By Yikai Wang, Mike Gullette and Sharon Whitaker
ABA DataBank
Headline news of office loan defaults has led some auditors and regulators to question whether small banks have set adequate credit loss reserves for their commercial real estate portfolios under the current expected credit loss, or CECL, accounting standard. Most non-SEC filers, including small banks, adopted CECL reporting in January 2023, whereas large banks have been operating under the CECL framework since January 2020. Despite topline numbers suggesting that small banks seem to be reserving systematically less than large banks for their respective CRE portfolios, we conclude that, in fact, their reserve levels are commensurate with corresponding risk and CRE portfolio performance.
Do small banks normally reserve less than large banks on their CRE portfolios?
Using historical data beginning in 2020, we find larger banks (with assets greater than $100 billion) and smaller regional, midsize and community banks (those with assets of less than $100 billion) have diverged in their CRE portfolio allowance for credit loss coverage ratios (the ratio of the ACL to total loans) since the first quarter of 2023. Even though large banks generally reserved more even at the beginning of their 2020 CECL adoption, the difference has become more apparent since early 2023. Global systemically important banks and larger regional banks (non-GSIBs with assets greater than $100 billion) have steadily increased their CRE reserves since 2023, while those with assets between $10 billion and $100 billion and community banks with assets of less than $10 billion) have kept theirs steady (Figure 1). This may have initially suggested that only the large banks had identified higher risk in their CRE portfolios since early 2023.
Figure 1: Median ACL Coverage Ratio for CRE Portfolio, bps
Source: FFIEC and ABA. Note: While dollar-weighted averages over-emphasize the coverage ratios of a few larger banks, we believe median ratios at the respective cohorts better indicate individual bank practices. With this in mind, we find the quarterly median ratios to be relatively consistent between cohorts, even up to the current period.
Are small banks’ current reserve levels commensurate with risk?
Can this CRE ACL coverage ratio divergence be justified? To answer this question, we need to better understand the different sectors within CRE and which banks are major participants in those sectors.
As a previous ABA DataBank has illustrated, office loans account for only 17 percent of banks’ CRE holdings, and some subcomponents within the office sector, such as suburban medical offices, are unaffected by the hybrid working model introduced after the pandemic.
The riskiest part of the CRE space — central business district office buildings — is more likely to be owned by large banks and nonbanks than by midsize or community banks. Recent Federal Reserve research, illustrated in Figure 2, also shows that GSIBs and nonbank CRE lenders have higher concentrations of office lending in the most at-risk geographic areas (CBDs and areas that experienced a greater shift to remote work). As coverage ratios are meant to reflect credit risk, this could be one reason why large banks have relatively higher ratios.
Figure 2: Exposures to at-risk office loans by lender
Source: Federal Reserve and Real Capital Analytics. Note: This figure plots the shares of loans in the RCA office property database that are in central business districts (red bars), in counties where the time at workplaces declined by at least one-third relative to before the pandemic (blue bars), or areas with both risk factors (purple bars).
In contrast to the ACL coverage ratio, which reflects anticipated losses, the charge-off rate reflects the percentage of value of loans and leases actually removed from the books and charged against loss reserves. Figure 3 shows the median CRE portfolio charge-off rates for banks of different sizes. In the third quarter of 2024, the median charge-off rate was 21.6 basis points for GSIBs, 11.6 basis points for regional banks, 1.8 basis points for midsize banks and 1.5 basis points for community banks. This is compared to the ACL coverage ratio of 297 basis points for GSIBs, 262 basis points for regional banks, 106 basis points for midsize banks and 111 basis points for community banks. Compared to charge-off activity, which normally results shortly after impairment is identified, midsize and community bank ACL coverage now appears robust.
For GSIBs and regional banks, despite the recent increase in charge-off rates on banks’ CRE portfolios, ACL coverage levels still appear reasonable when compared to the charge-off rates. For midsize and community banks, their charge offs are still stable, which reflects that the credit quality in these banks is more favorable compared to their larger peers. The charge-off rate calculation here does not consider any recovery, which may serve as another buffer, because even after loans are charged off, it is still possible for lenders to recover some or all of the value through negotiation or legal actions.
Figure 3: Median CRE Charge Off Rates, bps
Source: FFIEC and ABA
Do loan modifications and loan-to-value ratios tell a consistent story?
The CRE loan modification data from call reports tell a similar story. Figure 4 shows the percentage of banks with CRE loans under modification and past due, in which community and midsize banks have a much lower percentage than GSIBs and regional banks, indicating higher credit quality of small banks’ CRE loan exposures. Of particular note, less than 20 percent of community banks have recorded loans under modification and past due since late 2022.
Figure 4: CRE Portfolio Credit Quality (Percentage of Banks with CRE Loans in Modification and Past Due), ppt
When analyzing potential loss on CRE portfolios, loan-to-value ratio is a key metric to predict expected losses banks may incur if these loans default. The lower the LTV at origination, the larger the property price decline is needed before banks have to absorb loss. LTVs at origination across property types have been trending downward since 2015 and are currently below 60% (Figure 5). The current peak-to-trough price decline in core CRE sectors is around 20%, suggesting there is still a buffer before banks have to take losses. More encouragingly, CRE prices have reversed the downward trend since the beginning of the year (Figure 6). Further rate cuts by the Federal Reserve will also likely help boost CRE prices as borrowing becomes cheaper.
Figure 5: Real Estate Loan LTVs at origination
Source: DoubleLine and RCA
Conclusion
The narrative surrounding CRE risks, especially for small banks, has often been overstated without considering nuanced realities. Our analysis underscores that the divergence in ACL coverage ratios among banks of different sizes reflects varying risk exposures rather than inadequate reserving practices by smaller institutions. While larger banks have increased reserves due to heightened exposure to at-risk assets such as CBD office loans, smaller regional, midsize and community banks are maintaining steady reserve levels commensurate with their relatively lower-risk portfolios.
Figure 6: Green Street Core Sector Commercial Property Price Index
Source: Green Street and ABA
Furthermore, key indicators such as low charge-off rates, minimal loan modification and past due, and declining LTVs at origination demonstrate the resilience of smaller banks’ CRE portfolios. This resilience is reinforced by improving CRE property prices and the inherently conservative nature of U.S. banks’ credit reserve practices, which remain robust compared to their global peers.
Ultimately, a deeper understanding of the underlying risk factors and robust reserve methodologies and processes is critical to dispelling misconceptions and fostering confidence in the soundness of small banks’ CRE portfolios. As the economic landscape evolves, the ability to adapt CECL frameworks thoughtfully will remain pivotal in ensuring stability across the banking sector.
Yikai Wang is VP for banking and economic research in ABA’s Office of the Chief Economist. Mike Gullette is SVP for tax and accounting policy at ABA. Sharon Whitaker is VP for commercial real estate and mortgage finance policy at ABA. For additional research and analysis from the Office of the Chief Economist, please see the OCE website.
