On August 3, 2026, the Federal Reserve published the July 2026 Senior Loan Officer Opinion Survey, covering the second quarter. The commercial real estate headline was straightforward: banks generally reported easier standards and basically unchanged demand. A moderate net share of respondents reported having eased standards on loans secured by nonfarm nonresidential properties, a modest net share reported easing on multifamily, and standards on construction and land development were basically unchanged.
Then there is the footnote, and the footnote is the story. The easing was not distributed evenly across the panel. Large banks reported easier standards for all CRE loan types. Other banks reported basically unchanged standards on loans secured by multifamily properties and on construction and land development. The survey defines a large bank as one holding $100 billion or more in total domestic assets as of March 31, 2026.
Note what the footnote does not do: it reports the large-bank easing without sizing it. No net share is attached. So the split is a difference in direction, not a measured gap in degree, and this post does not treat it as one.
That date is why this is a finding rather than a news item.
The Join
March 31, 2026 is the exact measurement date of the current edition of our NJ Bank CRE Lending Tracker, which computes CRE exposure and concentration for every New Jersey-chartered bank from Q1 2026 FFIEC Call Reports. The Fed sorted its respondent panel on the same day's balance sheets that the tracker reads. So the two datasets can be joined without approximation, estimation, or a proxy — the cohort test is a single number on a single date, and the Call Reports answer it for every institution in the state.
Run the join and the result is categorical. As of March 31, 2026, all 50 New Jersey-chartered banks held total assets below $100 billion. Not most of them. All of them. There is no New Jersey-chartered institution on the easing side of the Fed's split.
The arithmetic on the closest case: the largest New Jersey-chartered institution by total assets as of March 31, 2026 is Valley National Bank, at $64.4 billion. Against the survey's $100.0 billion threshold, that is a gap of $35.6 billion — and every other NJ-chartered bank is further from the line, most of them by an order of magnitude. The state's largest bank is not sitting near the boundary of the cohort that eased.
The second derived figure follows directly. New Jersey-chartered banks held $74.7 billion of commercial real estate loans as of March 31, 2026, including $29.1 billion of multifamily. One hundred percent of that sits at institutions in the Fed's "other banks" bucket. The reported easing in multifamily credit standards last quarter came entirely from a group of banks that contains zero New Jersey-chartered institutions. (The concentration formula, the Call Report fields behind it, and the scope rules are on the tracker's methodology page.)
What "Unchanged" Actually Means
Unchanged is not the same as closed, and the survey is unusually clear on this point if you read past the quarterly deltas.
The July survey included a special question on where standards sit today relative to the midpoint of their range since 2005. Banks reported that levels are at the tighter end of that range for every category except commercial and industrial loans. For CRE specifically, a significant net share reported relatively tight standards on construction and land development, and moderate net shares reported relatively tight standards on nonfarm nonresidential and multifamily. So yes — tight, by the survey's own accounting.
But the comparison that matters is the one against last year. The net shares reporting standards at the tighter end were lower than in the July 2025 survey. Read those two results together and the correct interpretation of the "other banks" cohort is not that credit is unavailable. It is that this cohort's box is meaningfully wider than it was twelve months ago, and it did not move again in the second quarter. Level improved over the year; change was flat in the quarter.
That distinction is the whole credibility of this piece, and it should not be compressed in either direction. A sponsor who reads "unchanged" as "not lending" will not make a call that would have been answered. A sponsor who reads the aggregate headline as "loosening" will wait for a widening that, in their lender cohort, is not currently happening.
And the split is not a one-quarter artifact, which is the main reason this is worth publishing now rather than filing away. The April 2026 survey, covering the first quarter, reported aggregate CRE standards basically unchanged across all three categories — but underneath that aggregate, moderate net shares of large banks reported easing all three, while other banks reported tightening on construction and land development (a moderate net share) and on multifamily (a modest net share), with nonfarm nonresidential basically unchanged. That is two consecutive surveys with the divergence running the same way. For the smaller cohort, the second quarter was an improvement — from modest tightening to no change — just not the easing the headline describes.
The Demand Side Is the Opportunity
The demand data runs the same direction, and the sentence in the survey is explicit: changes in demand for nonfarm nonresidential and multifamily loans were mixed across bank size categories, as large banks reported stronger demand and other banks reported weaker demand.
Be precise about what a bank reporting weaker demand is reporting. It is reporting fewer applications, not more generosity. Nothing about a thinner pipeline widens a credit box.
But it changes the competitive position of a sponsor who shows up anyway. The cohort holding essentially all New Jersey bank CRE is the cohort reporting fewer inbound files — and per the Q1 2026 tracker data, it is also the cohort where balance-sheet room, not appetite, is the scarce input. As of March 31, 2026, sixteen of the 34 New Jersey-chartered banks with CRE books above $50 million sit at or above the 300% interagency monitoring level, and those sixteen hold 92% of all CRE on New Jersey bank balance sheets. We walked through what that inversion does to a renewal conversation in 92% of the Exposure Sits Where the Capacity Isn't; this post is its direct sequel, and the SLOOS adds the missing half of the picture. Capacity was already the constraint. Now we know that whatever room exists is not being consumed by a rising volume of competing applications either.
Less competition for a fixed amount of capacity is a real advantage, and it is available this quarter to sponsors who structure into the box that exists.
What a Sponsor Does With This
The operational conclusion is narrow and it is not "wait."
A $1M–$5M sponsor reading "bank credit is loosening" and deciding to hold two more quarters for a wider box is waiting on an easing that already happened, at institutions that were never going to quote the deal. The correct move is the opposite: build a file that clears an unchanged box now, while the queue in front of it is short.
In practice that means three things. Arrive pre-underwritten. A capacity-constrained credit committee is not spending scarce room on a file that requires work to evaluate — trailing-twelve operating statements, a current rent roll, the capex actually spent versus budgeted, and the story of the business plan in one page. Identify the exit before you ask for the loan. A bank at or near the monitoring level is underwriting how quickly the exposure leaves as much as whether it should go on. Know where your incumbent sits before you call. Concentration position is a public number, computed from a public filing, and it predicts the opening posture of a renewal conversation better than payment history does. Walking in knowing it is the difference between negotiating and being informed.
None of that requires the credit box to move. All of it is available this quarter.
The Counter-Cases
Five things cut against the argument above, and the second is the most important paragraph in this post.
The panel is small, this is a survey, and it is not a clean time series. The July domestic panel is 56 banks; "other banks" is a subgroup of that. The Fed publishes no confidence intervals on these net shares, and a modest net share is a single-digit percentage of a small sample. The panel also changed between the two surveys compared here — April drew responses from 64 domestic banks, July from 56, alongside 18 U.S. branches and agencies of foreign banks in each — so the two cohort reads are consecutive observations rather than the same institutions tracked across quarters. And everything here describes what surveyed banks reported, not what the credit market did. Those are different claims and the distinction is not decoration.
Sub-$100B is not community banking, and the SLOOS cannot see your lender. The under-$100 billion bucket contains multi-tens-of-billions regionals sitting alongside institutions that look nothing like a $3 billion New Jersey bank. The survey does not sample the community bank cohort in any meaningful way. So the honest version of this argument is not that the Fed measured your sponsors' lenders and found them unmoved. It is that the survey cannot see them at all — which is precisely why a Call Report-derived tracker exists. What the join proves is cohort membership: every NJ-chartered bank sits on the non-easing side of the line the Fed drew. What it cannot prove is that any individual New Jersey bank behaved like the cohort average.
Standards are the gate; terms are what is inside it — and terms eased. The Fed measures these on separate axes and says so explicitly: standards are the extensive margin, the approval decision, while terms are the intensive margin, what a borrower gets once approved. This post is a standards argument, and on terms the reported picture runs the other way. The April 2026 survey carries an annual special question, asked each April for the past ten years, and banks reported having eased or left basically unchanged almost every CRE term over the trailing year — significant to moderate net shares cited higher maximum loan sizes, narrower spreads, and longer interest-only periods, and modest net shares reported having lowered debt service coverage requirements on construction and land development and on multifamily. So the two-sided version is this: on the extensive margin, the sub-$100 billion cohort reported flat-to-tightening standards across two consecutive surveys, while on the intensive margin the panel reported easier terms over the trailing year. The caveat that keeps this honest is that the terms question is panel-wide and is not broken out by bank size, so none of it can be attributed to either cohort.
The tracker's scope is charter, not geography. New Jersey-chartered banks are not the only banks lending on New Jersey property. New York, Pennsylvania, and national institutions lend here and file as institutions of their home states, and some of those are above $100 billion — so the cohort that eased is present in this market. Our operating observation is that it is not the cohort quoting $1M–$5M value-add multifamily. That is a read on who competes for that ticket size, not something either dataset demonstrates, and it should be weighed as such.
Quarter mismatch, stated in the open. Q1 tracker data cannot demonstrate Q2 behavior. It establishes cohort membership exactly, because the measurement dates match, and it establishes the capacity position going into the quarter. It does not confirm what those banks did during the quarter. Every tracker figure above is stamped as of March 31, 2026 for that reason.
The September Test
Here is the question, with a date on it.
Q2 2026 Call Reports were due to the Central Data Repository on July 30, 2026, and the Q2 edition of the tracker publishes in September. 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.
One hard caveat, because it would be easy to overclaim in September. The tracker measures balance-sheet outcomes; the SLOOS measures credit standards. Those are different objects. A bank can hold its standards constant and still grow its book, or tighten and still grow through a merger, or ease and shrink because payoffs outran originations. A convergent Q2 result would be corroborative, not confirmatory.
We will publish the answer either way, including the version where the balance sheets grew and this framing needs revising.
FAQ
Did banks ease commercial real estate lending standards in the second quarter of 2026? In the aggregate, yes. The Federal Reserve's July 2026 Senior Loan Officer Opinion Survey, published August 3, 2026, reported that banks generally eased standards for commercial real estate loans over the second quarter, with a modest net share reporting easier standards on multifamily. But the survey's own footnote splits that result by institution size: large banks reported easier standards for all CRE loan types, while other banks reported basically unchanged standards on multifamily and on construction and land development. The same divergence appeared in the April 2026 survey covering the first quarter, when moderate net shares of large banks reported easing all three CRE categories while other banks reported tightening on construction and land development (moderate net shares) and on multifamily (modest net shares), with nonfarm nonresidential basically unchanged — two consecutive surveys of the same pattern.
How does the Federal Reserve define a large bank in the SLOOS? As an institution with $100 billion or more in total domestic assets as of March 31, 2026. Every respondent below that threshold is grouped as an "other bank." The domestic panel for the July 2026 survey was 56 banks, alongside 18 U.S. branches and agencies of foreign banks.
Are New Jersey banks easing multifamily lending standards? The SLOOS does not report state-level results and does not meaningfully sample community banks, so it cannot answer that question directly. What the Call Report data does show is that as of March 31, 2026 — the same measurement date the Fed used to sort its panel — all 50 New Jersey-chartered banks held total assets below $100 billion, which places every one of them in the cohort that reported no change rather than the cohort that reported easing. Bank-by-bank figures are on our NJ Bank CRE Lending Tracker.
What should a sponsor do if their lender's credit box didn't loosen? Structure to the box that exists rather than waiting for a wider one. That means arriving pre-underwritten, with the exit already identified, and knowing where the incumbent bank sits on the CRE concentration table before the renewal conversation starts. A sponsor whose lender cohort reported weaker demand last quarter is competing against a thinner queue than in 2024 — appetite is not the binding constraint at these institutions, capacity is.
Send Us the File, Not the Headline
If you have a New Jersey multifamily loan maturing inside the next eighteen months, the useful exercise this quarter is not monitoring credit-conditions coverage. It is finding out where your incumbent bank sits on the concentration table, then building the file that clears the box as it is currently written. 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.
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. NMLS information available upon request. Lending in all 50 states.
Survey characterizations, the large-bank definition, the panel composition, and the level-of-standards special question are from the Federal Reserve's July 2026 Senior Loan Officer Opinion Survey on Bank Lending Practices, published August 3, 2026. The prior-quarter cohort comparison, the annual CRE-terms special question, and the extensive-margin/intensive-margin distinction between standards and terms are from the April 2026 survey. The Fed's magnitude terms are conventions of that survey: "basically unchanged" denotes a net share of 0–5%, "modest" more than 5% and up to 10%, "moderate" more than 10% and up to 20%, "significant" more than 20% and less than 50%. Bank-level figures are Blue Sky Capital Advisors analysis of Q1 2026 FFIEC Call Report data — data as of March 31, 2026, retrieved from the FDIC BankFind Suite API July 14, 2026. The next SLOOS supersedes the quarterly read above; the Q2 2026 tracker edition is expected in September 2026. This is market commentary, not a commitment to lend or investment advice.
Loans are for business purposes only and are not subject to TILA, RESPA, or HOEPA. Not for primary residences. Equal Housing Opportunity. All loans subject to underwriting approval. Rates and terms shown for illustration; actual rates depend on deal specifics. We do not lend to borrowers with credit below 600 or on owner-occupied properties.