Swiss Federal Statistical Office FSO 

Division Economy – Section Prices 

Espace de l’Europe 10, CH-2010 Neuchâtel

The Swiss Residential Property Price Index (IMPI) is an economic indicator that measures quarterly market price changes in the transactions of single-family houses and flats bought and sold for residential purposes. The land component of the residential property is included. 

The total IMPI index is published as H.1. “Purchases of dwellings”. At the Swiss national level, in addition to the total, indices for single-family houses and flats and sub-indices for 5 types of municipalities are calculated. Separate indices for new and existing properties (H.1.1. and H.1.2.) are not available.

The IMPI covers a relatively large sample of all single residential properties purchased with a mortgage during the whole reference period and located in the entire country.

The IMPI uses in calculations all residential purchases with a mortgage no matter what the economic sector of the purchasing party is.

The IMPI measures the price evolution of dwellings in the reference period, compared to the base period. Quarterly and annual rates of change are calculated by comparison of the reference period index with the previous quarter respectively with the same quarter of the previous year.

Only market prices of residential properties are included, non-marketed prices are ruled out from the scope of the IMPI, so donations, anticipated inheritances and preferential prices are excluded. The IMPI records the transaction price, as entered in the purchase agreement. Market prices are the result of what are known as 'arm's length transactions'. This refers to a transaction where the transaction price is negotiated between a buyer and seller, who are both acting in their own self-interest.

The IMPI measures the price developments of all single residential properties purchased, independently of its final use, so dwellings bought for uses other than owner-occupancy are included (for investment, e.g. to rent it out).

The target are dwelling acquisitions, independently of their previous owner; so all purchases of new and existing dwellings are to be considered, including those existing dwellings transacted between households.

The value refer to the prices paid by buyers and include both the land and the structure. Other costs related to acquisition (notary fees, registry/cadaster fees, real estate agency commission, bank fees) are excluded.

The index is computed from the purchase values in Swiss francs.

Single residential property transactions on the open market.

The IMPI survey collects the transactions data from Switzerland's largest mortgage-lending institutions (lending banks). Through these banks, the IMPI covers around 90% of the market, measured in terms of total mortgage volume. Data on cash purchases respectively transactions not financed by mortgages are not collected.

Switzerland entire territory. Sub-indices are calculated for 5 types of municipalities: urban municipalities of large agglomerations; urban municipalities of medium-sized agglomerations; urban municipalities of small or outside agglomerations; intermediate municipalities; rural municipalities.

The index is produced quarterly since the first release on November 26, 2020 of the Q1 2019 – Q3 2020 data. August 7, 2023 the index series were extended backwards to Q1 2017.

The base period used by the FSO is Q4 2019 = 100.

Indices are equal 100 in the base period. They are published with a final 1-decimal rounding by the FSO (4-decimal rounding available on request) but they are not published by Eurostat which only publishes indices with base 2015=100.

Quarterly and annual percent change rates are calculated by the FSO in full precision and published with a final 1-decimal rounding. On the other hand, they are recalculated by Eurostat with a double rounding of indices (on 2 decimals) and change rates (on 1 decimal). Therefore, the change rates published by Eurostat could differ by +/- 0.1% from the change rates published by the FSO.

For every quarter, the IMPI is compiled by the FSO based on data for dwellings that were transacted during the whole calendar quarter. The determinant transaction date represents the mortgage deposit day. Data are compiled by the mortgage institutes and collected by the FSO during the first weeks immediately following the reference period and published by the FSO around 5 weeks after the end of the reference period.

The basic legal acts providing for the compilation of the IMPI at national level are the Federal Statistics Act (FStatA, SR431.01) and the related Ordinance about the execution of federal statistical surveys (SR 431.012.1, Annex Nb. 200).

The sharing of the IMPI between the FSO and Eurostat is ruled by the Agreement between the European Community and the Swiss Confederation on cooperation in the field of Statistics.

Data protection and individual privacy are fundamental principles of public statistical work, and the statistical secrecy is regulated in the Federal Statistics Act (FStatA, SR431.01). The FStatA and the related Ordinance about the execution of federal statistical surveys (SR 431.012.1) as well as the Federal Act on Data Protection (FADP, SR 235.1) and its related Ordinance (SR 235.11) rules the confidentiality of the collected data.

Individual data collected by the FSO through statistical surveys from the mortgage institutions for the IMPI are anonymized at the source and used for statistical purposes only. They will not be given to any third party (e.g. even for research purposes under non-disclosure agreements). 

The IMPI is released on a quarterly basis, around 5 weeks after the end of the reference period. 

The release calendar of the IMPI is publicly available in advance on the Prices Agenda on the FSO website.  

The IMPI results are announced by a newsmail and published on the FSO website on the basis of the European Statistics Code of Practice, which ensures the independence, integrity and accountability of national and community statistical offices. Privileged access is supervised and under embargo.

The total index is transmitted to Eurostat that disseminates it with the recalculated quarterly and annual change rates on the Eurostat's website.

Quarterly.

The IMPI quarterly newsmail can be publicly and freely subscribed on the FSO website.

The media releases are published on the FSO website.

Swiss residential property price index (IMPI). FSO, Neuchâtel 2020, 6 pages, FSO number 2074-2003

The IMPI tables and graphs are published on the FSO website.

None.

See the further information on the IMPI webpage.

Swiss residential property price index (Q4 2019 = 100): Methodological principles. FSO, Neuchâtel 2020, 44 pages, FSO number 2074-2001

Swiss residential property price index: Quality Adjustment Procedure. FSO, Neuchâtel 2020, FSO number 2074-2002

Swiss residential property price index (Q4 2019 = 100): Methodological principles; Chapter 6: Quality management. FSO, Neuchâtel 2020

As subsequent correction of a published index is not envisaged in the IMPI production process, the FSO will do everything it can to guarantee quality throughout the whole process. All prices collected undergo controls before being validated definitively and entered in the index calculation. Various control functions are integrated in the PRESTA IT production platform. At the end of each production cycle and before the publication of the results, a meeting is organised in order to carry out a structured and documented evaluation of the production and the initial results.

The FSO places great value on the quality of the information it produces as well as on its credibility and the trust placed in it by its users. In terms of quality management, the production of the IMPI adheres to the international standards (Eurostat, European Statistics Code of Practice; United Nations, Fundamental Principles of Official Statistics; Eurostat, Handbook on Residential Property Prices Indices) and to the internal quality manual (FSO quality, process and risk management policy). 

The IMPI, Swiss residential property price index, has multiple purposes for a variety of users:

a) Economic indicator

b) Monitoring financial market stability

c) Monetary policy and inflation driving

d) Decision-making basis

e) International comparison

f) Use in other official statistics

g) Analyses and forecasts

The IMPI was developed in collaboration with a sounding board of expert users from the public sector and the private economy as well as fellow statisticians from different Offices. A panel of such users is kept along the production of the index, similarly to other price statistics, giving to FSO valuable feedbacks about their satisfaction and advising the IMPI producers about desired improvements.

Indices on new and existing dwellings (H.1.1. and H.1.2.) are not available. Time series begins in Q1 2017 (with base period Q4 2019 = 100), time series going further back are not available.

The prices are from real transactions on the open market, as they are reported in the contracts. The data providers guarantee the correct delimitation of the relevant transactions by object, date and transaction type. The survey is limited to mortgage based transactions so cash based transactions are excluded. The data are collected by sampling and do not represent the whole universe of the transactions. The sample covers a large majority of mortgage-financed transactions, e.g. around 90%, measured in terms of total mortgage volume.

Sampling errors not quantified.

Non-sampling errors not quantified. 

The IMPI is released and transmitted to Eurostat about 5 weeks after the end of the reference quarter.

The IMPI results are transmitted to Eurostat at the release date, around 35-40 days after the end of the quarter. That means t+35-40, namely about 45-50 days before the deadline for the delivery to Eurostat.

The release calendar of the IMPI is available on the FSO website.

The IMPI newsmail can be subscribed on the FSO website.

The survey covers the entire Swiss territory: all regions and urban as well as rural areas. Data covers single family houses and flats, new and existing. The treatment of quality attributes not differ across properties for which prices have been collected over time. The quality adjustment has been developed according to the accepted methods and approaches as described in the OOH/HPI Technical Manual published by Eurostat and the IMPI released by the FSO can be considered as most suited for cross-country comparisons.

The IMPI results are calculated with a stable methodology and thus comparable over time. 

There is only one set of IMPI available.

The IMPI data are internal coherent. Higher-level aggregations are derived from detailed indices according to well-defined procedures.

The FSO is legally obliged to minimize the burden on data suppliers. It obtains the data required for the IMPI only electronically, from various sources and makes the greatest possible use of administrative data. 

The released IMPI results are definitive and not subject to revision. There is no publication of provisional data. Data could be revised only in the case of published errors.

There has been no case of data revision thus far for the IMPI. There is no publication of provisional data.

See below.

The data suppliers are Switzerland's largest mortgage-lending institutions (banks and insurances).

Weights are calculated on the basis of the volumes of the transactions of the previous calendar year collected by the same IMPI survey from the Switzerland's largest mortgage-lending institutions that provides the data for the calculation of the index.

No House Sales indicators are calculated/published.

Quarterly.

Data collection is carried out electronically by the FSO. The mortgage-lending institutions’ data on properties sold are supplemented by various location information from publicly accessible administrative geodata.

The enrichment of the banks' transaction and structural data with the location variables is carried out by the data providers themselves. For this purpose, the FSO provides them with a survey IT application that links the data via object address. After the enrichment process, the addresses are deleted. The resulting output file is then data protection-compliant and ready to be sent to the FSO.

An initial validation is carried out by the survey IT application when data suppliers enrich their input data with location variables. The application validates the formats and values of variables. Likewise, a check is carried out on the quality of address matching. By means of a log file, suppliers are informed about potential errors and have the option of correcting the data before they are sent to the FSO. 

In addition, he FSO has implemented a data validation procedure to detect missings, errors or outliers and a consequent data treatment procedure to impute missings when possible, verify and correct errors and exclude outliers. This procedure is applied directly after the data reception at the FSO.

See below.

See below.

Indices are calculated by chained Young type formulae with geometric means of the prices and of the hedonic values for quality adjustment.

Stratification weights are based on collected transaction volumes of the previous calendar year (datapool of Q1, Q2, Q3 and Q4 data for the weights). Chaining is performed yearly at Q4. For details, refer to the linked Methodological principles.

Average unadjusted prices and average quality hedonic values are calculated for each strata by geometric mean formulae.

Indices are quarterly calculated on a yearly base set in Q4, then chained with the previous series with base Q4 2019 = 100.

Elementar indices are aggregated with the weights calculated on the collected transactions volumes of the previous year.

Weights are calculated at the beginning of each year based on the value of the transactions as given by the collected mortgage data of the previous year.

Weights are not price updated, as the weight reference period is the previous calendar year.

Elementary aggregate weights are not adjusted with a price change between the previous year and the Q4 of the previous year.

Indices are quarterly calculated on a yearly base set in Q4, then chained with the previous series with base Q4 2019 = 100.

The FSO uses a two-level stratification and an hedonic repricing model in order to quality adjust the IMPI.

The indices in each stratification cell are quality-adjusted using the separate hedonic repricing models for single-family houses and flats. The hedonic models employs 20 respectively 17 among structural, use and micro/macro location variables and are yearly revised on the basis of the previous collected data.

The calculation process does not involve the use of SPAR method.

The 10 stratification items compose the product basket: two object types (single-family houses and flats) by five municipality types (urban of a large, medium or small/outside agglomeration, intermediate, rural).

See below.

Dwelling prices are recorded at the mortgage deposit payment.

VAT is included in the price of new dwellings.

Any other taxes included in the selling price are included in the price of dwellings.

Only transactions on the open market are considered for the calculation of the HPI.

The price of land is included in the dwellings prices.

Housing cooperatives are not included in the HPI.

"Non-market" transactions are identified by the respondents and excluded from the HPI Data transmission.

Multi-object transactions are identified by the respondents and are excluded from the calculation of the HPI.

Fractional transactions are identified by the respondents and are excluded from the calculation of the HPI 

Outliers are detected by a two stages process: a) a Cook's distance is performed for the actual transactions with the pool of transactions of the last 4 quarters and b) extreme values for the non categorized variables and other calculated combination of variables (as Price per m2 and others) are used for the outliers detection and eventually the exclusion from the calculation.

Not applicable.

No further statistical procedures other than quality adjustment methods are used for adjusting the IMPI data (no seasonal adjustment methods, nor time series decomposition, nor other similar methods).

Indices on new and existing dwellings (H.1.1. and H.1.2.) are not available. Time series begins in Q1 2017, time series going further back are not available. Base period is Q4 2019 = 100. First IMPI publication: November 26, 2020 (data up to Q3 2020). First IMPI transmission to Eurostat: data up to Q3 2020. August 7, 2023 the index series were extended backwards from Q1 2019 to Q1 2017. Comprehensive information, mostly available also in English, on the FSO website.