What indicators signal commercial real estate stress at community banks?

The earliest signals are a rising watch-list share of assets and past-due loans that climb ahead of charge-offs. Reserves that thin out against troubled balances confirm the pressure, and heavy CRE concentration magnifies it because the same percentage loss lands on a larger base. Any single indicator can mislead, so the useful reading is the combination.

Concentration and watch-list share set the first screen

High CRE exposure is one of four indicators of potential bank stress. The watch-list share of assets is the companion measure, because it counts loans flagged for closer attention before they turn into recorded losses.

In one regional comparison, community banks under $10 billion in assets carried a larger share of assets with watch-list flags than similar-sized banks nationally. A gap like that tells a reader where to look first and says nothing about which bank is in trouble.

Nonfarm nonresidential CRE mortgages run about one-fifth of total assets at banks below the largest regulatory categories. That average is a baseline, and the bank’s own ratio is the test. A bank above the baseline takes a given percentage loss on a larger base, so the hit to capital is proportionally bigger.

The screen is a pairing: watch-list share on one axis, CRE as a share of assets on the other.

“Importantly, some banks may have more concentrated exposures to CRE mortgages than average and therefore may experience higher-than-average losses should CRE conditions weaken.”

From past due to charge-off: the order signals arrive

A loan counted as 30 or more days delinquent is late on payments but still carried at full balance. A charge-off removes the portion judged uncollectible. Past-due status usually comes first, so the delinquency rate is usually the earlier reading and the charge-off rate the later one.

In one quarter of one reported series, nonfarm nonresidential loan delinquencies of 30 or more days rose 55 basis points year over year to 1.26 percent. One quarter is a single observation and not a market reading. What carries over is the bucket definition and the habit of comparing against the same quarter a year earlier.

The median CRE charge-off rate in one quarter was 1.5 basis points for community banks and 21.6 basis points for global systemically important banks. A median describes the middle institution. A community bank in the tail can sit far from it, and a low median says little about that bank’s pipeline of past-due loans.

Reserve coverage against realized loss: what two medians leave out

The reserve coverage ratio is the allowance for credit losses divided by noncurrent loans, measuring reserves relative to troubled balances. The charge-off rate is losses booked over a period divided by loans, a flow. They answer different questions.

In one quarter, median CRE allowance coverage was 111 basis points for community banks and 297 basis points for the largest banks, against median charge-off rates of 1.5 and 21.6 basis points. The cohorts differ in size, borrower base and loss experience, so neither median sets a target for an individual community bank.

The decision test is the direction of the bank’s own allowance against its own past-due and watch-list balances. Allowance dollars can rise while coverage per dollar of troubled loans falls, which is the pattern the worked example below puts in numbers.

Loan growth and demand move separately

Loan balances measure exposure. Demand measures borrower appetite for new credit. Growth enlarges the amount at risk even while past-due and charge-off rates stay low, and weaker demand slows originations without curing problem loans already on the books.

Community bank nonfarm nonresidential CRE loans grew $7.9 billion, or 1.4 percent, in one quarter. At the end of that quarter balances stood at $576,575 million, up 0.7 percent from a year earlier. Because a 1.4 percent quarter sits inside a 0.7 percent year, the other three quarters netted to a small decline, and one quarter’s pace says little about the annual trend.

Property type matters as well. In one lending survey, moderate net shares of banks reported weaker demand for construction and land development loans and for nonfarm nonresidential loans, while multifamily demand was basically unchanged on net. An aggregate CRE total can blend segments moving in different directions.

Worked example: two hypothetical banks with the same CRE book

Two hypothetical banks each hold $1,000 million in assets and $300 million in CRE loans, so CRE is 30 percent of assets. That sits above the roughly one-fifth average for banks outside the largest categories, which is why a given loss weighs more on capital. Each bank has $100 million of equity. The only difference is credit performance, so every ratio below uses the same $300 million denominator.

Base case: $15 million of CRE loans sit on the watch list, which is 5 percent of CRE loans. Loans 30 or more days past due total $3 million, or 1 percent. The CRE allowance is $3.6 million, or 1.2 percent of CRE loans, which covers 120 percent of past-due balances.

Stressed case: the watch list grows to $45 million, or 15 percent, and past-due loans reach $12 million, or 4 percent. That is a fourfold rise. The allowance is lifted to $4.5 million, or 1.5 percent of CRE loans, an increase of $0.9 million. Coverage of past-due balances falls to 37.5 percent.

Now apply an illustrative loss severity of 30 percent to past-due balances. In the base case the loss is $0.9 million, which uses up 25 percent of the allowance and equals 0.9 percent of equity. In the stressed case the loss is $3.6 million, which uses up 80 percent of the allowance and equals 3.6 percent of equity.

The allowance went up, yet the cushion shrank relative to the problem. The CRE allowance also backs performing loans, so the past-due ratio works as a stress lens and gives no loss forecast. These balances are hypothetical and illustrate arithmetic only.

Mistakes in reading community bank CRE stress

Treating a low charge-off rate as proof of sound underwriting mixes up a late reading with an early one. Past-due and watch-list balances can deteriorate for several periods before any loss is charged off, so a quiet charge-off line can sit beside a worsening pipeline.

Applying the system average as a bank-specific threshold ignores dispersion. About one-fifth of assets is the average for banks below the largest categories, and some banks hold more concentrated exposure and could see higher-than-average losses if CRE conditions weaken. The average cannot say which side of it a given bank sits on.

Comparing a community bank with the largest banks on charge-offs or coverage mixes cohorts. Median coverage of 297 basis points against 111, and charge-offs of 21.6 against 1.5, reflect different portfolios. A community bank with coverage below the larger-bank median has not thereby shown that it is under-reserved.

Reading balance growth as a statement about asset quality skips a step. A 1.4 percent quarterly gain and a 0.7 percent annual gain describe the size of exposure and say nothing about repayment performance, and annualizing one quarter overstates the year.

Pooling property types into one CRE total hides offsetting movements. Construction and land development loans, nonfarm nonresidential loans and multifamily loans showed different demand readings in one survey, so a flat aggregate can contain a weakening segment.

Questions for a community bank CRE stress review

Is the watch-list share rising while CRE concentration stays high? The pairing shows whether more assets are flagged on a balance sheet that is still heavily exposed.

Are 30-day past-due balances worsening before charge-offs move? Payment deterioration appears in the past-due bucket first, so a gap between the two rates points to losses not yet recognized.

Is the allowance keeping pace with past-due and watch-list balances? A higher allowance in dollars can still be thinner coverage if troubled balances grow faster.

Which property types account for loan growth and for weaker demand? Construction and land development, nonfarm nonresidential and multifamily loans can move in different directions.

Is the comparison group a community bank cohort or the largest banks? Median charge-offs and coverage describe groups and set no target for an individual institution.

Does the bank’s CRE share of assets sit above or below the system average? The average is a baseline, and distance from it indicates how much a given loss percentage would weigh on capital.

Is balance growth separated from asset quality? A larger book raises exposure without proving deterioration, and a single quarter does not set the annual pace.