When you check a grocery receipt and wonder why your total climbed higher than last month, the U.S. government is tracking those exact price shifts on a massive scale. Each month, the Bureau of Labor Statistics calculates the Consumer Price Index by gathering roughly 100,000 retail and service prices alongside 8,000 rental housing quotes across 32 urban geographic areas. Economists sort those prices into 7,776 specific item-and-city combinations, compute the price movement inside each bucket, and weight them using real household budgets to measure US consumer prices across the economy.
To find the inflation rate between any two months or years, you calculate the percentage change between their published index levels rather than subtracting the raw numbers. Understanding the mechanics behind that monthly report helps you read inflation headlines, lease escalations, and tax bracket updates with confidence, though you should speak with a licensed financial advisor before basing personal investment decisions on macroeconomic data.
How does the Bureau of Labor Statistics collect prices each month?
The index tracks out-of-pocket spending by urban households, which means every recorded price includes applicable sales and excise taxes. Instead of checking prices on a single day, field staff gather quotes across three pricing periods spread throughout the entire calendar month.
Price collection runs through two separate BLS pricing surveys that divide daily purchases from shelter costs. In the Commodities and Services survey, data collectors record about 100,000 price quotes every month. Roughly two-thirds of those quotes come from in-person visits to brick-and-mortar stores, while the remaining third are gathered by telephone, mobile apps, or outlet websites. Online retail websites account for about 8% of the monthly price sample, chosen based on where households actually report shopping.
For shelter, the agency runs a separate Housing survey that collects about 8,000 rental housing unit quotes every month. Those rental units are split into six rotating panels, with each apartment or house priced twice a year. The agency uses those rent quotes to measure both tenant rent and the implicit shelter cost of owning a home.
How do Consumer Price Index calculation weights and formulas work?
Once the raw price quotes arrive, the agency runs a two-stage index calculation that turns individual store tags into a single national number. The agency divides the urban United States into 32 geographic index areas and splits household purchases into 211 item strata—209 for commodities and services plus 2 for housing. Because health insurance retained earnings are broken down further during processing, those 211 strata expand into 243 basic items for aggregation.
Multiplying 32 geographic areas by 243 basic items produces 7,776 distinct item-area cells, such as apples in Chicago or electricity in the Philadelphia area. In the first stage of the math, the agency calculates a basic price index for each of the 7,776 cells. Most of these basic indexes use a geometric mean formula, which accounts for shoppers substituting one item for another within the same category, like buying Gala apples when Honeycrisp prices spike.
In the second stage, the agency combines those 7,776 basic indexes into regional and national totals using spending weights from the Consumer Expenditure Survey, which are updated annually. For the standard headline indexes, this upper-level step uses a Laspeyres formula that holds spending quantities fixed across categories. In the agency’s relative importance tables, you can see how an individual category drives the overall result: in October 2017, comparing the change in the energy component’s weight to the change in the All Items weight produced a ratio of 0.223, showing that energy accounted for 22.3% of the entire monthly index movement.
How do CPI-U, CPI-W, and Chained CPI differ?
The Bureau of Labor Statistics publishes three main national indexes for different audiences, each using the same underlying store prices but applying different population weights or aggregation math.
- The Consumer Price Index for All Urban Consumers (CPI-U) is the headline figure reported in financial news and covers about 93% of the U.S. population. It excludes people living in rural nonmetropolitan areas, farm households, military installations, and institutions such as prisons or nursing facilities.
- The Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W) covers roughly 30% of the U.S. population. To qualify for this subset, more than half of a household’s income must come from clerical or wage occupations, and at least one earner must have worked at least 37 weeks during the previous 12 months.
- The Chained Consumer Price Index (C-CPI-U) covers the same 93% urban population as the CPI-U, uses a December 1999 = 100 base period, and has been used since 2017 to adjust federal income tax brackets. Instead of holding category weights fixed in stage two, the C-CPI-U uses a superlative Törnqvist formula that blends expenditure data from adjacent periods to capture shoppers switching across categories, such as buying pork when beef gets expensive. Because complete spending data take time to collect, the C-CPI-U is released as a preliminary number and revised three times on a quarterly schedule over 10 to 12 months. From 2001 to 2023, the average December-to-December gap between the CPI-U and the C-CPI-U was about 0.2 percentage points per year.
How do you calculate the inflation rate from two index numbers?
Most CPI-U and CPI-W series are normalized to a reference base period of 1982–84 = 100, meaning the average prices across those 36 months equal 100. An index reading of 107.000 means prices are 7% higher than they were in 1982–84, but when you compare two later months, subtracting the numbers only gives you index points rather than the percentage inflation rate.
- Pick the matching series for your timeframe: use Seasonally Adjusted (SA) figures when comparing one month to the prior month, and Not Seasonally Adjusted (NSA) figures when calculating 12-month annual inflation or adjusting a contract.
- Subtract the earlier period’s index level from the later period’s index level. If an index moves from 300.000 to 309.000, the difference is 9.000 index points.
- Divide that point difference by the earlier period’s index level: 9.000 divided by 300.000 equals 0.03.
- Multiply the result by 100 to convert the decimal into a percentage rate, which equals 3.0% inflation over that period.
What do people get wrong about the Consumer Price Index?
One of the most frequent mistakes is comparing two regional index numbers to see which city has a higher cost of living. Under the official CPI cost-of-living concepts, an index of 200 in one city and 150 in another does not mean the first city is more expensive in dollar terms. It only tells you that prices in the first city have doubled since its 1982–84 base period, while prices in the second city have risen by 50% over the same span.
Many readers also assume that rising mortgage rates and home sale prices directly push the CPI higher. Because buying a house carries an investment component, the BLS strips out home purchase prices and mortgage interest entirely. Instead, it measures a homeowner’s shelter cost through Owners’ Equivalent Rent—an estimate of what homeowners would pay if they rented their own homes, derived from the monthly survey of 8,000 rental units.
Finally, people often confuse the CPI with the Personal Consumption Expenditures (PCE) price index published by the Bureau of Economic Analysis. Because the CPI only tracks what households pay out of their own pockets, while the PCE index also counts expenses paid on consumers’ behalf (such as employer-provided health insurance and government medical programs), their category weights look very different. According to a 2021 Brookings Institution comparison, housing makes up 42% of the CPI basket compared with 23% in the PCE index, whereas health care accounts for 9% of the CPI versus 22% of the PCE index.
Frequently asked questions
Does the Consumer Price Index include income taxes or stock investments?
No. The index excludes federal and state income taxes as well as financial investments like stocks, bonds, and life insurance policies. It only includes sales taxes, excise taxes, and government user fees that you pay directly when buying consumer goods and services.
Does the Bureau of Labor Statistics revise the CPI after it is published?
Not Seasonally Adjusted CPI-U and CPI-W index numbers are normally final as soon as they are published and are not revised for routine data lags. Only the Chained CPI (C-CPI-U) undergoes three scheduled quarterly revisions over 10 to 12 months as complete household expenditure data arrive.
Are subnational or city CPI numbers available for the Chained CPI?
No. While the BLS publishes regional and metropolitan indexes for urban areas under the CPI-U and CPI-W programs, the Chained CPI (C-CPI-U) is calculated and published only at the national level.