The seasonally adjusted business loan delinquency rate gets all the attention. It is the number that shows up in Federal Reserve press releases and Bloomberg terminals. But the not-seasonally-adjusted version -- the raw data -- tells you something the smoothed number cannot: when businesses actually struggle to make payments.
At 1.37% versus the SA reading of 1.34%, the current spread of +0.03 percentage points reflects the seasonal component of delinquency. This is not noise. It is a signal about the timing of business cash flow stress.
Looking at Q4 specifically: 2023 was 1.04%, 2024 was 1.30%, and 2025 was 1.37%. Same-quarter year-over-year comparisons with NSA data eliminate the seasonal question entirely. If Q4 2025 is higher than Q4 2024, that is real deterioration, not a calendar artifact.
Q1 readings tend to run higher because many businesses close their fiscal year in December and January audits often trigger loan reclassifications. Banks themselves conduct year-end portfolio reviews that can shift loans from current to past-due status. Q4 can show a bump for the same reason -- year-end reporting creates paperwork that surfaces delinquencies that may have been informally managed earlier.
Q2 and Q3 tend to run lower. Businesses generate stronger cash flows during the middle of the year (for most industries), and there is no fiscal-year reporting pressure to reclassify loans.
During turning points in the credit cycle, the seasonal adjustment model can lag reality. The model is built on historical seasonal patterns. But if the economy is entering a recession and delinquencies are rising for structural reasons, the adjustment model may still be subtracting seasonal effects that no longer apply. The result: the SA number understates the true deterioration for 1-2 quarters until the model catches up.
This happened in late 2007 and early 2008. The NSA readings were climbing faster than the SA readings suggested, because the seasonal adjustment was still assuming a normal economic environment. Bank credit officers who watched the raw data saw the turn earlier.
For business owners and investors, the NSA data is most useful for:
The year-over-year change in the NSA rate is +0.07pp (from 1.30% to 1.37%). That is a cleaner signal than the quarter-over-quarter move of +0.07pp, which is contaminated by seasonal effects.
Cross-category comparison using seasonally adjusted rates:
| Loan Category | Current | Prior Qtr | QoQ Change | Year Ago | YoY Change |
|---|---|---|---|---|---|
| Business Loans (C&I) | 1.34% | 1.33% | +0.01pp | 1.27% | +0.07pp |
| Commercial Real Estate | 1.58% | 1.56% | +0.02pp | 1.56% | +0.02pp |
| Consumer Loans | 2.62% | 2.71% | -0.09pp | 2.76% | -0.14pp |
| Credit Cards | 2.94% | 2.98% | -0.04pp | 3.08% | -0.14pp |
| All Loans (total) | 1.48% | 1.49% | -0.01pp | 1.53% | -0.05pp |
SA (seasonally adjusted) removes predictable calendar effects. NSA shows the raw numbers. The current SA rate is 1.34% while the NSA rate is 1.37%. The 0.03pp gap is the seasonal component.
Q1 and Q4 tend to run slightly higher due to fiscal year-end reporting and bank portfolio reviews. Q2-Q3 typically run lower as businesses generate stronger mid-year cash flows.
During economic turning points, the seasonal adjustment model can lag. NSA data may signal deterioration 1-2 quarters before the SA version catches up. Bank examiners and distressed-debt investors often prefer the raw numbers for this reason.
Use year-over-year comparisons (Q4 2025 vs Q4 2024) rather than sequential quarters. This controls for seasonal effects without relying on the adjustment model.
Yes. The year-over-year comparison shows an increase from 1.30% to 1.37%, a rise of 0.07pp. The trend is upward, with increases in 3 of the last 4 quarters.
FRED series DRBLACBN, published quarterly by the Federal Reserve Board of Governors. The SA version is DRBLACBS. Both cover all FDIC-insured commercial banks.