Why is commercial real estate a systemic risk?

The short answer

Commercial real estate is treated as a systemic risk category not because of its absolute size but because of how it is held. The U.S. CRE loan book sits primarily on small and regional bank balance sheets — banks that lack the diversified business models of the largest money-center institutions. When a CRE shock hits, it concentrates in a specific group of vulnerable banks rather than diluting across the financial system.

The 2023 collapse of Silicon Valley Bank and First Republic was not directly caused by CRE, but it illustrated the principle: a problem in one asset class on a regional bank balance sheet can trigger run dynamics quickly. CRE is the same pattern with different paper.

The post-pandemic CRE shock has three dimensions: declining office values (work-from-home), rising debt-service costs (rate hikes), and a $929 billion wall of refinancings due in 2024. Each amplifies the other.

→ New to systemic risk? How did the 2023 regional bank crisis differ from 2008?

What the data shows

Sources (FAU/Cole Banking Initiative, Wharton, Trepp, Federal Reserve, S&P Global, GAO 2024):

• Total U.S. bank CRE loans: roughly $3 trillion
• 1,788 small and regional U.S. banks held CRE exposure above 300% of equity capital in 2024 (FAU/Cole)
• 67 banks held CRE exposure exceeding 600% of capital
• Combined assets of these 67 banks: $216 billion
• $929 billion in CRE loans matured in 2024 (Mortgage Bankers Association)
• Office sector delinquency on CMBS reached 11.01% in February 2025 (Trepp)
• Office values down 32.7% from pre-Covid peaks (Yardi Matrix)
• Wharton research: extend-and-pretend lending hides true delinquencies, with reported figures understating real risk by factor 4

The exception that contextualizes the data: the largest U.S. banks have CRE exposure well below 100% of capital, so the systemic risk is concentrated in a specific tier of banking. The “too big to fail” institutions are not the worry here — the regional and community banks are.

→ Dataset: Credit spreads and recession risk

Why it happens — the macro mechanism

CRE systemic risk transmits through three channels that compound each other.

The valuation channel. Office property values have declined 32.7% from pre-Covid peaks (Yardi Matrix data through January 2026), driven by structurally lower demand from work-from-home. This valuation drop directly impairs loan-to-value ratios on CRE loans, pushing many into technical default zones.

The refinance wall channel. $929 billion in CRE loans matured in 2024. Borrowers face the choice of refinancing at sharply higher rates (often 200+ bp above origination) or selling at depressed prices to repay. Our analysis of REIT leverage applies this reasoning to listed real estate. Many cannot do either — leading to extend-and-pretend modifications that defer the recognition of losses.

The hidden-losses channel — the Wharton finding. Wharton research (Hinzen et al., 2024) documents that regional banks have been disproportionately extending troubled CRE loans rather than recognizing losses. This makes reported delinquency rates roughly four times lower than the true underlying credit deterioration. The implication: even the visible CRE stress (11% office CMBS delinquency) is the tip of an iceberg.

Synthesis by regime. In the pre-2020 stable regime, CRE was a high-yielding stable asset class with low default rates. In the pandemic dislocation regime (2020–2022), office demand structurally repriced lower while financial conditions stayed loose, masking the impairment. In the rate-shock regime (2022–2026), the impairment can no longer be masked by easy financing — refinance walls force decisions, and extend-and-pretend reaches its limits over multi-year horizons. The transition parameter is the duration of the high-rate environment: each year of rates above 5% adds to the unrealized losses on CRE loan portfolios.

The 2008 crisis taught everyone to watch the money-center banks. The CRE risk shows that the next stress could come from a thousand small ones.

→ Framework: Economic cycle phases and signals

What it means for different economic actors

Regional bank investors. Need to focus on individual bank-level CRE exposure, particularly concentration above 300% of capital. Aggregate banking sector statistics obscure the concentration of CRE stress in specific lenders. Office exposure as a share of CRE concentration is the relevant metric, not aggregate CRE alone.

Municipal bond investors. Face erosion of property tax base in cities with high pre-pandemic office concentration. New York, San Francisco, Chicago and Washington DC face multi-year fiscal headwinds as office values reprice lower. Some downtown-focused special tax districts and revenue bonds may face deeper impairment.

Small business owners downtown. Experience the most direct impact from reduced office utilization. The asymmetric recovery (Manhattan strong, San Francisco weak) means business decisions depend heavily on the metro-level dynamics rather than national averages.

A common error is to treat the office vacancy crisis as a uniform national phenomenon. The data shows extreme dispersion: some metros (Manhattan, Miami) are recovering quickly while others (Seattle, Austin, San Francisco) remain in deep stress. National aggregate statistics hide the systemic risk, which is concentrated in specific cities and specific bank balance sheets.

Practical observation

Go deeper

Related questions

• Why do office vacancies matter for the broader economy?
• How do REITs behave during inflation and recession?
• 2023 regional bank crisis vs 2008?
• Credit cycle vs economic cycle?

Frequently asked questions

How is CRE different from residential real estate as a systemic risk?

Two key differences. First, residential mortgages in the U.S. are mostly held by Fannie Mae, Freddie Mac and Ginnie Mae through agency MBS — federal backstops absorb the duration and credit risk. CRE loans, by contrast, sit primarily on private bank balance sheets without federal guarantees. Second, residential prices are supported by the lock-in effect freezing supply; CRE values, especially office, face structural demand decline from work-from-home that has no equivalent supply-side support. The result: CRE losses concentrate on bank balance sheets in a way residential losses largely do not.

What does “extend-and-pretend” mean and why is it concerning?

Extend-and-pretend is the practice of modifying troubled CRE loans — extending maturity, reducing payment requirements, or capitalizing missed interest — rather than declaring them in default. This keeps the loan classified as performing, avoiding the loss recognition that would impair bank capital. It is concerning because it accumulates losses off-balance-sheet: the impairment is real but unreported. Wharton research (Hinzen et al., 2024) finds that the gap between reported delinquencies and true credit deterioration has widened sharply since 2023. Eventually these positions must reconcile with reality: either through asset price recovery (unlikely in office given structural demand decline) or eventual loss recognition (possibly forced by regulators).

Could CRE trigger a 2008-style crisis?

The cleanest answer is: probably not in form, but possibly in some functional aspects. The reasoning is spelled out in this analysis of deposit concentration risk. The 2008 crisis was triggered by deeply leveraged residential MBS held across the global financial system, with structural opacity that hid risk concentrations. CRE loans are held more transparently on identified bank balance sheets, and the largest banks have manageable exposure. However, the regional banking crisis of 2023 (SVB, First Republic, Signature) showed that concentrated stress in a specific bank tier can trigger broader funding stress through deposit flight. CRE could plausibly create a similar regional bank crisis, particularly if interest rates remain elevated and refinance walls force loss recognition. The systemic dimension would depend on speed and clustering of failures.

Last updated — 18 June 2026