased in 2020 despite a decline in commercial real estate prices; the median value across economies reached about 3.6 percent.
  - The increase in misalignment largely reflects deterioration in fundamentals such as a drop in aggregate demand and net operating income.
  - Structural shifts in demand are considered via a scenario: a persistent increase in vacancy rates for five years.
    - A permanent increase in the vacancy rate of 5 percentage points would result in a median drop in fair values of about 15 percent after five years.
    - The scenario calibrates a 5 percentage point vacancy-rate increase to be equivalent to what the United States experienced during the global financial crisis and abstracts from potential property repurposing.
- Linkages to macro-financial stability and growth-at-risk:
  - CRE prices are highly procyclical: the short-term cross-correlation between changes in real CRE prices and real GDP growth is strongly positive across economies.
  - A misalignment in CRE prices can amplify adverse shocks and increase downside risks to future GDP growth.
  - Empirical growth-at-risk findings:
    - In advanced economies, a one standard deviation increase in CRE price misalignment—corresponding to a negative deviation of the capitalization rate from its long-term trend by 10 basis points—is associated with an increase in downside risk of ½ percentage point in GDP in the short term and ¼ percentage point in the medium term.
    - For emerging market economies, the impact is about 0.2 percentage point in the short term.
  - The smaller estimated impact for emerging market economies may reflect the smaller size of the CRE market and lower leverage in the sector relative to advanced economies.
- Key implications and uncertainties:
  - The COVID-19 pandemic represents a large shock to CRE market fundamentals affecting both supply and demand; some factors are conjunctural while others may reflect structural changes in demand.
  - CRE prices may not fully reflect these structural changes given high uncertainty, complicating assessments of price misalignment.
  - Continued price misalignments in a low-rate environment could create risks of sharp future price corrections, particularly if combined with negative shocks to income growth, vacancy rates, capital inflows, or premature withdrawal of policy or lender support (such as loan extensions and deferred payment options).

### Advanced Economies: Impact of CRE Price
- Key empirical findings on macro‑financial transmission:
  - CRE is highly procyclical and higher CRE price misalignment increases downside risks to GDP growth in the short and medium terms.
  - A one standard deviation cumulative decline in local CRE prices of 16 percent over eight quarters (mild adverse scenario) implies for banks with high CRE exposures (ex ante CRE-loans-to-total-assets ratio at the 75th percentile; corresponding to 43 percentage points higher exposure):
    - Cumulative 8 percentage point increase in the CRE nonperforming loan ratio (90+ days overdue) over eight quarters.
    - Cumulative 2.5 percentage point increase in the net charge-off rate of CRE loans over eight quarters.
    - 12 percent drop in net revenues before provisioning.
    - 4.9 percent decline in total regulatory capital (relative to banks with no CRE loan exposure).
  - Estimated losses relative to banks’ risk-weighted assets before the shock average 14 basis points; they exceed 1 percentage point for banks with very high CRE exposures (defined as those in the top 3 percent for the ratio of CRE loans to total assets—smaller banks and community banks).
  - A structural shock (permanent increase in vacancy rates by 5 percentage points) would roughly double the impact on bank capital compared with the mild adverse scenario.
  - CRE price declines materially affect nonfinancial corporations via the collateral channel:
    - A one standard deviation decrease in the market value of real estate assets (market value normalized by PPE; standard deviation of this ratio is 1.4) implies a 21 percent decrease in the ratio of investment to the value of property, plant, and equipment.
    - Each additional $1 of real estate collateral increases investment by $0.03.
    - The investment effect is larger for financially constrained firms (defined as firms with total assets less than the 30th percentile; firms that do not pay dividends; firms with no Moody’s risk rating or with expected default frequency higher than the 70th percentile).
- Bank-level distributional effects and scenarios:
  - Banks with larger CRE loan exposures experience significantly higher CRE nonperforming loan ratios and higher loan charge-offs over the subsequent eight quarters; net revenues before provisioning and total regulatory capital are also lower (coefficients statistically significant at 10 percent or lower).
  - Distributional scenarios:
    - Mild adverse scenario: 16 percent cumulative drop in CRE prices over eight quarters with slow recovery afterward produces a distribution of eight‑quarter‑ahead projected capital losses concentrated in smaller and geographically concentrated banks (small bank defined as total assets never exceed $5 billion during 2001:Q1–2020:Q3; medium-sized if assets exceed $5 billion at least once but never exceed $100 billion; large if assets exceed $100 billion at least once).
    - Permanent vacancy shock: permanent increase in vacancy rate by 5 percentage points amplifies capital losses (about twice the impact on bank capital relative to the mild scenario).
- Macroprudential and policy effectiveness on CRE downside risks:
  - Measures considered:
    - Targeted CRE measures: caps on loan-to-value or debt-service-to-income ratios specific to CRE; higher risk weights or sectoral capital buffers for CRE exposures; limits on banks’ concentration in CRE; supervisory guidance on CRE lending.
    - Broader borrower-based measures: borrower limits targeting residential mortgages (loan-to-value, debt-service-to-income caps) that can spill over to multifamily CRE.
    - Capital flow management measures: restrictions on investments by nonresidents (ownership restrictions, higher stamp duties) and indices of capital inflow restrictiveness.
  - Estimated effects on downside risks to CRE price growth (5th percentile of future CRE price distribution; sample covers 30 economies over 2000:Q1–2019:Q4):
    - Tighter targeted CRE measures reduce downside risks to CRE prices by 0.26 percentage point a quarter in the near term.
    - Broader borrower-based macroprudential tightening reduces downside risks by about 2 percentage points (cumulative) in the medium and long term.
    - Capital flow management measures (overall capital inflow restrictiveness index and CRE-specific inflow restrictiveness) are associated with lower downside risks to CRE prices in advanced-economy samples where such measures have been applied.
  - Caveats:
    - Macroprudential measures generally apply to domestic banks and can be circumvented if CRE debt funding is borrowed directly from abroad or via nonbank financial institutions.
    - Measures targeting nonbank financial institutions are limited in practice; one example is the 2014 US credit risk retention standards for asset-backed securities.
    - Use of capital flow management measures to address financial stability risks should be considered only under specific circumstances (as outlined in IMF 2012, 2017).
- Conclusions and policy recommendations:
  - The CRE sector’s large size, debt dependence, and interconnectedness with the real economy make it highly relevant for domestic macro‑financial stability and warrant enhanced supervisory attention.
  - Key channels identified:
    - Bank solvency channel: CRE price declines raise CRE nonperforming loans and charge-offs, lower bank revenues and regulatory capital, with outsized effects for banks with concentrated CRE exposures (notably smaller and community banks).
    - Corporate collateral channel: declines in firms’ CRE market value reduce investment, especially for financially constrained firms, and tighten borrowing constraints (notably for long-term debt).
  - Policy guidance:
    - Consider employing targeted macroprudential measures to reduce CRE-specific vulnerabilities (caps, higher risk weights, sectoral buffers, concentration limits, supervisory guidance).
    - Consider broader borrower-based measures (including residential borrower constraints) to mitigate spillovers to multifamily CRE.
    - Where appropriate and under specific conditions, consider capital flow management measures to limit foreign-funded CRE inflows.
    - Maintain policy support in the near term to preserve the flow of credit to nonfinancial corporations, while using macroprudential tools to reduce balance-sheet vulnerabilities and lessen the risk of large downward CRE price corrections.

### Impact of a CRE-Specific Macroprudential Tightening Measure on downside risks and City-level outcomes
- Macroprudential and capital flow measures:
  - CRE-specific macroprudential tightening measures are effective in limiting downside risks to CRE prices in the short term.
  - Broader borrower-based tightening measures reduce downside risks to CRE prices in the medium and long terms.
  - Capital flow management measures appear to limit tail risks to CRE prices, with CRE-specific measures having a more pronounced effect.
  - Methodological note:
    - Dependent variable in panels is the 5th percentile of the future (average) CRE price growth distribution.
    - Panel 1: CRE-specific measures categorical variable: –1, 0, or 1 for loosening, no change, or tightening in a quarter.
    - Panel 2: Borrower-based policies based on a two-year rolling sum (+1 = tightening, 0 = no change, –1 = loosening) and purged of the credit-to-GDP ratio.
    - Panels 3 and 4: Changes in capital flow measures correspond to overall and real-estate-specific capital inflow restriction indices, based on a two-year rolling sum and purged of the capital-flow-to-GDP ratio.
    - Dotted lines in the figure indicate 90 percent confidence intervals.
- City-level CRE price developments during the COVID-19 pandemic (2020:Q2 highlights):
  - Sample: 64 cities in 11 economies.
  - Largest quarter-over-quarter price decline in 2020:Q2: Winnipeg, Canada, about 5½ percent.
  - Among “first-tier” cities: London recorded –1.2 percent; New York recorded –1 percent.
  - Retail segment extremes:
    - Largest retail price declines in 2020:Q2: Minneapolis, Minnesota, and Baltimore, Maryland, up to 9½ percent.
    - Retail price increases in 2020:Q2: Austin, Texas, and Fukuoka, Japan, increased by 4 percent.
  - Office segment examples:
    - Worst performing cities in office: Halifax, Canada, and Houston, Texas.
    - Best performing cities in office: Melbourne, Australia, and Philadelphia, Pennsylvania.
  - Factors associated with larger price declines:
    - Cities with above-median containment stringency recorded about a 0.6 percentage point larger price decline than other cities in 2020:Q2.
    - Lower workplace mobility is associated with larger price declines (work mobility index sourced from Google).
    - Smaller cities, cities with lower CRE capital growth before the pandemic, and cities with a sharper decline in market liquidity suffered larger price declines.
    - Greater fiscal policy support at the national level (sum of equity, loans, guarantees, additional spending and forgone revenue as percent of GDP) was generally associated with smaller CRE price declines.
  - Urban versus suburban:
    - During 2010–19, urban CRE prices increased on average 1.4 percent larger than suburban areas; during the pandemic CRE price declines were slightly larger in urban areas than in suburbs.
- Policy guidance and supervisory actions:
  - Maintain borrower support measures until the economic recovery is firmly established; examples include debt repayment relief, credit guarantees, and direct support for viable firms.
  - Encourage restructuring or liquidation of nonviable firms with high solvency and liquidity risks.
  - Consider stress testing exercises embedding large declines in CRE prices to ensure banking sector resilience and inform adequacy of capital buffers for CRE exposures.
  - Supervisors should review banks’ CRE valuation assumptions and ensure provisions are adequate.
  - Deploy targeted macroprudential tools once structural changes from the pandemic are clearer to address excessive financial risk taking and prevent persistent large price misalignments; examples include borrower-based measures such as loan-to-value (LTV) and debt-service-to-income (DSTI) ratios.
  - Timing considerations:
    - Optimal timing depends on the economy-specific pace of the recovery and the degree of financial vulnerabilities.
    - Possible implementation lags argue for early action in some circumstances.
  - Capital flow measures:
    - Consider commercial-real-estate-specific capital flow management measures if a surge in capital flows into the sector poses systemic financial risks that cannot be addressed with other tools; phase out once risks subside.
  - Nonbank financial institutions:
    - Broaden the reach of macroprudential tools, grant macroprudential powers to relevant supervisors, and enhance data collection.
    - Where macroprudential powers are absent, supervisors can reduce structural vulnerabilities (for example, stricter rules for property investment funds or linking life insurers’ capital requirements to property type or LTV and DSTI ratios).

### Cross-border CRE investment developments and vulnerabilities
- Trends and magnitudes:
  - Cross-border commercial real estate investments averaged about $270 billion a year during 2014–19.
  - Cross-border flows declined sharply following the global financial crisis, recovered toward 2015, and dropped again in 2020 due to the COVID-19 crisis.
- 2020 outcomes:
  - Cross-border investments in 2020 fell, with office and retail segments declining the most: office down 48 percent and retail down 65 percent during the crisis.
  - Frontier markets in Africa and the Middle East experienced declines in 2020 ranging from 5 percent to 100 percent relative to 2019.
  - Economies that relied entirely on foreign investors recorded no CRE investment in 2020.
- Recipient patterns:
  - A large share of global cross-border CRE investment is in advanced economies; as a share of total CRE investment within economies, cross-border investment is relatively larger in emerging markets.
  - Top emerging market recipients in the 2018–20 period included China and Poland; both experienced a slowdown in 2020.
- Institutional investor influence and synchronization:
  - Institutional investors (pension funds and insurance companies) have increased their share in cross-border investment flows, especially in Europe and Asia and the Pacific.
  - Cross-border institutional investors tend to be more fickle in large global shocks than direct investors, increasing domestic market synchronization with global CRE cycles.
  - International price synchronization spiked during the pandemic, building on an increasing trend since the global financial crisis; the synchronization metric is normalized with a maximum value equal to 100.
- Policy implication:
  - If cross-border CRE investments increase market vulnerabilities and threaten financial stability, policymakers might consider policies that reduce demand by foreign buyers in some circumstances.

### Box: The US Commercial Mortgage-Backed Securities Market during the COVID-19 Crisis
- Market disruption and issuance collapse:
  - In March 2020, the CMBS market in the United States was severely disrupted as stress in funding markets reverberated through the CRE sector.
  - Funding costs increased sharply, with spreads on BBB-rated commercial mortgage-backed securities and these securities’ indices jumping sharply.
  - Monthly CMBS issuance fell from $14.8 billion in February to $0.3 billion in April 2020.
  - Issuance of agency CMBS rebounded during the second quarter of 2020, but the volume of year-to-date cumulative issuance at the end of June 2020 remained lower than for the corresponding period in 2019.
- Federal Reserve interventions:
  - The Federal Reserve bought almost $9.3 billion in securities issued by Fannie Mae, Freddie Mac, and Ginnie Mae during the second quarter (March–July 2020).
  - As a result, spreads of agency securities tightened significantly and returned to their precrisis level after a few weeks.
  - In early April 2020 the Federal Reserve included nonagency AAA CMBS in its Term Asset-Backed Securities Loan Facility (TALF 2.0) program, but did not intervene more broadly in the nonagency CMBS market.
- Agency versus nonagency segments and uneven recovery:
  - Agency CMBS (primarily securitizations of multifamily residential properties) saw a rapid tightening of spreads following Fed purchases and a resumption of credit flows to the multifamily housing sector.
  - Recovery in nonagency segments was more uneven: the spread between BBB-rated and AAA-rated securities continued to widen over the second half of 2020.
  - The CARES Act tied much of federal mortgage relief to residential mortgages (including the multifamily segment), but provided no explicit protection to nonresidential CRE borrowers.
  - Early in the Fed’s purchase program the total amount of bids submitted greatly exceeded the announced maximum purchase amount at the weekly auction; the difference declined rapidly thereafter, indicating market recovery.
- Policy considerations and financial-stability risks:
  - Key policy question in the nonagency CMBS market: to what extent should policies mitigate private sector losses that could pose systemic risk (similar to the 2007–09 financial crisis)?
  - Dodd-Frank credit risk retention requirements (launched in 2014) require issuers to retain at least 5 percent of any security they issue and may have reduced overall risk and improved lending standards for CMBS loans.
  - A sluggish recovery in CRE markets may result in greater losses than current initiatives can address.
  - Stress in the CMBS market could spill over to other financial market segments, leading to liquidity or potential solvency problems for banks and nonbank financial institutions, especially those with large exposures to CMBS.
  - Indirect support measures cited include the Main Street Lending Program and the Small Business Administration’s Paycheck Protection Program.

*Source: IMF staff summary of Chapter 3, “Commercial Real Estate: Financial Stability Risks during the COVID-19 Crisis and Beyond,” Global Financial Stability Report, April 2021.*

### Chapter 3 at a Glance

### Chapter 3 at a Glance

### Key near-term developments and segmental impacts
- The COVID-19 crisis has hit the commercial real estate sector hard and increased uncertainty about the outlook for some of its segments due to possible structural shifts in demand, warranting enhanced supervisory attention.
- Commercial property transaction volumes and prices plummeted globally in the second quarter of 2020 as containment measures eroded economic activity; the sector has recovered somewhat since then, especially in Asia, but generally remains depressed.
- Among major segments, retail, hotels, and offices have been the most affected; the industrial segment has fared relatively better.
- Indicators:
  - Prices of publicly listed mortgage and equity real estate investment trusts showed signs of recovery in line with prices of other listed securities, reflecting unprecedented policy support since March 2020.
  - Net operating income dropped sharply in the hotel and retail segments.
  - CMBS delinquency rates increased sharply during the crisis, with rates for the retail and hotel segments reaching all-time highs in 2020:Q2.

### Valuation dynamics and misalignments
- While there was little evidence of large price misalignments at the onset of the pandemic, signs of overvaluation have emerged in some economies as actual prices have not fallen as much as prices implied by fundamentals.
- In the run-up to the pandemic, the median commercial real estate price across many economies steadily increased; in Sweden and the United States, real commercial real estate prices almost doubled between 2009 and 2019.
- Some segments (notably retail) were already facing structural headwinds from a shift toward e-commerce, putting downward pressure on revenues and prices (capital growth).

### Financial stability relevance and transmission channels
- The chapter analyzes commercial real estate for 30 economies over a 20-year period, from 2000:Q1 to 2020:Q2. The core sample includes: Australia, Austria, Belgium, Canada, China, the Czech Republic, Denmark, France, Germany, Hong Kong SAR, Hungary, Indonesia, Ireland, Italy, Japan, Korea, Malaysia, The Netherlands, New Zealand, Norway, Poland, Portugal, Singapore, South Africa, Spain, Sweden, Switzerland, Thailand, the United Kingdom, and the United States.
- Three key channels through which a decline in commercial real estate prices affects financial stability:
  - Bank solvency channel: banks are exposed via CRE loans and CMBS holdings; borrower defaults and CMBS price drops weaken bank capital and may reduce credit supply.
  - Collateral channel: CRE properties pledged as collateral; lower CRE values raise loan-to-value ratios and risk-weighted assets, reducing regulatory capital ratios and curtailing corporate borrowing and investment.
  - Nonbank financial institution (NBFI) channel: insurers, pension funds, and investment funds hold CRE debt and equity; asset value declines, redemptions, and fire sales can amplify price declines and affect funding availability.

### Size and concentration of exposures (selected figures)
- The commercial real estate sector had total assets of about 20 percent of GDP as of the end of 2019, on average, across major advanced and emerging market economies, up from 17 percent a decade ago.
- Market shares by economy can be much higher: as high as 50 percent or more in Singapore, Sweden, and Switzerland.
- In the United States and some European economies (for example, Estonia and Poland), direct lending related to commercial real estate constituted more than 50 percent of total bank lending to nonfinancial corporations in 2019.
- In the United States, commercial real estate lending is highly concentrated among smaller banks (total assets < $100 billion): over 165 percent of their regulatory capital was committed to commercial real estate and construction lending in 2019, compared with 50 percent for large banks.
- Listed firm leverage: the median debt-to-total-assets ratio is 35 percent for listed real estate firms versus 20 percent for other firms as of the end of 2019:Q4.
- Loan-to-value ratios on new commercial real estate loans averaged about 60 percent in 2019 compared with 82 percent in 2007 in the United States and the European Union (per market contacts).
- U.S. CMBS market context: annual issuance in the run-up to the global financial crisis reached about $230 billion; issuance fell to a few billion dollars in 2008–09 and dropped again during the COVID-19 crisis.

### Historical experience and potential losses
- Past crises show substantial CRE-related losses (examples cited include Sweden in the early 1990s, the U.S. savings and loan crisis, Ireland 2008–11, and the U.S. financial crisis of 2007–09).
- During the U.S. 2007–09 crisis, the cumulative loss rate for CMBS was about 14 percent and for commercial real estate loans about 8 percent; banks with high CRE exposures had a much higher likelihood of failure.

### Policy implications and recommendations
- Near-term: Policy support to maintain the flow of credit to the nonfinancial corporate sector and to stimulate aggregate demand will help facilitate the recovery in the commercial real estate sector.
- Medium-term / targeted actions:
  - To the extent that large price misalignments persist, policymakers should swiftly deploy targeted macroprudential measures to contain vulnerabilities in the sector as warranted.
  - Capital flow management measures could be considered under specific circumstances to limit potential risks from excessive cross-border inflows.
- Supervisory priority: Enhanced supervisory attention is warranted given the increased uncertainty and potential for structural shifts in demand across segments (for example, retail, offices, hotels).

*Source: IMF staff summary of Chapter 3, “Commercial Real Estate: Financial Stability Risks during the COVID-19 Crisis and Beyond,” Global Financial Stability Report, April 2021.*

### CHAPTER 3 COMMERCIAL REAL ESTATE: FINANCIAL STABILITY RISkS duRING ThE COVId-19 CRISIS ANd BEYONd

### CHAPTER 3 COMMERCIAL REAL ESTATE: FINANCIAL STABILITY RISkS duRING ThE COVId-19 CRISIS ANd BEYONd

### Market size, returns, and valuation dynamics
- Nominal annual capital appreciation averaged about 3 percent globally.
- The capitalization rate (ratio of net operating income to commercial real estate prices) declined steadily since the global financial crisis to its lowest level since the global financial crisis.
- The decline in the capitalization rate has been broadly in line with the reduction in real long-term government bond yields; the spread between the two series has remained within a narrow range over the past 15 years.
- Real commercial property prices in most economies have risen above their levels before the global financial crisis.
- Retail performance was weak even before the pandemic.

### Pre-pandemic valuation assessment (fair-value model)
- For the core sample, the average deviation of commercial real estate prices from fair values before the pandemic is estimated at about minus 2 percent.
- By contrast, before the global financial crisis the estimated overvaluation was 8 percent.
- For individual segments (offices, retail) price misalignments appear to have been limited before the pandemic for most economies in the sample.
- In the United States, the sharp decline in aggregate demand and net operating income during 2020 put downward pressure on fair values, implying an overvaluation that large monetary easing did not fully offset.

### Pandemic-era misalignments and scenario analysis
- Commercial real estate price misalignments increased in 2020 despite a decline in commercial real estate prices; the median value across economies reached about 3.6 percent.
- The increase in misalignment largely reflects deterioration in fundamentals such as a drop in aggregate demand and net operating income.
- Structural shifts in demand (for example, a shift toward e-commerce and teleworking) are considered via a scenario: a persistent increase in vacancy rates for five years.
- A permanent increase in the vacancy rate of 5 percentage points would result in a median drop in fair values of about 15 percent after five years.
- The scenario calibrates a 5 percentage point vacancy-rate increase to be equivalent to what the United States experienced during the global financial crisis and abstracts from potential property repurposing.

### Linkages to macro-financial stability and growth-at-risk
- Commercial real estate prices are highly procyclical: the short-term cross-correlation between changes in real commercial real estate prices and real GDP growth is strongly positive across economies.
- A misalignment in commercial real estate prices can amplify adverse shocks and increase downside risks to future GDP growth.
- Empirical growth-at-risk findings:
  - In advanced economies, a one standard deviation increase in commercial real estate price misalignment—corresponding to a negative deviation of the capitalization rate from its long-term trend by 10 basis points—is associated with an increase in downside risk of ½ percentage point in GDP in the short term and ¼ percentage point in the medium term.
  - For emerging market economies, the impact is about 0.2 percentage point in the short term.
- The smaller estimated impact for emerging market economies may reflect the smaller size of the commercial real estate market and lower leverage in the sector relative to advanced economies.

### Key implications and uncertainties
- The COVID-19 pandemic represents a large shock to commercial real estate market fundamentals affecting both supply and demand; some factors are conjunctural while others may reflect structural changes in demand.
- Commercial real estate prices may not fully reflect these structural changes given high uncertainty, complicating assessments of price misalignment.
- Continued price misalignments in a low-rate environment could create risks of sharp future price corrections, particularly if combined with negative shocks to income growth, vacancy rates, capital inflows, or premature withdrawal of policy or lender support (such as loan extensions and deferred payment options).

*Source: CHAPTER 3 COMMERCIAL REAL ESTATE: FINANCIAL STABILITY RISkS duRING ThE COVId-19 CRISIS ANd BEYONd (IMF, April 2021).*

### 2. Advanced Economies: Impact of CRE Price

### ch3 - 2. Advanced Economies: Impact of CRE Price

### Key empirical findings on macro‑financial transmission
- Commercial real estate (CRE) is highly procyclical and higher CRE price misalignment increases downside risks to GDP growth in the short and medium terms.
- A one standard deviation cumulative decline in local CRE prices of 16 percent over eight quarters (mild adverse scenario) implies for banks with high CRE exposures (ex ante CRE-loans-to-total-assets ratio at the 75th percentile; corresponding to 43 percentage points higher exposure):
  - Cumulative 8 percentage point increase in the CRE nonperforming loan ratio (90+ days overdue) over eight quarters.
  - Cumulative 2.5 percentage point increase in the net charge-off rate of CRE loans over eight quarters.
  - 12 percent drop in net revenues before provisioning.
  - 4.9 percent decline in total regulatory capital (relative to banks with no CRE loan exposure).
- Estimated losses relative to banks’ risk-weighted assets before the shock average 14 basis points; they exceed 1 percentage point for banks with very high CRE exposures (defined as those in the top 3 percent for the ratio of CRE loans to total assets—smaller banks and community banks).
- A structural shock (permanent increase in vacancy rates by 5 percentage points) would roughly double the impact on bank capital compared with the mild adverse scenario.
- CRE price declines materially affect nonfinancial corporations via the collateral channel:
  - A one standard deviation decrease in the market value of real estate assets (market value normalized by PPE; standard deviation of this ratio is 1.4) implies a 21 percent decrease in the ratio of investment to the value of property, plant, and equipment.
  - Each additional $1 of real estate collateral increases investment by $0.03.
  - The investment effect is larger for financially constrained firms (defined as firms with total assets less than the 30th percentile; firms that do not pay dividends; firms with no Moody’s risk rating or with expected default frequency higher than the 70th percentile).

### Bank-level distributional effects and scenarios
- Impact on loan portfolio performance and revenues:
  - Banks with larger CRE loan exposures experience significantly higher CRE nonperforming loan ratios and higher loan charge-offs over the subsequent eight quarters; net revenues before provisioning and total regulatory capital are also lower (coefficients statistically significant at 10 percent or lower).
- Distributional outcomes under scenarios:
  - Mild adverse scenario: 16 percent cumulative drop in CRE prices over eight quarters with slow recovery afterward produces a distribution of eight‑quarter‑ahead projected capital losses concentrated in smaller and geographically concentrated banks (small bank defined as total assets never exceed $5 billion during 2001:Q1–2020:Q3; medium-sized if assets exceed $5 billion at least once but never exceed $100 billion; large if assets exceed $100 billion at least once).
  - Permanent vacancy shock: permanent increase in vacancy rate by 5 percentage points amplifies capital losses (about twice the impact on bank capital relative to the mild scenario).

### Macroprudential and policy effectiveness on CRE downside risks
- Types of measures considered:
  - Targeted measures specific to CRE: caps on loan-to-value or debt-service-to-income ratios specific to CRE; higher risk weights or sectoral capital buffers for CRE exposures; limits on banks’ concentration in CRE; supervisory guidance on CRE lending.
  - Broader borrower-based measures: borrower limits targeting residential mortgages (loan-to-value, debt-service-to-income caps) that can spill over to multifamily CRE.
  - Capital flow management measures: restrictions on investments by nonresidents (ownership restrictions, higher stamp duties) and indices of capital inflow restrictiveness.
- Estimated effects on downside risks to CRE price growth (5th percentile of future CRE price distribution; sample covers 30 economies over 2000:Q1–2019:Q4):
  - Tighter targeted CRE measures reduce downside risks to CRE prices by 0.26 percentage point a quarter in the near term.
  - Broader borrower-based macroprudential tightening reduces downside risks by about 2 percentage points (cumulative) in the medium and long term.
  - Capital flow management measures (overall capital inflow restrictiveness index and CRE-specific inflow restrictiveness) are associated with lower downside risks to CRE prices in advanced-economy samples where such measures have been applied.
- Caveats and implementation considerations:
  - Macroprudential measures generally apply to domestic banks and can be circumvented if CRE debt funding is borrowed directly from abroad or via nonbank financial institutions.
  - Measures targeting nonbank financial institutions are limited in practice; one example is the 2014 US credit risk retention standards for asset-backed securities.
  - Use of capital flow management measures to address financial stability risks should be considered only under specific circumstances (as outlined in IMF 2012, 2017).

### Conclusions and policy recommendations
- The CRE sector’s large size, debt dependence, and interconnectedness with the real economy make it highly relevant for domestic macro‑financial stability and warrant enhanced supervisory attention.
- Key channels identified:
  - Bank solvency channel: CRE price declines raise CRE nonperforming loans and charge-offs, lower bank revenues and regulatory capital, with outsized effects for banks with concentrated CRE exposures (notably smaller and community banks).
  - Corporate collateral channel: declines in firms’ CRE market value reduce investment, especially for financially constrained firms, and tighten borrowing constraints (notably for long-term debt).
- Policy guidance implied by the analysis:
  - Consider employing targeted macroprudential measures to reduce CRE-specific vulnerabilities (caps, higher risk weights, sectoral buffers, concentration limits, supervisory guidance).
  - Consider broader borrower-based measures (including residential borrower constraints) to mitigate spillovers to multifamily CRE.
  - Where appropriate and under specific conditions, consider capital flow management measures to limit foreign-funded CRE inflows.
  - Maintain policy support in the near term to preserve the flow of credit to nonfinancial corporations, while using macroprudential tools to reduce balance-sheet vulnerabilities and lessen the risk of large downward CRE price corrections.

*International Monetary Fund | April 2021 — Chapter 3 excerpt*

### 1. Impact of a CRE-Specific Macroprudential Tightening Measure on

### ch3 - 1. Impact of a CRE-Specific Macroprudential Tightening Measure on

### Impact of macroprudential and capital flow measures on downside risks to CRE prices
- CRE-specific macroprudential tightening measures are effective in limiting downside risks to CRE prices in the short term.
- Broader borrower-based tightening measures reduce downside risks to CRE prices in the medium and long terms.
- Capital flow management measures appear to limit tail risks to CRE prices, with CRE-specific measures having a more pronounced effect.
- Methodological note from source:
  - The dependent variable in all four panels is defined as the 5th percentile of the future (average) commercial real estate (CRE) price growth distribution.
  - Panel 1: CRE-specific measures are a categorical variable taking values –1, 0, or 1 for loosening, no change, or tightening in a quarter.
  - Panel 2: All borrower-based macroprudential policies are based on a two-year rolling sum of individual measures (+1 = tightening, 0 = no change, –1 = loosening) and are purged of the credit-to-GDP ratio.
  - Panels 3 and 4: Changes in capital flow management measures correspond to overall and real-estate-specific capital inflow restriction indices, based on a two-year rolling sum (+1 = tightening, 0 = no change, –1 = loosening) and purged of the capital-flow-to-GDP ratio.
  - Dotted lines in the figure indicate 90 percent confidence intervals.
- Sources cited for the analysis: Haver Analytics; MSCI Real Estate; and IMF staff calculations.

### City-level CRE price developments during the COVID-19 pandemic (2020:Q2 highlights)
- Overall city-level variation:
  - Sample: 64 cities in 11 economies.
  - Largest estimated quarter-over-quarter price decline in 2020:Q2: Winnipeg, Canada, about 5½ percent.
  - Among “first-tier” cities: London recorded –1.2 percent; New York recorded –1 percent.
- Retail segment extremes:
  - Largest retail price declines in 2020:Q2: Minneapolis, Minnesota, and Baltimore, Maryland, up to 9½ percent.
  - Retail price increases in 2020:Q2: Austin, Texas, and Fukuoka, Japan, increased by 4 percent.
- Office segment examples:
  - Worst performing cities in office: Halifax, Canada, and Houston, Texas.
  - Best performing cities in office: Melbourne, Australia, and Philadelphia, Pennsylvania.
- Factors associated with larger price declines:
  - Cities with above-median containment stringency recorded about a 0.6 percentage point larger price decline than other cities in 2020:Q2.
  - Lower workplace mobility is associated with larger price declines (work mobility index sourced from Google).
  - Smaller cities, cities with lower CRE capital growth before the pandemic, and cities with a sharper decline in market liquidity suffered larger price declines.
  - Greater fiscal policy support at the national level (sum of equity, loans, guarantees, additional spending and forgone revenue as percent of GDP) was generally associated with smaller CRE price declines.
- Urban versus suburban:
  - During 2010–19, urban CRE prices increased on average 1.4 percent larger than suburban areas; during the pandemic CRE price declines were slightly larger in urban areas than in suburbs.

### Policy guidance and supervisory actions
- Maintain borrower support measures until the economic recovery is firmly established; examples include debt repayment relief, credit guarantees, and direct support for viable firms.
- Encourage restructuring or liquidation of nonviable firms with high solvency and liquidity risks.
- Consider stress testing exercises embedding large declines in commercial real estate prices to ensure banking sector resilience and inform adequacy of capital buffers for CRE exposures.
- Supervisors should review banks’ CRE valuation assumptions and ensure provisions are adequate.
- Deploy targeted macroprudential tools once structural changes from the pandemic are clearer to address excessive financial risk taking and prevent persistent large price misalignments; examples include borrower-based measures such as loan-to-value (LTV) and debt-service-to-income (DSTI) ratios.
- Timing considerations:
  - Optimal timing depends on the economy-specific pace of the recovery and the degree of financial vulnerabilities.
  - Possible implementation lags argue for early action in some circumstances.
- Capital flow measures:
  - Given significant cross-border CRE investor presence in some jurisdictions, consider commercial-real-estate-specific capital flow management measures if a surge in capital flows into the sector poses systemic financial risks that cannot be addressed with other tools.
  - Such measures should be phased out once the risks subside.
- Nonbank financial institutions:
  - Address CRE-related systemic risks from nonbank financial institutions by broadening the reach of macroprudential tools, granting macroprudential powers to relevant supervisors, and enhancing data collection.
  - Where macroprudential powers are absent, supervisors can reduce structural vulnerabilities (for example, stricter rules for property investment funds or linking life insurers’ capital requirements to property type or LTV and DSTI ratios).

### Cross-border CRE investment developments and vulnerabilities
- Trends and magnitudes:
  - Cross-border commercial real estate investments averaged about $270 billion a year during 2014–19.
  - Cross-border flows declined sharply following the global financial crisis, recovered toward 2015, and dropped again in 2020 due to the COVID-19 crisis.
- 2020 outcomes:
  - Cross-border investments in 2020 fell, with office and retail segments declining the most: office down 48 percent and retail down 65 percent during the crisis.
  - Frontier markets in Africa and the Middle East experienced declines in 2020 ranging from 5 percent to 100 percent relative to 2019.
  - Economies that relied entirely on foreign investors recorded no CRE investment in 2020.
- Recipient patterns:
  - A large share of global cross-border CRE investment is in advanced economies; as a share of total CRE investment within economies, cross-border investment is relatively larger in emerging markets.
  - Top emerging market recipients in the 2018–20 period included China and Poland; both experienced a slowdown in 2020.
- Institutional investor influence and synchronization:
  - Institutional investors (pension funds and insurance companies) have increased their share in cross-border investment flows, especially in Europe and Asia and the Pacific.
  - Cross-border institutional investors tend to be more fickle in large global shocks than direct investors, increasing domestic market synchronization with global CRE cycles.
  - International price synchronization spiked during the pandemic, building on an increasing trend since the global financial crisis; the synchronization metric is normalized with a maximum value equal to 100.
- Policy implication:
  - If cross-border CRE investments increase market vulnerabilities and threaten financial stability, policymakers might consider policies that reduce demand by foreign buyers in some circumstances.

*Sources: Haver Analytics; MSCI Real Estate; IMF staff calculations; and text from the IMF chapter content provided.*

### Box 3.2 (continued)

### Box 3.3. The US Commercial Mortgage-Backed Securities Market during the COVID-19 Crisis

### Market disruption and issuance collapse
- In March 2020, the commercial mortgage-backed securities (CMBS) market in the United States was severely disrupted as stress in funding markets reverberated through the commercial real estate sector.
- Funding costs increased sharply, with spreads on BBB-rated commercial mortgage-backed securities and these securities’ indices jumping sharply.
- Monthly CMBS issuance fell from $14.8 billion in February to $0.3 billion in April 2020.
- Issuance of agency CMBS rebounded during the second quarter of 2020, but the volume of year-to-date cumulative issuance at the end of June 2020 remained lower than for the corresponding period in 2019.

### Federal Reserve interventions
- The Federal Reserve stepped into the agency CMBS market to prevent a collapse, buying almost $9.3 billion in securities issued by Fannie Mae, Freddie Mac, and Ginnie Mae during the second quarter (March–July 2020).
- As a result of these interventions, spreads of agency securities tightened significantly and returned to their precrisis level after a few weeks.
- In early April 2020 the Federal Reserve included nonagency AAA CMBS in its Term Asset-Backed Securities Loan Facility (TALF 2.0) program, but did not intervene more broadly in the nonagency CMBS market.

### Agency versus nonagency segments and uneven recovery
- Agency CMBS (primarily securitizations of multifamily residential properties) saw a rapid tightening of spreads following Fed purchases and a resumption of credit flows to the multifamily housing sector.
- Recovery in nonagency segments was more uneven: the spread between BBB-rated and AAA-rated securities continued to widen over the second half of 2020.
- The Coronavirus Aid, Relief, and Economic Security (CARES) Act tied much of federal mortgage relief to residential mortgages (including the multifamily segment), but provided no explicit protection to nonresidential commercial real estate borrowers.
- Early in the Fed’s purchase program the total amount of bids submitted greatly exceeded the announced maximum purchase amount at the weekly auction; the difference declined rapidly thereafter, indicating market recovery.

### Policy considerations and financial-stability risks
- The relevant policy question in the nonagency CMBS market differs from the residential mortgage market: to what extent should policies mitigate private sector losses that could pose systemic risk (similar to the 2007–09 financial crisis)?
- Previous regulatory reforms such as Dodd-Frank credit risk retention requirements (launched in 2014) require issuers to retain at least 5 percent of any security they issue on their books and may have reduced overall risk and improved lending standards for CMBS loans.
- Nonetheless, a sluggish recovery in commercial real estate markets may result in greater losses than current initiatives can address.
- Stress in the CMBS market could spill over to other financial market segments, leading to liquidity or potential solvency problems for banks and nonbank financial institutions, especially those with large exposures to commercial mortgage-backed securities.
- Indirect support measures cited include the Main Street Lending Program (offers loans with deferred repayments for smaller companies) and the Small Business Administration’s Paycheck Protection Program.

*Source: Box 3.3 (continued), ch3 - Box 3.2 (continued), Global Financial Stability Report: Preempting a Legacy of Vulnerabilities, April 2021.*

---


_Source: https://www.imf.org/-/media/files/publications/gfsr/2021/april/english/ch3.pdf_