Revision of the national accounts time series


Like in most European countries, Norway has completed a revision of the national accounts time series. The aim of this periodic revision was to incorporate new information. There have not been any changes in the definitions. The growth rate in macroeconomic key indicators, such as GDP, is approximately not changed.

The incorporation of a new data source for salaries and employment (a-ordningen) is the most important single cause of the revisions. Transfer of some specific units from market producers to government sector, as well as a change in how we use some existing sources, have caused other corrections.  

2016 is the last year with previously published final national accounts figures. For the years 2015 and backwards, the revisions of levels are not significant, and vary between minus 0.2 and plus 0.1 per cent. However, for 2016, GDP was revised down with 0.7 per cent.

The growth rates for GDP and GDP mainland-Norway are not revised, except for the years 2009 and 2016 where the growth rate of GDP mainland-Norway was revised with minus 0.1 and 0.2 per cent respectively. 

For 2016 the growth rate of employed persons is revised from 0.2 to 0.3 per cent. The revision of the figures for 2017 and 2018 can be explained by changes due to a normal revision cycle. Figures for 2017 is now based on more extended and updated data sources than when we published preliminary figures for this year. The figures for 2018 are still preliminary. The growth in GDP mainland- Norway is now calculated to 2.0 per cent for 2017 and 2.2 per cent for 2018. For both years this is the same growth rate as previously published. The increase in employed persons is revised from 1.1 to 1.2 per cent for 2017, and from 1.5 to 1.6 per cent for 2018.

Periodic revisions of time series and revisions due to a normal revision cycle

Final national accounts figures are based on comprehensive data sources through use of different available economic statistics. Many of these sources take some time to produce. Since governments and other users of national accounts data need updated figures, several versions of preliminary national accounts figures are published before final figures are published about a year and a half after the end of year t.

This gives continuous updates of the last published periods. In addition to the revisions due to the ordinary production and revision cycle, the historical time series are revised periodically. These periodic revisions of the time series, or main revisions, are caused by new international principles or guidelines, or by new substantial changes in the data sources.

A few series were revised backward in 2017, and a more significant main revision was published in 2014. The main revision in 2014 was a consequence of new international principles and was carried through and published the same year in most European countries. However, the more limited revision in 2017 was a result of changes in national data sources. Eurostat recommends periodic main revision about every five year and the countries are encouraged to publish the results from these revisions approximately at the same time. The revision of the time series in 2019 is caused by incorporation of new data sources and a new interpretation and use of a part of the existing data sources.

Toll road companies and governmental companies administrating public transport

One cause to some of the revisions is a regrouping of some units from market producers to general government. This applies to toll road companies as well as government-owned companies with task to administrate public transport. The companies last mentioned plan and coordinate the supply of public transport and enter into contract with transportation companies which supply the transportation services. The regrouping leads to reduced market activity and increased activity in the general government sector, however the net effect on GDP is insignificant. 

General government

In addition to the regrouping of some companies mentioned above, the compensation of employees in general government is reduced. The reason for this reduction is new calculations of the pension costs in health companies. As a consequence, the calculated value added in general government is reduced, as well as final consumption expenditure in general government. The effect of reduced pensions on valued added and final consumption expenditure in general government is more significant than the effect of the regrouping of companies mentioned above, so the total effects on value added and final consumption expenditure in general government are negative.

The gross fixed capital formation in central government is revised upwards in the period 2003-2010, since the gross fixed capital formation due to three road construction projects carried through as public-private partnerships (PPP) are now included in the figures for central government. In addition, there are some adjustments in connection to rent of land.

Trade through internet from abroad

A study of the sources for calculation of internet shopping from rest of the world from 2012 leads to a redistribution of import of goods and services. The level of internet trade was at the same time reduced from about 40 to 25 billion Norwegian kroner in 2017. For the year 2017 internet trade with goods are now calculated to 12 billion kroner, while the internet service trade is calculated to 13 billion kroner.  

As a consequence of the new figures for internet trade, the households’ final consumption expenditure of services was revised up, while the households’ final consumption expenditure of goods was not changed. However, new calculations of consumption of electricity caused a downward revision of the households’ final consumption expenditure.

Export and import except from internet trade

Import of traditional commodities as basic metals and metal products are revised downward for all years back to 2012. This correction is related to gross fixed capital formation of rigs and in extraction of oil and gas resources. Internal analyses showed that import related to investment in some specific oil rigs and some specific field development projects were doubled in the previous figures (in different periods). The size of the downward revision in value varies from year to year, between 2 and 3 billion Norwegian kroner, for the years 2013-2016 about 15 billion in total. At the same time the analyses showed that a bigger part of some projects were conducted abroad than earlier expected, with the isolated effect being that the figures for unspecified import for commodity costs related to gross fixed capital formation or operating costs in the oil- and gas extraction were adjusted upward. In summary these analyses have resulted in larger figures for import related to gross fixed capital formation in oil and gas extraction and in oil rigs.

Employment and compensation of employees

After the establishment of a-ordningen as a channel for employers to report figures for employment, salaries, and compensation of employees to the general government from January 2015, Statistics Norway has had access to more detailed and comprehensive data set concerning employment, salaries and compensation of employees in the local kind of activity units.  This information, together with detailed register data concerning self-employed persons, have been used to calculate new levels for salaries and employment figures. Employed persons in total for 2016 is calculated to 2 715 000 persons, which means a downward revision of 47 000 persons, or 1.7 per cent lower than earlier published. Among the industries with significant revisions are manufacturing industry and wholesale and retail trade. Employed persons in the manufacturing industry is calculated to 217 000 persons in 2016, which means a downward revision of 16 300 persons. In wholesale and retail trade the number of total employed persons are adjusted down with 15 100 persons. The corresponding downward revision in central government was 16 000 persons.       

The new data source has also given the opportunity, according to international guidelines, to reclassify employers’ cost connected to finance employees’ sick leave. These costs were earlier included in compensation of employees but are now included in social costs related to the job.  This reclassification, together with the effects of using the new sources, resulted in a reduction in total salaries for all industries with 21 billion kroner, or 1.7 per cent less than previous published figures for 2016. Compensations of employees for all industries in total is reduced with 8.4 billion kroner, or 0.6 per cent. For the industry electricity, gas and steam the figures was revised upward with 3.6 billion kroner. The corresponding figure for the transport industries in total was revised down with 3.4 billion kroner.

New calculations of pension costs in the regional health enterprises lead to a reduction from previous published figures for compensation of employees in central government with 4.6 billion.

The number of total hours worked for employees for 2016 was adjusted down with 3.9 per cent from previous published figures. The time series are revised back to 2001 to avoid a break in the time series.

Summary of the Revisions

Table 1. Revisions in final expenditure and GDP. 2011, 2013, 2015, 2017 and 2018. Billion kroner

To table

Figure 1 and Figure 2 show the effect of the revisions on the development in constant prices in GDP and GDP mainland-Norway, respectively. The figures show that the effect of the main revision did not had significant effect on the development in GDP for years up to 2016. In addition, updated figures for 2017 and 2018 have not given rise to essential changes in the annual growth rates for these years. The growth rate for GDP for 2017 is revised from 2.0 to 2.3 per cent, while the growth rate for GDP mainland-Norway has not changed. For 2018 the revision in the GDP growth rate was minus 0.2 per cent, while the growth rate in GDP mainland-Norway was not affected.

Figure 1. Annual growth rate in GDP in constant prices

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Previous 3.2 2.1 1.4 0.9 4.0 2.6 2.4 3.0 0.5 -1.7 0.7 1.0 2.7 1.0 2.0 2.0 1.2 2.0 1.5
Now 3.2 2.1 1.5 0.9 4.0 2.6 2.4 3.0 0.5 -1.7 0.7 1.0 2.7 1.0 2.0 2.0 1.1 2.3 1.3

Figure 2. Annual growth rate in GDP mainland-Norway in constant prices

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Previous 3.1 1.9 1.5 1.2 5.0 4.7 5.1 5.7 1.8 -1.7 1.9 1.9 3.7 2.3 2.2 1.4 1.1 2.0 2.2
Now 3.1 1.9 1.5 1.2 5.0 4.7 5.1 5.7 1.8 -1.8 1.9 1.9 3.7 2.3 2.2 1.4 0.9 2.0 2.2