Personal Consumption Expenditures Price Index

The Personal Consumption Expenditures Price Index is a monthly measure of the change in prices paid for consumer goods and services in the United States, produced by the Bureau of Economic Analysis as part of the National Income and Product Accounts. BEA describes it as "a measure of the prices that people living in the United States, or those buying on their behalf, pay for goods and services," and notes that it is "known for capturing inflation (or deflation) across a wide range of consumer expenses and reflecting changes in consumer behavior." It is released each month with the Personal Income and Outlays report, and quarterly and annual figures appear with the gross domestic product release.

The name needs explaining, because two different things share most of it. BEA titles the price measure the Personal Consumption Expenditures Price Index. Bare "personal consumption expenditures," which is what BEA's glossary defines, is a different statistic entirely: "the goods and services purchased by persons," measured in dollars of spending rather than as a price level. A single BEA release routinely reports both. The release covering July 2026, for instance, said in one paragraph that "personal consumption expenditures (PCE) increased $36.3 billion (0.2 percent)" and then, separately, gave a percentage change for "the PCE price index." One is how much was spent; the other is what things cost. A reader who treats the short form as interchangeable will eventually read a spending figure as an inflation rate. The full name, or "PCE price index," avoids it.

This entry covers how the index is built and why it gets revised. What the Federal Reserve does with it, including the 2 percent goal and why comparing a Consumer Price Index headline against that goal compares the wrong series, belongs to the entry on the Federal Reserve.

The differences from the Consumer Price Index are not a list of oddities; they are four identified mechanisms, and BEA reconciles them arithmetically. BEA publishes a table comparing the two indexes and groups the gap into a formula effect, a weight effect, a scope effect and "other effects." Each is a specific thing rather than a vague methodological difference.

The formula effect is the mathematics. In BEA's words, "the PCE price index is based on the Fisher-Ideal formula, while the CPI is based on a modified Laspeyres formula." The practical consequence is the one the entry on the Federal Reserve already notes, that the PCE price index's weights update as spending patterns shift rather than holding a basket fixed between scheduled reweightings.

The weight effect is the relative importance of the same items in the two baskets, and BEA's own worked comparison shows where it concentrates: in the quarter BEA uses as its example, the weight effect was accounted for by imputed rental of owner-occupied nonfarm housing and by gasoline and other motor fuel. Housing shows up there because BEA imputes a value for the services an owner-occupied home provides, which it identifies as the largest imputation in the GDP accounts.

The scope effect is the one most people have never heard and the one that changes how a reader should think about the number. BEA states it directly: "PCE measures spending by and on behalf of the personal sector, which includes both households and nonprofit institutions serving households; the CPI measures out-of-pocket spending by households." So medical care paid for by an employer's health plan or by Medicare is inside the PCE price index and largely outside the Consumer Price Index, and in that same worked comparison BEA identifies hospital and nursing home services as the largest contributor to the scope effect. The index the Federal Reserve's goal is written on therefore includes prices no household ever pays at a register, and that is deliberate rather than a flaw: it is measuring the price of consumption, not the size of a receipt.

Other effects covers seasonal adjustment differences, price differences and a residual.

The index is revised, and the two agencies run opposite policies on that. This is the most practically useful thing to know about the PCE price index. BEA's monthly release restates recent months as better source data arrive, notes in each release which months were updated, and warns that "with the next release, today's data will be superseded." On top of that comes an annual update of the National Economic Accounts, which explicitly includes monthly personal income and outlays and reaches further back. The Bureau of Labor Statistics runs the opposite policy for the Consumer Price Index: because the index is written into escalation agreements, pensions and tax brackets, "revisions can be costly for the users of these indexes, so these series are final when issued." BLS corrects an outright error, and it re-estimates the seasonal adjustment factors annually, which means published seasonally adjusted figures can move for up to five years. But the unadjusted index levels the law points at do not change.

The consequence is worth stating plainly. A PCE inflation figure quoted in a news story, an investment commentary or a financial plan is a snapshot of an estimate, and the same month can be reported differently a year later without anyone having made a mistake. A Consumer Price Index level is not like that. Anyone comparing a remembered figure against a current publication should check which index they are holding.

The production chain runs the opposite way from the Consumer Price Index, and that is the root of most of the differences. BLS builds the CPI upward from prices its own collectors gather from households and retail outlets. BEA builds the PCE price index inside the national accounts, drawing largely on business-side source data such as retail sales, service receipts and industry surveys, and folding in BLS price indexes as inputs. Because the accounts have to add up to total consumer spending, the PCE price index inherits the accounts' definition of what consumer spending is, which is where the scope difference comes from.

A hypothetical example of the scope effect, using round numbers to isolate the mechanism rather than to represent any real weight. Suppose a household's medical care over a year costs $10,000 in total. The household pays $2,000 of it out of pocket, and an employer plan and Medicare pay the other $8,000 directly to providers. Over that year, suppose out-of-pocket prices rise 3 percent while the prices of the care paid on the household's behalf rise 7 percent.

• An index built only on out-of-pocket spending records medical inflation of 3 percent.
• An index built on the whole $10,000 records 0.20 multiplied by 3 percent, which is 0.6, plus 0.80 multiplied by 7 percent, which is 5.6, for a total of 6.2 percent.

Same care, same year, same underlying prices, two different medical-inflation figures more than three percentage points apart. Neither number is an error. The difference is entirely a question of what each index was built to measure, and BEA's reconciliation table exists precisely so that the gap can be attributed rather than argued about.

Where to find the numbers: the monthly release and the archive of previously published estimates are on BEA.gov, which is also where the annual update schedule is posted. This entry states no reading, because they are revised monthly and revised again annually.

What the index does well

• Covers consumption comprehensively, including the large share of medical care that someone other than the household pays for, so it measures the price of what is consumed rather than the price of what is paid at a register.
• Its Fisher-Ideal formula lets the basket shift as spending shifts, without waiting for a scheduled reweighting.
• Comes with a published reconciliation against the Consumer Price Index, so a gap between the two can be decomposed into named effects.
• Is the series the Federal Reserve's inflation goal is written on, which makes it the relevant number for anticipating policy.

Its limits

• It is revised, both in each subsequent monthly release and in the annual update, so any single reading is provisional.
• Its comprehensiveness cuts both ways. A household cannot read its own cost of living off an index that includes prices it never sees.
• It relies more heavily on business-side and administrative source data, which is a large part of why it needs revising.
• Because "PCE" also abbreviates the dollar spending total, the short form invites confusing a spending figure with an inflation rate.