December 1, 2025 will be remembered as the day the Fed admitted it ran out of road. Not because they wanted to stop. Because they had to.

Let's perform the autopsy.

After draining $2.39 trillion from the financial system over three years, the Federal Reserve officially ended quantitative tightening on December 1, 2025. The balance sheet now sits at $6.57 trillion—frozen in place after the largest liquidity withdrawal in central banking history.

The headlines called it "prudent monetary policy management." The Fed's press release called it "consistent with the Committee's plans."

The plumbing tells a different story.

Bank reserves scraped $2.89 trillion—uncomfortably close to the $2.5-2.7 trillion floor where the repo market nearly broke in September 2019. The overnight reverse repo facility that once held $2.5 trillion? Near zero. Money market stress indicators flashing amber across the board. The Fed's Standing Repo Facility saw its largest draw since inception—$50 billion in a single day in late October.

This wasn't a choice. This was the Fed narrowly avoiding a repeat of September 2019 while pretending it was all part of the plan.

What happens next? Here's our probability-weighted outlook for Q1 2026:

•       60% probability: Temporary tightness from TGA rebuild (~$200B drain) and tax season. Annoying but manageable.

•       25% probability: Smooth sailing as organic balance sheet growth offsets seasonal pressures. The optimist's case.

•       15% probability: Renewed stress requiring Fed intervention. The "here we go again" scenario.

Key takeaways:

•       QT drained $2.39 trillion over 41 months—the largest balance sheet reduction in Fed history.

•       Bank reserves at $2.89 trillion sit just 7-15% above the estimated stress threshold ($2.5-2.7T).

•       The RRP facility—once the Fed's $2.5 trillion shock absorber—is functionally depleted.

•       SOFR-IORB spreads widened to 30 basis points in October, the widest since the facility's inception.

•       This validates our liquidity regime shift framework: the transition from "abundant" to "ample" reserves is now official.

•       What comes next matters more than what just ended.

Let's put the scale of this operation in perspective. The Fed's pandemic response between March 2020 and April 2022 created $4.81 trillion in new money. QT clawed back roughly half of it.

Balance Sheet Contraction:

•       Peak balance sheet (April 2022): $8.96 trillion

•       Final balance sheet (December 1, 2025): $6.57 trillion

•       Total reduction: $2.39 trillion (26.7%)

•       Duration: 41 months (June 2022 – December 2025)

Treasury Securities:

•       Reduction: $1.73 trillion (from $5.77T to $4.04T)

•       Monthly runoff cap (final phase): $5 billion

Mortgage-Backed Securities:

•       Reduction: $670 billion (from $2.74T to $2.07T)

•       Monthly runoff cap: $35 billion (continues post-QT)

•       Note: MBS runoff will continue until holdings reach zero, replaced by T-bills

The Fed shed 44% of the assets accumulated during the pandemic response. That's not nothing. But the circumstances of the ending tell you everything about where we are now.

The official narrative is that QT ended "as planned." The data tells a different story.

The Standing Repo Facility (SRF)—the Fed's backstop that allows banks to borrow cash overnight by pledging Treasury securities as collateral—was created in 2021 specifically to prevent a repeat of September 2019. It's the lender-of-last-resort for overnight funding.

For most of its existence, usage was essentially zero. Banks didn't need it because reserves were abundant.

Then came the fall of 2025:

•       Mid-September 2025: $18.5 billion drawn in a single day—the largest since inception

•       October 31, 2025: $50.35 billion borrowed—another record

•       First three days of November: $44 billion in cumulative SRF loans

When banks start tapping the Fed's emergency window instead of borrowing from each other, that's not normal market function. That's stress. The facility designed to prevent the next crisis was actively preventing the next crisis. Think about that.

The spread between SOFR (Secured Overnight Financing Rate—the benchmark rate for overnight borrowing backed by Treasury securities) and IORB (Interest on Reserve Balances—the rate the Fed pays banks for parking cash at the Fed) is the cleanest measure of funding market stress.

Normally, SOFR trades at or below IORB. Banks can earn IORB risk-free just by parking cash at the Fed, so there's no reason to accept less in the repo market.

When SOFR trades significantly above IORB, it means banks are willing to pay a premium to borrow cash rather than lend their own reserves. That's a liquidity shortage in action.

The progression:

•       Q1 2025: SOFR-IORB spread averaging 5-8 basis points (normal)

•       Mid-September 2025: Spread jumped to 15+ bps

•       Late October 2025: Spread hit 30 basis points—the widest in years

•       Warning threshold per our framework: >15 bps indicates excessive tightness

At 30 basis points, banks were paying significantly more to borrow cash than they'd earn leaving it at the Fed. That's like paying for storage at a warehouse that's offering you free rent. The system was screaming that reserves were too scarce. The Fed finally listened.

The overnight reverse repo facility (ON RRP)—where money market funds park excess cash at the Fed in exchange for Treasury securities overnight—was the Fed's secret weapon during QT. When reserves got tight, money could flow back from the RRP into the banking system. It provided cushion.

That cushion is gone. Poof.

•       RRP peak (December 2022): $2.55 trillion

•       RRP current (early December 2025): Near zero

•       Where did it go: Treasury bills (higher yields attracted money market funds)

Here's why this matters: During Phase 1 of QT, the balance sheet shrank but reserves stayed stable because the RRP did the heavy lifting. Money flowed out of the RRP—not out of bank reserves. Sweet deal while it lasted. But once the RRP hit zero, every dollar of additional tightening came directly out of reserves. No more buffer. No more shock absorber. Every liquidity hiccup now lands directly on the banking system. That's the difference between a car with airbags and one without.

Bank reserves as of November 2025: approximately $2.89 trillion.

Estimated stress threshold: $2.5-2.7 trillion.

That's a cushion of just $190-390 billion—or 7-15% above the floor where things start breaking.

For context: In September 2019, when repo rates spiked to 10% overnight and the Fed had to inject $75 billion in emergency liquidity, reserves were around $1.4-1.5 trillion. But the financial system was smaller then. Scaled to today's banking sector ($25 trillion in assets vs. $17 trillion in 2019), the comparable "minimum comfortable level" is significantly higher.

Fed Governor Waller has said reserves below 8% of GDP—roughly $2.7 trillion currently—could be "problematic." At $2.89 trillion, we're uncomfortably close.

How close to the edge are we? Depends who you ask:

•       Conservative floor ($2.7T, Waller's 8% GDP): Current cushion of $190B. That's 7%. Not much.

•       Moderate floor ($2.5T, 2019 scaled): Current cushion of $390B. That's 15%. Better, but...

•       Aggressive floor ($2.3T, minimum viable): Current cushion of $590B. That's 26%. Still not comfortable.

•       The math that matters: A typical $200B TGA rebuild would consume 34-100% of the buffer depending on your floor estimate. Yikes.

The Fed didn't stop QT because they reached their target. They stopped because continuing risked recreating September 2019.

The Fed officially operates in an "ample reserves" framework. But what does that actually mean?

Abundant reserves (2020-2023): So much liquidity that changes in reserve levels have essentially zero impact on short-term rates. The RRP absorbed the overflow. Banks didn't think about reserves—they had plenty.

Ample reserves (where we are now): Enough reserves for monetary policy to function smoothly, but changes in supply now have modest effects on rates. The Fed has to actively monitor conditions. Rate control requires attention.

Scarce reserves (pre-2008, and September 2019): Not enough reserves. Small changes in supply cause large, volatile rate movements. The Fed has to actively manage the market daily to maintain rate control. Things break.

The transition from abundant to ample is what we've been calling the "liquidity regime shift" in our previous pieces. The Fed just made it official.

In the abundant reserves era (2020-2023):

•       Liquidity was essentially free

•       Exit liquidity was always available

•       Even bad credits could refinance at low rates

•       Volatility stayed suppressed

•       Quarter-ends were non-events

In the ample reserves era (2025+):

•       Liquidity commands a premium

•       Quarter-end repo spikes become normal features

•       Funding costs matter again

•       Credit quality differentiation returns

•       Structural volatility is higher (VIX baseline 18-22 vs. low teens)

•       The Fed has less room to maneuver without risking disruption

Why 18-22 VIX? This range mirrors two previous ample reserves periods: 2018-2019 (when the Fed's first QT attempt hit the wall) and 2006-2007 (the last time monetary policy was genuinely tight). Both times, structural volatility ran 30-50% higher than during easy money periods. It's not a bug—it's the math of operating with thinner buffers. Get used to it.

This isn't a temporary condition. This is the new normal.

Forecasting is hard. Let's see how the Street did:

Wall Street's QT End Date Predictions:

•       Bank of America: December 2025 ✓ (nailed it)

•       Goldman Sachs: Q1 2025 ✗ (9 months early—too scared of 2019 repeat)

•       JPMorgan: Q1 2026 ✗ (1 quarter late—overestimated RRP buffer)

•       Deutsche Bank: Q1 2026 ✗ (same mistake as JPM)

•       Barclays: June 2026 ✗ (6 months late—way too optimistic on reserves)

•       TMS Capital Research: December 2025 ✓ (we'll take it)

The spread—from Q1 2025 to June 2026—shows how uncertain even the smartest analysts were. Goldman was spooked by 2019 memories; JPMorgan and Deutsche thought the RRP buffer would last longer. BofA got it right for the same reason we did: watching the RRP depletion timeline as the binding constraint. Sometimes the boring answer is the right answer.

In "The Liquidity Trap" (November 15), we identified the SOFR-IORB spread widening and RRP depletion as early warning signals. In "The Liquidity Regime Shift" (November 18), we outlined the framework for understanding the transition from abundant to ample reserves.

What happened:

•       QT ended exactly on our predicted timeline (December 1) ✓

•       Reserve levels hit the threshold we identified ($2.85-2.90T) ✓

•       Stress signals (SOFR-IORB, SRF usage) flashed as predicted ✓

•       RRP depleted to near-zero as anticipated ✓

The framework isn't magic. It's watching the plumbing while everyone else watches headlines.

Most macro analysis focuses on CPI prints, payroll numbers, and Fed dot plots. Important data, sure. But secondary to liquidity flows.

The Fed ended QT because the plumbing forced them to—not because inflation hit their target or the labor market softened. The data was inconclusive. The liquidity constraints were not.

The end of QT isn't the end of the story. If anything, it's the beginning of a more complex chapter.

Look, it's not all doom and gloom. Several factors could provide relief:

Rate cuts are helping. The Fed already cut twice (September and October 2025), putting the target at 3.75-4.00%. Markets price ~88% odds of another cut December 9-10. Lower rates reduce funding costs even as reserve scarcity persists. Mixed signals? Sure. But better than tighter quantity AND higher price.

The balance sheet is getting cleaner. The $35 billion monthly MBS runoff means the Fed is gradually exiting the mortgage market and rotating into T-bills. Less market distortion, more liquidity. Progress.

Crypto and risk assets get a breather. Our "Why Crypto Crashes First" analysis showed speculative assets sit at Tier 4 of the liquidity hierarchy—first to fall during tightening. With QT over, that structural headwind fades. Analyst Matthew Hyland notes non-QT periods historically precede 29-42 month altcoin rallies. Not a moon call—just less gravity.

The bottom line: Think of it as driving with both brake (tight reserves) and accelerator (lower rates) partially engaged. Manageable, but requires more attention than the 2020-2023 autopilot era.

Treasury runoff: Stopped. No more Treasury securities rolling off the balance sheet.

MBS runoff: Continues at $35 billion monthly. The Fed still wants out of the mortgage market. As mortgages prepay, the proceeds get reinvested into Treasury bills, not new MBS.

Balance sheet growth: Organic expansion begins. The Fed will start purchasing T-bills to maintain reserve levels as the economy grows. Estimated pace: roughly $20 billion monthly to match GDP expansion.

Net effect: The balance sheet should grow modestly in 2026 rather than shrink. Not QE—just keeping pace with the economy.

December 31 Quarter-End: The most severe quarter-end of the year coincides with the SEC Treasury clearing mandate taking effect. Regulatory window-dressing plus market structure change plus depleted RRP = maximum stress convergence. Watch SOFR closely.

Treasury General Account (TGA) Swings: With QT stopped, TGA volatility from tax dates and Treasury operations will be the primary driver of reserve fluctuations. Tax season in Q1 2026 could create temporary tightness.

Japan Repatriation Risk: Japanese Government Bond yields hit 1.82%—highest since 2013. If Japanese institutions start repatriating capital from U.S. Treasuries, it adds another drain on dollar liquidity. (See our detailed analysis: "The $1.1 Trillion Japan Question.")

Debt Ceiling (November 2026 estimated): When Congress hits the debt ceiling, Treasury draws down the TGA—injecting reserves. When the ceiling is lifted, Treasury rebuilds the TGA—draining reserves rapidly. These swings will be more impactful without QT as a predictable background drain.

Here are the metrics to track now that we've entered the ample reserves regime:

Bank Reserves (Fed H.4.1, Thursdays):

•       Current level: ~$2.89 trillion.

•       Warning threshold: Below $2.7 trillion.

•       Crisis threshold: Below $2.5 trillion.

RRP Facility (NY Fed, daily):

•       Current level: Near zero.

•       Implication: No buffer remaining—all stress hits reserves directly.

SOFR-IORB Spread (NY Fed, daily):

•       Normal range: 5-10 basis points.

•       Warning threshold: >15 bps.

•       Stress signal: >20 bps sustained.

Standing Repo Facility Usage (NY Fed, daily):

•       Normal: Near zero.

•       Elevated: >$10 billion.

•       Crisis signal: >$25 billion sustained.

VIX (CBOE, real-time):

•       New baseline (ample reserves era): 18-22.

•       Stress signal: >25 sustained.

•       Cascade risk: >30 sustained.

Fed Balance Sheet (H.4.1, Thursdays):

•       Current: $6.57 trillion.

•       Expected trajectory: Modest growth (~$20B/month) to maintain ample reserves.

Volatility is structural, not temporary. The VIX baseline of 18-22 isn't a spike to fade—it's the new normal. Volatility strategies that worked in the low-teens VIX environment need recalibration.

Liquidity premium matters again. Deep, liquid markets will outperform. Illiquid positions may be harder to exit during stress periods. Position sizing should account for reduced exit liquidity.

Quarter-ends are events now. Regulatory window-dressing combined with thin reserve buffers means predictable funding stress at March 31, June 30, September 30, and December 31. Trade around them or reduce risk into them.

Cash is competitive. 3.9% Treasury bill yields with positive real returns make cash a legitimate allocation, not a drag. Opportunity cost of holding dry powder is much lower than 2020-2023.

Quality still wins. Fortress balance sheets, consistent cash flows, and pricing power outperform in any liquidity regime. The ample reserves era just makes the quality premium more pronounced.

The Fed has tools. The Standing Repo Facility, potential targeted interventions, and the ultimate backstop of QE restart mean true systemic crises remain unlikely. But the bar for intervention is higher now.

Liquidity analysis beats narrative. The headlines said QT was proceeding "as planned." The plumbing said the Fed was running out of road. The plumbing was right. Keep watching the plumbing.

Let's be absolutely clear about what December 1, 2025 represents:

The era of free liquidity is officially over.

From June 2022 to December 2025, the Fed drained $2.39 trillion from the financial system. They stopped not because they wanted to, but because continuing risked breaking the repo market. The stress signals were unmistakable: SRF usage at record highs, SOFR-IORB spreads blown out, RRP cushion exhausted, reserves scraping the floor.

Now we enter a new phase. Not QE—the Fed isn't printing money again. Not QT—the aggressive drainage has stopped. Something in between: a steady state where the Fed actively manages reserve levels to keep them "ample" without letting them become "abundant" again.

This matters because it changes the rules of the game:

•       Volatility stays elevated

•       Liquidity commands a premium

•       Quarter-ends matter

•       Cash is competitive

•       Quality beats speculation

The 2020-2023 playbook is dead. Anyone still trading like liquidity is free will learn that lesson the hard way.

We called the regime shift. We identified when QT would end and why. The framework continues.

No doom. No hype. Just the plumbing.

This Week: "December FOMC Preview: What the 90% Rate Cut Odds Are Missing"

Next Week: "December Quarter-End Dynamics: Positioning for the Repo Spike"

Subscribe for liquidity-driven analysis that explains market moves before the headlines catch up.

•       Federal Reserve H.4.1 Weekly Balance Sheet (November 28, 2025)

•       FOMC Statement and Balance Sheet Developments (October 29, 2025)

•       New York Fed Standing Repo Facility Operations Data

•       NY Fed SOFR and Overnight RRP Data

•       Cleveland Fed Economic Commentary: "QT, Ample Reserves, and the Changing Fed Balance Sheet"

•       Federal Reserve Board FEDS Notes: "Market-Based Indicators on the Road to Ample Reserves"

•       Brookings Institution: "How will the Federal Reserve decide when to end quantitative tightening?"

•       Wolf Richter balance sheet analysis (Wolf Street)

•       CME FedWatch Tool

Disclaimer: This is analysis and commentary, not investment advice. TMS Capital Research shares our perspective on liquidity markets and positioning considerations. You bear full responsibility for your investment decisions. Markets can stay irrational longer than you can stay solvent. Size appropriately, use stops, and don't bet the farm on any single trade. We're sharing what we see—you're responsible for what you do with it.

•       The Liquidity Regime Shift: Where Smart Money Flows When Free Money Ends

•       The Liquidity Trap: Understanding Fed Reserve Requirements and RRP Depletion

•       Why Crypto Crashes First: Liquidity Analysis

•       Japan Treasury Holdings 2025: The $1.1 Trillion Repatriation Question

Twitter: @TMSCapResearch