MORTGAGE30US: US 30-Year Fixed Mortgage Rate Weekly Survey Since 1971

The MORTGAGE30US series, published weekly by Freddie Mac via FRED, tracks the average 30-year fixed-rate mortgage commitment rate offered to US prime conforming borrowers since April 1971 — the longest continuous record of US mortgage rates available. MORTGAGE30US is the reference price of homeownership credit for most American households and the single most important rate for the US residential mortgage market. The series is produced by Freddie Mac and distributed through FRED under a Freddie Mac copyright (citation required); Eco3min provides the chart and the analytical layer, and points to FRED and Freddie Mac as the distribution channels for the data itself.

Dataset: US 30-Year Mortgage Rate (1971–present)

Source: Federal Reserve Bank of St. Louis (FRED). Chart generated and served by FRED.

Macro Takeaway

Dataset Overview

Licensing note: MORTGAGE30US is © Freddie Mac and is flagged “Copyrighted: Citation Required” on FRED. Eco3min therefore does not redistribute the file; the complete weekly history since 1971 remains accessible from FRED and from Freddie Mac directly via the links below.

Dataset Variables

The FRED CSV contains the following columns. Each row represents one weekly survey release.

Missing values appear as “.” in the FRED CSV.

FRED Direct CSV Access

The complete weekly series since 1971 is available from FRED under series code MORTGAGE30US:

https://fred.stlouisfed.org/graph/fredgraph.csv?id=MORTGAGE30US

Freddie Mac also publishes the full PMMS history as a downloadable workbook on its own site (freddiemac.com/pmms), which is the primary channel for the pre-aggregated historical file.

The FRED URL returns the full history in CSV format, usable directly in pandas, R, curl, or any data tool, no download or API key required.

Using the Dataset in Python

import pandas as pd

url = "https://fred.stlouisfed.org/graph/fredgraph.csv?id=MORTGAGE30US"
df = pd.read_csv(url, parse_dates=["observation_date"], na_values=".")

print(f"Latest rate: {df['MORTGAGE30US'].dropna().iloc[-1]:.2f}%")
print(f"Historical average: {df['MORTGAGE30US'].mean():.2f}%")
print(f"Record high: {df['MORTGAGE30US'].max():.2f}% on {df.loc[df['MORTGAGE30US'].idxmax(), 'observation_date'].date()}")

Using the Dataset in R

library(readr)

url <- "https://fred.stlouisfed.org/graph/fredgraph.csv?id=MORTGAGE30US"
df <- read_csv(url, na = ".")

head(df)
summary(df$MORTGAGE30US)

Both examples load the data directly from FRED, no download or API key required.

Methodology

MORTGAGE30US comes from Freddie Mac’s Primary Mortgage Market Survey (PMMS). Since November 17, 2022, the weekly rate has been based on mortgage applications submitted to Freddie Mac by lenders across the country (Loan Product Advisor data), averaged over the prior Thursday through Wednesday; before that date it came from a weekly survey of lenders. The reference borrower has good to excellent credit and a 20% down payment on an owner-occupied single-family home purchase. The published value is reported each Thursday morning.

The series begins in April 1971, making MORTGAGE30US the longest continuous record of US mortgage rates available from any source. Prior to that date, no comparable continuous series exists.

A methodological note: on November 17, 2022, Freddie Mac replaced the lender survey with data from loan applications submitted to Freddie Mac (see the Freddie Mac note reproduced on the FRED series page). Pre-November-2022 and post-November-2022 values are conceptually consistent but reflect different data collection and aggregation procedures, which matters for any precise comparison across the methodology break.

The chart above is generated and served by FRED and always reflects the latest weekly release.

Data Quality & Provider Notes

MORTGAGE30US is a survey-based indicator rather than a transaction-tape series, which carries specific implications for interpretation and use. For the definitional layer behind that caveat, see how Freddie Mac builds the weekly PMMS rate.

• Licensing (important). The series is © Freddie Mac and carries a “Copyrighted: Citation Required” flag on FRED. Redistribution of the file itself requires the owner’s permission, which is why Eco3min links to FRED and Freddie Mac rather than mirroring the data.
• Release latency. Freddie Mac publishes the PMMS weekly on Thursdays at 10:00 ET; FRED ingests the release the same week.
• Revisions policy. MORTGAGE30US is not revised after release. Historical values remain unchanged once posted.
• Methodology break. In November 2022, Freddie Mac transitioned the PMMS from a lender survey to data on loan applications submitted to Freddie Mac. Pre-2022 and post-2022 observations are not strictly methodologically identical, though Freddie Mac maintains the series as continuous in FRED.
• Alternative sources. The Mortgage Bankers Association (MBA) publishes a competing weekly survey with separate methodology and slightly different coverage. Daily lender-level rate data is available from Optimal Blue, Mortgage News Daily, and Bankrate. These daily sources typically lead MORTGAGE30US by several days due to the PMMS reporting lag.
• Known gaps. No daily data exists — MORTGAGE30US is weekly only. The earliest observation is April 2, 1971. The series tracks conforming 30-year fixed loans only and does not cover jumbo, FHA, VA, or adjustable-rate mortgages.

For real-time mortgage rate analysis, daily lender data should supplement MORTGAGE30US during periods of rapid rate movement when the weekly lag becomes material.

Common Pitfalls When Using MORTGAGE30US

MORTGAGE30US is widely cited but several interpretation errors regularly appear in applied analysis.

1. Treating MORTGAGE30US as a daily quote. MORTGAGE30US is a weekly average released on Thursday, not a daily transaction rate. During periods of rapid rate movement (March 2020, 2022-2023), the weekly value can lag actual market conditions by several days. Daily lender data from Optimal Blue or Mortgage News Daily is more appropriate for real-time analysis.
2. Ignoring fees and points. Freddie Mac stopped publishing fees and points in November 2022, so MORTGAGE30US does not capture the full cost of a loan. Quotes from individual lenders depend on the points paid and can differ from the MORTGAGE30US figure. Comparing MORTGAGE30US directly with online rate-shopping quotes without adjusting for fees introduces a systematic bias.
3. Conflating MORTGAGE30US with the broader mortgage market. MORTGAGE30US covers conforming 30-year fixed-rate loans for prime borrowers only. It excludes jumbo, FHA, VA, and ARM products, which can carry materially different rates. Using MORTGAGE30US as a stand-in for “the US mortgage rate” misses meaningful dispersion across loan types.
4. Reading MORTGAGE30US in isolation from the Treasury curve. The mortgage-Treasury spread (typically 150-200 basis points) compresses or widens with MBS demand, prepayment expectations, and Fed balance sheet policy. Movements in MORTGAGE30US that look like rate shocks may simply reflect spread widening; conversely, stable MORTGAGE30US during falling Treasury yields signals deteriorating MBS demand.

Historical Regimes

1971–1981: Inflationary surge. MORTGAGE30US rose from approximately 7.3% in April 1971 to a record peak of 18.63% in October 1981, tracking the Volcker tightening that pushed the 10-year Treasury yield above 15%. Affordability collapsed and household formation slowed materially during this period.

1981–2003: Secular decline. MORTGAGE30US fell from 18% to roughly 5%, with cyclical oscillations around the secular downtrend. Each Fed easing cycle brought MORTGAGE30US to a lower trough. The mortgage-Treasury spread averaged approximately 170 basis points across this period, with structural compression as the MBS market deepened.

2003–2007: Housing boom. MORTGAGE30US ranged between 5.2% and 6.8% during the housing boom, with the spread compressing on strong MBS demand. The combination of low MORTGAGE30US and aggressive lending standards inflated the housing bubble that broke in 2007-2008.

2008–2021: Zero-bound era and QE. MORTGAGE30US dropped to a record low of 2.65% in January 2021, supported by Fed MBS purchases (QE1, QE3, QE4) that compressed the mortgage-Treasury spread to historic lows. The Fed’s balance sheet expansion was the dominant force on MORTGAGE30US during this period, often overwhelming Treasury yield movements.

2022–2023: Steepest proportional rise on record. MORTGAGE30US surged from about 3.1% to 7.79% in under two years, a move that crushed affordability and froze the resale market as existing homeowners refused to surrender their sub-3% loans. The relationship between rates, prices, and affordability during this period is detailed in Rates vs Prices: US Housing Affordability.

2024–2026: Plateau, then a renewed climb. MORTGAGE30US stayed between about 6.1% and 7.2% through 2024 and 2025, touched 5.98% in February 2026, then climbed to 7.40% in October 2026 (Freddie Mac, via FRED). The persistence of elevated rates relative to the locked-in mortgage stock continues to constrain housing turnover. The duration risk embedded in the 30-year segment is analyzed in US 30Y Duration Risk.

Related Macroeconomic Datasets

MORTGAGE30US is structurally driven by the 10-year Treasury yield and Fed MBS policy, and feeds directly into housing affordability and household balance sheet dynamics. Cross-reference with the following datasets to understand transmission and regime.

Macroeconomic Dataset Hub

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

Explore the Eco3min Dataset Hub

Sources

• Freddie Mac: Primary Mortgage Market Survey (PMMS)
• Federal Reserve Bank of St. Louis — FRED series MORTGAGE30US

Dataset Reference

Eco3min Research. (2026). US 30-Year Fixed Mortgage Rate — MORTGAGE30US Weekly Survey (1971–present).<br> https://eco3min.fr/en/us-30-year-mortgage-rate-dataset/