What’s new in the HICP? The 2026 classification update and its implications for inflation analysis

Prepared by Martin Eiglsperger, Katalin Bodnár, Rania Bouhaouita Haddad and Elisabeth Wieland

Published as part of the ECB Economic Bulletin, Issue 3/2026.

The January 2026 release of the HICP introduced a major change in how consumer products are classified. The HICP, which the European Central Bank (ECB) uses as its principal measure of inflation in the euro area, serves as a yardstick for assessing price stability, the ECB’s primary monetary policy objective. A key feature of the HICP is the availability of detailed price indices by product – with products classified according to their consumption purpose – and predefined aggregate indices, for example, for food (unprocessed and processed), energy, non-energy industrial goods and services. The HICP classification is based on the Classification of Individual Consumption according to Purpose (COICOP), which was introduced in 1999 and later refined at the European level as European COICOP (ECOICOP).^[1] A comprehensive update has been made to the HICP as of January 2026, aligning the European classification with the international COICOP 2018 standard to establish ECOICOP version 2. In addition, the index reference year for the HICP has changed from 2015 = 100 to 2025 = 100. The euro area aggregate has also been extended to include Bulgaria, following its entry into the euro area on 1 January 2026.^[2]

One major objective of reclassifying HICP items was to achieve a clearer separation between goods and services. The HICP’s new classification allows for finer delineation of services, some of which had previously been included in goods. For example, whereas delivery fees were previously included in the total price of delivered goods, they are now classified separately as a service. Separately invoiced delivery fees for goods are allocated to “Delivery of goods”.^[3] There is also a “Courier and parcel delivery services” category for online purchases delivered by a standalone courier or parcel delivery company.^[4] Repair, maintenance, rental and installation services are now listed separately, and IT hardware and software have moved from “Recreation and culture” to “Information and communication”. Insurance and financial services have been separated from other personal services. The new HICP classification has also been further aligned with international standards, with games of chance (such as lotteries, betting and gaming) being included as a new item.^[5]

The new classification has not affected headline inflation or led to major changes in HICP main aggregates. While, in principle, recalculations using ECOICOP version 2 would have led to adjustments in total inflation rates, European statisticians left headline HICP inflation rates unchanged.^[6] Some minor differences (up to 0.1 percentage points) materialised in the annual rates of change of the all-items HICP for the euro area. However, these differences only reflect the impact of rounding, which was applied in the process of updating the index reference year from 2015 to 2025.^[7] At the level of sub-components, the recalculation of historical data using the new classification led to some changes in price indices and weights, with the largest impact being seen in aggregates of food items. In the food category, some chilled and frozen food is now classified as unprocessed rather than processed. As a result, unprocessed food inflation based on the new classification is slightly less volatile (Chart A, panel a). The maximum absolute revision in monthly year-on-year rates amounts to around 1.2 percentage points for unprocessed food and 0.3 percentage points for processed food. For the other main components (energy, non-energy industrial goods and services), the maximum absolute revision ranges between as little as 0.1 and 0.2 percentage points (Chart A, panel b).

Chart A

ECOICOP version 1 versus ECOICOP version 2: revisions by main HICP components

a) Annual rates of change in HICP food under ECOICOP version 1 and ECOICOP version 2

(percentages)

b) Revisions in HICP annual rates of change under ECOICOP version 2, by HICP component

(percentage points)

The changes in the HICP classification have also affected measures of underlying inflation, triggering small revisions to permanent and temporary exclusion-based indicators while leaving their overall dynamics broadly unchanged. The range of underlying inflation indicators monitored by the ECB includes permanent and temporary exclusion-based indicators as well as model-based ones.^[8] Overall, the recalculation using the new classification has had only minor effects on permanent and temporary exclusion-based indicators. For example, historical data for HICP excluding energy and food (HICPX), HICP excluding energy, food, air travel-related items, clothing and footwear (HICPXX) and trimmed-mean indicators have changed very little.^[9] For these indicators, the maximum absolute revision is around 0.2 percentage points for monthly year-on-year growth rates. By contrast, the weighted median exhibits somewhat larger revisions, reaching up to 0.4 percentage points in some months. The resulting range of exclusion-based underlying inflation measures is barely affected by a recalculation of historical data (Chart B, panel a).

Chart B

Underlying inflation indicators

a) Range of exclusion-based measures

(annual percentage change)

b) Persistent and Common Component of Inflation (PCCI), Supercore and domestic inflation

(annual percentage change)

The PCCI and domestic inflation have shown little change in light of the new classification, while the larger revisions in Supercore mainly reflect an adjustment in the underlying methodology (Chart B, panel b). The PCCI captures the common and persistent components across the ECOICOP four-digit classes of HICP items from 12 euro area countries.^[10] It has not been significantly affected by the introduction of ECOICOP version 2, despite now being based on a larger set of components. Domestic inflation captures HICPX items with a low import content. In addition to the reclassification, this indicator has also been updated using more recent information on import shares. Together, the two updates result in a different threshold for inclusion, but the indicator has been revised only moderately for the past few years. By contrast, Supercore – a proxy for cyclical inflation – has been revised to a larger degree. While part of the change in the indicator can be attributed to the switch to ECOICOP version 2, the largest impact stems from an adjustment in the estimation method itself, which has been implemented alongside the change in classification.^[11] Previously, cyclical items were identified using the output gap in a forecast-based framework.^[12] In the revised approach, cyclical items are identified based on the sign and statistical significance of the coefficient on the business cycle stance in Phillips curve regressions, estimated using the unemployment gap in the sample period preceding the COVID-19 pandemic. These estimates are based on ECOICOP version 2 HICPX items. While the overall dynamics have not changed significantly (for example, the two versions peaked at the same time), the updated and reassessed Supercore measure shows somewhat higher peaks during periods of tight labour markets compared with the previous version.

Overall, the introduction of ECOICOP version 2 is beneficial for inflation monitoring and forecasting. The enhanced HICP provides a more accurate representation of consumer price developments, reflecting more recent changes in consumption patterns and market structures. By providing a clearer separation between goods and services, the new classification allows drivers of inflation to be identified more easily and captures consumption expenditure shares more accurately, for example in the case of delivery fees and games of chance. Most underlying inflation measures have been only marginally affected by the classification changes, and their overall dynamics remain broadly unchanged.

References

Bańbura, M. and Bobeica, E. (2020), “PCCI – a data-rich measure of underlying inflation in the euro area”, Statistics Paper Series, No 38, ECB, October.

Bańbura, M., Bobeica, E., Bodnár, K., Fagandini, B., Healy, P. and Paredes, J. (2023), “Underlying inflation measures: an analytical guide for the euro area”, Economic Bulletin, Issue 5, ECB.

Bodnár, K., Höynck, C., Fagandini, B., Healy, P. and Rousseau, F. (2026), “Domestic and cyclical inflation in the euro area”, Statistics Paper Series, ECB, forthcoming.

European Central Bank (ECB) (2025), “An overview of the ECB’s monetary policy strategy - 2025”, ECB, June.

Eurostat (2026), “Questions & Answers on the improvements in the Harmonised Index of Consumer Prices (HICP) effective January 2026”, Luxembourg, 25 February 2026.

O’Brien, D. (2018), “The Supercore measure of underlying inflation”, Economic Bulletin, Issue 4, ECB.