5. Publication in practice

At the start of the 2025=100 series, CPI figures from the January 2025 reporting month onwards are published on the basis of the new methods for holiday centres, package holidays and games of chance. In addition, figures dating back to the year 2010 have been calculated. For the HICP, figures have been calculated back to the year 1996. The figures for other European countries and for the Euro area and the EU as a whole, based on ECOICOPv2, are published by the countries and by Eurostat itself. In practice, not all groups, classes and subclasses in ECOICOPv2 are calculated and published. A group, class or subclass is excluded from publication if there is limited spending on it in the Netherlands and the group would therefore have a very low weighting. Neither is an ECOICOPv2 group, class or subclass published if spending falls outside the scope of the index. The annex contains an overview of alle sub(classes) and groups that will be published under the new ECOICOPv2. Correspondence tables between the previous ECOICOP and new ECOICOPv2 show how the translation from the old to new classification has been done.

Revision strategy

The figures for the CPI and HICP are published shortly after the reporting month. Sometimes these figures may be revised as additional information becomes available. In such instances, the revised figures are announced as definitive or revised. When there is a change in the method used to calculate the index, the figures published before and after that change will not always be fully aligned and comparable.

In the HICP, methodological changes are generally introduced in December of a given year, meaning that the price changes between December and January are the first to be calculated using the new method. The calculation of year-on-year changes in the index is then based on two methods for the subsequent 12 months. In some cases, a country may choose to recalculate the previous year’s figures using the new method and to revise the HICP series for a period of 12 months.  New inflation rates will then be based on the figures calculated using one method, but the previous year's revised indices will not be fully comparable to those of two years ago. This approach will be applied for holiday centres and package holidays. As a result, the monthly inflation figure will change over the course of 2025. In such instances, the CPI is available in two separate tables, whereby two sets of figures will remain available for (at least) one year, and year-to-year changes will be the result of indices calculated using the same method.

Recalculation of the results 

As described in the previous sections, in the 2025=100 series, certain methods and classifications have changed with respect to previous series. To enable comparability across longer time series, the CPI figures from 2010 to 2025 have been recalculated using the new methods and classifications wherever possible. The results based on the old methods and classifications (up to and including the year 2025) will remain available in the archive of the StatLine database.

For the 2025=100 series, the following applies:

Completely new figures based on the new methods and classifications have been calculated and published for the year 2025.

For the years 2010 to 2024, all items included in the CPI have been recoded from the old ECOICOP classification to the new ECOICOPv2 classification. The weighting factors and (unpublished) index series for each article are unchanged. Next, all items have again been aggregated according to the new ECOICOPv2 microclasses, subclasses, classes, groups and divisions. Finally, all the figures have been rescaled to 2025=100. The figures at the level of divisions, groups and classes may therefore vary, but the trend for the CPI as a whole between 2010 and 2024 in the new series is the same as in the old series, with the exception of some rounding differences.

No indices dating back further than 2010 have been calculated for the CPI. For the HICP, these series have been made available in order to deal with changes and time series consistently across the EU. The required elementary aggregate source data is no longer available for recoding and new aggregation. In order to calculate the HICP indices for the years 1996 to 2009, a different method have been applied due to the lack of data at this level of aggregation. The most detailed COICOP subclasses have been recoded based on the new ECOICOPv2 subclasses as best as possible according to the old classification. The time series published in early 2026, including at the most detailed level, can be considered the best possible estimate for the 1996-2009 period, but the less accurate method of classification does have a detrimental effect on quality.

Recalculation in detail

Changes to the CPI as described in chapter 3 have been introduced in the same manner wherever possible. In practice it means that games of chance, nursing homes and personal contributions for health care costs and net prices for child care are introduced in the index starting from 2025. The changes to the index of holiday centers and package holidays are also introduced in 2025, to leave unaffected the 2026 inflation rate. Earlier years have not been changed, because recalculation of the weights would take up too much effort or source data are lacking. Sewage treatment charges are an exception, these costs are part of COICOP 044310 Sewage collection through sewer systems for the entire time series.

The changes to sewage treatment charges and games of chance that are applied to the CPI cannot be applied to the HICP, due to European regulations. These expenditures are part of the HICP index starting from January 2026. Games of chance has an index reference period of December 2025. Sewage collection is not a separately published COICOP and therefore the index reference period of December 2025 is not applied.

The new method for holiday centers and package holidays is introduced in the HICP index through a revision of the 2025 figures. The indices from the old and new method are presented in the graphs below, and available in the attachment. Importantly, the average annual indices for these expenditure categories remain unchanged. As a result, the average annual index of the all items HICP is also unchanged.

For a number of (sub) classes the index reference year 2025 could not be calculated. In such cases a different index reference period has been chosen. In the attachment a separate table with different reference periods has been included. A special case are the so called “gaps”. When an index is not calculated for a consecutive period, but has been discontinued intermittently, a “gap” in the time series appears. In such cases for the period before the gap a diverging reference period has been chosen. The indices using that reference period cannot be compared to the indices after the gap. All indices belonging to the consecutive period before the gap are comparable.