Statistics Sweden

Data management, Registers

Solna Strandväg 86, 171 54 Solna, SWEDEN

The House Price Index (HPI) is a quarterly indicator that measures the changes in the dwellings transaction prices that households acquire on the market. The HPI covers all transactions of dwellings made by households regardless of its final use. This index thus covers not only the transactions that are new to the household sector but also all that are traded between households. Prices include land value, except for the index Purchases of newly built dwellings excluding land, which is to be considered a proxy.

The data on the HPI is provided in the form of an overall index number comprising two sub-indices: Purchases of newly built dwellings and Purchases of existing dwellings. Index figures for the purchase of existing dwellings are compiled on the basis of full transaction prices and for new dwellings on asking prices.

The HPI is a chain-linked Laspeyres-type price index. HPI data are presented in the form of index numbers (present index base year: 2015) and quarterly and annual rate of change.

The HPI is weighted by the value of dwelling transactions both for new and existing. Land value is excluded in the weights for new dwelling.

National Accounts is the source for the compilation of weights, which is complemented by other administrative sources (land registry and tax authorities).

The HPI is classified according to the following categories:

H.1. Purchases of dwellings

H.1.1. Purchases of newly built dwellings.
H.1.2. Purchases of existing dwellings.

ESA 2010 sector S.14 (Household sector).

HPI covers all residential properties purchased by households in the reference period.

The main statistical variables are price indices.

The basic statistical unit of the HPI is private households. The main statistical variable for purchases of newly built dwellings are asking price. The main statistical variable for purchases of existing dwellings are transaction price.

The target population is the set of all transactions of dwellings purchased by households.

Data are available for the country as a whole.

Data are available since the first quarter of 2005 for all index series.

The HPI is a chain-linked Laspeyres-type price index that use the year 2015=100 as index base year.

Index figures with index base year 2015=100 are available, as well as quarterly and annual growth rates in percentage. Weights are given as in parts per 1000 and in Swedish Crowns, SEK.

The compiled quarterly indexes cover the whole calendar quarter.

The basic act providing for the compilation of the House price index (HPI) and the Owner-occupied price index (OOHPI) is the European Parliament and Council Regulation (EU) 2016/792, of 11 May 2016.

The relevant implementing act is Commission Regulation (EU) 2023/1470 of 17 July 2023.

None.

Regulation (EC) No 223/2009 of the European Parliament and of the Council, of 11 March 2009, on the transmission of data subject to statistical confidentiality to the Statistical Office of the European Communities. The Swedish law "Offentlighets- och sekretessförordning (2009:641)" also restricts the usage of data from statistical producing authorities.

Not relevant.

The HPI for H.1 (Residential real estate prices) is published at the webpage of Statistics Sweden.

Release calendar at Statistics Sweden, Financial Market Statistics, Release calendar

The statistics are available to all users.

The HPI indices are published quarterly and the weights for the index is published annually.

None.

None.

A database with HPI index numbers is available in the Eurostat database and in Eurostat Statistics Explained

The H1 is published as a rate of change compared to the previous quarter in the Swedish Statistical Database as Residential real estate prices

Yes, microdata can be accessed anonymised after a confidentiality assessment, provided that Statistics Sweden considers that the user has the grounds to process the data.

Not relevant.

HPI methodology is available in Eurostat's internet site.

Not applicable.

All data are checked for completeness and consistency with statistical concepts and definitions. Statistics Sweden work according to Code of Practice.

The quality of the HPI index is routinely reviewed using a framework that is based on the OOH technical manual and the European Statistical System (ESS) definition of quality.  

The HPI responds, on the one hand, to the needs of all individual users interested in the behaviour of the domestic house price market and, on the other, to the needs of the European Central Bank and Commission for a set of indicators monitoring the stability of the house price market.

Not available.

All relevant expenditure categories defined in the Commission Implementing Regulation (EU) 2023/1470 of 17 July 2023.

The HPI is considered to be accurate and to reflect well the overall behaviour of the housing market.

Not applicable except for existing owner-occupied flats in multi-dwelling buildings.

Transactions with existing owner-occupied flats in multi-dwelling buildings in small towns, villages are not covered in the Swedish HPI to the full extent.

The first preliminary national publication of quarterly statistics for the first, third and fourth quarter are published approximately 1 month after the end of the quarter, while the second quarter has a time lag of approximately 2 months. For the House Sales the first preliminary statistics are published 6 months after the reference year.

House price index are sent to Eurostat within 85 days after the end of the reference quarter and weights are sent middle of June of the year following the year to which the weights relate.

There is no record of delays in the transmission and dissemination of the data.

No comparable geographical breakdown is available.

Data are comparable over time.

The Owner-Occupied Housing index (OOH) may partly overlap the scope of the HPI. However, both indices cover a different set of transactions and any comparison of the two indices should be done with this in mind.

All figures are checked for internal consistency.

The cost as hours worked at Statistics Sweden are unknown for the compilation of the statistics. The burden due to compiling the House Price Index is non existing due to using components that are published statistics.

HPI series are revisable under the terms set in Commission implementing regulation (EU) 2023/1470 of 17 July 2023. The published HPI data and value of housing transactions may be revised for mistakes, new or improved information, and changes in the system of harmonised rules.

The general public is informed about possible non-periodic revisions in press releases.

See below.

The index for new dwelling are mainly using data from survey and the index for existing dwellings are using different administrative data.

The source for the weights are the most recently available National Accounts data and actual values from existing dwelling transactions. 

The data source for the House Sales are the National Accounts data and actual values from existing dwelling transactions.

The raw data used in the compilation of HPI are collected on weekly, monthly, quarterly and annually basis. The raw data that are collected via administrative sources are only compiled to monthly statistics. The statistical aggregates used in the compilation of the HPI are based on monthly or quarterly statistics. The monthly statistics are aggregated to quarterly aggregates. The weights are collected on a annual basis.

Transaction prices for one- or two dwelling buildings and vacation homes are using administrative data. The data are collected through the use of an electronic protocol established between Statistics Sweden and Land registry. For owner-occupied flats (private data) electronic protocol is established where data is collected from real estate agents. Prices for new dwellings are collected via postal survey from construction firms. For the Construction Cost Index the collection is mainly electronically via the web-based collection tool SIV which is standard at Statistics Sweden.

In the near future there are no planned changes for the collection of the data.

The data are checked for internal consistency. Each of the surveys included in the production of the HPI uses different types of data validation, i.e. extreme values, missing values.

See below.

See below.

HPI is a Laspeyres-type price index.

For Newly built buildings, the Building Price Index is nowcasted with the Construction Cost Index for buildings. The resulting index is re-referenced to Q4 y-1 = 100. Each elementary index is aggregated arithmetically to higher-level indices using expenditure weights that are price updated to Q4 y-1 =100.

Weight figures are annually revised (on the basis of the most recently available National Accounts data and statistics from transaction data) and price updated to the last quarter of year t-1. A moving average of the last three years is applied for HPI sub-indices for newly build dwellings.

The weight data that we use refer to t-2 and under normal circumstances we assume that these weights can approximately be representative also for t-1. In exceptional situations when this is unlikely to hold we will investigate alternatives, such as using preliminary data for y-1 from national accounts (in consultation with Eurostat). The elementary aggregate weights are adjusted with the corresponding price change between the year t-2 and the fourth quarter of the year t-1.

The five quarters (Q4 to Q4) higher-level indices are chain linked to arrive at the final series 2015=100. Q4 is thus the overlap or link quarter in which the new and old basket are priced.

For the index on purchases of existing one- or two-dwelling buildings, a SPAR method is used.

The index for purchases of cooperative owned flats (existing dwellings) is compiled applying hedonic regression method.

The methodology used in the compilation for newly produced dwellings is based on hedonic method.

The hedonic method used in the compilation of the sub-indices for HPI are of average characteristics method. A semi-logarithmic equation is used in which the logarithm of price per square meter is used as dependent variable. The statistical model is reviewed annually. The independent variables are different for the specific type of building. The common independent variables are average price per square meter, number of toilets and baths, type of tenure, basement, elevator, balcony, garage and type of building.

The prices of existing one- and two-dwelling buildings uses a SPAR method. The valuation data is from the tax authority and the assessment is done for the whole stock of one- or two-dwelling buildings each third year.

Stratification is not used in the calculation of the HPI.

See below.

Both contract date and transfer of ownership is uses in the HPI. For existing one- or two-dwelling the transfer date of ownership is used. For existing dwellings in cooperative housing the contract date is used. For the new dwellings the date for building start is used.

VAT is not included in the price of dwellings.

Taxes other than VAT are not included in the price of dwellings.

The HPI includes only transactions at market prices. Government subsidies are not included in the prices in the compilation of HPI. 

The price of land is generally included in the compilation of HPI. For the new dwellings the land components is excluded. For the House Sales indicator the price of land is included.

Dwellings in housing cooperatives are considered to be owner-occupied housing and are included in the HPI and House Sales indicator. The housing cooperatives are sold on the open market and the price and number of transactions are used for the compilation of the HPI and House Sales indicators. Housing cooperative dwellings are included and constitute a sizable part of dwellings in Sweden.

The non-market transactions are excluded in the compilation of HPI and House Sales indicators. The non-market transactions are registered with a specific coding and are excluded.

Transactions for objects with multiple properties of dwelling-character are included. If at least one of the properties is a non-dwelling property (for example office property), we exclude this.

Fractional transactions (when only a share of a dwelling ownership is sold) are excluded. Such dwellings exist in for example Swedish ski-resorts.

The Swedish HPI is mainly based on sub-indices that have been published for other purposes and quality assured at an earlier stage. In the compilation of these sub-indices there are different ways to detect outliers. For the existing dwellings it is mostly different types of bounds that rule out outliers. For the new dwellings the prices are manually reviewed. In rare cases objects are excluded from the population due to extreme influence on the outcome.

There is no known under-coverage of data in the production of the House Sales indicator.

No adjustments other than quality adjustment are applied to the HPI data.

None.