## 2.1  Commercial Property Indices

## Source details

**Canonical URL:** [2.1  Commercial Property Indices](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf.json)

---

### Commercial property indices
- Green Street Commercial Property Price Index: appraisal-based index covering the period since 1998; focuses on properties owned by Real Estate Investment Trusts; sector weights: retail (20%), office (17.5%), apartment (15%), health care (15%), industrial (10%), lodging (7.5%), other sectors (15%).
- RCA US CRE property index (via MSCI CPPI US report): covers the period since 2000 and captures traded US commercial property in the apartment, retail, industrial and office segments.

### Commercial property transactions (ZTRAX)
- Data source: Zillow’s ZTRAX database with over 400 million public deed records across 2,750 U.S. counties; contains deed transfers, mortgages, foreclosures, property tax delinquencies, building square footage, land surface area, zip code, sale address, mortgage amount, sale price, loan amount.
- Coverage: data available for approximately 150 million parcels across 3,100 counties; final dataset includes 30-40 states that contain reliable, representative data and account for more than 80 percent of the U.S. population between 1994-2020.
- Transaction filtering: only transactions greater than $250,000 considered, resulting in around 1.3 million real estate sale transactions.
- Aggregation and classification: daily transaction-level information aggregated by quarter and zip-code; transactions identified by land usage and aggregated into six CRE types: retail, office, industrial, multi-family living units, lodging, other.
- Population density proxy: 2013 NCHS urban-rural classification aggregated at county level with six categories: (1) Large central metropolitan; (2) Large fringe metro; (3) Medium metro; (4) Small metro; (5) Micropolitan; (6) Noncore.

### Local economic activity and controls
- State and national indicators included: state-level GDP growth, population growth, inflation, imports and exports, cost of entry (general corporation license and franchise tax year-on-year growth), business applications growth, rental vacancy rates (year-on-year growth), private sector net job creation.
- Financial conditions: National Financial Conditions Index (NFCI) used to capture financial conditions; NFCI interacted with a dummy of household indebtedness to test heterogeneous state sensitivity.
- Household indebtedness interaction: HHDebt dummy constructed later as part of panel regressions.

### SafeGraph validation (local economic activity)
- Data source: SafeGraph business listings and footfall data covering over 6 million POIs in U.S. and Canada; raw dataset contains 150 million observations; after cleaning, almost 95 million observations representing 3,098 U.S. counties with an average of 125 POIs and a median of 21 POIs per county.
- Sector aggregation: POIs aggregated into five sectors using business activity codes: retail, auto, restaurants, manufacturing, wholesale trade.
- Normalization: coverage increased over time, data normalized to obtain accurate foot-traffic counts.
- Validation vs. Census: correlation of aggregated monthly visits (SafeGraph) with monthly sales (Census):
  - restaurant sector correlation: in excess of 90%;
  - retail sales correlation: around 70%;
  - manufacturing sector correlation: around 70%;
  - auto sector correlation: in excess of 60%;
  - wholesale trade correlation: in excess of 60%.

### Commercial Mortgage-Backed Securities (CMBX/CMBX spreads)
- CMBX contracts: typically CDS on an underlying portfolio of 25 CMBS deals; pricing available at daily frequency; data sourced through JP Morgan’s DataQuery.
- Tranche structure: CMBX AAA references super-senior CMBS with credit enhancement of about 30%; CMBX AJ, AA, A, BBB, BBB- refer to increasingly lower seniority tranches of same portfolio.
- Loan-pool exposure: earlier cohorts more heavily exposed to retail; collateralized retail asset share decreasing from around 40% to 25% in latest cohort.
- Delinquency differences: CMBX 6 delinquency rate in June 2020 is double compared to CMBX 12.
- Crisis reactions in spreads:
  - June 2008 to December 2008: average spread increased by roughly 12 percentage points, or 2.5 times higher than the long-run average.
  - December 2019 to June 2020: different impact pattern across tranches—financial crisis affected low-risk tranches most, Covid-19 mainly impacted higher-risk tranches; CMBX 6 spread adjustment slightly more than double that for CMBX 12.

### Prices and volumes — aggregate and by property type
- Validation: aggregated ZTRAX CRE price data compared to Green Street and RCA indices for 2000-2021 to verify similar behavior.
- Long-run price evolution (two decades): CRE prices generally doubled or tripled, with slightly stronger growth for industrial spaces.
- Post-GFC and pre-pandemic: prices corrected after the GFC (notably retail and multi-family) but recovered and reached new highs before the pandemic.
- Volumes: transaction counts drop substantially around the two major crises; transaction volumes peaked at the end of 2016 for most CRE types while prices continued to rise.
- Pandemic effects:
  - volumes dropped significantly for all segments during the pandemic;
  - lodging, retail, multifamily prices affected the most;
  - industrial and other segments remained relatively stable.
- Liquidity dynamics: during crises liquidity dries up, absorbing price shocks; when liquidity rebounds, prices recover.

### Spatial heterogeneity and gradients
- Estimation approach: regress ln(property price per built surface, per land surface, or ln price) on hedonic controls, UrbanCDC (1 to 6), interaction UrbanCDC*Γt, year (Γt) and county (Φc) fixed effects.
- UrbanCDC: ordered categorical variable 1 to 6 from highest to lowest population density.
- Findings:
  - Clear trend of higher prices close to urban core and lower prices outside across all property types.
  - Financial crisis (2008) and Covid-19 (2020) show similar short-term inversion patterns.
  - Industrial and office gradients: increasing in 2009, decreasing in 2020.
  - Multifamily: strong rebound of spatial gradient after pandemic, from -0.2 to roughly -0.16 within a single year.
  - Early 2000s vs. early 2010s: estimated coefficients suggest prices were around 30% higher in the most rural areas compared to the city core in the early 2000s; in the early 2010s they become 30% lower in the most rural areas relative to the city core.

### Risk pricing and market-perceived shock differences
- Crisis differential: 2008 crisis perceived as pervasive across debt holders (affected low-risk tranches most); Covid-19 perceived to impact more heterogeneously, increasing default risk mainly in vulnerable sectors and higher-risk tranches.
- Loan-cohort exposure and delinquencies: earlier cohorts concentrated in retail (higher exposure to lockdown impacts); CMBX 6 delinquency in June 2020 double that of CMBX 12, aligning with spread adjustments.

### Long-run determinants (state-level panel regressions)
- Panel specification: median price per square foot (P_st) regressed on local characteristics X_st, interaction δ HHDebt_s * NFCI_t, state (Γ_s) and quarter (Φ_t) fixed effects over 2002-2020; regressors lagged by one quarter except rental vacancy lagged four quarters.
- Key marginal effects and statistical findings:
  - Local output (GDP) growth: marginal positive effect of 0.6% to 1% for each percentage point change in local output.
  - Rental vacancy rate: marginal negative effect of 0.5% to 0.6% for a one percentage point change in rental vacancy rate.
  - Corporate license state tax (proxy for firm creation): statistically significant with a marginal effect of 1%.
  - NFCI (national financial conditions): negative and significant; a one-standard deviation tightening in NFCI in the previous quarter leads to a drop of about 2.5% in CRE prices.
  - HHDebt interaction: effects of NFCI milder in states where households are less indebted relative to incomes; impact of tighter financial conditions on CRE prices about 1/3 of the average effect (0.8%) in states with lower debt levels.
- Retail-specific results:
  - Local inflation: paramount driver of retail CRE valuation.
  - Rental vacancy rate: marginal effect of 0.9% price appreciation after a 1 percentage point drop in vacancy.
  - NFCI impact on retail: a one-standard deviation tightening in NFCI in the previous quarter leads to a drop of about 3% in retail CRE prices.
  - HHDebt interaction for retail: in states with lower indebtedness, retail CRE prices drop by 1.5% under a one-standard deviation tightening in NFCI.

*Source: wpiea2023015-print-pdf - 2.1  Commercial Property Indices (https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf)*

---

### 3.6  Local cash-flow variation

### Footfall as a cash-flow measure for CRE
- Footfall data captures the degree to which real estate space is actually being used (“consumed”) at any given point in time and location, providing a direct proxy for cash-flow variation in commercial real estate (CRE).
- For owner-occupied properties (industrial, healthcare), income is an imputed internal transfer price that depends on actual use; for retail and hospitality it is direct tenant income observed in profit and loss statements.
- Covid-19-related lockdowns provide clean exogenous variation in consumption of commercial real estate and associated local cash-flow variation.

### Pandemic-driven mobility and sectoral impacts (empirical observations)
- Year-over-year growth-rate analysis shows:
  - Visits to retail locations dropped by 30% to 50% during the initial phase of the Covid-19 shock.
  - Restaurants and hotels, along with healthcare and other contact-intensive services, experienced substantial declines in visits.
  - The recovery in contact-intensive sectors was slower and remained well below pre-pandemic levels, particularly for healthcare and other services.
  - Visits to industrial places were generally less affected and showed a much faster recovery, although some counties experienced substantial declines (lower band p25).
- Cross-sectional (cross-county) variation in footfall:
  - Despite a large overall level shift downward after March 2020, the cross-sectional variation of footfall across counties is not materially affected; no significant change in cross-county variation is observed throughout the sample.
  - Two opposing forces are proposed: national lockdowns could homogenize responses across locations, while differential local health impacts could magnify heterogeneity; empirically these effects offset, yielding stable cross-sectional dispersion.

### Local footfall and CRE prices (county-level regression results)
- Yearly regressions of average CRE prices on county-level visits for the retail sector find:
  - For a change in footfall of 10%, the marginal effect on prices is roughly 1.5%.
  - The magnitude of this effect is economically significant and remarkably stable across time, being only very modestly higher in 2020.

### Interpretation and implications
- The stable cross-sectional relationship between footfall and prices provides validation for the hypothesis that Covid-19 market dynamics were primarily driven by a demand shock.
- The observed stability also implies a direct path for recovery: as economic activity recovers and footfall returns to roughly pre-pandemic levels, prices recover as a pervasive feature across locations.

*Source: wpiea2023015-print-pdf - 3.6  Local cash-flow variation*

### 2.1  Commercial Property Indices

### wpiea2023015-print-pdf - 2.1  Commercial Property Indices

### Commercial property indices
- Green Street Commercial Property Price Index: appraisal-based index covering the period since 1998; focuses on properties owned by Real Estate Investment Trusts; sector weights: retail (20%), office (17.5%), apartment (15%), health care (15%), industrial (10%), lodging (7.5%), other sectors (15%).
- RCA US CRE property index (via MSCI CPPI US report): covers the period since 2000 and captures traded US commercial property in the apartment, retail, industrial and office segments.

### Commercial property transactions (ZTRAX)
- Data source: Zillow’s ZTRAX database with over 400 million public deed records across 2,750 U.S. counties; contains deed transfers, mortgages, foreclosures, property tax delinquencies, building square footage, land surface area, zip code, sale address, mortgage amount, sale price, loan amount.
- Coverage: data available for approximately 150 million parcels across 3,100 counties; final dataset includes 30-40 states that contain reliable, representative data and account for more than 80 percent of the U.S. population between 1994-2020.
- Transaction filtering: only transactions greater than $250,000 considered, resulting in around 1.3 million real estate sale transactions.
- Aggregation and classification: daily transaction-level information aggregated by quarter and zip-code; transactions identified by land usage and aggregated into six CRE types: retail, office, industrial, multi-family living units, lodging, other.
- Population density proxy: 2013 NCHS urban-rural classification aggregated at county level with six categories: (1) Large central metropolitan; (2) Large fringe metro; (3) Medium metro; (4) Small metro; (5) Micropolitan; (6) Noncore.

### Local economic activity (SafeGraph validation)
- Data source: SafeGraph business listings and footfall data covering over 6 million POIs in U.S. and Canada; raw dataset contains 150 million observations; after cleaning, almost 95 million observations representing 3,098 U.S. counties with an average of 125 POIs and a median of 21 POIs per county.
- Sector aggregation: using business activity codes, POIs aggregated into five sectors: retail, auto, restaurants, manufacturing, wholesale trade.
- Normalization: coverage increased over time, data normalized to obtain accurate foot-traffic counts.
- Validation vs. Census: correlation of aggregated monthly visits (SafeGraph) with monthly sales (Census):
  - restaurant sector correlation: in excess of 90%;
  - retail sales correlation: around 70%;
  - manufacturing sector correlation: around 70%;
  - auto sector correlation: in excess of 60%;
  - wholesale trade correlation: in excess of 60%.

### State- and national-level macroeconomic activity
- Model includes state and national indicators: state-level GDP growth, population growth, inflation, imports and exports, cost of entry (general corporation license and franchise tax year-on-year growth), business applications growth, rental vacancy rates (year-on-year growth), private sector net job creation.
- Financial conditions: National Financial Conditions Index (NFCI) used to capture financial conditions; NFCI interacted with a dummy of household indebtedness to test heterogeneous state sensitivity (expectation: higher-debt states more sensitive).
- Household indebtedness interaction: HHDebt dummy constructed later as part of panel regressions.

### Commercial Mortgage-Backed Securities (CMBX/CMBX spreads)
- CMBX contracts: typically CDS on an underlying portfolio of 25 CMBS deals; pricing available at daily frequency; data sourced through JP Morgan’s DataQuery.
- Tranche structure: CMBX AAA references super-senior CMBS with credit enhancement of about 30%; CMBX AJ, AA, A, BBB, BBB- refer to increasingly lower seniority tranches of same portfolio.
- Loan-pool exposure: earlier cohorts more heavily exposed to retail; collateralized retail asset share decreasing from around 40% to 25% in latest cohort; CMBX 6 delinquency rate in June 2020 is double compared to CMBX 12.
- Crisis reactions in spreads:
  - June 2008 to December 2008: average spread increased by roughly 12 percentage points, or 2.5 times higher than the long-run average.
  - December 2019 to June 2020: different impact pattern across tranches—financial crisis affected low-risk tranches most, Covid-19 mainly impacted higher-risk tranches; CMBX 6 spread adjustment slightly more than double that for CMBX 12.

### Prices and volumes — aggregate and by property type
- Validation: aggregated ZTRAX CRE price data compared to Green Street and RCA indices for 2000-2021 to verify similar behavior.
- Long-run price evolution (two decades): CRE prices generally doubled or tripled, with slightly stronger growth for industrial spaces.
- Post-GFC and pre-pandemic: prices corrected after the GFC (notably retail and multi-family) but recovered and reached new highs before the pandemic.
- Volumes: cycles more visible in volumes, with transaction counts dropping substantially around the two major crises; transaction volumes peaked at the end of 2016 for most CRE types while prices continued to rise.
- Pandemic effects: volumes dropped significantly for all segments during the pandemic; heterogeneous price dynamics:
  - lodging, retail, multifamily prices affected the most;
  - industrial and other segments remained relatively stable.
- Liquidity dynamics: during crises liquidity dries up, absorbing price shocks; when liquidity rebounds, prices recover.

### Spatial heterogeneity and gradients
- Estimation approach: regress ln(property price per built surface, per land surface, or ln price) on hedonic controls, UrbanCDC (1 to 6), interaction UrbanCDC*Γt, year (Γt) and county (Φc) fixed effects.
- UrbanCDC: ordered categorical variable 1 to 6 from highest to lowest population density.
- Findings:
  - Clear trend of higher prices close to urban core and lower prices outside across all property types.
  - Financial crisis (2008) and Covid-19 (2020) show similar short-term inversion patterns.
  - Industrial and office gradients: increasing in 2009, decreasing in 2020.
  - Multifamily: strong rebound of spatial gradient after pandemic, from -0.2 to roughly -0.16 within a single year.
  - Early 2000s vs. early 2010s: estimated coefficients suggest prices were around 30% higher in the most rural areas compared to the city core in the early 2000s; in the early 2010s they become 30% lower in the most rural areas relative to the city core.

### Risk pricing and market-perceived shock differences
- Crisis differential: 2008 crisis perceived as pervasive across debt holders (affected low-risk tranches most); Covid-19 perceived to impact more heterogeneously, increasing default risk mainly in vulnerable sectors and higher-risk tranches.
- Loan-cohort exposure and delinquencies: earlier cohorts concentrated in retail (higher exposure to lockdown impacts); CMBX 6 delinquency in June 2020 double that of CMBX 12, aligning with spread adjustments.

### Long-run determinants (state-level panel regressions)
- Panel specification: median price per square foot (P_st) regressed on local characteristics X_st, interaction δ HHDebt_s * NFCI_t, state (Γ_s) and quarter (Φ_t) fixed effects over 2002-2020; regressors lagged by one quarter except rental vacancy lagged four quarters.
- Key marginal effects and statistical findings:
  - Local output (GDP) growth: marginal positive effect of 0.6% to 1% for each percentage point change in local output.
  - Rental vacancy rate: marginal negative effect of 0.5% to 0.6% for a one percentage point change in rental vacancy rate.
  - Corporate license state tax (proxy for firm creation): statistically significant with a marginal effect of 1%.
  - NFCI (national financial conditions): negative and significant; a one-standard deviation tightening in NFCI in the previous quarter leads to a drop of about 2.5% in CRE prices.
  - HHDebt interaction: effects of NFCI milder in states where households are less indebted relative to incomes; impact of tighter financial conditions on CRE prices about 1/3 of the average effect (0.8%) in states with lower debt levels.
- Retail-specific results:
  - Local inflation: paramount driver of retail CRE valuation.
  - Rental vacancy rate: marginal effect of 0.9% price appreciation after a 1 percentage point drop in vacancy.
  - NFCI impact on retail: a one-standard deviation tightening in NFCI in the previous quarter leads to a drop of about 3% in retail CRE prices.
  - HHDebt interaction for retail: in states with lower indebtedness, retail CRE prices drop by 1.5% under a one-standard deviation tightening in NFCI.

*Source: wpiea2023015-print-pdf - 2.1  Commercial Property Indices (https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf)*

### 3.6  Local cash-flow variation

### 3.6  Local cash-flow variation

### Footfall as a cash-flow measure for CRE
- Footfall data captures the degree to which real estate space is actually being used (“consumed”) at any given point in time and location, providing a direct proxy for cash-flow variation in commercial real estate (CRE).
- For owner-occupied properties (industrial, healthcare), income is an imputed internal transfer price that depends on actual use; for retail and hospitality it is direct tenant income observed in profit and loss statements.
- Covid-19-related lockdowns provide clean exogenous variation in consumption of commercial real estate and associated local cash-flow variation.

### Pandemic-driven mobility and sectoral impacts (empirical observations)
- Year-over-year growth-rate analysis (to avoid seasonality) shows:
  - Visits to retail locations dropped by 30% to 50% during the initial phase of the Covid-19 shock (Figure 4a).
  - Restaurants (Figure 4b) and hotels (Figure 4c), along with healthcare (Figure 4e) and other contact-intensive services (Figure 4f), experienced substantial declines in visits.
  - The recovery in contact-intensive sectors was slower and remained well below pre-pandemic levels, particularly for healthcare and other services.
  - Visits to industrial places (Figure 4d) were generally less affected and showed a much faster recovery, although some counties experienced substantial declines (lower band p25).
- Cross-sectional (cross-county) variation in footfall:
  - Despite a large overall level shift downward after March 2020, the cross-sectional variation of footfall across counties is not materially affected; no significant change in cross-county variation is observed throughout the sample.
  - Two opposing forces are proposed: national lockdowns could homogenize responses across locations, while differential local health impacts could magnify heterogeneity; empirically these effects offset, yielding stable cross-sectional dispersion.

### Local footfall and CRE prices (county-level regression results)
- Yearly regressions of average CRE prices on county-level visits for the retail sector (Figure 5) find:
  - For a change in footfall of 10%, the marginal effect on prices is roughly 1.5%.
  - The magnitude of this effect is economically significant and remarkably stable across time, being only very modestly higher in 2020.

### Interpretation and implications
- The stable cross-sectional relationship between footfall and prices provides validation for the hypothesis that Covid-19 market dynamics were primarily driven by a demand shock.
- The observed stability also implies a direct path for recovery: as economic activity recovers and footfall returns to roughly pre-pandemic levels, prices recover as a pervasive feature across locations.

*Source: wpiea2023015-print-pdf - 3.6  Local cash-flow variation*

### References

### References

### Cited works
- Adjemian, S.,Bastani, H.,Juillard, M.,Mihoubi, F.,Perendia, G.,Ratto, M.andVillemot, S.(2011). Dynare: Reference manual, version 4.
- Agarwal, S.,Ambrose, B. W.,Lopez, L. A.andXiao, X.(2020). Pandemic risk factors and the role of government intervention: evidence from covid-19 and cmbs mortgage performance.
- Alter,  A.andDernaoui,  Z.(2020). Non-primary home buyers, shadow banking, and the us housing market.IMF Working Papers,2020(174).
- —andMahoney, E. M.(2021). Local house-price vulnerability: Evidence from the us and canada.Journal of Housing Economics,54, 101791.
- An,  X.,Deng,  Y.andGabriel,  S.  A.(2011). Asymmetric information, adverse selection, and the pricing of cmbs.Journal of Financial Economics,100(2), 304–325.
- Bergeaud, A.,Eyméoud, J.-B.,Garcia, T.andHenricot, D.(2021). Working from home and corporate real estate.Available at SSRN.
- Buchak, G.,Matvos, G.,Piskorski, T.andSeru, A.(2020). iBuyers: liquidity in real estate markets.Available at SSRN.
- Chernozhukov,  V.,Kasahara,  H.andSchrimpf,  P.(2020). Causal impact of masks, policies, behavior on early covid-19 pandemic in the us.Journal of Econometrics,220(1), 23–62.
- Dale-Johnson, D.,Brzeski, W. J.et al.(2001). Spatial regression analysis of commercial land price gradients. InAsian Real Estate Society Sixth Annual Conference at Keio University, Tokyo, Japan, Citeseer.
- Deghi, A.,Mok, J.andTsuruga, T.(2021). Commercial real estate and macrofinancial stability during covid-19.IMF Working Papers,2021(264).
- Diaz,  A.andJerez,  B.(2013). House prices, sales, and time on the market: A search-theoretic framework.International Economic Review,54(3), 837–872.
- D’Lima, W.,Lopez, L. A.andPradhan, A.(2020). Covid-19 and housing market effects: Evidence from us shutdown orders.Real Estate Economics.
- Driessen, J.andVan Hemert, O.(2012). Pricing of commercial real estate securities during the 2007–2009 financial crisis.Journal of Financial Economics,105(1), 37–61.
- Duca, J. V.andLing, D. C.(2020). The other (commercial) real estate boom and bust: the effects of risk premia and regulatory capital arbitrage.Journal of Banking & Finance,112, 105317.
- —,Muellbauer, J.andMurphy, A.(2021). What drives house price cycles? international experience and policy issues.Journal  of  Economic  Literature,59(3), 773–864.
- Genesove,  D.andHan,  L.(2012). Search and matching in the housing market. Journal of Urban Economics,72(1), 31–45.
- Ghent, A. C.(2021). Whats wrong with pittsburgh? delegated investors and liquidity concentration.Journal of Financial Economics,139(2), 337–358.
- —,Torous, W. N.andValkanov, R. I.(2019). Commercial real estate as an asset class.Annual Review of Financial Economics,11, 153–171.
- Gupta, A.,Mittal, V.,Peeters, J.andVan Nieuwerburgh, S.(2021). Flattening the curve: Pandemic-induced revaluation of urban real estate.Journal  of Financial Economics.
- Herbst,  E. P.andSchorfheide,  F.(2015). Bayesian estimation of dsge models. InBayesian Estimation of DSGE Models, Princeton University Press.
- Kennedy, G.,Killeen, N.,Skouralis, A.,Wosser, M.andVelasco, S.(2021). Covid-19 and the commercial real estate market in ireland.Financial Stability Notes.
- Levitin,  A.  J.andWachter,  S.  M.(2013). The commercial real estate bubble. Harv. Bus. L. Rev.,3, 83.
- Ling, D. C.,Wang, C.andZhou, T.(2020). A first look at the impact of covid-19 on commercial real estate prices: Asset-level evidence.The Review of Asset Pricing Studies,10(4), 669–704.
- Ramani, A.andBloom, N.(2021). The donut effect of covid-19 on cities.National Bureau of Economic Research Working Paper Series.
- Rosenthal, S. S.,Strange, W. C.andUrrego, J. A.(2022). Jue insight: Are city centers losing their appeal? commercial real estate, urban spatial structure, and covid-19.Journal of Urban Economics,127, 103381.
- Titman,  S.andTsyplakov,  S.(2010). Originator performance, cmbs structures, and the risk of commercial mortgages.The  Review  of  Financial  Studies,23(9), 3558–3594.
- Tuzel, S.andZhang, M. B.(2017). Local risk, local factors, and asset prices.The Journal of Finance,72(1), 325–370.

### Appendix figures, tables, and captions
- Appendix section heading: 6  Appendix
- Figure A.1 — Ztrax Data Validation
  - Data sources are Ztrax, Green Street, and the Federal Reserve Board.
  - Panels: (a)Green St. Vs ZTRAX; (b)FRB vs. Ztrax
- Figure A.2 — Additional Summary Statistics
  - Panels: (a)Number of Visits by Month (All CRE types, in millions); (b)Number of Visits by Sector (Normalized, in millions)
- Figure A.3 — Ztrax Data County-level Coverage
  - We report the number of total commercial real estate transactions by county in Panel (a) and the number of retail visits in Panel (b).
  - Panels: (a)(2000-2020); (b)Number of normalized retail visits by county (January 2020)
- Figure A.4 — County-level relationship between visits and CRE prices in 2019 and 2020 (Removing outliers and states lacking observations)
  - Panels: (a)2019; (b)2020
- Figure A.5 — State-level relationship between visits and CRE prices in 2019 and 2020 (Dropping outliers and states lacking observations)
  - Panels: (a)2019; (b)2020
- Table A.1 — Summary Statistics (state-level)
  - Source: The table reports additional summary statistics on the US Census, ZTRAX, and Chicago Fed data sets.
  - The low-debt dummy variable is created based on state-level debt-to-income data from the Federal Reserve. A state with an average debt-to-income ratio over the time period below 1.5 is considered a state with low debt.
  - The median CRE prices were winsorized by 1 percent.
  - The CPI index is recorded at the level of regions.
- Figure A.6 — Illustration of identification
  - Panel A — Supply shocks: Charts showing Quarters after impact 0 2 4 6 8 10 12 and year-on-year changes for Prices, Volumes, CMBX Spreads with labelled percent ranges (-0.5% to 1.5% for Prices; -3% to 2% for Prices in alternate chart; -4% to 4% for Volumes; -15 to 10 for CMBX Spreads in Panel C charts).
  - Panel B — Demand shocks: Charts showing Quarters after impact 0 2 4 6 8 10 12 and year-on-year changes for Prices, Volumes, CMBX Spreads with labelled percent ranges (-0.5% to 1.5% for Prices; -2% to 6% for Prices in alternate chart; -10% to 5% for Volumes).
  - Panel C — Risk shocks: Charts showing Quarters after impact 0 2 4 6 8 10 12 and year-on-year changes for Prices, Volumes, CMBX Spreads with labelled percent ranges (-4% to 2% for CMBX Spreads; -15% to 5% for Prices and Volumes).

*Commercial Real Estate in Crisis: Evidence from Transaction-Level Data — Working Paper No. WP/23/15*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023015-print-pdf.pdf_