Methodology
Data sources, geographic scope, and caveats
Geographic scope
This analysis covers four small towns in the West Texas Edwards Plateau and Permian Basin fringe region. County-level data is the finest granularity available from federal statistical agencies for these locations.
| Town | County | FIPS | Pop. (est.) | Primary economy |
|---|---|---|---|---|
| Sonora | Sutton | 48435 | ~3,000 | Ranching, oil/gas services, IH-10 corridor |
| Eldorado | Schleicher | 48413 | ~2,800 | Ranching, oil/gas, agriculture |
| Ozona | Crockett | 48105 | ~3,400 | Ranching, oil/gas, IH-10 corridor |
| Junction | Kimble | 48267 | ~4,400 | Ranching, tourism (Llano River), agriculture |
State-level data (Texas) and national data (United States) are included as benchmarks for comparison.
Data sources
| Source | Dataset | What | Frequency | Lag |
|---|---|---|---|---|
| BLS LAUS | Local Area Unemployment Statistics | County unemployment rate, labor force, employment | Monthly | ~1 month |
| BEA Regional | CAINC1 | County per-capita personal income | Annual | ~6–9 months |
| BEA Regional | CAGDP1 | County GDP (total market value of production) | Annual | ~6–9 months |
| FRED | UNRATE | US civilian unemployment rate | Monthly | ~1 week |
| FRED | TXUR | Texas unemployment rate | Monthly | ~1 month |
| FRED | TXRGSP | Texas real gross state product | Annual | ~6 months |
All data is public domain (US government work). Attribution is provided as courtesy.
Pipeline
fetch_fred.R → data/economy/cache/unemployment.csv, tx_unemployment.csv, tx_rgsp.csv
fetch_bls.R → data/west-texas/cache/bls_laus_{county}_ur.csv
fetch_bea.R → data/west-texas/cache/bea_income.csv, bea_gdp.csv
build_west_texas.R → data/west-texas/unemployment_monthly.csv
data/west-texas/income_annual.csv
data/west-texas/gdp_annual.csv
data/west-texas/west_texas_summary.csvCaveats and limitations
Small-county volatility. Counties with populations under 5,000 exhibit high variability in economic statistics. A single business opening or closing can swing the unemployment rate by several percentage points. The four-county regional average smooths some of this volatility but does not eliminate it.
BEA data suppression. The BEA may suppress county-level GDP or income data for confidentiality when too few establishments contribute to a given industry. Missing values are shown as gaps rather than interpolated.
Lagged data. BEA county income and GDP are published with a 6–9 month lag. BLS LAUS county data is published with a ~1 month lag. FRED state-level series are typically available within 1–2 months.
Energy-economy link. These counties are in or adjacent to the Permian Basin, the largest US oil-producing region. Local economic conditions are strongly influenced by oil and gas prices. See the Energy section for WTI crude and natural gas price trends.
Not causal. The comparisons presented are descriptive. Differences between county, state, and national data reflect many factors beyond local policy — including industry composition, demographic shifts, and commodity price cycles.
License
Code: MIT. Data: public domain (US government sources).