About the statistics
Name and topic
Name: Post and telecommunications, structural business statistics
Topic: Technology and innovation
Transport, Tourism and ICT Statistics
Definitions of the main concepts and variables
Enterprise In the Standard Industrial Classification (SIC) an enterprise is the smallest combination of legal entities that is an organisational unit producing goods or services and that benefit from a certain degree of autonomy in decision-making.
Local kind-of-activity unit (local KAU) The SIC defines a local KAU as a functional unit, which at a single physical location is engaged primarily in activities within a specific activity group.
Location The location is determined in accordance with municipal borders as of 1 January 2002. In some industry divisions an enterprise can conduct business in several municipalities and counties without being divided into several KAUs. In such cases, the combined operations of the enterprise are registered at the head office address.
Owners without regular pay who work for the enterprise on a daily basis Owners include owners of sole proprietorships or general partnerships who work for the enterprise on a daily basis without receiving regular wages etc. Co-owners of limited liability companies and cooperatives who receive regular wages are not included.
Employees Employees include everyone who works for the employer, have an employment contract and receive compensation in the form of wages, fee, bonuses, payments-in-kind etc. Employees who are temporarily absent due to illness, holiday, labour disputes etc. are included, but employees absent because of military service are excluded.
Employment Employment is the sum of owners and employees. Persons with more than one employment may be counted as an employee in several industries. The employment figures show the average number of people employed during the year.
Turnover Turnover is defined as the sum of remuneration for rendering services to customers, sales of merchandise and gross income from other activities. Rental income and commissions are included, while special taxes, public grants and profit on the disposal of fixed assets are not. VAT is not included in the statistics.
Operating income Turnover as defined above in addition to special taxes, state and municipal grants, and profits on the disposal of fixed assets. VAT is not included.
Operating costs All costs of the enterprise excluding financial costs, depreciation of financial assets, exchange rate loss, corporate taxes, and extraordinary costs.
Operating margin Operating income minus operating costs as a percentage of operating income.
Merchandise Merchandise is goods bought and resold with no value added.
Wage costs Wage costs include wages, holiday allowance, remuneration, employers' national insurance contribution, reportable pension costs and other personnel expenses. Wage costs do not include remuneration to owners of sole proprietorships or partnerships or to family members without regular pay.
Production value Production value means turnover adjusted for changes in inventory, and goods and services purchased for resale.
Value added (at factor cost) Value added is the total of production value less the purchase of goods and services (for other goods and services than those purchased for resale) and corrected for changes in stocks of raw materials and consumables. Special public grants for manufactured/sold merchandise and other public grants/reimbursements are included.
Total purchases of goods and services Total purchases of goods and services include the value of all goods and services purchased during the year for resale ment for use in the enterprise's own production process or for stock. Investments are not included in these figures.
Gross investments Gross investments are the total value of new capital goods such as buildings and plant (except housing), machinery, tools, fixtures and fittings and vehicles and other means of transport (except for personal use), both new and used. Improvements are added whereas sales of used capital stock are deducted. Investment figures are exclusive of incoming value added tax.
Acquisitions Acquisitions include all fixed assets acquired and completed in the course of the year. In assessing the value of the asset, cost price is used for purchases and production cost for own manufacturing. Acquisitions through financial leasing are included when entered as assets in the balance sheet.
Sale of assets The sale of assets is estimated at sales value (replacement value) upon realisation of used assets over the course of the year. The figures include investment tax. Assets are regarded as sold at the time of delivery.
Improvement of own assets This includes the value of all capitalised improvements and major repairs purchased and executed by others and/or undertaken by own employees on own business assets. It is required that such improvements increase the efficiency/value of and/or extend the lifetime of the asset. The improvement is estimated at cost price or at production cost for own manufacturing.
Operating margin Operating income minus operating costs as a percentage of operating income.
National level, some main figures at county level
Frequency and timeliness
Timeliness: Data pertains to the statistical year of 2007. Final figures are normally published 17 months after the last month of the statistical year. Preliminary figures on employment and turnover are published 10 months after the last month of the statistical year.
Primary data and information on sample units and population are stored temporarily in the programming language SAS, and stored permanently as text-files.
Background and purpose
The structural business statistics for post and telecommunications are part of Statistics Norway's industrial statistics, and present detailed information about the activities in this sector based on survey data and financial information. The statistics are compiled in accordance with the EU regulation on structural business statistics, and were first published for the statistical year 1998.
Users and applications
Users include public and private sector agencies, private organisations and individuals. The statistics are also used by the Division for National Accounts and in other research and analysis in Statistics Norway.
Coherence with other statistics
The structural business statistics for transport and tourism conform to the EU regulation on structural business statistics, as do Statistics Norway's structural business statistics for wholesale and retail trade, business activities and construction. These statistics are compiled for all EU and EEA countries.
In comparisons with other statistics for employment, for instance Statistics Norway's Labour Force Survey or statistics based on the Register of Employees and Employers, it must be kept in mind that definitions and methodology vary between different statistics. The structural business statistics are compiled according to definitions given by the EU regulation on structural business statistics (see section 4.2).
The structural business statistics for transport and tourism are used as a basis in the compilation of the national accounts.
The Statistics Act of 16 June 1989, §§ 2-1, 2-2, 2-3
Council Regulation No. 58/97 of 20 December 1996
The structural statistics for post and telecommunications are organised according to the NACE standard (see section 4.3) and include the following industry division:
64 Post and telecommunications
The structural business statistics for the industry in question comprise most of the enterprises and local kind-of-activity (KAU) units registered, provided that the enterprise or local KAU is registered with activity in Norway in the relevant statistical year.
The exception is enterprises and local KAUs within public service sectors 110, 510 and 550. Such units are not included in the structural business statistics for land transport even if they should happen to be involved in the abovementioned industry.
However, all other enterprises and local KAUs in the state sector or state-owned enterprises, i.e. units with sector codes 610, 630, 635, 660 and 680, are included in the statistics if they are active in the industry.
Data sources and sampling
For a sample of the enterprises in the population, the statistics are based on tax returns and a supplementary questionnaire. The tax returns are either enclosed with the questionnaires or submitted electronically via the Directorate of Taxes.
Financial statements and number of employees for enterprises with only one local KAU outside the sample is obtained from the Register of Company Accounts, the VAT Register and/or the Registrar of Employees and Employers (AA-Register). The remaining enterprises with only one local KAU receive a questionnaire to report turnover through the structural survey.
The Central Register of Establishments and Enterprises is used to obtain necessary information about the population. The VAT Register, the Central Coordinating Register for Legal Entities, information from trade organisations and direct input from the enterprises are used to update the Central Register of Establishments and Enterprises.
The population consists of all active enterprises in the relevant industry in the statistical year. The population is divided into subpopulations, called strata, according to criteria such as industrial classification and number of employees. In some strata all enterprises are included in the sample. In the remaining strata, a representative selection of enterprises is drawn. All enterprises in this sample are asked to report a full set of tax returns and to complete a questionnaire.
The detailed financial data from the sample are combined with additional information from the various registers and Statistics Norway's structural survey. This is the basis for the estimation of the economic and financial structure of the different industries of the transport and tourism sector and makes up the final figures reported.
Collection of data, editing and estimations
The questionnaires are sent out during April in the year following the statistical year, with a response deadline of three or four weeks. Enterprises that do not respond to the first letter are sent reminders for approximately six months from the first deadline.
More detailed information on sample percentages and coverage is published in the series Norway's Official Statistics (NOS).
Revisions and corrections of the information obtained from the sample population and the structural survey are carried out routinely. The revision of industry codes, employment, turnover, wage costs, other operating costs, operating result, investments and cost of goods sold is particularly important.
The data are checked against last year's figures, the Central Register of Establishments and Enterprises, the Register of Company Accounts and other available sources including contact with the respondents if necessary.
Enterprises in the sample, with a complete tax return and questionaire, will have their actual numbers represented. The main source of data is the questionaire. Enterprises not in the sample, and with no quesionaire, will have their figures estimated. The estimation is based on the enterprises' tax return, the relationship between qestionaire-data and taxreturn-data in the sample, and other sources of data mentioned under 3.2.
Employment figures are collected from the Registrar of Employer and Employees (AA-registret). The numbers are subject to automated and manual audits before beeing published.
If data from less than three statistical units are represented in a table Statistics Norway will withold from publishing the actual number. The reason for this is to maintain the anonymity of our respondents.
Comparability over time and space
The definition of investments has been changed as of the 2001 statistical year (see section 4.2). The investment figures for previous years have been changed accordingly and will therefore deviate from previously published figures.
Up to and including 2001, the Standard Industrial Classification 1994 (SIC94) was used. As of 2002, the Standard Industrial Classification 2002 (SIC2002) is used. This means that not all industries are comparable over time on 4- and 5-digit level. This is noted with explanatory notes and/or breaks in the tables affected. The change in classification has no influence on 2- and 3-digit level.
Starting in 2004 employment numbers is gathered directly from the Registrar of Employers and Employees ("Arbeidsgiver-/ og arbeidstakerregisteret" a.k.a. "A/A-registeret"). Previously employment numbers was assembled through a combination of forms, sent out to enterprises in our sample, and the Registrar of Employers and Employees. Enterprises that were not in our sample, and without a record at the Registrar of Employers and Employees, had their employment numbers estimated using variables like personnel costs and turnover. The change in main source represents a methodolocigal break in the time series for employment numbers. At the same time with this change in source, the method for estimating number of employees at enterprises not recorded with the Registrar of Employers and Employees was changed. This change adds to the methodolocigal break. An analysis conducted on the effect of this transition concluded that employment numbers fell on average by approximately 2 %. This rate varied greatly between industry subclasses. For a more detailed analysis, see publication (As of yet - only in Norwegian).
Sources of error and uncertainty
Measurement errors are defined as data errors due to the measurement instrument. Examples of such errors are vague or ambiguous questions, misunderstandings or incorrect data supplied by the respondent. A typical error can be the use of incorrect units of measurement, for instance NOK instead of NOK 1 000. Processing errors are defined as errors due to Statistics Norway's data processing. Examples of such errors are misinterpretation of the answers on printed questionnaires when these are read optically, for instance if "1" is read as "7", or if correct figures are considered incorrect and changed.
Non-response errors are either due to unit non-response, i.e. a respondent does not respond to the survey, or partial non-response, i.e. that the respondent has not answered at least one of the questions of the survey. Surveys based on questionnaires will normally have some non-response. Unit non-response can be a result of company closures, shutdowns, mergers, demergers, holidays, refusals etc. Partial non-response can be a result of omissions, lack of data or other circumstances.
Non-response by particularly important units is examined in manual checks. Other enterprises that fail to return the questionnaires are treated in the same way as enterprises that are not part of the survey. Although the use of compulsory fines has a considerable effect on reducing non-response, it may increase the statistical uncertainty if the respondent chooses to fill in the questionnaire at random in order to avoid the compulsory fine.
For variables other than turnover and employment the statistics are based on estimated figures from sample surveys. When basing the statistics on a sample instead of a full census, there may be deviations between the properties of the sample and the population. This imbalance in the sample is normally reduced by increasing the size of the sample and coverage for various key variables.
There are no fixed rules as to what are acceptable sampling rates or coverage. In a homogeneous population a relatively low sampling rate or coverage rate can be satisfactory, whereas in heterogeneous populations the sampling rate and coverage rate should be higher.
Groups based on relatively few observations will easily be affected by observations that deviate strongly from the group average. To identify and help determine what to do with such extreme observations Statistics Norway has developed their own model.
Sample skewness Sample skewness may arise when the distribution of some variables in different parts of the sample is not the same as the corresponding distribution in the total population. Stratification is used to divide the population into several unique subpopulations according to criteria such as industrial classification, turnover and number of employees. This is done to ensure that each subpopulation consists of units that are as homogenous as possible. The sample is then drawn from all the subpopulations.
The structural business statistics use the Central Register of Establishments and Enterprises, the Register of Company Accounts, the VAT Register and the Registrarr of Employees and Employers to define the population and collect the necessary data.
The most common errors are those caused by delayed registration. Such delays can be a result of a delay in reporting to the registers or the fact that changes are normally recorded some time after they have occurred. As a result, the registers are not updated at all times, which may lead to sample skewness or the use of old data.
When drawing samples, characteristics such as industry code, status etc in the Central Register of Establishments and Enterprises are also important. The quality of these characteristics is imperative when dividing the population into adequate strata, and may affect the quality of the sample.