What does the survey cover?
The investment statistics describe annual investments in fixed assets by enterprises. Statistics Netherlands looks at the value of investments from various perspectives:
• the Standard Industrial Classification 2008 (SIC’2008) of Statistics Netherlands (CBS)
• region: province and group of provinces
• type of fixed asset
• new or existing/used asset
In addition, the overview includes yields from the sale/exchange of tangible fixed assets.
All enterprises which are economically active during the reporting period and whose main activity takes place in one of the following SIC2008 sectors: 016, 06-09, 10-33, 35, 36-39, 41-43, 45-47, 49-53, 55-56, 58-63, 68, 69-75, 77-82, 84, 85, 86-88, 90-93 and 95-96. The sectors 85, 86-88 and 90-93 are added from reporting period 2021 and onwards. For the reporting periods until 2021 no figures are available for these sectors.
Business sectors and industries are common terms for groups of enterprises classified by their main activity. Until 2009, Statistics Netherlands used the so-called Standard Industrial Classification 1993 (SBI'93) to classify enterprises by main activity. In 2008, the international classifications ISIC and NACE, and consequently the SIC based on these classifications, underwent a thorough revision. As a result, these classifications are more in line with today's business activities.
The above also means that the SIC codes of the SIC'93 and SIC2008 differ in many cases. A SIC code from the 2008 version often represents a different activity than the same SIC code from the 1993 version. The composition of a business sector or branch may have changed as well; for example, new branches may have been added and branches may have disappeared or been allocated to another business sector or industry.
Enterprise. The actual transactor in the production process, characterised by self-sufficiency with respect to the decisions about that process and by offering its products to third parties. An enterprise comprises one or several legal entities.
It follows from this definition and especially from the self-sufficiency element that an enterprise can have more than one establishment, but also more than one legal unit. (Legal units include both natural and legal persons). This is the case when individual establishments or legal units do not operate independently. Conversely, it is possible to distinguish between different parts within a legal unit that operate independently in terms of production. These then constitute as many enterprises as defined. The latter occurs especially in the case of larger groups with diverse activities. When a unit thus defined extends over several countries, the Dutch part is regarded as a whole enterprise for the purpose of national statistics.
In official CBS terminology the enterprise as defined here is called the business unit (BE), in order to avoid confusion with the term enterprise used in common parlance, which is not very precise in this respect.
The statistical unit enterprise is an operationalisation of the kind-of-activity unit, as defined by Eurostat. This definition combines two requirements that may conflict: contributing to one activity versus corresponding to one or more operational units. The Netherlands gives priority to the second requirement when operationalizing to the statistical unit enterprise.
Commencement of survey
The survey in its present form started in 1995. Data on the years 1995-1999 relate to concern investments realised based on the usage criterion (i.e. including operational lease and hire). From 2000 onwards, the investment data are based on the criterion of ownership (i.e. excluding operational lease and hire). Investments were also studied during the period 1949-1997, but this does not form a sequentially comparable series of figures with the survey in the current set-up.
The investment data of year T have the following publication timetable:
• at the end of year T+1, provisional investment data (total investments without specifications) of year T are published for five sectors, namely:
-Mining and quarrying;
-Electricity, gas, steam and air conditioning supply;
-Water supply, sewerage, waste management and remediation activities;
• in the middle of year T+2, the definite data of year T for all sectors in the survey (including the three above-mentioned sectors) are published. The data for the five above-mentioned sectors are often published somewhat earlier than those for other sectors
How is the survey conducted?
Type of survey
The survey is a sample survey. On the basis of the SIC and enterprise size, a stratified sample is drawn from the survey population. Enterprises with one employee or more are all included in the survey.
Respondents may submit the required data electronically.
The sampling fraction of the investment survey is usually around 8 percent, i.e. one in eleven enterprises in the survey population are included in the survey. The sample thus comprises slightly over 53,000 enterprises.
Detection and correction methods
Statistics Netherlands conducts plausibility, completeness and consistency checks on data received from the respondents. Estimations are conducted for enterprises not responding (on time).
The population estimation is conducted by means of a quotient estimator. In order to raise the response data to the entire population, an assistance variable is used which is known for the entire enterprise population, namely the number of persons employed. Based on the response, the average of investments per employed person is calculated. With this result, the total investments for the rest of the enterprise population can be estimated.
Quality of the results
Based on research into the confidence margins of the investments at the various SIC levels, a safe level of detail is determined for the publication. The same applies to the regional breakdown.
The method applied in the statistical processing of the sample data is aimed at ensuring comparability of the results over time.
The statistic is compiled using error detection and correction methods, performing accuracy studies and ensuring the highest possible response rate.