Three weeks ago this publication made a prediction with a date on it. Here is the sentence, verbatim, from The Banks That Eased Multifamily Credit Last Quarter Are Not the Banks Financing Your Deal, published August 11, 2026:
If the cohort split is real and the New Jersey banks behaved like the group they belong to, aggregate Q2 CRE growth at NJ-chartered banks should be flat to negative, with the flattest results concentrated at the sixteen institutions that were at or above the 300% level as of March 31, 2026.
Second-quarter Call Report data is now in. The prediction was wrong on the dollars and right on the banks, and the gap between those two answers is more useful than either one would have been alone.
First, the Acquisition
Anyone who pulls the raw file will see a 6.93% quarterly increase in New Jersey bank multifamily balances and stop reading. That number is an artifact and it needs disposing of before anything else.
OceanFirst Bank, National Association of Toms River acquired Flushing Bank of Uniondale, New York — $8.86 billion in total assets at March 31, 2026, carrying $2.39 billion of multifamily loans. The holding-company merger closed June 1, 2026; Flushing Bank merged into OceanFirst Bank effective June 2, 2026, inside the quarter. OceanFirst's total assets went from $14.5 billion to $23.2 billion, and its multifamily line rose $1.26 billion.
That is a New York multifamily book changing charter. No New Jersey bank made a loan for it. Every figure below excludes OceanFirst entirely.
Here is the headline the exclusion is protecting you from, all 50 New Jersey-chartered banks:
| Q1 2026 | Q2 2026 | Change | |
|---|---|---|---|
| Multifamily | $29.13B | $31.15B | +6.93% |
| Non-owner-occupied nonfarm nonresidential | $37.58B | $39.83B | +5.99% |
| Construction & land development | $7.98B | $7.97B | −0.07% |
| Total CRE | $74.69B | $78.96B | +5.71% |
Do not lead with those. One event is most of them.
The Like-for-Like Result
Dropping OceanFirst from both quarters leaves 49 banks measured identically at both dates. This is the conservative cut — it removes OceanFirst's own organic lending along with the acquisition, so if the bank grew anything on its own account in the quarter, the figures below understate the cohort.
| Q4 → Q1 | Q1 → Q2 | $100B+ cohort, Q1 → Q2 | |
|---|---|---|---|
| Multifamily | +0.61% | +2.69% | +1.67% |
| Total CRE | +0.77% | +1.63% | +1.35% |
Excluding OceanFirst's acquisition of Flushing Bank, New Jersey-chartered banks grew multifamily loans 2.69% in the second quarter of 2026, faster than the 1.67% growth at banks with $100 billion or more in assets, per FDIC Call Report data. In dollars: multifamily $28.04 billion to $28.79 billion, total CRE $69.20 billion to $70.32 billion. Construction and land development went the other way, $7.32 billion to $7.24 billion, a decline of 1.20%.
So the prediction failed, and it failed on both lines. The cohort the Fed's survey said reported unchanged multifamily standards did not merely fail to shrink its multifamily book — it grew it more than four times faster than the prior quarter, and faster than the cohort that reported easing. Total CRE ran the same way: +1.63% against +1.35% at the large banks. There is no version of this quarter's balance-sheet data in which the New Jersey cohort behaved like a group whose credit box sat still while someone else's widened.
The Breadth Result
Now count banks instead of dollars.
Of the 46 New Jersey-chartered banks holding multifamily loans at June 30, 2026, 23 reduced their multifamily balances during the second quarter and 20 increased them. Three were unchanged, using a ±0.5% band. The median bank in that group saw its multifamily balance fall 0.55%. Two of the twenty increases are de novo charters booking a first small multifamily loan, not an existing lender expanding.
The growth is concentrated to an unusual degree. The five largest multifamily increases outside OceanFirst were Valley National (+$0.45B), ConnectOne (+$0.10B), Columbia Bank (+$0.10B), Cross River (+$0.10B), and Provident (+$0.08B). Those five added $0.83 billion between them against a cohort-wide net increase of $0.76 billion — the other 44 banks are net negative as a group. Valley National alone is about 60% of the entire ex-acquisition increase.
The $100B+ cohort was mixed in exactly the same way: 32 banks, 13 growing multifamily, 11 shrinking, 8 flat. Both cohorts have a majority that isn't participating. Only one of them has a $66 billion bank inside it doing the arithmetic's heavy lifting.
This is the sentence the piece exists for. The Senior Loan Officer Opinion Survey polls institutions and reports what a net share of them did. Call Reports total dollars. In New Jersey in the second quarter those two measurements pointed opposite directions, and a sponsor reading either one alone gets the wrong answer about their own lender.
The second half of the prediction did hold
The pre-registered sentence had a clause the aggregate result obscures: the flattest results should concentrate at the sixteen banks that were at or above the 300% concentration level on March 31, 2026. That clause scores better than the first one.
Excluding OceanFirst — which was one of the sixteen — the fifteen remaining high-concentration institutions grew multifamily 2.32% and total CRE 1.21%, both below the 49-bank cohort's 2.69% and 1.63%. Nine of the sixteen shrank their multifamily book over the quarter, six grew it, one was flat. The other thirty-four banks grew multifamily 8.07%, but on a combined multifamily book of $1.82 billion, and two-thirds of that increase is one institution. Percentage growth on a small base is not evidence of a lending cycle.
Concentration itself moved the way the constraint story predicts. Seventeen New Jersey-chartered banks held CRE concentrations at or above the 300% interagency monitoring level at June 30, 2026, up from sixteen at March 31, and those seventeen held 92% of all CRE on New Jersey bank balance sheets. Being at or above that level is not a violation, a cap, or a supervisory finding about any institution — it is the point at which examiners expect heightened CRE risk management, which in practice makes a bank more selective about what it adds. The NJ Bank CRE Lending Tracker now carries the second-quarter edition, so the ratio behind that count is checkable bank by bank. The capital-treatment split that determines how expensive that CRE is to hold is a separate question, covered in CBLR Banks Are Excluded From the 2026 CRE Risk-Weight Cut.
What This Does to the Original Argument
The August post said the credit box at New Jersey banks was unchanged and a sponsor should structure into it rather than wait for it to widen. That advice survives, with one correction that matters operationally.
The box is unchanged at most New Jersey banks and open at a few. Twenty banks grew multifamily last quarter. Five of them grew it meaningfully. The sponsor's job this quarter is not deciding whether "credit is easing" — it is knowing which side of that split their lender is on, because the two groups will answer the same file differently. A sponsor who takes an easing headline to a bank that shrank its multifamily book in the second quarter is still waiting on something that is not coming there, and the headline gave them no way to know it.
That is also the correction to make on our own August framing. We treated the sub-$100 billion cohort as a group that would behave like a group. It didn't. It contained both the banks that added a billion dollars of multifamily and the banks that ran off what they had, and the survey's cohort split has no resolution below "large" and "other." The tracker does, which is the argument for reading it bank by bank rather than reading the state total.
What Would Make This Read Wrong
Four things, all of which survive into the next edition:
One quarter is one data point. Ex-acquisition multifamily growth was +0.61% in the first quarter and +2.69% in the second. That is a real acceleration and it is also two observations. Nobody should call it a trend before the third.
The like-for-like cut is deliberately conservative. Dropping OceanFirst removes its organic lending alongside the acquisition. If OceanFirst's ex-Flushing book grew in the quarter, every cohort figure above is understated. We cannot separate the two from Call Report data, so we took the version that works against our own result.
Balances are net of payoffs. A bank whose multifamily line fell may have originated steadily and simply seen more run off. Call Reports report a residual, not an origination volume, and cannot distinguish the two. The SLOOS measures standards and demand; balances measure what is left after both. Neither one is a substitute for the other, and this post's whole finding is what happens when you treat one as the other.
The cohort has a scale problem. Valley National at $66.2 billion sits in a group whose median member is a fraction of that size. Any aggregate weighted by that balance sheet tells you less about a $3 billion bank than a reader naturally assumes it does. That is the same objection this post raises against the SLOOS aggregate, and it applies to ours.
The Next Test
Two scheduled events will distort the third-quarter aggregate before any lending decision does, so the prediction has to be stated on the like-for-like basis rather than the headline.
Blue Foundry Bank ($1.89 billion at June 30, 2026) merged into Fulton Bank, N.A. effective July 11, 2026. It was still a separately chartered New Jersey bank filing its own Call Report at quarter-end, which is why it is in this quarter's cohort of 50; it will not be in the third quarter's, which drops to 49.
Northfield Bank of Staten Island, New York — roughly $5.5 billion in assets — merged into Columbia Bank of Fair Lawn, New Jersey on July 20, 2026, alongside Columbia Financial's second-step conversion. Columbia is a New Jersey-chartered bank already in this cohort, and the combined institution has pro forma total assets of about $18 billion. So the third quarter brings a second New York multifamily book across a charter line into New Jersey, and the like-for-like adjustment has to be made again — this time by excluding Columbia rather than OceanFirst.
With that stated, the pre-registration for the third quarter:
Excluding acquisitions, the New Jersey cohort's multifamily growth in Q3 2026 will be below Q2's 2.69%, and fewer than half of the banks holding a multifamily book will grow it.
Both halves are checkable. Third-quarter Call Reports are due to the Central Data Repository on October 30, 2026, and the Q3 pull lands in early December. If the first half is wrong again, that is two consecutive quarters of acceleration in a cohort that told the Federal Reserve its standards were unchanged, and it stops being an artifact of one bank.
FAQ
Did New Jersey banks grow CRE lending in Q2 2026? Yes, on the aggregate. New Jersey-chartered banks held $78.96 billion of commercial real estate loans at June 30, 2026, up from $74.69 billion at March 31 — but $4.27 billion of that headline increase is dominated by a single acquisition, OceanFirst Bank's purchase of Flushing Bank. Excluding OceanFirst entirely, the remaining 49 New Jersey-chartered banks grew total CRE 1.63% and multifamily 2.69% over the quarter. The typical bank did not participate: of the 46 New Jersey banks carrying a multifamily book at June 30, 2026, 23 reduced it during the quarter, 20 increased it, and 3 were unchanged.
Why did the SLOOS prediction about New Jersey banks turn out wrong? Because it was a prediction about dollars and the survey it came from measures banks. In August 2026 we predicted that aggregate second-quarter CRE growth at New Jersey-chartered banks should be flat to negative if those banks behaved like the sub-$100 billion cohort the Federal Reserve's July 2026 Senior Loan Officer Opinion Survey said reported unchanged multifamily standards. Balances instead grew 2.69% on multifamily excluding the quarter's one acquisition. The count went the other way: 23 of the 46 New Jersey banks with a multifamily book shrank it and the median bank's multifamily balance fell 0.55%. Five institutions produced the growth. An aggregate weighted by the state's largest bank describes what happened to the state's dollars, not what happened at the typical lender.
Which New Jersey bank acquired Flushing Bank? OceanFirst Bank, National Association, of Toms River, New Jersey. Flushing Bank of Uniondale, New York merged into OceanFirst Bank effective June 2, 2026, one day after the holding-company merger of OceanFirst Financial Corp. and Flushing Financial Corporation closed on June 1, 2026. Flushing held $8.86 billion in total assets and $2.39 billion of multifamily loans at March 31, 2026. OceanFirst's total assets rose from $14.5 billion to $23.2 billion over the second quarter, and its multifamily line rose $1.26 billion. Those are New York loans changing charter, not New Jersey banks originating more.
How many New Jersey banks are above the 300% CRE concentration level? Seventeen New Jersey-chartered banks held total CRE at or above 300% of total risk-based capital at June 30, 2026, up from sixteen at March 31, 2026. Those seventeen hold 92% of all commercial real estate on New Jersey bank balance sheets, unchanged from the first quarter. The 300% figure is the 2006 interagency concentration monitoring criterion; being at or above it is not a violation, a cap, or a supervisory finding about any institution, and the bank-by-bank figures behind it are on our NJ Bank CRE Lending Tracker, which now publishes the second-quarter 2026 edition.
Is any New Jersey-chartered bank a large bank under the Fed's SLOOS definition? No. The Federal Reserve defines a large bank in its Senior Loan Officer Opinion Survey as an institution holding $100 billion or more in total domestic assets as of March 31, 2026. The largest New Jersey-chartered bank, Valley National Bank, held $66.2 billion in total assets at June 30, 2026 — $33.8 billion below the line, and further from it than any acquisition completed this year would close. Every New Jersey-chartered bank remains in the survey's "other banks" cohort, which means 100% of New Jersey bank CRE sits outside the group that reported easing.
Which Side of the Split Is Your Bank On?
The useful question this quarter is not whether New Jersey bank credit loosened. It is whether the specific balance sheet holding your loan grew its multifamily book last quarter or ran it off, and where it sits on the concentration table going into the renewal. Both are public numbers computed from public filings. Send us the rent roll, the trailing twelve, and the maturity date, and we will tell you which lenders have room, what the file needs to clear at each of them, and whether the incumbent is the right conversation at all. Our New Jersey multifamily lending practice does this work every week, and the NJ Bank CRE Lending Tracker — refreshed to June 30, 2026 alongside this piece — is where the bank-by-bank figures live.
Dominick Prevete — Founder, Blue Sky Capital Advisors. 31 years in real estate finance; 100+ bank and private-lender relationships across the bridge, agency, and private-credit spectrum. 4 Sutton Ct, Hamburg, NJ 07419 · (908) 220-6404. Financing arranged in all 50 states.
Bank-level figures are Blue Sky Capital Advisors analysis of FFIEC Call Report data retrieved from the FDIC BankFind Suite API on September 3, 2026, covering the 50 New Jersey-chartered banks active at June 30, 2026 — a cohort that includes Blue Foundry Bank, which was still separately chartered at quarter-end and merged into Fulton Bank, N.A. effective July 11, 2026. First-quarter figures are re-pulled from the same retrieval as the second quarter so both dates come from one snapshot. CRE is the 2006 interagency concentration measure as the tracker computes it — construction and land development, multifamily, and non-owner-occupied nonfarm nonresidential — which is the definition every figure on this property uses; the tracker's methodology page explains why the measure's fourth component, commercial real estate loans not secured by real estate, is excluded and how that makes published concentration ratios slightly conservative. Concentration ratios are computed against total risk-based capital, or, for the twelve banks that elected the Community Bank Leverage Ratio framework and therefore report none, against tier 1 capital plus the allowance for credit losses; the tracker's methodology page sets out both. The $100 billion-plus comparison cohort is the group of U.S. banks at or above the Federal Reserve's SLOOS large-bank threshold, measured on the same Call Report basis. Merger facts are from the acquirers' own announcements. The August 2026 prediction quoted at the top is from this publication's July 2026 SLOOS analysis; the survey characterizations it relies on are the Federal Reserve's. This is market commentary, not a commitment to lend or investment advice.