Methodology

Every dataset on InflationVault traces back to an official, publicly available source. We do not use estimates, projections, or proprietary models. The primary sources include:

• Bureau of Labor Statistics (BLS) — CPI data, average consumer prices, employment statistics
• U.S. Energy Information Administration (EIA) — Retail fuel prices, energy consumption data
• U.S. Census Bureau — Housing prices, construction data, demographic information
• Bureau of Economic Analysis (BEA) — National accounts data, personal consumption expenditures
• Department of Labor — Minimum wage rates and labor standards data

Each dataset page includes a direct link to its primary source. We encourage verification — the whole point of this site is transparency.

We use the CPI-U (Consumer Price Index for All Urban Consumers) to adjust nominal prices for inflation. The CPI-U is published monthly by the Bureau of Labor Statistics and covers approximately 93% of the U.S. population.

The formula is straightforward:

The CPI uses a base period of 1982–84 = 100. So if the CPI was 82.4 in 1980 and 313.0 in 2024, a dollar in 1980 is equivalent to about $3.80 in 2024 (313.0 / 82.4 = 3.798).

We use annual average CPI values rather than monthly values for consistency, since most of our price data is reported annually.

Most datasets are updated annually when the relevant government agency publishes its year-end figures, typically in the first quarter of the following year. CPI data is updated as soon as the BLS publishes the annual average.

Some datasets (like gasoline prices) have more frequent source data available, but we standardize to annual averages for consistency and comparability across datasets.

Historical data is not always available for every year, especially before the 1970s when modern statistical collection methods were established. When data is missing for specific years:

• We include only years for which we have reliable source data.
• We do not interpolate or fill gaps with estimates unless explicitly noted.
• Charts connect available data points, which may span multiple years.

If a dataset has significant gaps, it is noted in the methodology section on that dataset's page.

Honest data presentation means being upfront about what the numbers can and cannot tell you:

• CPI is not perfect. It measures a fixed basket of goods and does not fully account for quality improvements, product substitution, or changes in consumer behavior.
• National averages mask variation. Gas in California costs significantly more than gas in Texas. A median home price means half of homes cost more and half cost less. Regional and demographic differences are real.
• Older data is less precise. Government price collection before the 1970s was less systematic than it is today. Figures from the 1950s and 1960s are estimated from historical reports and carry wider margins of error (±10–15%). Treat pre-1970 values as informed approximations, not exact measurements. Post-1980 data aligns closely with official BLS, EIA, and Census publications.
• Nominal vs. real comparisons require context. A price that looks flat in nominal terms may actually represent a decrease after accounting for inflation, and vice versa.
• Prices are not the whole story. A new car costs more than it did in 1970, but it also has airbags, GPS, fuel injection, and antilock brakes. Whether the price increase is “justified” depends on how you value those improvements.