M2SL: US Money Supply M2 Monthly Data from FRED (1959–2026)

The M2SL series, published monthly by the Federal Reserve via FRED, measures the US M2 money supply — currency, checkable deposits, savings deposits, retail money market mutual funds, and small-denomination time deposits, seasonally adjusted in billions of US dollars. Since January 1959, M2SL has expanded from 287 billion to over 21 trillion. M2 growth has historically led CPI inflation by roughly 12–18 months in some regimes: the 41 percent expansion of M2SL between February 2020 and April 2022 preceded the 9.1 percent CPI inflation peak of June 2022, reviving monetarist debate after a decade of weak transmission.

Dataset: US Money Supply M2 (1959–2026) · Updated 2026-06-01

Source: FRED series M2SL · Federal Reserve Bank of St. Louis

Macro Takeaway

M2SL Dataset Overview

CSV & Excel Variables

The CSV and Excel files contain the following columns, ready for analysis in Python, R, or any data tool.

Column names match the CSV headers exactly.

Download the Complete M2SL Dataset

The full M2 money supply dataset is available as instant CSV and Excel download — no signup, no API key.

FRED Direct CSV Access

The underlying data is available from FRED under series code M2SL:

https://fred.stlouisfed.org/graph/fredgraph.csv?id=M2SL

Eco3min Structured CSV — Direct Access

https://eco3min.fr/dataset/us-m2-money-supply.csv

This URL returns the complete M2SL dataset in CSV format with pre-calculated YoY growth. Use directly in pandas, R, curl, or any data tool.

Load M2SL Data in Python

import pandas as pd

url = "https://eco3min.fr/dataset/us-m2-money-supply.csv"
df = pd.read_csv(url, parse_dates=["date"])

print(df.head())
print(df["m2_billions"].describe())
print(df["m2_yoy"].describe())

Load M2SL Data in R

library(readr)

url <- "https://eco3min.fr/dataset/us-m2-money-supply.csv"
df <- read_csv(url)

head(df)
summary(df$m2_billions)
summary(df$m2_yoy)

Both examples load the M2 money supply CSV directly from the URL — no download or API key required.

Methodology

M2SL is published monthly by the Federal Reserve Board of Governors as part of the H.6 statistical release (“Money Stock Measures”). It aggregates M1 (currency in circulation, demand deposits, traveler’s checks, and other checkable deposits) plus savings deposits, small-denomination time deposits below 100,000 USD, and retail money-market mutual fund balances.

The series is seasonally adjusted and expressed in billions of US dollars. The Federal Reserve revises the methodology periodically: a notable redefinition in May 2020 reclassified savings deposits from the M1/M2 boundary, mechanically increasing M1 substantially without changing the M2 total. M2SL is published monthly with a roughly four-week lag from the reference month.

The year-over-year growth column (m2_yoy) is calculated by Eco3min from the raw M2SL levels. This Eco3min dataset is updated monthly (15th of each month, 08:00 UTC) via automated pull from the FRED API.

Data Quality & Provider Notes

M2SL is the canonical broad-money aggregate for the United States, with a continuous monthly series back to January 1959 and quarterly variants extending further. The Eco3min mirror updates monthly with the H.6 release plus a pre-calculated year-over-year change.

• Release latency. The Federal Reserve publishes the H.6 release on the fourth Tuesday of each month with M2 data for the preceding month. FRED ingests within hours; Eco3min mirrors within 72 hours.
• Revisions policy. Recent observations are subject to small revisions over the following 1–2 months as deposit and money-market fund data are finalized. Historical values are rarely revised after the initial 90-day window. Major redefinitions (such as the May 2020 reclassification of savings deposits) create structural breaks worth flagging in component-level analysis.
• Alternative sources. Bloomberg (M2 Index), Refinitiv/LSEG, and the BIS Monetary and Financial Statistics database provide parallel series. ALFRED (Archival FRED) preserves vintage M2SL values, useful for real-time-data analysis of how policymakers viewed money supply at the moment of past policy decisions.
• Known gaps. No gaps in the monthly cadence since 1959. The May 2020 methodology revision is the most consequential structural break in the modern series: comparing M2 components pre- and post-revision requires careful reconstruction, though the M2 total remains continuous.

For analysts focused on real-time monetary policy signals, the year-over-year growth rate (m2_yoy column) carries substantially more signal than the level. The level rises mechanically with nominal GDP; deviations of the growth rate from its long-run average are the analytically informative metric.

Common Pitfalls When Using M2SL

M2SL is widely cited in monetary analysis, but several recurring interpretation errors distort the signal.

1. Reading the level rather than the growth rate. M2SL trends upward mechanically with nominal GDP and population growth. The level alone carries little signal. Year-over-year change, change relative to nominal GDP, or deviation from a trend reference are the analytically meaningful transformations. The 21 trillion USD figure in 2026 is not “high” without that normalization context.
2. Applying the monetarist transmission mechanically. The empirical 12–18 month lag between M2 growth and CPI inflation has held in some periods (1970s, 2020–2022) but broken in others — the 2009–2019 period saw substantial M2 expansion with persistent below-target inflation, attributed to depressed velocity and reserves trapped in the banking system. The relationship is conditional, not deterministic, and depends critically on velocity dynamics.
3. Confusing M2 with reserves or the monetary base. M2 measures money held by the public (deposits and currency). Bank reserves (TOTRESNS) are claims of banks on the Federal Reserve. The two can move in opposite directions: QE creates reserves but does not mechanically create M2 unless those reserves support new bank lending. Confusing the two leads to systematic misreading of QE eras.
4. Comparing M2 across regimes without adjusting for methodology breaks. The May 2020 reclassification of savings deposits altered M1 substantially and changed the internal composition of M2. Comparing M1-to-M2 ratios across that break, or using older M1 series to extrapolate, introduces a methodology-error component independent of any underlying monetary dynamics.

Historical Regimes

M2SL history since 1959 traces the principal monetary regimes of the modern US economy: Bretton Woods, the Volcker disinflation, the Great Moderation, the post-2008 QE era, and the pandemic surge that revived monetarist debate.

• 1959–1979 — Bretton Woods and post-Bretton Woods inflation buildup. M2SL expanded from 287 billion to 1.5 trillion. The acceleration of M2 growth from 7 percent annualized in the 1960s to over 12 percent by the late 1970s preceded the inflation crisis of 1979–1981, an episode central to monetarist theory and the empirical case for a money-supply leading indicator.
• 1980–1990 — Volcker disinflation. M2SL growth decelerated to 6–8 percent through the 1980s as Paul Volcker’s Federal Reserve formally adopted M2 targeting in October 1979. CPI inflation declined from 13.5 percent in 1980 to roughly 4 percent by 1986, validating the M2-inflation link at the cost of the deepest recession since the 1930s.
• 1990–2007 — Great Moderation. M2SL grew at 5–7 percent annually with stable single-digit inflation and steady velocity. The Federal Reserve shifted away from explicit money-supply targeting toward interest-rate-based policy under Alan Greenspan, and academic debate intensified over whether M2 retained signal value in the new operating framework.
• 2008–2019 — QE era with weak transmission. Despite massive Fed balance sheet expansion via QE1, QE2, and QE3, M2SL growth remained modest at 5–7 percent. The textbook money-multiplier transmission broke down: Fed assets (WALCL) expanded roughly 5-fold, but bank lending and broad money did not respond proportionally. CPI inflation stayed below the 2 percent Fed target for most of the decade.
• February 2020 – April 2022 — Pandemic surge. M2SL expanded from 15.4 trillion in February 2020 to 21.8 trillion by April 2022 — a 41 percent increase in 26 months, the largest monetary expansion in the modern series. The mechanism combined fiscal transfers credited directly to household deposits, PPP loan flows into business deposits, and elevated household saving. The lagged effect on CPI inflation peaked at 9.1 percent in June 2022, an episode analyzed in the M2 and inflation Friedman-test study.
• 2023–2026 — Post-pandemic contraction and stabilization. M2SL contracted year-over-year for the first time in the modern series during 2023 as deposits flowed into money-market funds (capturing higher overnight rates) and the Fed conducted QT. M2 growth returned to the 3–5 percent range by 2025 as deposit flows stabilized and Fed policy normalized. The episode confirmed that velocity dynamics remain a first-order driver of the M2-inflation relationship.

Related Macroeconomic Datasets

M2SL is the broadest commonly reported money-supply measure, but it operates on a different timescale from reserve-level liquidity. M2 growth has historically led inflation by 12–18 months in active-velocity regimes, while reserve-level liquidity tracks shorter-term financial-system plumbing. Cross-referencing these dimensions provides a dual-horizon view of monetary conditions.

Macroeconomic Dataset Hub

This M2SL dataset is part of the Eco3min macro-financial data repository. Browse all available FRED datasets with instant CSV download.

Explore the Eco3min Dataset Hub

Sources

• Federal Reserve Bank of St. Louis — FRED database, series M2SL
• Federal Reserve Board of Governors — H.6 statistical release (“Money Stock Measures”)

Dataset Reference

Eco3min Research. (2026). US Money Supply M2 — FRED M2SL (1959–2026).<br> https://eco3min.fr/en/us-m2-money-supply-dataset/