Construction, housing and property;Svalbard

Building statistics


Next update

Key figures

13.9 %

change in number of building start permissions for dwellings from August to September 2020

Registered building permissions (seasonally adjusted, trend and unadjusted)
September 2020
Seasonally adjustedTrendUnadjusted
Percentage change from the previous monthPercentage change from the previous monthTotalThe last 12-months periodPercentage change from the last 12-months period
Dwellings13.9-0.12 51428 321-16.4
Utility floor space (1 000 ms)
Dwellings26.20.3342.53 653.3-11.7
Other buildings-15.6-0.2434.35 308.0-8.2

See selected tables from this statistics

Table 1 
Registered building permissions. Last month and so far this year.

Registered building permissions. Last month and so far this year.
September 2020So far this yearChange in per cent. MonthChange in per cent. So far this year
September 2020September 2019 - September 2020September 2019 - September 2020
Number of dwellings2 51421 396-8.5-13.9
Utility floor space, 1000 m2
Total776.86 706.5-8.3-8.3
Dwellings342.52 762.30.5-9.3
Other buildings434.33 944.2-14.3-7.6
Industrial buildings302.72 961.7-19.5-8.0
Holiday property(cottages), private garages and other non-industrial activities131.6982.51.0-6.3

About the statistics

The statistics give figures for changes in the number of dwellings as a result of rebuilding existing buildings, and for buildings that have been demolished, burnt down or otherwise ceased to exist.


Definitions of the main concepts and variables

Approved building . When project application is granted it is considered as an approved building in the statistics. A granted project application gives permission to start up preparetory work. In cases where the municipality uses a one-step-application (applications where for different reasons there are no need for a two-stepwise executive work; first a general permission and then a project starting permission), the date for general permission is set to be equal to date for project starting permission.

Building work started . Building work started. Until 1999 building work was counted as started when the work with piles, the laying of the foundation wall and founding of footing begins. From 2000 the starting date used is the date when starting permission is given.

Buildings under construction . Buildings are considered under construction when the work is reported started but not yet reported completed even when the construction is stopped in the meantime. If a building is still under construction after a mighty long period, an investigation is made to see if the construction is stopped, or if the building is supposed to be completed. The registration is cancelled when a building that is started is not expected to be completed, and the building will not be considered in the statistics. This may affect the numbers.

Buildings completed . A buildings is considered completed when either temporary permission to use is given or when certificate for building accomplished is given.

Rebuilding. New dwellings as a result of rebuilding of existing buildings (i.e. new dwellings due to reconstruction of utility floor space from industry to dwellings). When an old building is being demolished to the foundation wall, it is regarded as a new building. When the building framework is not demolished, it counts as rebuilding.

Buildings demolished. Decrease in number of buildings due to demolition, fire etc. This includes number of dwellings and number of buildings. Buildings with more than one building units will be counted as more than one building when demolished. This also goes for vertically divided houses with 2 dwellings and rowhouses. A part of the decrease in number of buildings is caused by revisions of the register. This goes for private garages in particular.

Registered dates . For approved buildings, building work started and buildings completed, dates are used from when the buildings where registered by the municipalities.

Utility floor space is the floor area measured within the outer walls, defined in Norwegian Standard NS 3940 Area and volume calculations of buildings.

Dwellings . A dwelling is defined as one or more rooms and it must be possible to have access to the room(s) without having to go through another dwelling. Both dwelling units and single rooms are counted as dwellings.

Dwelling units  is a dwelling with kitchen.

Single rooms is a dwelling without kitchen or with a shared kitchen.

Dwellings equal dwelling units plus single rooms.

Room . A room must satisfy the room requirements of the Building Act. Kitchen, bath, hallway and the like are not counted as rooms.

Floor . The number of dwellings and area in the basement, lower floor and loft are registered. The difference between basement/lower floor and lower floor/regular floor is stipulated in the building regulations and in principle is determined by how high the panelled ceiling is above level terrain. A regular floor shall have a height of 2.10 metres or more above a breadth of minimum 1.20 metres. An accessible volume above the upper regular floor is called a loft. A free height of 1.90 metre with a minimum width of 0.60 metres is required.

Standard classifications

Type of building is established according to function, combined buildings, for instance combined dwelling and business buildings and storage and production buildings, are grouped by the function that occupies the main part of the utility floor space.

Industrial classification . In addition to building type, it is also registered what industry the user of the building belongs to. The Industrial Classification is attached to the user and is therefore independent of the building type. At the time of filling out the statistical form it is not always known who the user of the building will be. Because of this the industrial classification is considered to be more uncertain than the classification of the building type. The industrial classification is in accordance with the EU Industrial Classification NACE.

Administrative information

Name and topic

Name: Building statistics
Topic: Construction, housing and property

Next release

Responsible division

Division for Housing, Property, Spatial and Agricultural Statistics

Regional level

Municipal, county and national level.

Frequency and timeliness

Monthly, quarterly and annually.

International reporting

The statistics are reported to Eurostat.


Micro data are stored as text files on UNIX.


Background and purpose

The statistics measure the developments in building activities for all types of buildings. Statistics Norway has published the building statistics since 1967.

Users and applications

Users of building statistics include the building and construction industry, the Ministry of Finance, Norges Bank, Statistics Norway's national accounts section, municipalities, various domestic and foreign organizations and institutions.

Equal treatment of users

No external users have access to the statistics and analyses before they are published and accessible simultaneously for all users on ssb.no at 8 am. Prior to this, a minimum of three months' advance notice is given in the Statistics Release Calendar.

See also Principles for equal treatment of users in releasing statistics and analyses.


Coherence with other statistics

Information on completed detached houses is used in the price index for new detached houses. Building statistics are used for the compilation of national accounts statistics.

Legal authority

The Statistics Act, Section 3-2

EEA reference

EU Council regulation No 1165/98 May 19, 1998 regarding short-term statistics.



The population is licensed, started and completed buildings during the month, quarter and year. The observation unit is building. 

From 2010 and onwards the statistics include buildings with a utility floor space of 16 square metres and over. In addition buildings with a utility floor space of less than 16 square metres are included if there is a new dwelling unit. Before 2010 the treshold value for utility floor space was 30 square meters.

From 2014 and onwards the statistics is based upon registered buildings from 6 th of January to 5 th of January the year after. Before 2014 the statistics is correspondingly based upon a statistical year from 11 th of January one year to 10 th of January the next year.

From 2017 and onwards the statistics is based upon registered buildings from 6 th of January to 31 st of December.

From 2018 and onwards the statistics is based upon registered buildings from 1 st of January to 31 st of December.

Data sources and sampling

Matrikkelen is a computer register containing information about ground properties and addresses in Norway. The register also contains information on all buildings under construction at 31 December 1982 and all buildings that have been built or changed since 1 January 1983. Data to Matrikkelen are collected pursuant to the act relating to the division of landed property and provisions of this act.

AS from 2009 the municipalities have been transferred from the Ground Parcel, Address and Building Register (GAB) to the new property register, Matrikkelen.

The owner of the register is the Ministry of Environment, with the Norwegian Mapping and Cadastre Authority professionally responsible. The County Mapping Offices administrate the register system in the districts and together with each of the municipalities they are responsible for entering the data.

"Norsk Eiendomsinformation as" runs the register, and the municipalities provide the necessary information for Matrikkelen, based on data supplied by the investors and authorities.

The building statistics are a complete census.

Collection of data, editing and estimations

Data from Matrikkelen.

Every building is checked, and editated if necessary, according to editing procedures; a set of fixed machine and manual checks. The variables checked are date, type of building, industry, utility floor space, number of dwellings and dwelling units. The editing procedures ensure that obvious errors in the register are corrected. The procedures ensure that there is correspondence between the various variables for the same building.

Seasonal adjustment

Totals are calculated by summation. Monthly and quarterly series started are seasonally adjusted, while trend-cycle data are published both for utility floor space other than in dwellings and for dwelling units and utility floor space in dwellings. The numbers are adjusted for seasonal variations applying the X13ARIMA method with non-fixed seasonal effects and multiplicative model. Pre-correction of trading day effects, Easter effects and extreme values if they are significant.


Not relevant

Comparability over time and space

Agriculture, forestry and fishing building have been included again in the statistics since 1993. From 1971 to 1992 these buildings were excluded from the statistics due to the fact that the registration of these buildings failed after the obligation to submit reports was discontinued in January 1970. Beginning 1 March 1991 new registration instructions were imposed on the municipalities to report new buildings in agriculture, forestry and fishing to the previous GAB register.

Before the establishment of the GAB register in 1983, the useful floor space of buildings was used. Statistics Norway has calculated conversion factors between useful space and utility floor space for dwellings. Based on figures from 2 500 buildings, the ratio between useful floor space and utility floor space was measured, and these ratios are as follows:

Detached houses


Semi-detached houses


House with 2 dwelling units


Multi-dwelling houses




A figure for total floor space in 1982 is achieved by multiplying the useful floor space by 1.56.

Before 1983, the gross floor space was used for other buildings than dwellings. The gross floor space includes the outside of outer walls, while the utility floor space is measured from the inside of outer walls.

The industrial classification was formerly in accordance with the UN International Standard Industrial Classification ISIC. Since 1996 the classification has been in accordance with the EU Industrial Classification NACE. Utililty floor space divided by division can because of this not be compared with the previous years.

Due to new preparation routines there is a break in the statistics for buildings under construction from 1991. Buildings reported under construction more than five years ago, but not yet completed, are not included in the statistics as buildings under construction. This change means that the average figure for all buildings will be 9 per cent lower from 1991.

From 2000, new triple figures were used for type of building, while until 1999 double figures were used. Thus there is a break in type of building from 1999 to 2000.

As from 2007 Svalbard is included in the annual building statistics. The figures only include the building activity in Longyearbyen, while other areas of Svalbard are not included in the statistics. This is because only buildings in Longyearbyen are registered in Matrikkelen.

Accuracy and reliability

Sources of error and uncertainty

Municipalities can make mistakes in registering data in Matrikkelen.  There are also some municipalities which for various reasons do not always follow the current registration rules for Matrikkelen. Several munipalities cooperate on registrering data in Matrikkelen, hence some municipalities have out sourced the registrering of data in Matrikkelen to other municipalities. 

Registered building permissions . Figures for buildings in the building statistics rely on dates for when permits are registered by the municipalities in the date register(Matrikkelen). A building permit does not always mean that construction will be started at once. Especially in low conjunctures it might be that construction project are not realised, or may be postponed after a building pemit has been granted. When the building activity trend turns from a low to a high level, and vice versa, this could attect the interpretation of the statistics.

Rebuilding. Any rebuilding that takes part in connection with extension of existing buildings, are registered as extensions and are not included in the statistics for rebuild dwellings. They are however included in the figures for new dwellings.

The building statistics are a complete census, but with larger delays in registrations of buildings these buildings not be included in the statistics.

The building statistics are a complete census, and the figures in the statistics therefore have no sample variance.

The municipalities are responsible for entering building cases in Matrikkelen. There is a certain amount of lag in the registration of building cases by the municipalities. The registration delay means that not all buildings registered as started or completed during the statistical year were also actually started or completed during the statistical year. The longer the registration delay the smaller this share will be. Tables 3 and 4 show how large a percentage of the buildings which each year were registered as started and completed in the period from 1993 to 2019 which were also actually started and completed these years.

Table 1. Average annual registration delay. Residential buildings. 1993-2019. Months

Table 2. Average annual registration delay. Non-residential buildings. 1993-2019. Months

Table 3. Percentage of residential buildings that actually belongs to the statistical year. 1993-2019

Table 4. Percentage of registered non-residential buildings that actually belong to the statistical year. 1993-2019

Consequences and interpretation of registration delays

Registration delays entail two main problems with respect to interpreting the registered figures:

  • Registered building activities do not necessarily coincide with actual building activities.
  • Registered change in building activities from one year to the next does not necessarily coincide with the actual change in building activities.

The longer the registration delay is, the greater the chance of major deviations between registered and actual building activities.

Table 5. Number of started dwellings. Registered as started during the statistical year, actually started as of December in the statistical year and actually started as of December 2019

Table 6. Started utility floor space for dwellings. Registered as started during the statistical year, actually started as of December in the statistical year and actually started as of December 2019

Table 7. Started utility floor space of buildings other than dwellings. Registered as started in the statistical year, actually started as of December in the statistical year and actually started as of December 2019

Exactly how much building activity there actually is during a certain year or month can never be known until long after the end of the statistical period. For example, buildings that were actually started in 2000 have been registered every year in Matrikkelen until now.

Moreover, no one knows exactly how many building projects have gone unregistered in municipalities in the course of a year. As long as no one knows whether the delay in registration will increase or decline in the future, it is not easy, such a short time after the end of the statistical year, to say whether the registered building activities in a given year are too high or too low in relation to the actual building activities.

Table 8. Comparison of the number of dwellings registered started and registered completed. 1983-2019. Number of dwellings

Table 8 shows that over a period of time the number of dwellings registered started is higher than the number of dwellings registered completed. The main reason for this is lack of or late registration of completed dwellings in the municipalities. According to Statistics Norway’s routines for dwellings registered completed more than 3 years after actual completion are not included in the statistics.

Provisional - final figures

The same principals and definitions are used in monthly, quarterly and yearly statistics.

However, corrections made in Matrikkelen (because of errors in registration), on which the statistics are based, make the total figures of three monthly statistics differ from figures in the quarterly statistics. In the same way total figures of four quarterly statistics will differ from figures in the yearly statistics.