Consumer Price Index for All Urban Consumers

The **Consumer Price Index for All Urban Consumers (CPI-U)** is a monthly measure produced by the U.S. Bureau of Labor Statistics (BLS) of the average change over time in prices paid by urban consumers for a representative basket of goods and services. It is the broadest of the BLS’s principal consumer price indexes, covering more than 90 percent of the U.S. population. The CPI-U is widely used as an indicator of consumer inflation and as a basis for adjusting selected payments and economic measures. ^[1][2]

Scope and population

The CPI-U is designed to represent the spending and price experience of the CPI-U population, rather than that of every U.S. resident or a particular household. Its population includes urban households in core-based statistical areas and urban places with 10,000 or more inhabitants. It spans households with varied employment and income situations, including wage earners, salaried workers, self-employed people, unemployed people, retirees, and people outside the labor force. ^[2]

Some groups are outside the target population: people living in rural nonmetropolitan areas, farm households, people on military installations, people in religious communities, and people in institutions such as prisons and mental hospitals. As a result, the CPI-U does not directly measure price changes for those populations. Its broad coverage also does not mean that it precisely describes any one included household’s inflation experience. Household spending patterns differ, and therefore the effect of a price change on one household may be higher or lower than its effect on the CPI-U. ^[2][3]

The CPI-U is distinct from the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W). The CPI-W covers a narrower population: households meeting specified employment and earnings criteria in eligible occupations. The CPI-W population is a subset of the CPI-U population and represents approximately 30 percent of the total U.S. population. The chained CPI for all urban consumers (C-CPI-U) uses the CPI-U population but is a separate index with a formula designed to account for consumer substitution among item categories as relative prices change. ^[2][4]

Basket and included costs

The index covers goods and services purchased for consumption by its reference population. BLS organizes expenditure items into more than 200 categories within eight major groups: food and beverages; housing; apparel; transportation; medical care; recreation; education and communication; and other goods and services. The specific goods, services, and expenditure weights are intended to represent the consumption of the CPI-U population, not a fixed list that never changes. ^[2]

The CPI-U includes certain charges directly associated with purchases, such as sales and excise taxes, and some government user fees, including water and sewer charges, vehicle registration fees, and tolls. It excludes taxes not directly tied to buying consumer goods and services, such as income and Social Security taxes. It also excludes investment items—including stocks, bonds, real estate, and life insurance—because the index is intended to measure consumption prices rather than the value of savings or investments. ^[1][2]

Housing is a major part of the basket. For owner-occupied homes, the CPI measures the housing service provided to occupants using owners’ equivalent rent (OER)—an estimate of the rent homeowners would pay to rent their homes, without furnishings or utilities. The home itself is treated as a capital or investment good rather than a consumption item. Consequently, the CPI does not include the purchase price of an owner-occupied home as a direct price of shelter, nor does it include mortgage interest, property taxes, real-estate fees, or most maintenance and improvement costs as owner-occupied housing consumption. Rent of primary residence and OER are the principal measures of shelter costs for renters and homeowners, respectively. ^[5]

Price collection and calculation

BLS collects prices for the goods and services used in the CPI in sampled urban areas and retail and service establishments; it collects housing-rent information through a separate survey of rental units. The detailed data are combined into indexes for individual items and locations, then aggregated into broader categories and geographic indexes. The expenditure weights used in the aggregation come from the Consumer Expenditure Surveys, which estimate how the index population distributes spending among categories. ^[1][4]

The CPI-U is an index number, not a dollar amount or a direct measure of the cost of every household’s personal basket. For most CPI series, the reference base is 1982–84 = 100: the average index level across 1982, 1983, and 1984 is set to 100, and later index values express price levels relative to that base. Changing the reference base rescales index levels but does not, by itself, change percentage movements between periods. ^[2]

BLS updates price observations monthly and updates spending weights annually, using Consumer Expenditure Survey data. The weights determine how much each category contributes to the aggregate index; a category with a larger expenditure weight generally has greater influence on the all-items index than one with a smaller weight, all else equal. Weight updates help the basket reflect changes in consumer spending over time, although expenditure data and the resulting weights relate to a reference period rather than continuously tracking every household’s current purchases. ^[6][7]

Published measures and interpretation

BLS publishes CPI-U indexes for the United States and for various geographic areas, along with indexes for expenditure groups, individual items, and selected special aggregates. The national all-items index is often called the “headline” CPI. The “all items less food and energy” index is another commonly reported aggregate; it excludes those two categories from its scope and is not the same as the all-items CPI-U. BLS also publishes both seasonally adjusted and not seasonally adjusted measures where applicable. ^[1][8]

A percentage change in the CPI-U over a stated interval is commonly used to express inflation over that interval. The comparison period matters: a month-to-month change and a 12-month change answer different questions. Seasonally adjusted data remove estimated recurring seasonal effects to make underlying movements easier to analyze. They are generally used for short-term analysis of price trends. Unadjusted data reflect observed prices without that seasonal adjustment and are commonly used in payment escalation arrangements. ^[8]

Seasonally adjusted CPI series may be revised when seasonal factors are recalculated. BLS recalculates factors annually with the release of January data, and the process can revise seasonally adjusted indexes for the previous five years. Older seasonally adjusted indexes are considered final under the stated process, while unadjusted indexes are not subject to this annual seasonal-factor revision. Data users should therefore distinguish the adjustment status of a series and consider the vintage of seasonally adjusted data when comparing historical figures. ^[9]

Uses

The CPI-U is used as an economic indicator and as a deflator for converting some economic series into inflation-adjusted terms. It is also used in a range of adjustment mechanisms. BLS identifies uses that include adjusting income eligibility levels for government assistance, federal tax brackets, federally mandated cost-of-living increases, private-sector wages and salaries, poverty measures, and consumer and commercial rent escalations. Specific programs or contracts may refer to a particular CPI population, geographic area, item index, and adjustment convention, so “CPI-linked” does not necessarily mean that every arrangement uses the same series or calculation. ^[1][4]

CPI-U data are also used by researchers, businesses, labor organizations, government agencies, and households to describe changes in consumer prices. The index does not itself identify the cause of a price movement, predict future inflation, or measure changes in every component of household welfare. It is a statistical summary for a defined population and basket, constructed under published scope and methods. ^[1][3]

Limitations and distinctions

The CPI-U is an average across a broad population, so it may not match an individual’s experience. A household that spends a larger share of its budget on a category rising faster than the overall index may face a different rate of price change from the CPI-U; a household with a different location, housing situation, or consumption mix can also experience different movements. BLS makes research indexes for some population subgroups, but identifies methodological caveats for those series, including that shared sampled areas, stores, rental units, items, and prices may not capture all cohort-specific differences. ^[3]

The CPI-U should also be distinguished from other price measures, including the CPI-W, C-CPI-U, and the personal consumption expenditures (PCE) price index. These measures differ in population coverage, formula, data sources, or expenditure scope. They may therefore report different rates of price change over the same period without being direct contradictions: each reflects its own statistical design and measurement target. ^[2][4]

History and data access

BLS began publishing separate consumer price indexes for 32 cities in 1919, following earlier work on family expenditures and price collection. Regular publication of a national U.S. city average began in 1921, with estimates extending back to 1913. The CPI has been revised over time to update samples and spending weights, expand coverage, and improve methods. ^[10]

BLS releases CPI data monthly through its news releases, tables, and online databases. Series are available at different geographic and item levels, though the amount of detail varies by location. Users comparing observations should identify the population (CPI-U or another index), item category, area, period, reference base, and seasonal-adjustment status to ensure that the selected figures answer the intended question. ^[1][8]