IS THE US$ LIQUIDITY CRUNCH A SIGN OF IMPENDING FINANCIAL MELTDOWN?
On the sidelines of the annual Treasury Market Conference in New York on 12 Nov 2025, Federal Reserves Bank NY (FRBNY) met with representatives of banks who are primary dealers. That triggered off media reports of a "secret emergency" meeting that suggests a liquidity crisis is brewing, the fear being a liquidity crunch often precedes a financial meltdown, such as the one in 2008
Managing the U.S. financial system is like operating the world’s most complex water system — one that has to self-correct in real time, respond to local and global shocks, and never run dry or overflow. The financial market is complex and dynamic where imbalance in one part can trigger stresses elsewhere. Just like the water system where various valves and faucets have to be tweaked constantly for correct pressure, reserve levels, meeting dynamic demands, etc., someone has to do plumbing constantly to keep the flow of money steady. That responsibility falls on FRBNY, one of 12 regional Federal Reserve Banks.
WHERE DID ALL THE MONEY GO?
When there is low liquidity, it means there is a shortage of money in the system. So where did all the cash go? The feature image above tells the story. The Treasury General Account (TGA) is the US government's operating account at the Fed. The extreme right of the chart shows a spike in the TGA building up in the later part of 2025. Huge amount of USD has been sucked up by the government which is currently sitting on about US$1 trillion.
There is a convergence of factors that caused this massive cash inflow into the TGA in 2025 -- massive securities issuance, QT (Quantitative Tightening), Trump tariff shock, September tax collections, government shut-down.
Securities issuance:
Treasury Dept normally issues massive bills/notes/bonds when there is an increase in debt ceilings and lifting of suspension on debt ceilings. Prior to lifting of limit suspension, Treasury usually front loads on debt. The chart shows co-relation of TGA spikes:
* December 2021, the debt ceiling was increased by $2.5 trillion to $31.38 trillion. Treasury increased bill issuance leading to early 2022 TGA spike.
* Debt ceiling was suspended on 3 June 2023 till 31 December 2024. Treasury front-loaded bill issuance before ceiling was re-imposed, causing a mid 2024 TGA spike.
* July 2025, the debt ceiling was raised by US$5 trillion to $41.1 trillion as part of the "Big Beautiful Bill". Treasury increase bill issuance causing current TGA spike.
The only exception where TGA spike was not due to increase in debt ceilings and lifting of suspension on debt ceilings was 2020. This was due to massive debt issuance caused by the Covid pandemic.
QT (Quantitative Tightening)
From time to time the Fed may do Quantitative Easing (QE) which is the buying of government securities from the market. This way, liquidity is pumped into the market, which is the purpose of QE. Fed Balance Sheet ↑ Bank Reserves ↑ Market Liquidity ↑ This is strictly a Fed monetary policy action so it does not impact TGA. Once these notes/bonds/bills are purchased, the Fed holds them to maturity. On run-off date (maturity) of bills/notes/bonds, the Fed's balance sheet is decreased and TGA is reduced to cancel out the debt. Fed Balance Sheet ↓ TGA ↓.
QT is when bills/notes/bonds run-off (matures) at a time of too much liquidity in the market with inflationary tendencies and the Fed does roll-overs of the matured securities. Treasury will issue new securities to replace the maturing ones. TGA ↑ Bank Reserves ↓ Market Liquidity ↓. QT has been in effect since Jun 2022 and from April 2025 on, it was capped at US$60b per month for foth treasuries and mortgage-backed securities.
Tax
The timing for corporate tax payment is complicated and collection is spread throughout the year. But generally around the month of September each year, a substantial sum is collected by the government. A rough estimate of US$30b-US$50b came in Sep 2025. TGA ↑ Bank Reserves ↓.
Tariffs
There has been a lot of figures bandied about regarding revenue from Trump's tariff. Figures have ranged from US$0.5 to US$1 trillion. These are just rhetorics. According to USA Facts, excise duties in 2024 was US$77b and from Jan-Aug 2025 was US$165b. So the tariff war has brought in only about US$88b. TGA ↑ Bank Reserves ↓.
Government shutdown:
The government shutdown, the longest in history, lasted 43 days from October 1 to November 12, 2025. This deferred government spending of about US$11.5 billion daily means no outflows to bank reserves which exacerbated liquidity. In aggregate, a sum of US$495 billion which should have been spent into circulation was held back due to govt shutdown.
WHERE DOES MONEY COME FROM?
All USD cash transactions end up as entries in the Banks Reserve Accounts at the Fed, basically one bank paying another. The US$1,000 you in Singapore secretly remit to your mistress in Timbuctoo, ya that too, after winding its way through networks of correspondent banks, digitised clearing platforms or Fedwire, end up reflected in the bank reserves with the Fed. Another way of looking at it is the aggregate Bank Reserve Balances support all the daily USD settlement transactions of the entire world.
So whenever cash is sucked up by the government into the TGA, the bank reserves are reduced. It ended up as entries in the Fed's books, TGA account increase and bank reserves reduce. That means there is less cash to support all the USD transactions. You can appreciate the sensitive nature of this realisation if you can grasp the macro manifestation.
In the context of the current liquidity crisis, the short-term wholesale funding market is key. This is the web of short-term, unsecured or secured, dollar-denominated funding transactions that happen outside of retail deposits. Almost everything that funds banks, broker-dealers, money-market funds, hedge funds, corporates, and the Treasury market transact on a daily or overnight basis. It is currently about US$25-US$26 trillion. These markets comprise, ranked by market size - FX swaps (the USD leg), repos, commercial papers + CDs, FHLB (Federal Home Loan Advances) Fed funds and Overnight Eurodollar. FX swaps is about US$8 trillion, Repos is US$5 trillion.
About 40%-50% of these entire stack rolls, matures, or has to be refinanced every single business day. That is, about US$11-US$13 trillion (on a netted basis) change hands daily. And this gargantuan number is supported by the bank reserves, not forgetting the millions of retail transactions such as your US$1,000 remittance to the mistress.
Here's a snapshot data to understand the leverage involved:
The velocity provides a snapshot view of the liquidity situation. Based on Fed's reserve demand models and historical market data, there are certain thresholds:
Stable/Ample Regime:
Velocity is low (~2.5 to 3.5x per day). There is abundant reserves relative to turnover. Banks lend freely, money market rates (e.g., SOFR) stay stable and close to the interest on reserve balances (IORB, currently ~5.40%), and there's minimal rationing or volatility.
Transition/Scarce Regime: Velocity rises (~3.5 to 4.5x), as reserves tighten. Banks hoard more cash, funding spreads widen (e.g., SOFR-IORB >10-15 bps), and usage of facilities like the Standing Repo Facility (SRF) increases.
Red Line (Stress Threshold): Velocity exceeds ~4.5 to 5x, signaling acute scarcity. Reserves are insufficient for smooth recycling, leading to sharp rate spikes (SOFR-IORB >20-30 bps), collateral (securities) hoarding, and potential market seizures (echoing 2019 repo crisis).
What I have done here is to give a clear sense of the macro scale The aggregate bank reserves is "fuel" pool for the entire system. The crucial part that it supports is the daily turnover in the core rate-sensitive segments of the wholesale funding market. This highlights the "leverage" dynamic: Reserves are only ~20-25% of daily turnover volume, meaning the system relies on rapid recycling of reserves via intraday and overnight settlements. When the government sucks cash out of the system, the bank reserves is reduced $ for $. As reserves shrink, even small percentage drops create outsized stress—banks ration harder, repo rates spike (e.g., SOFR-IORB from ~8-9 bps in August to 32 bps now), and rollover risks amplify.
As you can see, the velocity was in "stable" region at 3.22 in January 2024, and it has risen to 3.63 in the Scarcity region by January 2025. It has continued to rise to current level of 4.15, still in the "scarcity" region, and the trajectory is towards the "stress" threshold of 4.5. In short, the liquidity status is serious, but not in the danger zone yet.
The US$610 billion reserves slide (from Jan 2024 to current date) amid stable $12T turnover in wholesale financing markets underscores the problem's mechanical scale. It is not a solvency crisis, but a "quantity crunch" where fiscal drags outpace monetary normalization, risking broader spillovers if unchecked.
THE MACRO EXPLANATION
Fiscal drags outpace monetary normalization, risking broader spillovers if uncheckedThat line basically explains the status quo. Let me try to translate the central bank-speak to English.
Fiscal drag:
'Fiscal' refers to what the Treasury does - all about revenue and spending;
'Fiscal drag' refers to the times when the government's cash inflow is more than its cash outflow This happens every time the government (a) spends less than it takes in, or (b) deliberately collects more cash than it immediately spends, that cash is sucked out of bank reserves and parked in the TGA. The fiscal drag is thus represented by the increase in TGA levels.
Monetary normalisation:
'Monetary' refers to what the central bank or Fed does - all about interest rates (for MAS it's about exchange rates) and the money supply within an economy.
The 'normal' situation for the system is envisaged by the Fed to fall within what it called a "Ample Regime" where certain quantitative criteria are met in a given size of economy.
* Quantity Easing - QE is where Fed buys govt securities in secondary market by "printing" money into the reserves. Balance Sheet ↑ Reserves ↑ Liquidity ↑. All securities purchased are held to maturity. The Fed pays for these with money they do not have, by simply crediting the reserves accounts of the banks. Liquidity is loosened.
* Quantity Tightening w/o re-investing - When securities run-off, the govt pays Fed. Balance Sheet ↓ TGA ↓. This is QT without re-investing which is in effect, the government paying off national debt. Has no impact on liquidity.
* QT w/re-investing - Securities run-off, the govt pays Fed. Balance Sheet ↓ TGA ↓.. The government issues new securities to replace the maturing one. TGA ↑ Bank Reserves ↓ Liquidity ↓ . In effect, QT with re-investing is a roll-over of maturing securities. Generally, this is what the market means when they talk of QT -- tightening of liquidity.
To have context, let's go back to 2008 housing-mortgage financial crisis and the 2020 Covid-19 pandemic.
The bail-out programme in 2008 saw the Fed bloated its balance sheet from US$0.9 trillion to US$4.9 trillion. It had 'printed' money and boosted bank reserves to purchase MBS (mortgage-backed securities). Balance Sheet ↑ Reserves ↑ Liquidity ↑.
From 2015-2019 it started normalising by letting MBS run-off. Fed Balance Sheet ↓ Bank Reserves ↓ Market Liquidity ↓. By 2019 it had brought its balance sheet down to US$3.8 trillion which means decreasing bank reserves by the same amount. When MBS matures, the Fed is paid and bank reserves is reduced. Liquidity is soaked up and in 2019 short term repo market was stressed out. That was the 2019 repo crisis. So the normalisation effort paused.
Then came the pandemic in 2020. With lockdowns and unemployment hitting new lows, and repo (short-term interest money) stressed out, the Fed reversed gear to pump liquidity into the market. It went on massive QE exercise, 'printing' money to buy securities which pumped money into the bank reserves. Fed's balance sheet bloated to US$9 trillion and the market was flushed with cash, short-term interest rates plummeted. Balance Sheet ↑ Reserves ↑ Liquidity ↑.
From 2022 Fed started normalising again with QT which was targeted to end on 1 Dec 2025. The normalisation efforts brought the situation to a good place by Jan 2025. 15% of the pandemic expansion had been cleared as Fed balance sheet decreased to US$7.1 trillion. Many other metrics were looking good in Jan 2025. Bank reserves was US$3.32 trillion, short-term rate SOFR–IORB spread was just 6-9 basis points. So for the Fed, Jan 2025 the ship was in calm waters and looking to a comfortable landing by 1 Dec 2025 when the QT ends and the situation should be normalised.
Let's circle back to the statement : "Fiscal drags outpace monetary normalization, risking broader spillovers if unchecked".
The Fed's monetary action of QT timed to end on 1 Dec 2025 was meant to reduce reserves to a certain "ample" level that would have normalised the system. Treasury's fiscal actions in 2025 (massive securities issuance, tariff shock, tax payments, and government shutdown) caused a fiscal drag, that is, more cash inflows than outflows at TGA, which means sucking cash out of bank reserves. The fiscal drag "outpaced", or takes out more cash off reserves than, the monetary normalisation (the on-going QT). The result is the reserves level goes from "ample" to "scarcity" level and the liquidity tightening may spillover into the broader market, that is, the risk spreads into other markets. For example, forced asset sales with huge losses, markets deleverage, margin calls, crypto/equities liquidate, hedge funds unwind, etc.
SHORT TERM INTEREST RATES
In normal economic conditions, the USD interest rate hierarchy follows the typical upward-sloping yield curve -- shorter-terms : lower rate; higher-terms : higher rate. Short-term rates are anchored by the Federal Reserve's policy rate (the fed funds rate), while longer-term rates are market-determined. Deviations in the hierarchy signifies stress in the system. An inversion indicates the market's expectations of coming economic slowdown.
The current situation sees spikes in short-term rates, widening spreads, curve dynamics of government securities shifted - yields of short-term T- Bills is higher than 10-year bonds.
Hierarchy of US interest rates:
(1). ON RRP (Overnight Reverse Repo Rate)
This is a facility that allows Feds to take deposits from MMFs (mutual funds), GSEs (Fannie Mae etc), banks and primary dealers. Since Fed is sole taker of funds, it controls the rate. This is the absolute floor rate for all money market ON trades. It is set 10 basis point below IORB.
2 Jan 2025 : 4.25%
This is a facility that allows Feds to take deposits from MMFs (mutual funds), GSEs (Fannie Mae etc), banks and primary dealers. Since Fed is sole taker of funds, it controls the rate. This is the absolute floor rate for all money market ON trades. It is set 10 basis point below IORB.
2 Jan 2025 : 4.25%
12 Nov 2025 : 3.75%. Usage down 13% as funds seek higher repo yield.
(2). IORB (Interest on Reserves of Banks)
It is a major terminology for secured overnight funding and serves as a key benchmark for the broader financial market, it is not a specialized market unto itself. Meant to replace LIBOR. Published daily by Fed based on actual market data.(2). IORB (Interest on Reserves of Banks)
Fed pays overnight (ON) rates on reserve balances of depository institutions. As the interest payer, Fed controls this rate. This sets the ceiling for most rates
2 Jan 2025 : 4.00%
12 Nov 2025 : 3.90%. When SOFR breach this, it signals stress.
(3). SOFR (Secured Overnight Financing Rate)2 Jan 2025 : 4.00%
12 Nov 2025 : 3.90%. When SOFR breach this, it signals stress.
2 Jan 2025 : 4.32%
12 Nov 2025 : 3.95%. Spiked to 4.01%. Hit upper limit multiple days, widest IORB spread since 2020.
.
(4). EFFR (Effective Fed Funds Rate)
The Fed funds is the market where depository institutions (commercial banks, GSEs) borrow/lend each other without collaterals. Primary Dealers are not allowed to participate here. Trades are predominantly overnight, but 2-90 days possible). The market is not very big -- about US$20-US$30 billion daily. The EFFR is the weighted average rate. Fed funds rate are market rates, the Fed has no direct control, but can only influence it by tweaking the IORB and ON RRP.
No bank will lend in Fed funds below IORB (they would rather keep the reserves and earn IORB). No bank will lend much above IORB because the borrowing bank can always get reserves from the Fed’s standing repo facility or discount window if desperate. Result: EFFR almost never trades more than 2–3 bps away from IORB in normal times, and even in stress it stays inside the Fed’s target range.
2 Jan 2025 : 4.37%
12 Nov 2025 : 3.98%. Near midpoint but firming (up 18 bps vs. RRP); less volatile than SOFR but trending higher.
12 Nov 2025 : 3.98%. Near midpoint but firming (up 18 bps vs. RRP); less volatile than SOFR but trending higher.
This is the interbank-bank money market outside of US involving offshore dollar deposits. The Eurodollar market has largely transitioned from LIBOR (phased out by September 30, 2024) to proxies like the overnight SOFR + a small credit spread (~5–10 bps historically, reflecting unsecured risk). Since no direct "overnight Eurodollar" rate is published post-LIBOR, I've used the standard fallback: Overnight SOFR + 6.44 bps (the ISDA fixed spread for overnight LIBOR transitions). This aligns with market conventions for legacy contracts and current benchmarks.
The overnight Eurodollar rate spikes first and hardest in crunches due to its unsecured, global exposure. It acts as an early warning for broader dollar scarcity. It is a volatility amplifier for the broader global USD market to what is happening in US. It only pulls away noticeably (20–200+ bps) when a real dollar liquidity crunch hits and unsecured offshore funding becomes expensive or scarce. This is exactly what we are seeing again in November 2025. Euro$ has no access to Fed relief and help comes in the form of central bank utilising their FX swap lines to provide liquidity.
The overnight Eurodollar rate spikes first and hardest in crunches due to its unsecured, global exposure. It acts as an early warning for broader dollar scarcity. It is a volatility amplifier for the broader global USD market to what is happening in US. It only pulls away noticeably (20–200+ bps) when a real dollar liquidity crunch hits and unsecured offshore funding becomes expensive or scarce. This is exactly what we are seeing again in November 2025. Euro$ has no access to Fed relief and help comes in the form of central bank utilising their FX swap lines to provide liquidity.
2 Jan 2025 : .4.46%
12 Nov 2025 :4.04%
(6).GFC repo
These are tri-party repos, traded through an exchange.
2 Jan 2025 : 4.42%
12 Nov 2025 : 4.007%. ~2 bps over SOFR; blind-brokered, saw 5–8 bps rise in early Nov on liquidity drain.
12 Nov 2025 :4.04%
(6).GFC repo
These are tri-party repos, traded through an exchange.
2 Jan 2025 : 4.42%
12 Nov 2025 : 4.007%. ~2 bps over SOFR; blind-brokered, saw 5–8 bps rise in early Nov on liquidity drain.
(7). Private bilateral repo (ON)
Mostly banks and hedge funds. This is most sensitive and the first to spike.
2 Jan 2025 : 4.43%
12 Nov 2025 - 4.01%. ~3 bps over SOFR; credit risk amplifies spikes (up to 10 bps wider in stress).
(8). FHLB short-term advances
There are 11 Federal Home Loan Banks and they are really the unsung heroes in times of liquidity stress that most people don't realise. Basically what they do is accept private-label or illiquid mortgage collateral (whole loans, non-agency MBS, CRE loans, etc.) from member banks at very low haircuts. They lend to member banks and fund themselves by issuing new debt (Consolidated Obligations) that the market prices almost identically to Treasury or Agency debt (typically 3–12 bps over SOFR, sometimes tighter than Fannie/Freddie). The FHLB system quietly transforms private-market credit and liquidity risk into de facto U.S. government risk, and it does so on a gigantic scale during a liquidity stress. FHLB does not create reserves, in effect it redistributes reserves from “cash-rich” entities (prime money funds, foreign banks, large corporates) to “cash-poor" banks with extreme efficiency and with almost no stigma. FHLB is basically a 'too-big-to-fail' institution and their debts are held as HQLA (high quality liquid asset) for regulatory liquidity maintenance purposes.
When banks are short of reserves or facing funding pressure, they have three main ways to improve their LCR (liquidity coverage ratio) instantly:
(9).SRF (Secured Repo Facility)
12 Nov 2025 : 4.00%.
12 Nov 2025 : 3.82%. ~8 bps below EFFR; yields firming slightly on auction demand but still down YTD.
Eurodollar-type rates become extremely important as they are usually the first and most violent movers upward in a dollar funding squeeze. They decouple sharply upward and become one of the clearest stress indicators in the entire USD rate complex, often moving faster and farther than repo rates or T-bills.
2 Jan 2025 : 1-month - 4.30%; 3-month - 4.4.32%.
12 Nov 2025 : 1-month - 4.32%; 3-month - 3.92%. 1-month .~5 bps below EFFR; unsecured, so more sensitive to crunch (spread widening). 3-month - ~3 bps below EFFR; forecasts show avg. 3.883% for Nov, but end-month uptick expected.
(13) 1-3 month Commercial Papers/ CDs
2 Jan 2025 : 4.35%
12 Nov 2025 :3.93%. ~5 bps over T-bills; unsecured risk premium up ~5 bps amid funding hoarding.
(14) Discount Window
This is Fed's "lender of last resort" facility for banks only. Primary dealers excluded.. It is meant for banks in distress so banks avoid this due to credit stigma.
2 Jan 2025 : 5.00%
12 Nov 2025 :4.50%. Fixed ceiling (+50 bps over upper target); SRF usage signals banks avoiding it due to stigma.
HOW THE FED CONTROLS LIQUIDITY:
In a liquidity crunch banks tend to hoard cash, moving away from higher risk assets. This tends to exacerbate the situation. Short term interest rates tend to rise.
Yet, in the above descriptions of various types of short term rates, a comparative of January 2025 and November 2025 shows interest rates have dropped! It's an anomaly. The reason is simple. This is due to the Fed's intervention. The US is an interest rate regime (Singapore is exchange rate regime) where the Fed uses interest rates as a monetary tool. The Fed sets interest rate ranges which guide short-term rates like the EFFR and SOFR, influencing borrowing costs economy-wide. Recent rate reductions were :
* 18 Sep 2024: From 5.25%-5.50% down to 4.75%-5.00%.
* 7 Nov 2024: down to 4.50%-4.75%
* 18 Dec 2024: down to 4.25%-4.5%
* 16 Sep 2025: down to 4.00%-4.25%
* 28 Oct 2025: down to 3.75%-4.00%.
The Federal Reserve operates under a statutory "dual mandate" from Congress: promote maximum employment and stable prices (targeting 2% inflation). For this, the Fed pays close attention to employment data and inflation trends.
The Fed's monetary tools to manage liquidity is primarily balance sheet actions like QE and QT (which ended December 1, 2025) or the Standing Repo Facility (SRF). Rate cuts are more about the broader economy. It is not explicitly a "liquidity injection" tool but used to calibrate the Fed's open market operations.
A GAUGE OF LIQUIDITY STRESS:
There are 2 key ways to see a liquidity stress developing -
a. The short term interest rates -- not the rates per se, but the spreads.
b. The usage of the SRF.
SRF
The Standing Repo Facility (SRF, min bid 4.00%) acted as the "ceiling enforcer," with usage quantifying stress—low/zero on calm days, spiking when SOFR exceeded 4.00% (arbitrage opportunity). November saw elevated but contained activity, totaling about $250–300B over the month (avg. ~$10–15B/day), up from near-zero pre-October but below 2019 peaks (about $100B+ daily). Key actions: NY Fed's Roberto Perli (Nov 12 'emergency' meeting) urged "large-scale" use to redistribute reserves, followed by John Williams' closed-door' dealer meeting (Nov 15) to reduce stigma. QT's Dec 1 end helped cap it, pulling usage down post-month-end.
When banks are short of reserves or facing funding pressure, they have three main ways to improve their LCR (liquidity coverage ratio) instantly:
1. Sell or repo Treasuries/reserves (Level 1)
2. Borrow from the FHLB (the advance itself is funded by issuing more Level 2A paper)
3. Use the Fed’s Standing Repo Facility (also against HQLA)
That is why FHLB advances explode in every stress episode. Borrowing from the FHLB simultaneously gives the bank cash and replaces the pledged mortgages (which were usually not HQLA) with an advance funded by new HQLA-eligible FHLB debt on someone else’s balance sheet.
In short: HQLA = the assets regulators trust banks to survive a crisis with.
FHLB debt is officially part of that trusted club — which is exactly why the market prices it only a few basis points away from Treasuries and why the FHLB system can act as such a gigantic liquidity backstop.
The actual canary in the coal mine is spikes in the aggregate FHLB advances which the Fed watches closely. FHLB advances have advanced gradually but in November there was no spike. This indicates the liquidity stress. although serious, is not yet a red flag state. There is great concern, but not panic yet.
2 Jan 2025 : 4.50%-4.60%
12 Nov 2025 :3.96%-4.10%. Fed funds +10–20 bps; floating, so upticks from EFFR firming add ~5–10 bps pressure. Outstanding FHLB was US$1.4 trillion rising only 7% in November.The actual canary in the coal mine is spikes in the aggregate FHLB advances which the Fed watches closely. FHLB advances have advanced gradually but in November there was no spike. This indicates the liquidity stress. although serious, is not yet a red flag state. There is great concern, but not panic yet.
2 Jan 2025 : 4.50%-4.60%
(9).SRF (Secured Repo Facility)
This is a Fed backstop for eligible institutions (primary dealers, banks, etc.) to borrow cash overnight against high-quality collateral via repo transactions. It's designed as a liquidity ceiling, capping repo rates to prevent spikes like in 2019. The SRF caps overnight rates by offering repo lending at the top of the target range (5.00%). Market participants won't borrow privately above this, as the Fed's facility is always available. The SRF offering rate is fixed by the FOMC and acts as the upper bound for money market rates. For example, in November, private repo rates were pushing beyond 4.00%. Arbitrage opportunities existed as banks can borrow from Fed's SFR against a private repo.
2 Jan 2024 : 4.50%.12 Nov 2025 : 4.00%.
(10). 1 month T-bills
2 Jan 2025 : 4.22%.12 Nov 2025 : 3.82%. ~8 bps below EFFR; yields firming slightly on auction demand but still down YTD.
(11) 3-month Treasury Bill 2 Jan 2025 : 4.25% 12 Nov 2025 : 3.85%. ~7 bps below EFFR; issuance surge absorbing liquidity, pushing secondary yields up 3–5 bps.
(12) 1-3 month Euro$Eurodollar-type rates become extremely important as they are usually the first and most violent movers upward in a dollar funding squeeze. They decouple sharply upward and become one of the clearest stress indicators in the entire USD rate complex, often moving faster and farther than repo rates or T-bills.
2 Jan 2025 : 1-month - 4.30%; 3-month - 4.4.32%.
12 Nov 2025 : 1-month - 4.32%; 3-month - 3.92%. 1-month .~5 bps below EFFR; unsecured, so more sensitive to crunch (spread widening). 3-month - ~3 bps below EFFR; forecasts show avg. 3.883% for Nov, but end-month uptick expected.
(13) 1-3 month Commercial Papers/ CDs
2 Jan 2025 : 4.35%
12 Nov 2025 :3.93%. ~5 bps over T-bills; unsecured risk premium up ~5 bps amid funding hoarding.
(14) Discount Window
This is Fed's "lender of last resort" facility for banks only. Primary dealers excluded.. It is meant for banks in distress so banks avoid this due to credit stigma.
2 Jan 2025 : 5.00%
12 Nov 2025 :4.50%. Fixed ceiling (+50 bps over upper target); SRF usage signals banks avoiding it due to stigma.
HOW THE FED CONTROLS LIQUIDITY:
In a liquidity crunch banks tend to hoard cash, moving away from higher risk assets. This tends to exacerbate the situation. Short term interest rates tend to rise.
Yet, in the above descriptions of various types of short term rates, a comparative of January 2025 and November 2025 shows interest rates have dropped! It's an anomaly. The reason is simple. This is due to the Fed's intervention. The US is an interest rate regime (Singapore is exchange rate regime) where the Fed uses interest rates as a monetary tool. The Fed sets interest rate ranges which guide short-term rates like the EFFR and SOFR, influencing borrowing costs economy-wide. Recent rate reductions were :
* 18 Sep 2024: From 5.25%-5.50% down to 4.75%-5.00%.
* 7 Nov 2024: down to 4.50%-4.75%
* 18 Dec 2024: down to 4.25%-4.5%
* 16 Sep 2025: down to 4.00%-4.25%
* 28 Oct 2025: down to 3.75%-4.00%.
The Federal Reserve operates under a statutory "dual mandate" from Congress: promote maximum employment and stable prices (targeting 2% inflation). For this, the Fed pays close attention to employment data and inflation trends.
The Fed's monetary tools to manage liquidity is primarily balance sheet actions like QE and QT (which ended December 1, 2025) or the Standing Repo Facility (SRF). Rate cuts are more about the broader economy. It is not explicitly a "liquidity injection" tool but used to calibrate the Fed's open market operations.
A GAUGE OF LIQUIDITY STRESS:
There are 2 key ways to see a liquidity stress developing -
a. The short term interest rates -- not the rates per se, but the spreads.
b. The usage of the SRF.
Spreads
While rate cuts lower the overall level, spreads like SOFR minus IORB reveal the intensity of liquidity frictions within the corridor. Normally, this spread is negative (about -5 to -10 bps), as SOFR trades below IORB (3.90% in November). A positive or widening spread signals banks are paying more to borrow in repos than they earn on reserves, indicating cash scarcity and hoarding.
The "upward pressure within the range" is the liquidity crunch's hallmark, raising fears of broader stress if it persists into year-end.
While rate cuts lower the overall level, spreads like SOFR minus IORB reveal the intensity of liquidity frictions within the corridor. Normally, this spread is negative (about -5 to -10 bps), as SOFR trades below IORB (3.90% in November). A positive or widening spread signals banks are paying more to borrow in repos than they earn on reserves, indicating cash scarcity and hoarding.
The "upward pressure within the range" is the liquidity crunch's hallmark, raising fears of broader stress if it persists into year-end.
In November 2025, the spread was moderately serious but not catastrophic. It was elevated and persistent, hitting +32 bps at its October 31 peak (widest since 2020 COVID turmoil), then averaging about +10–15 bps mid-month and about +20–25 bps at month-end (e.g., +22 bps on Nov 28 when SOFR hit 4.12%). This compression (SOFR hugging the ceiling) is serious but nothing compared to the 2019's +390 bps blowout. Volatility rose (SOFR swung 10–20 bps daily), but no systemic panic. Markets absorbed it via arbitrage, with the Fed viewing reserves (about US$2.85T) as still "somewhat ample."
SRF
The Standing Repo Facility (SRF, min bid 4.00%) acted as the "ceiling enforcer," with usage quantifying stress—low/zero on calm days, spiking when SOFR exceeded 4.00% (arbitrage opportunity). November saw elevated but contained activity, totaling about $250–300B over the month (avg. ~$10–15B/day), up from near-zero pre-October but below 2019 peaks (about $100B+ daily). Key actions: NY Fed's Roberto Perli (Nov 12 'emergency' meeting) urged "large-scale" use to redistribute reserves, followed by John Williams' closed-door' dealer meeting (Nov 15) to reduce stigma. QT's Dec 1 end helped cap it, pulling usage down post-month-end.
Overall, spreads and SRF signaled "creeping tightness" in November (not crisis), with the Fed's tweaks preventing escalation. Reserves held, no discount window stigma. If spreads stay > +15 bps into December, watch for policy signals.
THE CANARIES IN THE MINE:
THE CANARIES IN THE MINE:
If you want to know whether a liquidity squeeze is mild/technical or heading toward systemic, watch these two birds first:
a. Leveraged players - are they getting forcibly unwound across multiple markets at the same time?
b. Primary dealers’ - are their treasury inventories starting to balloon while liquidity metrics deteriorate? I liken this situation to primary dealers caught holding the grenade.
Forced deleveraging in risk assets (equities, crypto, credit, carry trades, basis trades)
Margin desks and prime-brokerage units are raising haircuts, issuing calls, or liquidating positions because collateral values or funding costs are moving against levered players. This is almost always the first place scarce dollars show up — long before CPI prints or payroll numbers move.Aug 2022 (crypto/ARKK), March 2023 (regional banks + credit), Aug–Nov 2025 (Nasdaq-100 + BTC wipeouts) all happened weeks before official recession or banking-stress signals appeared.
In November - $19 bn crypto liquidations + $150 bn equity ETF outflows + yen-carry unwind in first three weeks. This confirms wholesale funding was getting expensive and a market risk-off wave.
In November - $19 bn crypto liquidations + $150 bn equity ETF outflows + yen-carry unwind in first three weeks. This confirms wholesale funding was getting expensive and a market risk-off wave.
Primary Dealers get caught “holding the grenade” in a liquidity crunch.
In liquidity crunch the 25 PDs are stuck with inventories as end-buyers hold back. Meanwhile, they are still obliged to pick up new issuance by treasuries. Their balance sheet explodes and regulatory requirements force them to hold higher reserves. They are hit with 3 problems.:
- they need massive funding for the high inventories.
- they need to capitalise to build higher reserves to comply with regulatory requirements..
- treasury yields drop and sales means realised losses, thus better to hold than sell.
As ultimate buyers of last resort, PDs are being forced to warehouse huge amounts of “risk-free” assets they cannot finance or hedge under current regulation, intermediation is slowed as buyers 'wait out' the volatility. When dealer inventories start rising fast while off-the-run liquidity evaporates, the Treasuries market itself flashing red --- the deepest, most important funding market on earth.
March 2020 is the textbook: dealer positions went from $150 bn → $650 bn in ten trading days while on-the-run/off-the-run spreads (difference between newer vs older treasury issues) blew out to 100 bps — before the Fed even announced unlimited QE. The same early inventory spike appeared in Sep–Oct 2022 (UK LDI crisis spillover) and again in miniature in Oct–Nov 2025.
In November : Net dealer positions rose only ~5 % and bid-ask spreads widened just 2–3 bps → grenade stayed in hand, but safety pin still on. Confirmed stress was real but not yet systemic — Fed could lean on SRF and QT pause instead of panic QE
Summary:
These two canaries sing ahead of other indicators of liquidity problems -
* VIX, MOVE, credit spreads : These can stay calm until the last minute. Deleveraging shows up instantly in liquidation data.
* Bank CDS, regional-bank stocks : These only flare when banks are already in trouble. Dealer Treasury inventory rises before banks feel it.
* Recession indicators (yield-curve, PMI): These take months of lag before they show up. On the other hand, both deleveraging and primary dealer canaries scream within days or weeks.
* Official reserve numbers or TGA: These are revealed in weekly data, revised later. Whereas dealer positions (deleveraging) and crypto/equity liquidations are real-time
When both deleveraging and primary dealer canaries start singing loudly at once, it is usually only days or weeks away from either (a) a Fed capitulation or (b) something actually breaking. In November 2025 they chirped noticeably, but never reached full-throated alarm --- which is why this November episode stayed “creaky plumbing” rather than like the “2020 redux”.
These two canaries sing ahead of other indicators of liquidity problems -
* VIX, MOVE, credit spreads : These can stay calm until the last minute. Deleveraging shows up instantly in liquidation data.
* Bank CDS, regional-bank stocks : These only flare when banks are already in trouble. Dealer Treasury inventory rises before banks feel it.
* Recession indicators (yield-curve, PMI): These take months of lag before they show up. On the other hand, both deleveraging and primary dealer canaries scream within days or weeks.
* Official reserve numbers or TGA: These are revealed in weekly data, revised later. Whereas dealer positions (deleveraging) and crypto/equity liquidations are real-time
When both deleveraging and primary dealer canaries start singing loudly at once, it is usually only days or weeks away from either (a) a Fed capitulation or (b) something actually breaking. In November 2025 they chirped noticeably, but never reached full-throated alarm --- which is why this November episode stayed “creaky plumbing” rather than like the “2020 redux”.
PRIMARY DEALERS END UP HOLDING THE GRENADE
Primary dealers often hoard securities during liquidity crunches to protect their balance sheet, such as during March 2020 or November 2025. This isn't deliberate market sabotage but a rational response to risks and regulations. Key reasons include:
* Counterparty and Solvency Fears: In stress, dealers worry about other firms defaulting on trades or repos, leading them to hoard liquid assets (e.g., Treasuries) as buffers against potential losses. For instance, during the 2007–2008 crisis, banks hoarded liquidity anticipating securities write-downs, increasing holdings by ~30% in precautionary mode.
* Portfolio Risk and Future Illiquidity: Rising asset volatility (e.g., yield spikes) prompts hoarding to avoid forced sales at losses. Fears of "future illiquidity" create a self-reinforcing cycle - one dealer hoards, restricting cash for others, prompting more hoarding.
* Regulatory Constraints -- Rules like the Supplementary Leverage Ratio (SLR) penalize balance-sheet expansion, forcing dealers to hoard rather than intermediate when risks rise. In 2020, this led to ~$650B in unwanted Treasury holdings as dealers absorbed supply but couldn't offload amid volatility.
How Primary Dealers Reduce Securities Liquidity:
Hoarding stabilizes individual dealers but exacerbates system-wide stress, as seen in November 2025's modest crunch (reserves ~$2.85T), where it contributed to SOFR spikes without full-blown crisis.
PDs are market makers, typically providing two-way quotes to buy/sell securities, ensuring smooth trading. In stress, they reduce liquidity through:
* Inventory Reduction and Position Cuts -- Tighter capital rules (e.g., SLR, G-SIB surcharges) force dealers to shrink holdings, rejecting client flows and widening bid-ask spreads (e.g., +70–100 bps in 2020 Treasuries). This thins turnover—e.g., aggregate Treasury trading volume dropped ~20% in March 2020.
* Withdrawal from Intermediation -- Dealers stop financing client positions via repo, leading to failures-to-deliver (~$500B in 2020) and cascading illiquidity. In corporate bonds, inventories fell sharply, halting transactions.
* Hoarding and Risk Limits -- Internal Volcker Rule caps trigger automatic pullbacks, reducing market depth and amplifying price swings.
Why hoarding securities is bad
As primary dealers hoard, treasuries liquidity (the treasuries circulating in the market) shrinks.
Reduced treasuries liquidity impairs price discovery (e.g., distorted yields), raises borrowing costs (e.g., repo rates +300 bps over SOFR in 2020), and risks contagion --- thin markets lead to fire sales, deleveraging cascades, and broader economic harm (e.g., higher corporate funding costs slow growth). In extreme cases like 2020, it threatened the Treasuries market's role as a global safe haven, forcing Fed bailouts.
Reduced treasuries liquidity impairs price discovery (e.g., distorted yields), raises borrowing costs (e.g., repo rates +300 bps over SOFR in 2020), and risks contagion --- thin markets lead to fire sales, deleveraging cascades, and broader economic harm (e.g., higher corporate funding costs slow growth). In extreme cases like 2020, it threatened the Treasuries market's role as a global safe haven, forcing Fed bailouts.
Why Primary Dealers Are Crucial in Financial Markets and the Broader Economy
PDs are pivotal intermediaries, acting as the Fed's operational arm and the Treasury market's backbone. Their roles include:
* Monetary Policy Execution -- Direct Fed counterparties for open market operations (e.g., QE, repo facilities), implementing rate changes and liquidity injections. Ensures policy transmission to banks/economy; without them, rate signals fail, disrupting lending/growth.
* Treasury Auction Participation -- Obligated to bid pro-rata (~4% each) in auctions, absorbing ~$29T+ U.S. debt issuance. Stabilizes government funding; failures raise yields, increasing deficits and crowding out private investment.
* Market Making and Securities Liquidity Provision -- Provide quotes, finance inventories via repo (~$1.2T net positions in 2025), and resell to investors. | Maintains Treasury liquidity (world's safest asset), supporting global finance; breakdowns (e.g., 2020) spike volatility, harming pensions, trade.
* Price Discovery and Risk Management -- Use expertise to attract demand, hedge risks, and clear through CCPs (e.g., significant clearing members). Enables efficient capital allocation; without it, mispricing ripples to mortgages, loans, and GDP.
In the financial ecosystem, PDs are indeed a "critical cog" -- perhaps the most important for the $26T Treasury market, which underpins global rates, dollar dominance, and economic stability. They're not the only one (e.g., hedge funds now provide ~40% liquidity), but their failure (as in 2020) can halt everything, making them indispensable for policy, issuance, and resilience.
* Treasury Auction Participation -- Obligated to bid pro-rata (~4% each) in auctions, absorbing ~$29T+ U.S. debt issuance. Stabilizes government funding; failures raise yields, increasing deficits and crowding out private investment.
* Market Making and Securities Liquidity Provision -- Provide quotes, finance inventories via repo (~$1.2T net positions in 2025), and resell to investors. | Maintains Treasury liquidity (world's safest asset), supporting global finance; breakdowns (e.g., 2020) spike volatility, harming pensions, trade.
* Price Discovery and Risk Management -- Use expertise to attract demand, hedge risks, and clear through CCPs (e.g., significant clearing members). Enables efficient capital allocation; without it, mispricing ripples to mortgages, loans, and GDP.
In the financial ecosystem, PDs are indeed a "critical cog" -- perhaps the most important for the $26T Treasury market, which underpins global rates, dollar dominance, and economic stability. They're not the only one (e.g., hedge funds now provide ~40% liquidity), but their failure (as in 2020) can halt everything, making them indispensable for policy, issuance, and resilience.
REAL END INVESTORS HOLD BACK ON TREASURIES
Real end buyers of treasuries such as central banks, wait-out the volatility caused by the liquidity problem. They are not abandoning the US Treasuries due to geo-politics and dedollarisation, but merely riding out the volatility.
Over the past decades (from ~71% in 2000 to ~58% in 2025), USD share as reserves currency has declined. However, the decline has been gradual and more a factor of changing trade patterns rather than perceptions of geo-political risks and dedollarisation, which in my opinion, are over-hyped. The absolute value of USD reserves has still grown significantly (from ~$1.4T in 2000 to ~$7.5T in 2025), as total global reserves expanded from ~$2T to ~$13T, underscoring the USD's enduring dominance despite relative losses.
China's total holdings of US treasuries stood at approximately $759 billion, down from a peak of ~$1.3 trillion in 2013, representing a cumulative reduction of about $541 billion over more than a decade. Many indicated this as China's flight from US treasuries to gold. What's seldom commented on is China's gradual reduction of US treasuries is mirrored in increases of holdings by Luxembourg and Cayman Islands, two locations where the People's Bank of China has offshore custodian entities. It is very likely China moved inventories to the two offshore custodians for operational efficiencies as well as to reduce vulnerability through opacity.
CONCLUSION
The Fed's normalisation efforts by QT to shrink its balance sheet post-Covid pandemic was targeted to be achieved by 3 December 2025 when the tightening programme ends. Rate cuts were used to tweak the plumbing in response to employment figures and inflation. The fiscal drag in 2025 from massive treasury issuance, exacerbated by ongoing QT, September tax payments, and tariffs drained reserves. However, right up to October, the financial health seems to be safely within the "Ample Regime". The liquidity stress hit in November as a result of the government shut-down forced upon the Trump administration by the Democrats.
The liquidity stress is mechanical in nature, not structural. It is concerning, but not yet a red flag. The Fed feels the end of QT by December should provide relief. If this is not good enough, watch out for a fresh round of QE.
While fiscal drag of treasury issuance, QT, tax, tariff, government shut-down caused the liquidity stress, the primary dealers are the chokepoint. The primary dealers are left holding the grenade, but in November, the safety pin was still on, unlike 2020 where the safety pin had been pulled and almost detonated.
Further down the road, it is very likely the Fed will have to address the excessive burden of regulatory compliance placed on primary dealers, more specifically (1) the Supplementary Leverage Ratio (SLR) which forces dealers to hold extra Tier-1 capital against Treasuries and reserves which makes holding U.S. government debt “expensive” on the balance sheet; (2) the G-SIB surcharge & resolution planning which cause big dealers to face higher capital charges the larger their balance sheet became; and (3) the Liquidity Coverage Ratio (LCR) which ironically encouraged dealers to hoard reserves instead of lending them into repo.
Real end buyers of treasuries such as central banks, wait-out the volatility caused by the liquidity problem. They are not abandoning the US Treasuries due to geo-politics and dedollarisation, but merely riding out the volatility.
Over the past decades (from ~71% in 2000 to ~58% in 2025), USD share as reserves currency has declined. However, the decline has been gradual and more a factor of changing trade patterns rather than perceptions of geo-political risks and dedollarisation, which in my opinion, are over-hyped. The absolute value of USD reserves has still grown significantly (from ~$1.4T in 2000 to ~$7.5T in 2025), as total global reserves expanded from ~$2T to ~$13T, underscoring the USD's enduring dominance despite relative losses.
China's total holdings of US treasuries stood at approximately $759 billion, down from a peak of ~$1.3 trillion in 2013, representing a cumulative reduction of about $541 billion over more than a decade. Many indicated this as China's flight from US treasuries to gold. What's seldom commented on is China's gradual reduction of US treasuries is mirrored in increases of holdings by Luxembourg and Cayman Islands, two locations where the People's Bank of China has offshore custodian entities. It is very likely China moved inventories to the two offshore custodians for operational efficiencies as well as to reduce vulnerability through opacity.
CONCLUSION
The Fed's normalisation efforts by QT to shrink its balance sheet post-Covid pandemic was targeted to be achieved by 3 December 2025 when the tightening programme ends. Rate cuts were used to tweak the plumbing in response to employment figures and inflation. The fiscal drag in 2025 from massive treasury issuance, exacerbated by ongoing QT, September tax payments, and tariffs drained reserves. However, right up to October, the financial health seems to be safely within the "Ample Regime". The liquidity stress hit in November as a result of the government shut-down forced upon the Trump administration by the Democrats.
The liquidity stress is mechanical in nature, not structural. It is concerning, but not yet a red flag. The Fed feels the end of QT by December should provide relief. If this is not good enough, watch out for a fresh round of QE.
While fiscal drag of treasury issuance, QT, tax, tariff, government shut-down caused the liquidity stress, the primary dealers are the chokepoint. The primary dealers are left holding the grenade, but in November, the safety pin was still on, unlike 2020 where the safety pin had been pulled and almost detonated.
Further down the road, it is very likely the Fed will have to address the excessive burden of regulatory compliance placed on primary dealers, more specifically (1) the Supplementary Leverage Ratio (SLR) which forces dealers to hold extra Tier-1 capital against Treasuries and reserves which makes holding U.S. government debt “expensive” on the balance sheet; (2) the G-SIB surcharge & resolution planning which cause big dealers to face higher capital charges the larger their balance sheet became; and (3) the Liquidity Coverage Ratio (LCR) which ironically encouraged dealers to hoard reserves instead of lending them into repo.
This is not financial advice, but I'll wager that as deleverage hit cryptos hard during a liquidity crunch, once the plumbing is done and the stress is over, cryptos will bounce back and recover substantially more than it had lost, as it traditionally has.
This platform has withdrawn it's subscriber widget. If you like blogs like this and wish to know whenever there is a new post, click the button to my FB and follow me there. I usually intro my new blogs there. Thanks.