Total Construction Spending: Highway and Street in the United States (TLHWYCON) | FRED | St. Louis Fed
- Rows
- 293
- Columns
- 2
- Data range
- 2002-01-01 to 2026-05-01
Graph and download economic data for Total Construction Spending: Highway and Street in the United States (TLHWYCON) from Jan 2002 to May 2026 about expenditures, construction, and USA.
AI analysis
Dataset Analysis: Total Construction Spending — Highway and Street (TLHWYCON)
1. What this dataset represents
This series, sourced from the St. Louis Fed's FRED database (via fredgraph.csv?id=TLHWYCON), tracks monthly total construction spending on highway and street projects in the United States, denominated in millions of dollars (unadjusted or seasonally adjusted, per Census Bureau construction spending surveys that underlie FRED's data). The data spans January 2002 to May 2026 — notably extending into the future, which suggests this snapshot includes forecast, provisional, or placeholder values near the edge of the series, a common trait of FRED's rolling release windows. As a publicly maintained, high-frequency, long-horizon macroeconomic indicator, it holds strong value for correlation studies involving infrastructure investment cycles, fiscal policy impact (e.g., stimulus bills, infrastructure acts), and broader construction/economic activity.
2. Data quality observations
Data quality appears excellent at face value: zero null cells across both columns and zero nulls reported individually for observationdate and TLHWYCON. The date column has 293 distinct values matching 293 total rows, confirming no duplicate timestamps — a good sign for time-series integrity, though the formal duplicate-row recomputation is still pending ("Phase D"). One caveat: TLHWYCON shows only 287 distinct values against 293 rows, meaning a handful of repeated spending figures across different months — plausible for a slow-moving construction series but worth verifying isn't a data-carry-forward artifact (e.g., placeholder/estimated values for the most recent, not-yet-finalized months given the dataset extends to 2026).
3. Key columns and distributions
The TLHWYCON column ranges from 2,716 to 16,893 (units likely millions USD), with a mean of 7,608.98 and median of 7,352 — the mean sitting modestly above the median, consistent with the reported positive skew of 0.75, indicating a right-tailed distribution with a subset of higher-spending months pulling the average up. The interquartile range (Q1=5,108, Q3=9,547) spans roughly 4,400 units, suggesting substantial month-to-month variability typical of a series that has grown considerably over two decades (raw, non-inflation-adjusted dollar figures compounding with inflation and real investment growth). Only 2 outliers were flagged, suggesting the extremes (likely the historical low near 2002 and peak near recent years) are not numerous but do represent genuine structural shifts rather than noise. The standard deviation of 3,108.92 relative to the mean (a coefficient of variation ~41%) confirms this is a highly dynamic series over its 24-year window.
4. Recommended join key
observationdate is the clear and only viable join key — it's a clean, fully-populated, monthly-frequency date field with no duplicates. For cross-dataset correlation, it should be joined against other monthly (or resampled quarterly/annual) time series on their corresponding date fields, with attention to aligning month-start conventions and any seasonal adjustment differences.
5. Suggested pairings for correlation discovery
Strong candidates include: (a) other FRED construction spending subcomponents (e.g., TTLCONS total construction, residential/nonresidential spending) to decompose sector contributions; (b) macro indicators like the Producer Price Index for asphalt/concrete inputs, steel prices, or the Highway Trust Fund revenue/federal transportation appropriations to study cost-driver relationships; (c) employment data (e.g., construction sector payrolls, BLS CES series) to examine labor-spending correlation; (d) GDP and fixed investment series to contextualize highway spending against broader economic cycles; and (e) policy-event datasets (e.g., Infrastructure Investment and Jobs Act timelines) to test for structural breaks or lagged spending responses following major legislation.
Columns
- observation_date (date)
- TLHWYCON (integer)