How Inflation Is Measured: CPI, PPI and the GDP Deflator

When a news report says inflation was 3 percent last year, it is summarising the change in one specific price index over twelve months. Other indexes would give different answers, because each one tracks a different set of prices with different weights. Understanding how inflation is measured explains why the headline figure moves, why economists watch several gauges, and why a small technical choice can change a policy debate.

Inflation as a percentage change in an index

Inflation is the sustained rise in the general price level. It is measured by choosing a price index, which is a single number summarising the prices of a large set of goods and services, and then tracking how that number changes. The inflation rate over a period is the percentage change in the index: pi = (index_t - index_t-1) / index_t-1 x 100 percent.

Suppose an index stands at 300 in one month and 306 the next. The change is 6, and 6 divided by 300 is 0.02, so the monthly inflation rate is 2 percent. Price levels are never observed directly for the whole economy, so the index is the working tool for every inflation figure.

What a price index is

A price index starts with a basket, a fixed or periodically updated list of goods and services with a weight assigned to each. The weight reflects how much of total spending the item represents. The index is set to 100 in a base period. If the basket costs 100 in the base period and costs 104 later, the index reads 104, meaning prices have risen 4 percent since the base.

The choice of basket, weights and base period all affect the result. A basket heavy in food will respond strongly to food prices, while one heavy in services will not. This is the main reason different indexes diverge.

CPI: the consumer basket

The Consumer Price Index measures prices paid by households for a basket of goods and services that urban consumers buy. It is the most cited inflation measure and is used to adjust many contracts, pensions and tax brackets. Its basket is built from household spending surveys, and the weights are updated periodically to reflect changing habits.

CPI includes imported goods, so exchange-rate changes can affect it. Many statistical agencies also publish a core measure that excludes food and energy. Core inflation is often used to see underlying trends, because food and fuel prices can swing sharply for reasons that do not last.

PPI: prices received by producers

The Producer Price Index measures the prices that domestic producers receive for their output. It covers goods, services and construction, depending on the country and the release. Because producers often face cost changes before they reach consumers, PPI can sometimes give an early signal of consumer price movements.

PPI does not map neatly onto consumer prices. Many goods pass through several stages, and margins between stages change with competition. A rise in producer prices does not always show up at the checkout, and a fall may not either.

The GDP deflator

The GDP deflator is the ratio of nominal GDP to real GDP, multiplied by 100. Unlike CPI, it covers every good and service produced domestically, including investment goods, government purchases and exports. It does not include imports directly, because they are not part of domestic output.

Its weights change each year with the mix of output, not with a fixed household basket. This makes the deflator the broadest measure of domestic price change, but it is less familiar to the public and is reported less often than CPI.

PCE: the measure the central bank watches

The Personal Consumption Expenditures price index measures prices for the goods and services that households consume, using spending data from national accounts. It is broader than CPI in some respects, because it covers spending paid for by others, such as employer-provided health care. Its weights also adjust more quickly when households change their buying patterns.

In the United States, the Federal Reserve sets a longer-run inflation goal of 2 percent measured by PCE. Because of that target, PCE figures get close attention even though CPI is more widely quoted in the press.

Comparing the four measures

Measure	Whose prices	Basket or weights	Main use
CPI	Household purchases	Consumer spending survey, updated periodically	Public reporting, indexing contracts
PPI	Domestic producer output	Output by industry and stage	Early signal of cost pressure
GDP deflator	All domestic output	Mix of domestic output, updated each year	Economy-wide price changes
PCE	Household consumption	National accounts spending, adjusts faster	Central bank target measure

Each measure answers a slightly different question. CPI tells you what consumers face. The deflator tells you what happened to the prices of everything the economy produced. PCE is the closest to the Fed's view of consumer prices, and PPI is closest to the cost side of business.

Why measurement details matter

Some technical choices are not neutral. A fixed-weight index can overstate price growth over time, because it does not capture shoppers switching to cheaper alternatives when a good becomes more expensive. Chained indexes, which update weights frequently, reduce this substitution bias. Statistical agencies have made such changes over the years, and each change alters the headline series slightly.

Quality adjustment is another source of disagreement. When a product improves, statisticians must decide how much of any price rise reflects better quality rather than inflation. Getting this wrong for electronics or medical devices can change measured inflation noticeably.

Substitution and quality matter for policy too. Pensions and tax brackets indexed to CPI depend on those technical choices, so a method change can shift real incomes even when no prices have moved differently.

Types of inflation behind the number

The index tells you how much prices rose. Economists also ask why. Demand-pull inflation happens when spending grows faster than the economy can produce goods, so prices are bid up. Cost-push inflation comes from shocks to supply, such as an energy price spike or a poor harvest, which raise costs for producers. Built-in inflation, sometimes called a wage-price spiral, occurs when workers and firms expect continued inflation and set wages and prices accordingly, making the expectation self-fulfilling.

The distinction matters because the policy response differs. Central banks can restrain demand directly, but they cannot easily offset a supply shock without raising unemployment. Expectations are central to the built-in case, which is why central banks pay attention to surveys of what households and firms expect.

Key takeaways

Inflation is the percentage change in a price index over time, and the rate depends on which index is used. CPI tracks consumer prices, PPI tracks producer prices, the GDP deflator covers all domestic output, and PCE is the Federal Reserve's preferred gauge. Technical choices about weights, substitution and quality can change the figures. Reading the headline number well means knowing which index produced it and what that index is designed to capture.