Total Private Construction Spending: Nonresidential in the United States (PNRESCONS) | FRED | St. Louis Fed
- Rows
- 401
- Columns
- 2
- Data range
- 1993-01-01 to 2026-05-01
Graph and download economic data for Total Private Construction Spending: Nonresidential in the United States (PNRESCONS) from Jan 1993 to May 2026 about nonresidential, expenditures, construction, private, and USA.
AI analysis
Analysis: Total Private Construction Spending — Nonresidential (PNRESCONS)
1. Dataset Overview and Context
This dataset, sourced from the Federal Reserve Bank of St. Louis's FRED database (via fredgraph.csv?id=PNRESCONS), tracks monthly total private nonresidential construction spending in the United States, expressed in millions of dollars (unadjusted or seasonally adjusted, per FRED convention). The data spans January 1993 through May 2026 — notably extending ~2.5 years into the future, which suggests either forecasted/projected values or that the "as-of" pull date is near the dataset's terminal point and FRED has placeholder rows; this is worth verifying before treating recent observations as actuals. As an official government-adjacent macroeconomic series (FRED is a well-vetted, authoritative source), it's highly reliable for correlation and time-series studies involving construction activity, capital investment, industrial production, and broader economic cycles.
2. Data Quality Observations
The dataset is exceptionally clean at the ingestion level: zero null cells across both columns, and both observationdate and PNRESCONS show 401 distinct values matching the 401 total rows — indicating no duplicate timestamps and a consistent one-observation-per-month structure with no gaps. The duplicate-row check is flagged as pending recomputation ("Phase D"), so that should be confirmed before final use, but there's no evidence of missingness issues. No column shows disproportionate nulls. The one caveat is the sample-rows placeholder being empty, meaning actual value inspection (e.g., verifying units, checking for revision flags, or confirming whether trailing months are estimates) isn't yet possible from this summary alone.
3. Key Column Distributions
The PNRESCONS column ranges from a minimum of 144,221 to a maximum of 807,746 (units likely $ millions), with a mean of 367,972 and median of 302,812 — the mean exceeding the median by a wide margin signals a right-skewed distribution, confirmed by the skewness coefficient of 1.11. This makes sense economically: nonresidential construction spending has trended strongly upward since the 1990s (with cyclical dips), so recent decades' higher nominal values pull the mean above the historical median. The interquartile range (Q1=242,566, Q3=458,428) spans roughly 216,000, and the standard deviation of 175,350 is large relative to the mean, reflecting substantial variance driven by long-term growth trends, inflation, and recession-era contractions (2001, 2008-09, 2020). The 19 flagged outliers likely correspond to the most recent high-spending months (post-2021 construction boom, possibly inflated further by CHIPS Act/infrastructure-driven investment) or crisis-trough months — either tail of a long, trending series can trigger outlier flags in standard statistical tests, so these shouldn't be treated as data errors but as legitimate cyclical/trend extremes.
4. Recommended Join Key
The observationdate column (monthly, 401 distinct values, zero nulls) is the clear and only viable join key. Since it's a clean, gapless monthly time series from 1993–2026, it pairs naturally with any other FRED series or macroeconomic dataset using the same monthly date convention. Analysts should align on month-start dates and consider resampling if joining against quarterly, weekly, or daily datasets (e.g., aggregating daily data to monthly averages/sums).
5. Suggested Complementary Datasets
Strong correlation candidates include: Total Private Construction Spending: Residential (PRRESCONS) for a direct residential/nonresidential comparison; Producer Price Index for construction inputs or PPI: Construction Materials to study cost-driven spending changes; 10-Year Treasury Yield or Federal Funds Rate to examine interest-rate sensitivity of commercial investment; Nonresidential Fixed Investment (private) from GDP components and Industrial Production Index for broader capital-expenditure cycles; Employment in Construction (NAICS 23) from BLS/CES; and ISM Manufacturing/Non-Manufacturing PMI or CPI/PPI inflation series to contextualize spending against price and demand cycles. Regional or state-level construction permits/starts data would also allow drill-down analysis beneath this national aggregate.
Columns
- observation_date (date)
- PNRESCONS (integer)