brain/
sourcestock-market

Autoresearch: As of 19 Sep 2026, have CME, ICE, STT, BK, and NTRS printed later-than-Q1 2026 NII, deposit-beta, or clearinghouse collateral evidence? Has the Fed cut? Did exchange collateral/OI fall? Did a custodian disclose that >100% deposit beta offset reinvestment so NII failed to expand?

As of 19 Sep 2026: all five issuers printed Q2 2026 NII/collateral evidence; Q3 is unprinted; Fed hiked 25 bp on 16 Sep (no 2026 cut); custodian NII expanded; ICE collateral rose; CME cash-collateral averages drifted modestly.

Source

Autoresearch: As of 19 Sep 2026, have CME, ICE, STT, BK, and NTRS printed later-than-Q1 2026 NII, deposit-beta, or clearinghouse collateral evidence? Has the Fed cut? Did exchange collateral/OI fall? Did a custodian disclose that >100% deposit beta offset reinvestment so NII failed to expand?

Generated by /autoresearch on 2026-09-19. Synthesized across 3 rounds from 18 successfully retrieved web pages (plus a Grokipedia primer that is not used for any 2026 claim). See Provenance. Treat as raw material — review before promoting into a project or thread. Context: vault/projects/stock-market Research-only. No buy/sell/size.

Summary

Yes — later-than-Q1 primary evidence exists for all five names, and it is Q2 2026 (period ended 30 Jun 2026; prints 15–30 Jul 2026). Q3 2026 is unprinted as of 19 Sep 2026; official calendars put the five prints between 14 Oct and 29 Oct. The 2026 FOMC path through 16 Sep is five holds at 3-1/2 to 3-3/4 percent, then a unanimous 25 bp hike on 16 Sep to 3-3/4 to 4 percent — the “surprise easing” falsifier did not fire; policy moved the other way. Custodian NII expanded at STT, BK, and NTRS (none disclosed that a >100% deposit beta offset reinvestment so NII failed). Exchange evidence is mixed but not a collapse: ICE original margin and pledged collateral rose vs year-end 2025, and Q2 cash-margin NII rose vs Q2 2025; CME period-end performance bonds were slightly below year-end, Q2 cash-collateral earnings fell YoY on a lower average reinvestment rate, and rolling three-month cash collateral drifted $150.7B → $147B, while Q2 ADV stayed near last year’s record and July–August ADV rose. Existing-chain Q1 facts are not re-litigated here.

Findings

Q3 2026 is unprinted; the new primary is Q2

No Q3 2026 Form 10-Q or earnings 8-K was located for CME, ICE, STT, BK, or NTRS as of 19 Sep 2026. Issuer calendars that were retrieved schedule the next prints after this date:

  • STT — third-quarter 2026 results at approximately 7:30 a.m. ET on Wednesday, 14 Oct 2026, with the call at 11:00 a.m. ET (State Street remaining-2026 dates release, 5 Aug 2026).
  • BK — third-quarter 2026 report date Thursday, 15 Oct 2026, results ~6:30 a.m. ET, call 11:00 a.m. ET (BNY 2026 earnings-call schedule).
  • NTRS — “Third Quarter 2026 - Wednesday, October 21, 2026 at 8:00 a.m. CT” (Northern Trust IR).
  • ICE — Q2 release scheduled the Q3 2026 earnings call for 29 Oct 2026 at 8:30 a.m. ET (ICE Q2 2026 release on IR).
  • CME — no official HTML earnings-date page was successfully retrieved this pass. No Q3 2026 results page or 10-Q appears in the issuer/SEC search set; the next CME financial print is therefore treated as unprinted.

What is new versus the 20 Jul 2026 Q1-era chain is the full Q2 2026 10-Q / 8-K exhibit set for all five names (BK 15 Jul; STT 16 Jul; CME 22 Jul; NTRS 22 Jul; ICE 30 Jul).

Falsifier 1 — Fed easing in 2026: did not fire

Through 19 Sep 2026 the FOMC did not cut. It held 3-1/2 to 3-3/4 percent at the 28 Jan (Jan 28 statement), 18 Mar (Mar 18 statement PDF), 29 Apr (Apr 29 statement), 17 Jun (Jun 17 statement), and 29 Jul (Jul 29 statement) meetings. On 16 Sep it raised the target range by 1/4 percentage point to 3-3/4 to 4 percent, 12–0 (Sep 16 statement). The implementation note directs the Desk, effective 17 Sep 2026, to keep fed funds in 3-3/4 to 4 percent and sets IORB at 3.90 percent (Sep 16 implementation note).

July’s 9–3 hold had three dissenters who preferred a hike, not a cut (Jul 29 statement). January’s hold had two dissenters who preferred a cut (Jan 28 statement); they did not carry the Committee. This is the opposite of “surprise easing that would falsify step 1.”

Exchanges — CME Q2 2026: cash-collateral interest still earned; balances slightly down vs year-end; OI/ADV not collapsed

CME’s Q2 2026 10-Q (period ended 30 Jun 2026) reports performance bonds and guaranty-fund contributions of $158,110.7 million vs $159,656.1 million at 31 Dec 2025 — a small period-end decline (CME Q2 2026 10-Q). Cash on deposit at the Federal Reserve Bank of Chicago was $138.5 billion at 30 Jun 2026 (same 10-Q).

Q2 2026 earnings from cash performance bonds and guaranty-fund contributions were $1,388.8 million vs $1,487.6 million in Q2 2025; related distribution expense back to clearing firms was $1,265.7 million vs $1,374.5 million (same 10-Q). Implied net keep is about $123.1 million in Q2 2026 vs about $113.1 million in Q2 2025. MD&A states the YoY drop in gross earnings was “mainly due to a lower average rate of return despite higher overall reinvestment balances,” and the lower distribution expense was “due to lower interest income earned on our reinvestment balances as a result of a lower Federal Funds rate compared with the same period in 2025” (same 10-Q). H1 2026 cash-collateral earnings were $2,759.8 million vs $2,361.2 million in H1 2025 — higher balances more than offset the lower rate on a six-month basis (same 10-Q).

Q2 2026 aggregate ADV was 29,843 thousand contracts vs 30,217 in Q2 2025 (−1%) and H1 ADV was 33,011 vs 29,994 (+10%) (same 10-Q). The Q2 earnings release called H1 2026 the strongest in CME history and Q2 the second-highest Q2 volumes ever (CME Q2 2026 PR).

Post-Q2 monthly operational prints (not a 10-Q): July 2026 ADV was a record 27 million contracts, +23% YoY; rolling three-month customer cash collateral ending June was $150 billion (non-cash $170.4 billion) (CME July 2026 volume PR). August ADV was 29.7 million, +6% YoY; rolling three-month cash collateral ending July was $147 billion (non-cash $166 billion) (CME August 2026 volume PR). That is a modest cash-collateral drift, not an OI/volume collapse.

Exchanges — ICE Q2 2026: cash-margin NII up YoY; clearing collateral up vs year-end

ICE’s Q2 2026 10-Q footnote on cash-margin NII (included in total revenues) is $38 million (Exchanges) + $28 million (Fixed Income and Data Services) = $66 million for the three months ended 30 Jun 2026, vs $27 million + $28 million = $55 million in Q2 2025. Six-month 2026 cash-margin NII was $66 million + $54 million = $120 million vs $52 million + $54 million = $106 million in H1 2025 (ICE Q2 2026 10-Q via OpenCapital reprint of accession 0001571949-26-000011). Consolidated “Interest income” on the income statement (a different line) was $27 million in Q2 2026 vs $31 million in Q2 2025 (ICE Q2 2026 IR release) — do not confuse that residual with clearing-margin NII booked in revenues.

Clearing-house cash and invested deposits at 30 Jun 2026: original margin $106,430 million, guaranty fund $9,725 million, total $116,912 million, vs original margin $71,684 million and total $81,226 million at 31 Dec 2025 (ICE 10-Q collateral schedule, SEC R33). Cash and cash-equivalent margin deposits and guaranty funds were $114,598 million vs $76,782 million at year-end (same R33). Cash and non-cash original-margin plus guaranty-fund collateral pledged was $235.3 billion at 30 Jun 2026 vs $170.1 billion at 31 Dec 2025 (ICE Q2 2026 10-Q reprint). The exchange-collateral/OI-falls falsifier does not fire on ICE’s printed Q2 balance sheet.

MD&A attributes the Q2 OTC-and-other revenue increase “primarily [to] higher net interest income and fees on collateral balances” (same 10-Q reprint).

Custodians — STT, BK, NTRS Q2 2026: NII expanded; no >100% beta / NII-failure disclosure

BK. GAAP NII was $1,446 million in Q2 2026 vs $1,370 million in Q1 2026 (+6%) and $1,203 million in Q2 2025 (+20%); H1 2026 NII was $2,816 million vs $2,362 million (+19%) (BK Q2 2026 earnings release Ex. 99.1; BK Q2 2026 10-Q). The 10-Q: “Net interest income increased 20% compared with the second quarter of 2025, primarily reflecting the reinvestment of investment securities at higher yields and balance sheet growth, partially offset by deposit margin compression” (BK Q2 2026 10-Q). Compression is acknowledged; NII still expanded. Average deposits were $314 billion, +5% YoY, −1% QoQ (Ex. 99.1). On the Q2 call, CFO Dermot McDonogh said deposit betas would be “largely in line with the last cycle, and that was 80% for dollars and 60% to 70% for euros and sterling” (Motley Fool BK Q2 2026 transcript) — not >100%, and not framed as an NII failure.

STT. GAAP NII was $860 million in Q2 2026 vs $835 million in Q1 2026 (+3%) and $729 million in Q2 2025 (+18%); NIM (FTE) was 1.13% vs 1.16% in Q1 and 0.96% in Q2 2025 (STT Q2 2026 earnings release Ex. 99.1; STT Q2 2026 presentation Ex. 99.3). The release: “Net interest income increased 18% compared to 2Q25, reflecting an increase of 17 basis points in NIM. Net interest income increased 3% compared to 1Q26, primarily driven by a 4% increase in average interest-earning assets” (Ex. 99.1). Sequential NIM −3 bp is attributed to “deposit mix given growth in interest-bearing client deposit balances”; average total deposits were about $270 billion (interest-bearing $241 billion) (Ex. 99.3). FY2026 NII outlook was raised to +14–15% from +8–10%, “primarily reflecting stronger average deposit balances” (Ex. 99.3). No numerical deposit-beta figure and no “beta >100% offset reinvestment / NII failed” disclosure appear in the Q2 release, addendum, or slides (Ex. 99.2 addendum).

NTRS. FTE NII was $683.1 million in Q2 2026 vs $661.6 million in Q1 2026 (+3%) and $615.2 million in Q2 2025 (+11%); FTE NIM was 1.81% vs 1.75% and 1.69% (NTRS Q2 2026 earnings release Ex. 99.1). GAAP NII was $675.5 million vs $654.0 million and $610.5 million (same Ex. 99.1). The release: NII (FTE) rose sequentially on “lower funding costs and a favorable deposit mix, partially offset by lower deposits,” and rose YoY on “higher deposits and lower funding costs” (same Ex. 99.1). On the Q2 call, CFO Dave Fox: “The beta we like to say is a combination of the wealth. Beta is much lower than the institutional beta. We round it out to about 80% in total on dollars” (Motley Fool NTRS Q2 2026 transcript). That is not >100%, and NII expanded.

Falsifier 3 (custodian NII fails despite higher rates / >100% beta offsets reinvestment) did not fire on Q2 prints. All three reported higher NII YoY and QoQ. Where beta was spoken, it was ~80% dollars (BK, NTRS), not >100%. Deposit-margin compression is real at BK and shows up as a STT sequential NIM dip; it did not reverse NII expansion.

What this does and does not do to the existing chain

  • Step 1 (elevated-rate interest on client cash / mandatory collateral): Q2 primary updates and does not overturn the Q1 confirmation. BK Q2 NII $1.446B (+20% YoY); CME still earned $1.388.8B of cash-collateral interest (net keep ~$123M after rebate); NTRS FTE NII $683.1M (+11%); ICE cash-margin NII $66M in Q2 (up vs $55M). The Fed hiked in September rather than cutting.
  • Step 2 (near-costless float → high-margin NII): stronger than the prior single-source blog for exchanges, because CME’s 10-Q now shows the rebate mechanics in dollars and ICE’s 10-Q books cash-margin NII inside revenues. Still not a new independent mechanism write-up; Q3 unprinted.
  • Step 3 (YTD NII / betas / ICE collateral): Q2 prints replace the “Q2 unfetched as of 21 Jul” gap for these five names. STT +18%, BK +20%, NTRS +11% YoY NII; ICE clearing collateral up vs year-end. Spoken betas ~80%, not >100%. SCHW is out of scope for this pass.

Contradictions and open questions

  • CME period-end performance bonds ($158.1B) are slightly below 31 Dec 2025 ($159.7B) while ICE pledged collateral is sharply higher vs year-end. Rolling three-month CME cash collateral also drifted $150.7B (3m to May, from the Q2-era volume release cited in search) → $150B (3m to June) → $147B (3m to July). That is not an OI collapse (July/August ADV up), but it is the only printed sequential down print on the exchange-collateral falsifier and should be checked again at the Q3 10-Q.
  • CME Q2 gross cash-collateral earnings fell YoY even as H1 rose and net keep ticked up. MD&A blames a lower average reinvestment rate (lower fed funds vs Q2 2025), not a balance collapse. The September hike, if sustained into Q4, is a rate tailwind that Q2 does not yet contain.
  • ICE “Interest income” $27M ≠ cash-margin NII $66M. The load-bearing float number is the revenue footnote, not the other-income interest line.
  • No Q2 numerical deposit-beta table in STT’s 8-K exhibits; BK/NTRS betas are call commentary (~80%), not 10-Q line items. The prior-chain “custodian betas >100%” claim is not restated in these Q2 prints.
  • Q3 2026 NII, beta, and collateral are unknown until mid-to-late October. The 16 Sep hike lands inside Q3 and will first show in those prints.
  • CME official Q3 earnings-date HTML was not fetched this pass (secondary reprints say 21 Oct). Treat as unprinted, not as a dated commitment from an unretrieved page.
  • Grokipedia’s NII primer is definitional only and is not a source for any 2026 figure above.

Provenance

Rounds run: 3 of 3 (full). No early exit — Q3 calendars, post-Q2 monthly collateral, and the full 2026 FOMC path were material.

Sub-questions by round:

Round 1 (broad survey):

  1. Has the FOMC cut the funds rate in 2026 through 19 Sep, and what is the current target range?
  2. What did CME print in Q2 2026 (or later) for cash-margin interest, collateral, and OI?
  3. What did ICE print in Q2 2026 (or later) for clearing collateral and interest income?
  4. What did STT, BK, and NTRS print in Q2 2026 (or later) for NII and deposit beta?
  5. Have any of the five issuers released Q3 2026 earnings as of 19 Sep 2026?

Round 2 (drill-down):

  1. Confirm the full 2026 FOMC path (Jan–Jul) — targeting “no-cut” vs a single September observation.
  2. Did any custodian disclose >100% beta that offset reinvestment so NII failed? — targeting falsifier 3.
  3. CME sequential collateral / OI vs year-end and vs Q1 — targeting the exchange-collateral falsifier.
  4. Official Q3 calendars from issuer IR/SEC rather than aggregators.

Round 3 (resolve remaining uncertainty):

  1. Post-Q2 (Jul–Aug) CME monthly ADV / cash-collateral prints — targeting whether collateral kept drifting after the 10-Q date.
  2. Remaining official Q3 calendars (BK, NTRS, ICE) and CME date-page fetch.
  3. ICE Q2 cash-margin NII footnote on the 10-Q (not the $27M interest-income line).

Anchor source (Grokipedia, fetched before round 1; primer only — not used for 2026 claims):

  • Net interest income — 4,017 chars extracted via _lib/grokipedia.py — definitional NII/NIM vocabulary only. Encyclopedia-as-source forbidden for 2026 issuer/Fed facts; none of the Findings cite it.

X sources: not requested (--include-x off).

URLs fetched (18 successful, 6 failed/blocked):

Round 1:

Round 2:

Round 3:

Tools used: WebSearch, WebFetch, curl (browser UA for Fed HTML and selected IR pages after WebFetch timeouts), grokipedia-fetch helper (_lib/grokipedia.py). Consulted vault/_meta/SOURCE_RELIABILITY.md (prefer www.sec.gov, www.federalreserve.gov, www.prnewswire.com, www.fool.com; *.gov PDFs allowed). Generated: 2026-09-19 21:30 UTC

Referenced by
Mechanisms
Client float interest → custodian & exchange NII re-rate