France - House Price Index

France: House Price Index

Mnemonic HPI.IFRA
Unit Index 2015Q1=100, NSA
Adjustments Not Seasonally Adjusted
Quarterly 1.71 %
Data 2020 Q2 113
2020 Q1 111.1

Series Information

Source French National Institute of Statistics and Economic studies (insee)
Release House Price Index (Existing Homes) - Initial and Detailed Release
Frequency Quarterly
Start Date 3/31/1996
End Date 6/30/2020

France: Real Estate

Reference Last Previous Units Frequency
Dwelling Stocks 2020 36,997 36,615 Ths. # Annual
Vacancy 2020 3,116 3,054 Ths. # Annual
Building Permits Jul 2020 38,578 31,440 #, NSA Monthly
Housing Starts Jul 2020 27,266 31,275 #, NSA Monthly
Residential Building Permits Jul 2020 33,811 29,440 #, NSA Monthly
Residential Housing Starts Jul 2020 24,548 28,715 #, NSA Monthly
House Price Index 2020 Q2 113 111.1 Index 2015Q1=100, NSA Quarterly
House Price Index for Existing Homes 2020 Q2 113.7 112.1 Index 2015Q1=100, SA Quarterly
House Price Value for New Homes 2020 Q2 307,940 280,968 EUR, NSA Quarterly
Non-residential Building Permits Jun 2020 3,150 1,831 Ths. sq. m, NSA Monthly
Non-residential Housing Starts Jun 2020 2,305 1,385 Ths. sq. m, NSA Monthly
House Price Index for New Homes 2020 Q1 114.09 114.3 Index 2015=100, NSA Quarterly

Release Information

For France, this house price index is a transaction price index measuring pure price changes between two consecutive quarters for dwellings sold. Quarterly from 1994.

The Paris Notaries  Services manages a real estate database specifically for Ille-de-France. The source reports average sales price for houses for the region Ile-de-France and its departments. The departments that are included in the data set are Paris, Hauts-de-Siene, Siene-Saint-Dennis, Val-de-Marne, Seine-et-Marne, Yvelines, Essone, and Val-d’oise.

Petite Couronne consists of the departments of Hauts-de-SeineSeine-Saint-Denis, and Val-de-Marne. Grande Couronne consists of Seine-et-MarneYvelinesEssonne, and Val-d'Oise

For a given quarter, the index is calculated as the weighted mean of the following two indices:

  • Notaires - INSEE second-hand housing price index
  • New housing price index

The Notaires - INSEE second-hand housing price index is based on a methodology validated by the Scientific Board of the Notaires-INSEE indices (Conseil scientifique des indices Notaires-Insee - CSIN). The new housing price index is based on an econometric model linking the new housing price logarithm to different variables that characterise the property: its type (house or apartment), its physical characteristics (floor space, number of rooms) and the geographic characteristics of the municipality in which it is located (size of the urban unit, ZEAT , coastal municipalities, winter sports and mountaineering resorts, etc.). This econometric model (in this case a hedonic model) is estimated over two consecutive quarters. There is also a time dummy variable among the explanatory variables which is used to adjust the average change in transaction prices between the two quarters, all other things being equal (i.e. with constant dwelling characteristics). The quarterly change in the index is obtained from the exponential of the coefficient associated with the time dummy. The model is said to be of  “adjacent periods” as it is based on the price transition between two consecutive periods.

The housing price index for quarter Q of year Y is calculated in two steps: 1) weighted mean of the new and second-hand housing price indices, base 100 in Q4 of year Y-1; 2) chain-linking of the index to obtain an index expressed in base 100, as an annual average for 2010. The weightings of each of the two indices correspond to the respective share of new and second-hand housing in household expenditure on the purchase of dwellings in the course of year Y-2. This expenditure is obtained from the housing accounts published by the Observation and Statistics Service (Service de l’Observation et des Statistiques - SOeS) of the Ministry for Ecology, Sustainable Development, Transport and Housing.


Since October 2018, the source has advanced the base year to 2015=100.