PERMIT: US Building Permits Monthly SAAR Data from FRED (1960–2026)

The US Building Permits series counts new privately-owned housing units authorized by permit, monthly since 1960 at a seasonally adjusted annual rate, from the Census Bureau. It is the front of the construction pipeline and one of the ten components of the Conference Board’s Leading Economic Index: builders file permits before breaking ground, making this the most forward-looking hard-data series in the housing bloc. The record high is 2.42 million (December 1972), the GFC trough 513,000 (March 2009); mid-2026 runs at 1.41 million.

Dataset: US Building Permits (1960–2026) · Updated 2026-06-01

Source: US Census Bureau · FRED series PERMIT

Macro Takeaway

Dataset Overview

Dataset Variables

The CSV and Excel files contain the following columns.

A value of 1,410 means permits were being issued at a pace of 1.41 million units per year.

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 PERMIT:

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

Direct CSV Access — Eco3min Structured Dataset

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

print(f"Latest pace: {df['building_permits'].iloc[-1]/1000:.2f} million units/year")
print(df.tail())

Using the Dataset in R

library(readr)

url <- "https://eco3min.fr/dataset/building-permits.csv"
df <- read_csv(url)

tail(df)
summary(df$building_permits)

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

Methodology

Permits are compiled from the Census Bureau’s Building Permits Survey of permit-issuing places, covering jurisdictions that account for essentially all US residential construction. The headline is total privately-owned units authorized — single-family plus multifamily — at a seasonally adjusted annual rate.

The permit-issuing universe has been expanded periodically (from 14,000 to 19,000 to 20,000 places over the decades); the Census bridges these transitions, but very long-run comparisons span slightly different coverage.

This dataset is updated daily (Mon–Sat, 08:00 UTC) via automated pull from the FRED API; new observations appear with each source release.

Data Quality & Provider Notes

• Release latency. ~3 weeks after the reference month, with the residential construction release; a preliminary figure is finalized the following month.
• Revisions policy. One routine monthly revision plus annual benchmarking — materially smaller than starts revisions.
• Statistical precision. Administrative-count basis gives permits the tightest confidence intervals in the housing bloc; monthly changes are more often significant than for starts.
• Coverage changes. Universe expansions (14k→19k→20k places) create minor breaks bridged by Census methodology.
• Known gaps. None; continuous since 1960.

Common Pitfalls When Using Building Permits

1. Assuming every permit becomes a start. Permits can be delayed, abandoned, or re-filed; in stressed markets the permit-start conversion rate falls — precisely when the gap matters most.
2. Ignoring the single/multi split. Multifamily permits swing far harder than single-family and drove most of the 2021–2024 boom-bust inside the total.
3. Double-counting the signal. Permits lead starts mechanically; treating a permit drop and the subsequent starts drop as two independent confirmations overstates the evidence.
4. Cross-era level comparisons without per-capita scaling. The 1972 record of 2.42M served a far smaller population; today’s 1.4M is not directly comparable slack.

Historical Regimes

1960–1979 — The demographic surge. Permits repeatedly spiked past 2 million — the all-time high of 2.42M in December 1972 — as boomer household formation collided with volatile financing conditions in violent boom-bust cycles.

1980–1991 — Rate-shock whipsaws. The Volcker era crushed permits below 900k twice; the 1980s recovery topped near 1.8M before the S&L bust ended the cycle.

1992–2005 — The long expansion. A steady climb through two decades to the bubble peak of 2.26M (September 2005) — permits peaked a full quarter before starts and two years before the recession, the leading indicator performing on schedule.

2006–2009 — The collapse. Permits fell 77% to 513k (March 2009), bottoming one month before starts — the front of the pipeline turning first, in both directions.

2010–2019 — Underbuilding codified. The slow recovery mirrored starts: a decade to return to ~1.4M, locking in the 2010s construction deficit.

2020–2026 — Boom, repricing, and the gap. Permits surged to 1.92M (January 2022), led the rate-shock decline, and settled near 1.41M by mid-2026 — running well above actual starts, a pipeline of authorized-but-delayed projects awaiting financing clarity.

Related Macroeconomic Datasets

Permits open the construction pipeline; the series below track its later stages and its financial drivers.

Macroeconomic Dataset Hub

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

Sources

• US Census Bureau — Building Permits Survey, New Privately-Owned Housing Units Authorized
• Federal Reserve Bank of St. Louis — FRED series PERMIT

Dataset Reference