Statistics Norway

Division for Housing, property, spatial and agricultural statistics

Statistics Norway P.O. Box 8131 Dep. NO-0033 Oslo NORWAY

The House Price Index (HPI) is a quarterly indicator that measures the changes in the transaction prices of dwellings that households acquire on the housing market. Prices include land value. 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.

The data is based on harmonised definitions and concepts included in the Commission Implementing Regulation (EU) 2023/1470 and described in the OOH Technical Manual.

HPIs give comparable measures for changes in the prices of dwellings. They are calculated according to a harmonised approach and a single set of definitions.

Index numbers, weights and growth rates are made available at the level of detail established by the Commission Implementing Regulation (EU) 2023/1470.

ESA 2010 sector S.14/Household sector.

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 include transactions between households.

Index figures for Purchases of existing dwellings are compiled on the basis of full transaction prices (For co-operative dwellings, both deposit and joint debt is included in the price). The sales prices are provided by Real Estate Norway (prices recorded at the time of the first binding contract). This covers about 70 per cent of all purchased existing dwellings sold in Norway.

Transaction prices include the value of land for existing dwellings and new multi-dwellings. For new detached houses, the land is not included.

The house price indices are chain-linked Laspeyres-type price index. HPI data are presented in the form of index numbers with 2015 as the reference year and quarterly growth rates.

The weights for the two sub-indices are equal to the total value of dwelling transactions for new and existing dwellings, respectively. Both prices and weights include land value, as recommended by the Manual.

The basic statistical unit of the HPI is dwellings sold on the Norwegian housing market.

The index is based on market prices. 

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

Data are available for the country as a whole.

The HPI for existing dwellings has been published since the 1991q1.The HPI for new detached houses has been published since 1989q1, and new multi-dwellings since 2011q1.

The HPIs are referenced to 2015=100 for delivery to Eurostat.

Index values and quarterly and annual growth rates in percentage. Weights are given as parts in 1000 and in national currency.

The compiled quarterly indices represent the whole calendar quarter. The index reference period is 2015=100.

Commission Implementing Regulation (EU) 2023/1470 of 17 July 2023 laying down the methodological and technical specifications in accordance with Regulation (EU) 2016/792 of the European Parliament and of the Council as regards the house price index and the owner-occupied housing price index, and amending Commission Regulation (EU) 2020/1148.

The Norwegian Statistics Act regulates the activity of Statistics Norway.

Not applicable for HPI

Data is published up to the level of detail described in the Commission Implementing Regulation (EU) 2023/1470.

Single units cannot be identified in the dissemination. The collected data is handled according to the Norwegian Statistical Act.

The HPI for Purchases of existing dwellings is published quarterly at the national webpage, usually within 15 days after the end of the quarter. The HPIs for new dwellings are released about 80 days after the end of the quarter. The precise date of the release is available on the release calendar (see 8.2)

For national release, see calendar 

The data is disseminated to all users through the Statistics Norway's website.

Quarterly for the indices and annual for the index weights.

On-line press releases.

The HPI is published quarterly at Statistics Norway's website.

A database with HPI indices for new and existing dwellings is available on Statistics Norway’s National database: StatBank. The compilation of a total index is only available at the Eurostat webpage.

Existing dwellings, New dwellings 

Micro-data is not disseminated.

Not relevant.

Description of the methodology for existing dwellings and new dwellings is available on Statistics Norway's website.

Existing dwellings

New dwellings

In 2006 a quality document were published for Statistics Norway’s construction price indices and house and property price indices.

However, the section for our HPI in the above mentioned document  is not fully updated. Since 2009 there has been some adjustment to the index, concerning variables and more detailed geographical regions. Another change since 2009 is that we don’t publish separate sub-indices for ownership (owner occupier and co-operatives) anymore. Ownership is however adjusted for in the regression model. Documentation is available in Norwegian. Not available in English version.

Documentation of the HPI for new dwellings can be found here.

The framework for the quality assurance of official statistics is based on the European Statistics Code of Practice. All data are checked for completeness and consistency with statistical concepts and definitions. Automatic quality controls have been implemented within the HPI production process, in order to detect and correct errors in the data files that are received. Outliers, such as dwellings with particularly small or large floor area or very high or low prices per m2 are excluded from the index.

Not available.

The HPI is useful to all users that are interested in monitoring the development in the house price market. The main users include public sectors, National Central Banks and other financial institutions; economic analysts, media and public at large. The index is an important indicator for sellers and buyers of dwellings as well as for valuation and estate agencies.

No systematic measures of user satisfaction.  

All relevant expenditure categories defined in the Commission Implementing Regulation 2023/1470.

The HPI is perceived to be accurate and to reflect well the overall behavior of the housing market.

The sample of transactions of existing dwellings covers about 70 per cent of all dwelling-sales on the open market in Norway. Systematic skewness is still possible in the sample, regarding housing standard and location. Sampling errors are not calculated. 

The index for detached houses is based on transactions of detached houses in the Register of Deeds. Only transactions that contains both houses and land are included. Houses built on already transacted plots are excluded. The number of these are unknown.  

For new multi-dwelling houses the share of the total population that is covered by the sample, is estimated to vary between 22 and 36 per cent from 2011 to 2014. However, this estimate is uncertain because the exact number of transactions of new dwellings is unknown.

The hedonic model used in the compilation of the HPI does not include explanatory variables which adjust for housing standard. The models adjusts for dwelling size, location, age and ownership (owner occupier vs. co-operatives).

Data are sent according to Commission Implementing Regulation 2023/1470.

The national price index for existing dwellings is published approximately 2 weeks after the end of the quarter. The price index for new dwellings is published appr. 2 and a half months after the end of the quarter.

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

House price index numbers are sent to Eurostat within 85 days after the end of the reference quarter and weights are sent before 15 June of the year following the year to which the weights relate. 

The HPI for all dwellings (H1) consists of HPI for newly built dwellings (H11) and existing dwellings (H12). Only H12 are available on regional level.

Existing dwellings

Changes in data collection

Until the first quarter of 2002, the price index for existing dwellings was produced using a quarterly survey. A questionnaire was sent to all registered purchasers of second-hand freeholder dwellings. Information from the survey was combined with information collected from the Norwegian land register. The compilation method was the same, but the number of explanatory variables in the regression model was more comprehensive.

New weighting

The method for calculating weights has been changed since Statistics Norway started compiling the HPI in 1991. From 1991 to 1997, the total number of sales in each category was used as a weight. From 1997 to 2002, each category’s share of the total housing stock was used as a weight. From 2002, each category’s share of the value of the housing stock is used as a weight.

In accordance with the Commission Implementing Regulation (EU) 2023/1470, starting with 2012q2, the HPI provided to Eurostat is weighted by the value of dwellings purchased last year (flow). However, the official HPI of Stat Norway is still weighted by stock values.

Change in floor space concept

Until 2002, utility floor space was used to calculate square meter prices. Following the change in data collection in 2002, useful floor space is used. Square meter prices are therefore not comparable before and after the first quarter of 2002.

New dwellings

Changes in data collection

The index for new detached houses has previously been based on a survey as part of the data collection. As of 2021q1, the index is based on data from the Cadaster and Register of Deeds as a substitute for the form survey. The number of observations is approximately halved due to this change in data collection.  This is mainly because only houses where the land is bought together with the building structure is included. However, the quality of the data has probably improved.

The OOHI for Purchases of newly built dwelling and HPI for Purchases of newly built dwelling will overlap to a large degree, but theoretically there are some differences (The OOHI should only cover dwellings bought by household for own occupancy while the HPI should cover dwellings irrespectively of its final use).  

All figures are checked for internal consistency.

Appr. 2 full time equivalent working on the compilation of house price statistics (incl. both new and existing dwellings). 

The HPI data is not subjected to periodic and systematic revisions.

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

See below.

An important source is the national property register and the register of deeds. In addition Statistics Norway receives register data from Real Estate Norway, covering transactions of existing dwellings executed by a majority of the estate agencies in Norway. Surveys provide data for new multi-dwellings.

The regional sub-indices for existing dwellings are weighted with the same data as used for prices. For the aggregation of new and existing dwellings estimates of new dwellings are based on construction statistics (number) and mean prices (value) from the surveys. Estimated transaction value of existing dwellings is based on information from the deed registry.

The register of registered properties and land register.

Register data are collected monthly. The survey is conducted quarterly.

Monthly electronic reports of sales. Quarterly surveys via Internet.

Various procedures ensure that errors are minimised in the data. The following validation checks are made on each record in the source data:

• Extreme values are identified and eliminated if thought necessary

• Data cleaning is done to remove cases of obviously erroneous data;

• Minimum and maximum values are investigated and removed if believed to be suspect;

• A total of around 0,6% of records are removed each period.

See below.

See below.

HPI is a Laspeyres-type price index.

The regional sub-indices for existing dwellings are weighted with the same data as used for prices. 

Data are available for the country as a whole, excluding Svalbard. 

The elementary indices can be aggregated so that, for example, you get your own index for one region, for one housing type or an overall index. To create price indices at an aggregated level, we calculate a weighted average of the sub-indexes. The weights are the estimated value shares in the base period, similar to Laspeyres indices. The value shares are based on the average sales price from previous year multiplied by the number of transactions from the last housing statistics for each individual housing type in each of the regions. In that we assume that there is a certain bias when it comes to prices for owner-occupied homes and housing cooperatives we calculate an average for these separately. Data is broken down into 11 regions, and then aggregated. For the aggregation of new and existing dwellings estimates of new dwellings are based on construction statistics (number) and mean prices (value) from the surveys. Estimated transaction value of existing dwellings is based on information from the deed registry. 

The aggregation of new and existing dwellings estimates of new dwellings are based on construction statistics (number) and mean prices (value) from the surveys. 

The index numbers and weights conforms to the classification details required by the regulation. But, we also publish a national index with stock weights in addition to the one we transmit to Eurostat (flow-weighted). The price index for existing dwellings are published by Statistics Norway are at a more detailed level than that prescribed in the Regulation. For instance, we distinguish between dwelling types and regions. The national HPI is a weighted average of the regional HPIs, which are also published.

The aggregation formula is given by (H11=new dwellings; H12=existing dwellings):

The sub-index and elementary weights are based on data on household expenditure on transacted dwellings.

The HPI weights are updated annually.

The weight reference period is t-1. The aggregate weights are adjusted with an appropriate price change between the year t-1 and the fourth quarter of the year t-1.

After carrying out separate index calculations for each of the 11 regions for detached houses, detached houses and blocks of flats, we have 33 sub-indices, also called elementary indexes. There are fewer regions in the price index for new dwellings (5).

The elementary indices are calculated as chained indices with annual links. The base is changed in the 4th quarter, with the previous year as the new base year. The price coefficients revised every year based on the last two years' data basis. The coefficients are assumed to be constant over the one-year period.

The index is first calculated using the old basis. It is then recalculated with new basis. The chained index is thus multiplied by the new index the ratio between old and new index. Any updates in the calculation method, such as a change in the limit values, is also added the chaining time.

The price indices are calculated by hedonic method and by multivariable regression analysis. 

Price is the independent variable in the regression analysis and the main dependent variables are: living floor space, geographic zone, age and ownership. The hedonic model used in the compilation of the HPI does not include explanatory variables which adjust for housing standard. The models adjusts for dwelling size, location (dummies), age and ownership (owner occupier vs. co-operatives). We use a loglinear model. We use the average characteristics method.

We do not use the SPAR method.

We use a combination of the hedonic method with stratification. We construct the indices for new and existing dwellings. Both indices are regionally stratified in 11 regions and then aggregated. At last, we aggregate the indices for new and excising dwellings.

See below.

The price of the existing dwelling is the price in the sale contract and it is collected when the contract is signed. This is also the case for new multi-dwellings. For new detached houses, on the other hand, the prices are recorded when the dwelling has been completed.

VAT is included in the price of new dwellings.

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

The HPI includes only transactions at market prices. As transactions which include government subsidies are normally below market prices, they are excluded from the HPI.

The price of land is included in both prices and weights.

Transaction of cooperative dwellings are included in the source data. The price is measured as the deposit and the dwellings part of the joint debt.

Dwellings transacted by gifts or bequests are excluded from HPI coverage since there is no actual transaction price. Transactions which are between connected persons (e.g. members of a family) are excluded where known. The HPI includes only transactions at market prices. As transactions which include government subsidies are normally below market prices, they are excluded from the HPI. 

The properties are registered so that each dwelling gets its own number. When buying several dwellings, the price will be equalized for the different units.

The price is registered per dwelling, so it does not matter if there are several buyers of a dwelling.

Observations with supposedly incorrect square meter prices, floor space or price are excluded from the selection through the use of filters. The limit values for exclusion are different for the different housing types.

We could have used data from property land registers which are more complete, but these will have a time lag.

Apart from the register, currently, there are no other possible data sources that can solve the coverage issue.

The housing market displays seasonal effects on house prices. For example, prices tend to be higher in the second quarter than the first. These seasonal effects are estimated and adjusted for in the HPI in calculating quarter-on-quarter price changes.

Seasonally adjusted figures are provided for each region and for each dwelling type, and for the total index. Seasonal adjustment is done each quarter, using the standard X-12-ARIMA software.

None.