TOTRESNS: US Bank Reserves at the Federal Reserve Monthly Data (2001–2026)

The TOTRESNS series, published monthly by the Federal Reserve via FRED, measures total reserves held by US depository institutions at the Federal Reserve. Before September 2008, TOTRESNS averaged around 45 billion USD. QE programs expanded the series to 2.8 trillion by October 2014 and to a peak of 4.27 trillion in December 2021 — a roughly 90-fold expansion in thirteen years. TOTRESNS anchors the operational floor of the US monetary system: reserves determine banks’ lending capacity, the plumbing of the overnight funding market, and the practical lower bound below which Fed quantitative tightening triggers funding stress.

Dataset: US Bank Reserves at the Federal Reserve (2001–2026) · Updated —

Source: FRED series TOTRESNS · Federal Reserve Bank of St. Louis — FRED (TOTRESNS)

Macro Takeaway

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns.

Column names match the CSV headers exactly.

Download the Complete Dataset

The full dataset is available in CSV and Excel formats.

FRED Direct CSV Access

The underlying data is available from FRED under series code TOTRESNS:

https://fred.stlouisfed.org/graph/fredgraph.csv?id=TOTRESNS

Direct CSV Access — Eco3min Structured Dataset

https://eco3min.fr/dataset/us-bank-reserves.csv

This URL returns the complete dataset in CSV format. It can be used directly in pandas, R, curl, or any data tool.

Using the Dataset in Python

import pandas as pd

url = "https://eco3min.fr/dataset/us-bank-reserves.csv"
df = pd.read_csv(url, parse_dates=["date"])

print(df.head())
print(df["total_reserves_billions"].describe())

Using the Dataset in R

library(readr)

url <- "https://eco3min.fr/dataset/us-bank-reserves.csv"
df <- read_csv(url)

head(df)
summary(df$total_reserves_billions)

Both examples load the dataset directly from the URL — no download or API key required.

Methodology

TOTRESNS aggregates total reserves of depository institutions held at the Federal Reserve, not seasonally adjusted, published monthly as part of the Federal Reserve’s H.3 statistical release (“Aggregate Reserves of Depository Institutions and the Monetary Base”). The series combines what were historically classified as required reserves and excess reserves; the regulatory distinction was eliminated in March 2020 when the Fed set reserve requirements to zero across all deposit categories.

The reserve total reflects Fed asset purchases (QE creates reserves through securities purchases credited to bank accounts at the Fed), liquidity drains via the Treasury General Account and reverse-repo balances, and Treasury issuance dynamics. H.3 data are released monthly with approximately a one-week lag from the end of the reference period.

This Eco3min dataset is updated monthly via automated pull from the FRED API.

Data Quality & Provider Notes

TOTRESNS is among the most stable FRED monetary series: minimal revisions, no holiday gaps, monthly cadence aligned with the H.3 release. The Eco3min mirror updates monthly via automated FRED API pull.

• Release latency. The Federal Reserve publishes the H.3 release around the second Thursday of each month with reserve data for the preceding month. FRED ingests on release; Eco3min mirrors within 72 hours.
• Revisions policy. Historical values can be revised when the underlying H.3 components are restated, but revisions to TOTRESNS are typically small after the initial publication month. The series is generally considered final 60 days after first release.
• Alternative sources. Bloomberg (ARBLRSVS Index), Refinitiv/LSEG, and Haver Analytics republish the same H.3 data. The Federal Reserve’s H.4.1 release includes a closely related “Reserve balances with Federal Reserve Banks” line item at weekly frequency, useful for higher-resolution tracking between monthly H.3 prints.
• Known gaps. The March 2020 elimination of reserve requirements created a structural break in interpretation: pre-2020 and post-2020 reserve levels are not directly comparable in terms of “required” vs “excess” composition. The aggregate TOTRESNS level remains continuous.

For analysts tracking reserve-scarcity thresholds, the weekly H.4.1 “reserve balances” series provides finer granularity than monthly TOTRESNS at the cost of slightly higher short-term noise from intra-month TGA and RRP volatility.

Common Pitfalls When Using TOTRESNS

TOTRESNS is widely cited in monetary-policy and liquidity analysis, but several recurring interpretation errors distort the signal.

1. Confusing total reserves with “excess reserves” frameworks. Pre-2008 monetary analysis frequently invoked “excess reserves” as a forecasting signal under fractional-reserve banking. Post-QE, with reserves vastly exceeding any binding requirement, that framework no longer applies. The required-vs-excess distinction itself was formally abolished in March 2020.
2. Reading reserve levels as a measure of bank lending capacity. Banks extend credit based on capital constraints, loan demand, and risk-adjusted returns — not directly off reserve balances. The 90-fold reserve expansion since 2008 did not produce a proportional credit expansion, contradicting the textbook money-multiplier model still found in many macroeconomics textbooks.
3. Identifying QT impact through TOTRESNS alone. QT mechanically reduces reserves, but the Treasury General Account and reverse-repo balances also drain reserves independently. A 500 billion decline in TOTRESNS may reflect QT, a TGA build, an RRP increase, or any combination. The Net Liquidity decomposition (WALCL − TGA − RRP) provides the consolidated signal that isolates QT from autonomous factor swings.
4. Conflating reserve-scarcity thresholds across cycles. The September 2019 repo crisis began at roughly 1.4 trillion in total reserves. That figure is not a universal “scarcity threshold” — it reflects 2019-specific demand from regulatory liquidity requirements, money-market dynamics, and Treasury-market structure. The 2024–2026 operational threshold may differ given the Standing Repo Facility and ongoing Basel III adjustments to bank balance-sheet capacity.

Historical Regimes

TOTRESNS history since 2001 traces the structural transformation of US monetary policy: from a binding-requirements regime, through three rounds of QE, into the abundant-reserves operating framework that now defines Fed implementation.

• 2001–August 2008 — Pre-crisis baseline. TOTRESNS averaged 40–60 billion USD. Reserves were a binding constraint via reserve requirements, and the federal funds market cleared at a positive rate above the required-reserve floor. The federal funds rate was the operational lever and reserve management was a daily Fed-desk activity.
• September 2008 – October 2014 — QE expansion. Reserves surged from 45 billion in August 2008 to 2.8 trillion by October 2014 — driven by QE1, QE2, and QE3. The Federal Reserve implemented Interest on Excess Reserves (IOER) in October 2008 to control the policy rate in the new abundant-reserves regime, decoupling the Fed funds rate from reserve scarcity.
• 2014–2019 — Normalization and reserve scarcity. Following the end of QE3 and the start of balance-sheet runoff in October 2017, TOTRESNS declined from 2.8 trillion to 1.4 trillion by September 2019. The September 2019 repo crisis revealed that what had been considered “abundant” reserves had become operationally scarce relative to growing collateral demands and post-Basel III liquidity requirements, forcing the Fed to halt QT and restart bill purchases.
• March 2020 – December 2021 — Pandemic peak. Emergency QE in March 2020 added roughly 3 trillion to the Fed balance sheet within 90 days. TOTRESNS climbed from 1.6 trillion to a record 4.27 trillion by December 2021. The reserve flood saturated the system to the point that the overnight reverse repo facility rose to record balances absorbing the excess.
• June 2022 – 2024 — QT phase. With QT resuming in June 2022, TOTRESNS declined from 4.0 trillion toward 3.0–3.3 trillion. The Fed slowed the QT pace in 2024 and announced its end during 2025, citing concerns about repeating the 2019 reserve-scarcity episode. The operational history of this period is detailed in the Fed Balance Sheet Operational History study, which uses TOTRESNS alongside WALCL to reconstruct the regime sequence.
• 2025–2026 — Post-QT stabilization. Following the end of QT, TOTRESNS has stabilized in the 3.0–3.3 trillion range, well above the 2019 scarcity threshold but below the 2021 peak. The Fed continues to monitor the level relative to bank demand for reserves, with the Standing Repo Facility providing a backstop against a repeat of September 2019.

Related Macroeconomic Datasets

TOTRESNS is the liability-side counterpart to the Fed’s asset holdings (WALCL) and a core variable in any liquidity decomposition. The most useful cross-references combine reserve balances with the asset side, the autonomous drains (TGA, RRP), and the broader monetary aggregates that respond at longer horizons.

Macroeconomic Dataset Hub

This dataset is part of the Eco3min macro-financial data repository.

Explore the Eco3min Dataset Hub

Sources

• Federal Reserve Bank of St. Louis — FRED (TOTRESNS)
• Federal Reserve Board of Governors — H.3 statistical release (“Aggregate Reserves of Depository Institutions and the Monetary Base”)
• Federal Reserve Board of Governors — H.4.1 statistical release for higher-frequency reserve-balance cross-checks

Dataset Reference

Eco3min Research. (2026). US Bank Reserves at the Federal Reserve — FRED TOTRESNS (2001–2026).<br> https://eco3min.fr/en/us-bank-reserves-dataset/