Calculation of value indice:

The gross turnover indices are calculated at the most detailed level of the French classification of activities : the NAF rev. 2 sub-classes. At this level, the index for month m is calculated by linking the index for month m-12 to the change in turnover between m-12 and m. Business demographics (start-ups, closures, entries and exits, changes in sector of activity) are taken into account on a continuous basis to accurately reflect the economic reality. Restructurings that may affect the sectoral outline of VAT returns are treated separately.

The turnover indices change their base and reference every 5 years, except for the next reference year being 2021 in reason of the particularities of year 2020. The indices now published have a base and reference year of 2015. The average value of the indices is 100 in 2015 (previously 100 in 2010).

Aggregation to upper levels:

The estimation of annual weights is based primarily on output by industry at base prices per branch calculated by the annual national accounts. By definition, output by industry at basic prices excludes taxes on products but includes subsidies on products. It does not include transport services. At the finest levels of the classification, weights are also based on INSEE's Esane statistics (structural business statistics).

The 2015 rebasing implements an innovation, with the introduction of a two-weighting system (instead of a constant-weight system) in order to better take into account the structural evolutions. Until now, aggregation of elementary indices was done on the basis of constant weights representing the reference year - i.e. 2010 previously. From now on, 2010 weights are used for the aggregation of elementary indices between 2005 (respectively 1999 for turnover indices) and 2012, whereas 2015 weights are used for the period after. The calculation on the whole period is then achieved by chaining both series (computation of a link coefficient on the year 2013). This change responds to a recommandation of Eurostat in order to increase the robustness of indices over a long period.