Price Index
A price index is a measure of the average level of prices of a specified set of goods and services at one time relative to a base period. It is used to summarise changes in prices across a broad range of items, to adjust nominal values for the effects of inflation, and to compare price levels across regions or countries. Price indices are central to the measurement of inflation, to the indexation of contracts and benefits, and to the analysis of real magnitudes in the economy.
Construction
Constructing a price index begins with the selection of a basket of goods and services that represents the purchases of the population of interest. The prices of the items in the basket are collected at regular intervals, and the index is computed by comparing the cost of the basket in the current period with its cost in the base period. Weights are assigned to each item according to its share in total expenditure, so that items on which more is spent have a larger influence on the index.
The choice of weights and of the basket determines the properties of the index. A Laspeyres index uses base-period quantities as weights, while a Paasche index uses current-period quantities. The Laspeyres index tends to overstate price increases, because it does not account for substitution away from items whose prices have risen, while the Paasche index tends to understate them. Chain indices, which link period-to-period changes, and superlative indices, which approximate a cost-of-living index more closely, address these biases.
Types
Consumer price indices measure the prices of goods and services purchased by households, and they are the most widely used measure of inflation. Producer price indices measure the prices received by producers for their output, and they provide an early indication of price pressures in the production chain. Wholesale price indices, export and import price indices, and construction price indices serve specialised purposes.
The GDP deflator is a price index for all the goods and services produced in an economy, and it is broader than the consumer price index. Specialised indices measure prices in particular sectors, such as housing or energy, and underlying or core indices exclude volatile items to reveal the trend in prices. Each type has a specific purpose, and the choice depends on the question being addressed.
Uses
Price indices are used to measure inflation, which is the rate of change of a price index. Central banks use consumer price indices to set and to monitor monetary policy, and governments use them to index benefits, tax thresholds, and pensions. Businesses use them to adjust contracts and to compare prices over time, and analysts use them to convert nominal values into real values, allowing comparisons across periods.
Price indices are also used for international comparisons. Purchasing power parities, which compare the prices of comparable baskets across countries, are used to convert incomes and expenditures into a common unit and to compare living standards. The construction and use of price indices therefore extends across macroeconomic policy, contract design, and international comparison.
Limitations
Price indices have limitations. They measure the average change in prices, but they do not capture the effects of changes in quality, the introduction of new products, or changes in the availability of goods. Quality adjustment, which attempts to account for changes in the characteristics of products, is difficult and introduces measurement error. Substitution bias, in which consumers switch away from items whose prices rise, is imperfectly captured by fixed-weight indices.
The coverage and the timeliness of price indices also vary. Household surveys and price collection may not capture the full range of purchases, and the indices may be revised as more information becomes available. These limitations mean that measured inflation is an approximation, and that the interpretation of price indices requires attention to the methods used and to the sources of bias.
Weighting and index number theory
The choice of weights and formula is governed by index number theory, which examines the properties of alternative indices. A good index should satisfy tests such as proportionality, in which doubling all prices doubles the index, and the reversal test, in which the index computed in one direction is the reciprocal of that computed in the other. No single formula satisfies all desirable tests, and the choice involves trade-offs.
Superlative indices, including the Fisher and Tornqvist indices, approximate a cost-of-living index more closely than fixed-weight indices, because they account for substitution by using both base and current quantities. They are used in some statistical systems, particularly for the deflation of national accounts, while consumer price indices often use a modified Laspeyres formula for timeliness and ease of communication. The choice affects the measured rate of inflation and the interpretation of price developments.
Quality adjustment and new products
Price indices must account for changes in the quality of goods and for the introduction of new products, since a higher price for a better product is not the same as inflation. Statistical agencies use methods such as hedonic regression, which relates prices to the characteristics of products, to adjust for quality change, and they introduce new products into the basket as they become significant.
These adjustments are difficult and imperfect. The measurement of quality for complex products, including services and information technology, is particularly challenging, and the resulting estimates are subject to error. The treatment of new products, which often fall in price after introduction, and the treatment of goods that disappear, raise further issues. The accuracy of price indices depends substantially on the quality of these adjustments, and improvements in methods are a continuing priority.
Regional and international comparisons
Price indices can be constructed for regions, for social groups, and for countries, and the appropriate index depends on the question. Regional price indices reveal differences in the cost of living across areas, and group-specific indices reveal differences in inflation faced by different households, which arise from differences in consumption patterns. These differences matter for the assessment of changes in living standards and for the indexation of benefits.
International comparisons use purchasing power parities, which compare the prices of comparable baskets across countries. The construction of these parities requires extensive price collection and careful matching of products, and the results are used to compare incomes and expenditures in a common unit. The choices made in constructing the parities affect the comparisons, and they are a subject of ongoing methodological work by international organisations.
Uses in policy
Price indices are used across policy. Central banks use consumer price indices to define and to pursue their inflation targets, and the design of the index affects the measurement of the target and the assessment of performance. Governments use indices to index benefits, tax thresholds, and pensions, and the choice of index determines the protection of real incomes. Contracts often include indexation clauses, tying payments to a price index to preserve their real value.
The use of price indices for indexation raises issues about which index is appropriate and about the effects of indexation on inflation. Indexation can protect real incomes and reduce the costs of inflation, but it can also entrench inflation by making it automatic, and it can reduce the incentive to adjust relative prices. The design of indexation arrangements therefore involves a trade-off between protection and flexibility, and it is a consideration in the management of inflation.