DGS10: Daily 10-Year US Treasury Constant Maturity Yield from FRED (1962–2026)

The DGS10 series, published daily by FRED from US Treasury source data, tracks the constant-maturity 10-year US Treasury yield since January 1962 — over 16,000 daily observations. DGS10 is the global pricing reference for long-term US dollar debt and the anchor against which mortgage rates, corporate bond spreads, and sovereign debt valuations are calibrated worldwide.

Dataset: US 10-Year Treasury Yield · Updated 2026-07-31

Source: FRED series DGS10 · Federal Reserve Bank of St. Louis

Macro Takeaway

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns. Each row represents one business day.

Column names match the CSV headers exactly.

Download the Complete Dataset

The full dataset is available in CSV and Excel formats — over 16,000 daily observations spanning more than six decades.

FRED Direct CSV Access

The underlying data is published in the Federal Reserve Economic Data (FRED) database under the series code DGS10. FRED datasets are commonly accessed using the fredgraph.csv query format:

https://fred.stlouisfed.org/graph/fredgraph.csv?id=DGS10

The raw FRED CSV provides the same daily yield series. The Eco3min dataset mirrors FRED’s DGS10 data with identical values, but packages it in a stable, versionable CSV with consistent column names — designed for direct ingestion in Python, R, or any data pipeline. The URL never changes, making it suitable for use in automated scripts.

Direct CSV Access — Eco3min Structured Dataset

https://eco3min.fr/dataset/us-10y-treasury-yield.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-10y-treasury-yield.csv"
df = pd.read_csv(url, parse_dates=["date"])

# Display first rows
print(df.head())

# Basic statistics
print(df["dgs10"].describe())

# Plot
df.plot(x="date", y="dgs10", title="US 10-Year Treasury Yield", legend=False)

Using the Dataset in R

library(readr)

url <- "https://eco3min.fr/dataset/us-10y-treasury-yield.csv"
df <- read_csv(url)

head(df)
summary(df$dgs10)

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

Methodology

The primary data source is the US Treasury Department’s daily yield curve, as reported by the Federal Reserve Board of Governors in its H.15 statistical release. DGS10 — the 10-year constant-maturity rate — represents the interpolated yield on a theoretical Treasury security with exactly 10 years remaining to maturity. This interpolation method, maintained by the Treasury since 1962, ensures consistency even as individual bond issues mature and roll off. Related reading: the routes to buying bonds.

“Constant maturity” means DGS10 is not the yield on a single bond but a derived rate: the Treasury Department fits a yield curve to outstanding securities and reads off the 10-year point. This methodology, documented in the Treasury’s own publications, avoids the distortions that would arise from tracking any individual bond as its remaining maturity shrinks over time.

The dataset reflects nominal yields. To derive the real 10-year yield (net of inflation expectations), the Real 10-Year Treasury Yield composite dataset subtracts CPI year-over-year from DGS10. Alternatively, the market-implied real rate is observable via TIPS (Treasury Inflation-Protected Securities), though TIPS liquidity premia introduce their own measurement issues — a point documented in Federal Reserve research (Kim and Wright, 2005).

This dataset is updated weekly (Saturday 08:00 UTC) via automated pull from the FRED API. Daily observations reflect the previous business week’s closing rates. Weekend and holiday dates are excluded from the series.

Data Quality & Provider Notes

DGS10 is one of the most heavily monitored financial series globally and benefits from a transparent publication chain through the Federal Reserve H.15 release. The data is sourced directly from US Treasury closing yields and finalized end-of-day.

• Release latency. The Federal Reserve publishes DGS10 in its daily H.15 release at approximately 16:15 ET each business day. FRED ingests the H.15 data with a T+1 morning cadence. Eco3min mirrors FRED with a weekly Saturday pull at 08:00 UTC.
• Revisions policy. DGS10 is not revised. The constant-maturity yield is derived from end-of-day Treasury closing yields and finalized at publication. Unlike vintage-sensitive macro series, DGS10 has a single canonical print per date.
• Alternative sources. US Treasury Department daily yield curve (treasury.gov), Bloomberg ticker USGG10YR, Refinitiv US10YT=RR. Note that DGS10 reports the par yield curve constant-maturity rate; market participants often quote the on-the-run 10-year note yield, which can differ from DGS10 by 1–3 basis points due to liquidity premia on the most recently issued security.
• Known gaps. Weekends and US holidays (no Treasury market activity). Brief gaps during exceptional closures: the Treasury market closed partially during Hurricane Sandy (October 29–30, 2012) and other rare events. DGS10 has no observation for those specific dates.

For comparisons with intraday or tick-level data, DGS10 captures only the end-of-day closing rate — intraday high/low ranges are not in the series. For benchmark-grade yield curve work, the full set of constant-maturity rates (DGS1MO, DGS3MO, DGS6MO, DGS1, DGS2, DGS3, DGS5, DGS7, DGS10, DGS20, DGS30) on FRED provides the complete picture.

Common Pitfalls When Using DGS10

DGS10 is one of the most cited financial series in research and journalism, but several recurring interpretation errors distort the signal.

1. Confusing yield and price. DGS10 reports the yield; bond prices move inversely. A rise from 4% to 5% in DGS10 corresponds to roughly an 8% decline in the price of a 10-year zero-coupon Treasury. Users sometimes report “10-year up” without specifying whether they mean yield or price — a meaningful ambiguity in any portfolio-impact analysis.
2. Mixing nominal DGS10 with the real 10-year rate. DGS10 is a nominal yield. The real 10-year yield (DGS10 minus expected inflation, or directly via TIPS) is a different series with different macro signaling. A rising DGS10 with stable breakeven inflation signals tightening real conditions; the same nominal move with rising breakevens signals little about real rates and more about inflation expectations.
3. Reading short-term moves outside historical context. A 100-basis-point move in DGS10 over 6 months can feel extreme in a 2% environment but is unremarkable against a 60-year distribution that ranges from 0.52% (August 2020) to 15.84% (September 1981). Z-scores against rolling 10-year windows are more informative than absolute spreads measured against recent memory.
4. Ignoring the term-premium component. DGS10 decomposes into expected average short rates over 10 years plus a term premium compensating for duration risk. Federal Reserve Bank of New York ACM model estimates show the term premium swinging by 200+ basis points across cycles. A DGS10 move can come entirely from term premium, entirely from rate expectations, or from both — the macro reading differs sharply depending on which component drove the move.

Historical Regimes

1962–1981 — The Great Inflation. DGS10 rose from roughly 4% to nearly 16%, tracking the acceleration of US inflation driven by fiscal expansion (Vietnam War, Great Society programs), oil shocks (1973, 1979), and a Federal Reserve that consistently underestimated inflation persistence. By 1981, with Paul Volcker raising the fed funds rate above 19%, DGS10 peaked at 15.84% — a level that priced in both realized inflation and a substantial credibility premium for a central bank that had repeatedly failed to contain prices.

1981–2000 — Secular disinflation. As Volcker’s monetary tightening broke inflation expectations, DGS10 entered a two-decade decline from 16% toward 5–6%. This period saw a progressive compression of the inflation risk premium as CPI fell from double digits to the 2–3% range. Each business cycle produced a lower peak in long rates — a pattern sometimes described as the “great moderation” in academic literature.

2000–2019 — The zero-bound era. DGS10 dropped from 6.5% to below 1.5%, driven by three forces: the Fed’s zero interest rate policy after 2008, quantitative easing programs (which directly compressed term premia), and global demand for safe assets from foreign central banks and institutional investors. According to Federal Reserve Bank of New York estimates, QE alone may have reduced the 10-year term premium by 100–150 basis points at peak effect. The 2s10s yield curve and the 30-year mortgage rate moved in lockstep with DGS10 during this period.

2020–2022 — Pandemic shock and reversal. DGS10 briefly touched 0.52% in August 2020 — its all-time low — as the Fed cut rates to zero and launched unlimited QE in response to COVID-19. The subsequent reversal was the fastest in modern history: from 0.52% to above 4.2% in barely two years, as post-pandemic inflation surged past 9% (June 2022, BLS data) and the Fed executed its most aggressive tightening cycle since the early 1980s. The transmission of this move into housing affordability is documented in our study on rates versus prices in US housing.

2023–present — Structural repricing. With DGS10 trading in the 4–5% range, the market appears to be pricing a regime of higher neutral rates than the 2010s. The term premium — negative for much of the post-GFC period — has turned positive according to ACM model estimates from the New York Fed. Whether this repricing is permanent or cyclical remains the central question in fixed income markets, addressed in our complete history of yield curve inversions and recession signaling.

Related Macroeconomic Datasets

DGS10 is the anchor of the global fixed income system. It decomposes into two components: the real rate and inflation expectations. Cross-referencing with other maturities reveals the yield-curve shape — one of the strongest recession predictors available. Comparing with the mortgage rate shows how Treasury yields propagate into the real economy. Related material: the yield-curve time machine.

Macroeconomic Dataset Hub

This dataset is part of the Eco3min macro-financial data repository. Explore all available datasets including inflation, interest rates, equity returns, credit spreads, and currency indicators.

Explore the Eco3min Dataset Hub

Sources

• Board of Governors of the Federal Reserve System — H.15 Selected Interest Rates, 10-Year Treasury Constant Maturity Rate
• Federal Reserve Bank of St. Louis — FRED database, series DGS10
• US Department of the Treasury — daily yield curve, par yield interpolation methodology

Dataset Reference

Eco3min Research. (2026). DGS10 — Daily 10-Year US Treasury Constant Maturity Yield (1962–2026).<br> https://eco3min.fr/en/us-10y-treasury-yield-dataset/