Macro Mechanics #3: Bank Reserves & The “Ample” Floor
Pierce & Pierce Research | January 2026
Part of the Macro Mechanics Series
⏱️ 12-15 minutes • Beginner-friendly • No prior Fed knowledge required
Connection to Previous Briefs: In Brief #1, we covered the TGA—where government cash sits. In Brief #2, the ON RRP buffer—now exhausted. Bank reserves are where those flows ultimately land. This is the liquidity that actually matters for market plumbing.
Why This Matters
In September 2019, overnight lending rates spiked to 10%. The repo market seized up. Banks that needed to borrow cash for a single day suddenly could not find it at any reasonable price. The Federal Reserve had to intervene with emergency injections of $75 billion per day.
The cause? Bank reserves had fallen too low.
In October 2025, something similar started happening. Usage of the Fed’s Standing Repo Facility spiked to $50 billion. Money market rates drifted higher within the Fed’s target range. The Fed announced the end of quantitative tightening and prepared to restart balance sheet expansion.
Current Reserve Status:
| Metric | Value | Status |
|---|---|---|
| Reserve Balance | $2.88 trillion | 🔴 Below floor |
| Estimated Floor | $2.9-3.0 trillion | - |
| Position vs Floor | ~$20-120B below | 🔴 Critical |
Reserves are now below the estimated floor range. This is why the Fed pivoted to active T-bill purchases in December 2025. The plumbing is under stress.
Section 1: What Are Bank Reserves?
Bank reserves are the cash that commercial banks hold in their accounts at the Federal Reserve. Think of it as the bank’s own checking account at the central bank.
Banks use reserves to settle transactions with each other. When you transfer money from your Chase account to a friend’s Wells Fargo account, the underlying settlement involves Chase’s reserve account sending funds to Wells Fargo’s reserve account at the Fed.
Pre-2008 vs Post-2008:
| Era | Reserve Level | How Fed Controlled Rates |
|---|---|---|
| Pre-2008 | <$50 billion | Adjusted supply daily |
| Post-2008 (QE) | $2-4 trillion | Pays interest on reserves |
Before 2008, the Fed operated with a small amount of reserves. Banks borrowed and lent reserves constantly in the federal funds market. The Fed controlled rates by adjusting supply.
Then came the financial crisis. The Fed flooded the system with reserves through quantitative easing. Reserve balances exploded from under $50 billion to over $2.8 trillion by 2014.
The key takeaway: Reserves went from scarce (banks needed every dollar) to abundant (banks had more than they could use). The Fed now pays interest on reserves to keep rates where it wants them.
Section 2: The Reserve Framework
The Fed currently operates under an “ample reserves” framework. This sounds simple but contains crucial nuance.
Three Reserve Regimes:
| Regime | Reserve Level | Rate Behaviour | Status |
|---|---|---|---|
| 🟢 Abundant | >$3.2T (well above floor) | EFFR at bottom of range | Excess liquidity |
| 🟡 Ample | $2.9-3.2T (above floor with buffer) | EFFR within range | Target zone |
| 🔴 Scarce | <$2.9T (at or below floor) | Rates drift higher, stress signs | Current |
The Problem:
There is no precise number that defines “ample.” The threshold depends on:
| Factor | Why It Matters |
|---|---|
| Reserve distribution | Some banks may be short even if aggregate is high |
| Regulatory requirements | LCR, NSFR—Basel III liquidity requirements that force banks to hold minimum reserve buffers—trap some reserves |
| Bank risk preferences | Post-2019, banks hold larger precautionary buffers |
| Payment system structure | Intraday needs vary |
Floor Estimates:
| Source | Estimated Floor | Confidence |
|---|---|---|
| Fed research | $2.7-3.0 trillion | Medium |
| Market consensus | $2.8-2.9 trillion | Medium |
| October 2025 stress | ~$3.0T (stress emerged) | High |
The only way to truly discover the floor is to approach it and see what breaks.
Section 3: September 2019 - What Happens When Reserves Are Scarce
The clearest illustration of reserve scarcity came on September 16-17, 2019.
The Setup:
| Factor | Condition |
|---|---|
| Reserves | Declined to ~$1.4 trillion (lowest since 2011) |
| What drained them | QT1 balance sheet runoff since Oct 2017 |
| Trigger events | Corporate tax payments + Treasury settlement |
What Happened:
| Date | Repo Rate | Fed Target Range | Status |
|---|---|---|---|
| Sept 15 | ~2.0% | 2.00-2.25% | 🟢 Normal |
| Sept 16 | 5.0% | 2.00-2.25% | 🔴 Stress |
| Sept 17 | 10.0% (intraday) | 2.00-2.25% | 🔴 Crisis |
| Post-intervention | ~2.1% | 2.00-2.25% | 🟢 Resolved |
The federal funds rate itself moved above the target range. This represented a loss of monetary policy control. Banks could not access reserves at reasonable prices.
The Fed’s Response:
| Action | Amount | Timeline |
|---|---|---|
| Emergency repo injection | $75 billion/day | Sept 17 onwards |
| Continued injections | Daily | Months |
| T-bill purchases | Announced Oct 2019 | Ongoing |
The Lesson:
Reserves at $1.4 trillion were too low. The system could not handle even normal seasonal pressures. The Fed had overshot on QT and had to reverse course.
Section 4: October 2025 - Warning Signs Return
Fast forward to October 2025. Reserves had declined from post-pandemic peaks above $4 trillion to approximately $3.0 trillion through three years of QT2.
The Warning Indicators:
| Indicator | Normal | October 2025 | Status |
|---|---|---|---|
| EFFR position | Lower half of range | Drifting higher | 🟡 Warning |
| SRF usage | Near zero | $50 billion spike | 🔴 Stress signal |
| Money market volatility | Low | Elevated | 🟡 Warning |
Comparison: 2019 vs 2025:
| Factor | September 2019 | October 2025 |
|---|---|---|
| Reserves | ~$1.4T | ~$3.0T |
| Preceding action | QT1 drainage | QT2 drainage |
| Warning signal | Repo rate spike | SRF usage spike |
| Fed response | T-bill purchases | End QT + T-bill purchases |
| Severity | 🔴 Market seized | 🟡 Contained stress |
The Fed’s Response:
| Date | Action | Effect |
|---|---|---|
| Oct 2025 | Announced end of QT (effective Dec 1) | Stopped drainage |
| Dec 10, 2025 | Announced T-bill purchases ($40B/month) | Started injection |
| Dec 12, 2025 | Purchases begin | Active reserve management |
The Fed learned from 2019. Rather than waiting for markets to break, it acted preemptively when warning signs appeared.
Section 5: The “Ample” Floor Estimates
How low can reserves go before stress emerges?
Official Guidance:
Fed Chair Powell and other officials have described the target as maintaining reserves “somewhat above” the level where scarcity pressures emerge. They explicitly avoid naming a specific number.
Why the Uncertainty:
| Factor | Complication |
|---|---|
| Distribution | Aggregate can be high while some banks face scarcity |
| Regulation | LCR, NSFR, RLAP requirements trap reserves |
| Behaviour change | Post-2019 banks hold larger precautionary buffers |
| Intraday needs | Payment timing creates peaks and troughs |
Current Assessment:
| Metric | Value | Status |
|---|---|---|
| Current reserves | $2.88T | 🔴 |
| Distance to low estimate | +$180B above $2.7T | 🟡 |
| Distance to high estimate | -$20B below $2.9T | 🔴 |
November 2025 data now confirms reserves have fallen below the $2.9T floor estimate. The October 2025 stress at roughly $3.0 trillion was an early warning. The Fed’s pivot to T-bill purchases was timely.
Section 6: Where We Are Now
Current Reserve Dashboard (January 2026):
| Metric | Level | Status |
|---|---|---|
| Bank Reserves (TOTRESNS) | $2.88 trillion | 🔴 Below floor |
| Fed Target Range | 3.50% - 3.75% | - |
| IORB Rate | 3.65% | Interest paid on reserves |
| EFFR | ~3.58% | 🟢 Within target range |
| SRF Usage | Low | 🟢 Fed managing actively |
Reserve Trajectory:
| Date | Reserves | Driver |
|---|---|---|
| Jan 2022 | ~$4.0T | Post-QE peak |
| Jan 2023 | ~$3.5T | QT drainage |
| Jan 2024 | ~$3.3T | Continued drainage |
| Oct 2025 | ~$2.94T | SRF stress emerged |
| Nov 2025 | ~$2.88T | Below floor, QT ended |
| Jan 2026 | ~$2.88T | Fed T-bill purchases ongoing |
Projection (Fed operations only, assumes TGA neutral):
| Date | Expected Reserves | Status |
|---|---|---|
| Jan 2026 | ~$2.88T | 🔴 Below floor |
| Mar 2026 | ~$2.96-3.00T | 🟡 Approaching floor |
| Jun 2026 | ~$3.08-3.12T | 🟢 Rebuilding buffer |
Note: This projection isolates Fed T-bill purchases (~$40B/month) and doesn’t account for TGA movements or other balance sheet dynamics which could accelerate or offset these flows.
The Fed has indicated that initial T-bill purchases will be “elevated” through April 2026 and then reduced. The urgency is real—reserves are below floor.
Section 7: Thesis Connection
For Pierce & Pierce readers, reserve levels provide the fundamental constraint on the liquidity thesis.
Why Reserves Matter for Risk Assets:
| Reserve Condition | Effect on Markets |
|---|---|
| 🟢 Abundant | Banks have excess liquidity to deploy. Funding cheap and available. Supports risk-taking. |
| 🟡 Ample | Normal functioning. Adequate for system needs. |
| 🔴 Scarce | Banks hoard liquidity. Credit tightens. Market-making suffers. Risk appetite declines. |
Combined Injection Channels:
| Source | Monthly Impact | Status |
|---|---|---|
| Fed T-bill purchases | +$40B | 🟢 Ongoing since Dec 12 |
| TGA drawdown (variable) | +$40B | 🟢 Ongoing |
| ON RRP buffer | $0 (exhausted) | 🔴 No absorption |
| Net to reserves | ~$80B | 🟢 Direct injection |
⚠️ Transmission Caveats:
More reserves do not automatically mean higher asset prices. The relationship requires several transmission steps:
| Factor | Why Transmission Can Break |
|---|---|
| Bank deployment | Reserves must be deployed, not hoarded |
| Basel III constraints | LCR, NSFR can trap reserves |
| Credit demand | Must exist for reserves to become loans |
| Investor sentiment | Risk appetite must support allocation to crypto |
But the structural setup is supportive: multiple injection channels, no absorption buffer, reserves below floor with Fed actively managing to rebuild.
Section 8: What To Watch
Primary Data Sources:
| Source | What It Shows | Frequency |
|---|---|---|
| FRED: WRESBAL | Reserve balances | Weekly (Thursdays) |
| NY Fed SRF Data | Standing Repo usage | Daily |
| FRED: EFFR | Fed funds rate | Daily |
| FRED: SOFR | Repo market rate | Daily |
Reserve Level Interpretation:
| Level | Status | Interpretation |
|---|---|---|
| >$3.2T | 🟢 | Comfortable buffer above floor |
| $2.9-3.2T | 🟡 | Near floor, requires active management |
| $2.7-2.9T | 🔴 | At floor, stress likely without Fed support |
| <$2.7T | 🔴 | Below floor, money market stress expected |
Warning Signs to Monitor:
| Indicator | Normal | Warning Level | Current Status |
|---|---|---|---|
| SRF usage | Near zero | >$20B sustained | 🟢 Normal |
| EFFR position | Lower half of range | At/above top | 🟢 Normal |
| Repo rate vs IORB | Below IORB | Spikes above | 🟢 Normal |
The Reserve Pressure Formula:
Reserve Pressure = EFFR - (Bottom of target range)
Current: 3.58% - 3.50% = +8 bps → Ample, no stress
| EFFR Position | Interpretation | Status |
|---|---|---|
| Near bottom of range | Reserves abundant | 🟢 |
| Middle of range | Reserves ample | 🟡 |
| Near/above top of range | Reserves tightening | 🔴 |
Quick Reference
| Item | Detail |
|---|---|
| The Mechanic | Reserves are cash banks hold at the Fed. Too few reserves and money markets seize up (September 2019). The Fed targets “ample” reserves to maintain rate control. |
| The Nuance | No magic number for “ample.” Distribution matters as much as aggregate level. Banks can be reserve-constrained even when system-wide levels look adequate. |
| Current Level | $2.93 trillion (December 10, 2025) |
| Estimated Floor | $2.7-2.9 trillion (significant uncertainty) |
| Buffer Status | 🟡 Minimal. Reserves near the floor with Fed actively injecting. |
| Recent Stress | October 2025 SRF spike at ~$3.0T reserves prompted Fed to end QT and begin purchases. |
| Transmission | Reserve abundance supports but does not guarantee risk asset appreciation. |
| Monitor | FRED series WRESBAL (weekly), NY Fed SRF usage (daily), EFFR position in target range |
What’s Next
Brief #4: QT → QE (Balance Sheet Cycles)
How Fed balance sheet policy actually works, what changed in December 2025, and why “reserve management purchases” are different from QE.
Brief #5: Global M2 & The Liquidity Lag
Why Bitcoin correlates with money supply, why the lag typically runs 70-90 days, and what the current divergence tells us.
Sources:
Federal Reserve Economic Data (FRED) | NY Fed Open Market Operations | FOMC Meeting Minutes and Statements | Federal Reserve Staff Working Papers | Brookings Institution Repo Market Analysis | Office of Financial Research | NY Fed Liberty Street Economics
Methodology Note:
Reserve data from FRED series WRESBAL through December 10, 2025. September 2019 analysis from Federal Reserve staff notes and NY Fed research. October 2025 analysis from FOMC statements and Fed operations data. All figures represent publicly available data from official sources.
This publication is for educational purposes only. Not financial advice. Cryptocurrency investments carry substantial risk of complete loss.
Pierce & Pierce Research
Democratising Institutional-Grade Crypto Research