Mapping monetary transmission through funding markets

Players, horizons, frictions — from Pozsar's Global Money Notes #20–26 (2019) to the June 2026 landscape, with an estimation agenda

Working document · 5 June 2026 · companion interactive map: the transmission map · results: estimation results

Purpose and approach

Goal: build a usable map of how changes in the policy rate actually propagate through US dollar funding markets — who the players are, on what clock they act, where the chain can jam — as the foundation for a trading/market framework that anticipates funding stress and rate dynamics. The map is built in three layers: (i) Pozsar's 2019 micro-plumbing baseline (GMN #20–26, the seven notes in the Funding folder); (ii) everything that changed institutionally between 2019 and mid-2026; (iii) for each mechanism, the quantity we could estimate once we go get data. Data work is deliberately deferred; every estimation row names its likely sources.

The 2019 baseline: the causal arc of GMN #20–26

The seven notes form one argument. The Fed's 2017–18 hikes pushed 3-month FX-hedging costs above the entire Treasury curve, so the marginal buyer — FX-hedged foreign real money rolling 3-month swaps — went on strike (#20). Carry traders exited in sequence (FX-hedged Oct-18, Libor-funded Jan-19, classic May-19), leaving primary dealers as forced residual buyers with record inventories financed overnight (#23). Meanwhile taper, TGA rebuilds and the uncapped foreign repo pool destroyed or sterilized reserves (#21, #24), until the lendable “free float” of reserves was roughly zero against the pace of collateral supply (#22). The system's shock absorber — a few G-SIBs, above all JPMorgan, toggling HQLA between reserves and repo — hit intraday-liquidity and G-SIB-score limits (#22, #26). September 17, 2019 (SOFR 5.25%, EFFR outside the band) was the predicted collision; the proposed fixes were a standing o/n repo facility priced at the top of the band, open to banks without stigma (#25), and ultimately outright purchases (#26).

The players: role, clock, constraint

Transmission is a relay race between balance sheets. Each runner has a different clock and a different reason to drop the baton.

The transmission chains

Five chains carry the policy rate outward. Each is listed with its weak link.

• Administered-rate chain. Target range → IORB / ON RRP / SRF → EFFR and SOFR. Pass-through is mechanical while reserves are ample; when scarce, SOFR detaches upward from IORB (late 2025). Weak link: reserve distribution — averages mislead, the marginal bank matters.
• Collateral chain. Treasury supply (settlement days, bill share, TGA swings) → dealer inventories → repo rates → SOFR → everything indexed to it. Weak link: dealer balance-sheet capacity at reporting dates; nonlinear once utilization is high (Duffie et al.).
• Leverage chain. MMF cash → FICC sponsored repo → hedge-fund basis books → cash Treasuries vs futures → term yields and swap spreads. Weak link: margin spirals and the absence of any backstop for a forced unwind.
• Global dollar chain. o/n rates → t/n and 3m FX swap implied yields (OIS + XCCY basis) → hedged UST carry vs home bogeys → foreign participation → spot/forward FX. Weak link: the year-end turn premium and episodic re-hedging panics; Pozsar's 3FXs/10s gauge still organizes this chain.
• Credit chain (new since 2023). SOFR → floating-rate loans, FRNs, securitizations. LIBOR's death means policy and repo squeezes now pass directly into borrowing costs through collateral markets, and bank credit stress no longer shows in the benchmark. This raises the stakes of ceiling integrity — a key reason the SRF went to full allotment.

Frictions catalogue

What changed, 2019 → 2026

Estimation agenda — how to measure transmission

Each row turns a piece of the map into something estimable. Methods favor what a trading framework needs: real-time, spread-based, event-driven. Data column names sources to be collected later.

Identification episodes

The same shock class hit different institutional regimes — the cleanest identification we will get:

• Sep 16–17, 2019: tax date + settlement day, no ceiling, free float ≈ 0 → SOFR 5.25%, EFFR breaks band. The full Pozsar mechanism.
• Mar 2020: dash-for-cash; dealer capacity overwhelmed; basis unwind; Fed restores via QE + temporary facilities (FIMA born).
• Mar 2023: SVB; FHLB-before-window pecking order confirmed; FIMA $60bn draw; stigma intact.
• Apr 2025: tariff shock; 10y ±50bp intraday, swap spreads collapse, partial basis unwind, no Fed intervention — fragility revealed, backstop question unanswered.
• Oct–Dec 2025: scarcity 2.0 with the new toolkit: SOFR > IORB, record SRF use ($50bn Oct 31; $74.6bn year-end), QT stopped, cap removed, RMPs launched. Same shock as 2019, ~20bp of ceiling leak instead of ~300bp of spike.

Open questions the framework should trade on

• Redistribution: can full-allotment SRF cash reach non-counterparties fast enough at year-ends, and does FICC settlement of SRF trades (discussed, not done) close the gap?
• Does eSLR recalibration (Apr-26) measurably compress turn premia? First clean read: June and December 2026 turns.
• Where is LCLoR? Waller's 8–9% of GDP vs Logan's lean view; late-2025 says ~$2.9T was already tight. RDE is the live gauge.
• Who eats a forced basis unwind after mandatory repo clearing (Jun-27) — CCP margin, dealers, or a Kashyap–Stein-style facility that doesn't exist yet?
• Stablecoins at $1–2T: does front-end demand become a stabilizer (deep bill bid) or a new run channel with no lender of last resort?
• Fiscal dominance of the front end: with ATI persisting, bills, RMPs and stablecoin demand interact — who controls the bill rate, Treasury or the Fed?

Key sources

Primary: Pozsar, Global Money Notes #20–26, Credit Suisse (Feb–Dec 2019). The full source register is on the sources page.

• Fed framework & facilities: FOMC/NY Fed operating statements — SRF launch (Jul 28, 2021), morning ops (May 28, 2025), QT end (Oct 29, 2025), full allotment & RMPs (Dec 10, 2025); Perli speeches Nov-25–May-26; Waller (Jul-25); Logan (Oct-25); Williams (Sep-25).
• Regulation: eSLR final rule (Fed/OCC/FDIC, Nov 25, 2025; eff. Apr 1, 2026); Basel III endgame & G-SIB reproposals (Mar 2026); SEC Treasury clearing rule (Dec 2023) and extension (Feb 25, 2025 → cash Dec-26, repo Jun-27); SEC MMF reform (Jul 2023); GENIUS Act (Jul 18, 2025).
• Research: Afonso et al., “Market Events of Mid-September 2019” (SR918); Kahn et al., “Anatomy of the Repo Rate Spikes” (OFR 23-04); Copeland–Duffie–Yang, “Reserves Were Not So Ample After All” (QJE 2025); Afonso et al., Reserve Demand Elasticity (NY Fed, published monthly since Oct-24); Lopez-Salido & Vissing-Jorgensen; Duffie, “Resilience Redux” (Jackson Hole 2023) and SR1070; Barth–Kahn basis-trade papers; Kashyap–Stein–Wallen–Younger basis-purchase facility (BPEA 2025); OFR “Sizing the US Repo Market” (Dec 2025); Fed Notes “Cross-Border Trail of the Treasury Basis Trade” (Oct 2025).
• Market structure & players: OFR Hedge Fund Monitor; Liberty Street Economics on sponsored repo (Oct 2025) and intraday demands (Oct 2024); BIS Quarterly (Dec 2025); CFR on Taiwan lifers; TBAC charges 2025–26; Hudson Bay “Activist Treasury Issuance” (2024).