Coordinator: Hans-Eduard Hauser
European Commission - Eurostat
Phone number:
00352 4301 37795

This project falls under objective 2 of the Modernisation of European Enterprise and Trade Statistics which aims at the "achievement of a streamlined framework for business-related statistics".


1. Background of the project

1.1. Legal background

Decision No 1297/2008/EC of the European Parliament and the Council of 16 December 2008 established the Programme for the Modernisation of European Enterprise and Trade Statistics (hereinafter referred to as the MEETS programme).

This project falls under objective 2 of the MEETS-programme, which aims at the "achievement of a streamlined framework for business-related statistics". Action 2.1 foresees the "integration of concepts and methods within the legal framework".  

1.2. Need for the project

The European Statistical System has gradually evolved over the past decades, with a considerable increase in content and coverage taking place in the past 10 -15 years. This was accompanied by a similar increase in European statistical legislation where the requirements of the single statistical domains are defined as well as the duties of the Member States in order to provide the respective data. The development of the European Statistical System did not occur “in one pour”: different stakeholders, interests and user needs were involved; the Member States also had different views, history and experience. A comprehensive coordination of all these developments was not possible. Furthermore, the concepts and methods also evolved over time so that one specific statistical domain developed in the past might now be seen from a different perspective. Finally, not all statistical development took place under the ESS umbrella. Other organisations and stakeholders within the European Union as well as outside the European Union also develop and produce certain official statistics or – at least – are dealing with issues that have influence on the production of official statistics, such as accounting standards.

It is thus not surprising that concepts, definitions and methodologies as well as practices vary to some degree over the different statistical domains. This leads to a situation where the statistical outputs of these various domains cannot be compared because of the application of definitions, concepts and methodologies which are partially or even totally different. The user – who cannot (fully) compare the statistical data – will criticise this fact as a lack of (full) “coherence”.

However, not only the users are not fully satisfied but also the statistical institutes and the respondents having higher costs. The statistical institutes cannot use existing statistical data or otherwise already available data from administrative, accounting or other sources, if the requested data for the other statistical domain is defined differently. The same holds true for the respondents that have to deliver data more than once according to differing definitions and contexts. They will have higher costs in order to specify the data out of their book-keeping, accounting or other internal databases or have to deliver the data again even if similar data might already be accessible to the statistical institute.

Not all differences in definitions, concepts and methodologies that one might recognize in the current ESS, occurred in the past as a matter of insufficient coordination, differing stages of development over time or any other such reasons. In a number of cases differences were set deliberately because of the user needs or the specific basic concept of the respective statistical domain. Such differences might of course increase the production costs, however, the user needs in this domain might be better met. An example is the number of active enterprises. These are normally to be counted with reference to a certain date. However, according to the business demography concepts, the number of active enterprises should reflect the number of enterprises that were active during the calendar year and thus this number also includes enterprises which ceased their activity during the year.

From a general point of view both types of differences – those that were set deliberately as well as those that evolved less deliberately – result in the same situation: lack of coherence and probably higher production costs. However, a strict distinction between deliberate and not deliberate differences cannot be made and thus both kinds need to be identified and analysed. Not all differences have the same effect with respect to the infringement of comparability of the resulting data or the inevitable production costs.

Further reasons for inconsistencies are due to the concrete implementation by the Member States. This is especially the case, when the data requirements are not fully specified or are ambiguous, and the Member States have a certain degree of freedom which concept or definition they might use. This can also be observed in cases where the European concepts are less or insufficiently elaborated and are not based on given and agreed standards.

It should be noted that data quality – even if primarily not an issue of inconsistency – has much relation with this topic. If the requested data have to follow different concepts and definitions, more efforts are needed to fulfil the requirements and also the respondents might not be willing to provide the data with the requested specification. Furthermore, it is clear that comparability may also be reduced in case of diverging product quality, even if the concepts and definitions might be consistent. However, the main relation between quality and consistency refers to the coherence dimension of the European quality concept. This is certainly not fulfilled if there are differences in concepts, definitions and methodologies between the respective statistical domains.

If the statistical domains are elaborated using divergent concepts, definitions and methodologies, the resulting data – if compared – do not depict economic reality alone, but are also the result of the differences in concepts, definitions and methodologies. However, the user – and normally also not the producer – is not in the position to separate between theses two causes. Therefore, the user cannot really use the data or might draw the wrong conclusions.

As the ultimate goal is the establishment of an integrated system of economic and business statistics within the ESS, and that, furthermore, this system should be implemented in an integrated way with low costs and a minimum of response burden, any existing differences in the concepts, definitions and methodologies between the various statistical domains is an indicator that this situation might not yet be achieved. However, some differences might be justified through their analytical orientation. 

1.3. Types of consistency

From a conceptual point of view two kinds of consistency are relevant within the European context.

Horizontal consistency:

Horizontal consistency refers to the comparability between the various statistical domains. Data between statistical domains can be compared if they are elaborated using the same statistical unit, the same coverage, the same classifications, the same definitions, the same frame and the same reference time and period. This is also valid as concerns the relations between monthly or quarterly data and the respective annual data.

Vertical consistency:

Vertical consistency is the issue of comparability between the sum of Member States data and the European aggregate. Concepts developed for the national implementation may not be suited to derive the consistent European aggregate on the basis of such Member States data. This may occur in statistical domains where the statistical objects are of a cross-border nature.

2. Scope of the project

The work to be done in the whole ESSnet is grouped according to three main crucial issues, which are the sources for inconsistency and where harmonisation is rather possible on ESS level, into three ESSnet projects:

  • Statistical units - 2010 ESSnet project
  • Target population, frames, reference period, classifications and their applications (breakdowns, special aggregates) – 2011 ESSnet project
  • Characteristics and definitions – 2012 ESSnet project

For each project a Multi-Beneficiary Grant Agreement (MBGA) will be signed. The co-ordination between the three projects will be done by Eurostat.

The statistics involved can be grouped into three different types:

1. Legal acts providing a methodological/ conceptual fundament for surveys and data compilations:

  • Business Registers / EuroGroups Register
  • Statistical Unit Regulation
  • Economic activities
  • Products by activity
  • Prodcom as far as the ‘Prodcom list’ of industrial products is concerned
  • Further classifications mentioned in other legal acts such as for example in External Trade.

2. Legal acts governing surveys, data collections and statistical indicators

All these acts refer to/and make use of the fundamental regulations, adding additional concepts by defining target populations, time schedules, defining variables, etc.:

  • Structural Business Statistics and Business Demography
  • Short-Term Statistics
  • Prodcom
  • Foreign-Affiliates Statistics
  • Tourism
  • Energy
  • Environment
    • Waste Statistics
    • Environmental Expenditure
  • Statistics on Information and Communication Technology
  • External Trade
  • Research and Development
  • Employment statistics
    • Labour Force Statistics
    • Structure of Earnings Statistics / Labour Cost Statistics
    • Job Vacancy Statistics
    • Labour Cost Statistics
  • Vocational Training Statistics
  • Balance of Payments and Foreign Direct Investment (as far as Foreign Direct     Investment and International Trade in Services is concerned)

All these acts refer to/and make use of the fundamental regulations, adding additional concepts by defining target populations, time schedules, defining variables, etc.:

3. Legal acts referring to statistical projects which might be characterized as ‘Tertiary Statistics’

  • National Accounts (European System of Accounts – ESA 95 and forthcoming ESA 2008)
  • Balance of Payments and Foreign Direct Investment

These rely on the results of surveys and are crucially dependent on the consistency of the underlying concepts. Balance of Payments and Foreign Direct Investment