2.4    Scale of the problem

Currently a number of key cross-cutting variables are used in more than one statistical domain. These variables (such as e.g. turnover or employment) have often different meanings according to the various business statistics domains. This strongly hampers the comparability of the data, prevents their optimal use and can even lead to misleading policy input.

Moreover, the inconsistency also leads to inefficient data production methods as synergies between data collections cannot be used and data collected for one domain cannot serve the production of indicators in another domain.

The public consultation of users shows that 95% of respondents believe that inconsistencies have a negative impact on the quality of their work whereas 81% indicate this increases significantly the time needed for exploiting the data. Moreover, 70% of respondents indicate having already experienced difficulties in combining data from different business statistics domains.

The quality and the usability of the business statistics disseminated at present would benefit from further efforts, as foreseen by FRIBS, to remove not only inconsistencies, but also from collaborative efforts to remove e.g. the asymmetries in the international trade in goods data mentioned above.

Based on the results from the consultations of the relevant stakeholders, the estimated burden on businesses of the collection of the current business statistics is around EUR 689 million. Although the burden of collecting business statistics corresponds to just about EUR 1.4 per person in the European Union, the subjective perception of statistical burden is higher. One of the aims of FRIBS is to reduce the burden on businesses.

The Treaty on the Functioning of the European Union requires the co-legislators to adopt measures for the production of statistics where necessary for the performance of the activities of the Union. In this regard and to overcome the problem as described above, the existing legal provisions would have to be harmonised with each other for reaching more consistency between them. FRIBS aims at securing this consistency over time, notably by avoiding the risk that changes will be made in just one of the statistical areas in the future. Last but not least, the adjusted legal acts must be implemented in a harmonised way in the Member States.

When considering the above-mentioned constraints it looks impossible to produce more and better information in using the output of the current non-integrated business statistics domains. In any case, the importance of business statistics goes much beyond their own statistical area and serving their own purposes. Namely, business statistics play an integral role for the production of National Accounts key figures such as GDP. The significance of European business statistics is also substantiated by the major number of extractions registered in the Eurostat databases: over 20 % of the around 31 million extractions from Eurostat's databases during the period May 2015 - April 2016 concerned business statistics.

2.5    What would happen without an intervention

If no policy action is undertaken, all the above-mentioned problems of a scattered stovepipe system for statistical production of European business statistics will persist. The current non-harmonised legal frame and the sub-optimal legal decision-making process cannot be significantly improved under the context of the currently dispersed and non-harmonised legal acts for European business statistics.

Continuing in this way could even lead to deterioration of European business statistics in terms of relevance, consistency, quality or costs over time. Many users would remain dissatisfied. Additional user requirements could only be fulfilled when additional surveys are launched or when additional non-integrated data sources are exploited. This makes the system not only more costly, but also rather inflexible. As changes would be done in an isolated manner in each statistical area and as these changes will add on to a non-harmonised system, the inconsistency of the production output would increase. Moreover, the costs and burden for producing any additional business statistics would increase over-proportionally over time.

Due to the present limited cooperation in data production in the European Statistical System, the measurement of complex phenomena like globalisation would be less consistent and informative, namely in the domains where "mirror data" is produced such as Intrastat and FATS. European business statistics are already losing relevance and would increasingly do so..

The loss of relevance make users look for alternative data sources covering their informational needs outside official statistics. This has several consequences for all stakeholders. Apart from the additional costs for users, the quality related commitments of these alternative data sources differ from official statistics and as such may form a less secure basis for decision-making. Private data providers may launch data collections to produce their figures without regard to burden created on businesses. This leads to an actual increase in burden on businesses perceived as administrative burden. Lower relevance of official statistics could lead to further budget constraints for NSAs. Possible cost savings and reduction of burden e.g. resulting from an increased use of administrative data, merging questionnaires or from the exchange of micro-data cannot fully be realised with the current system.

Table 2: risks and potential impacts to the current European business statistics system

Furthermore, some of the existing legal frameworks, such as the framework governing the production of ICT statistics is subject to a sunset clause which will expire in 2018.

3.    WHY SHOULD THE EU ACT: RESPECT OF SUBSIDIARITY PRINCIPLE

Article 338 of the Treaty on the Functioning of the European Union provides the legal basis for the EU legislators to adopt measures for the production of statistics where necessary for the performance of the activities 10 of the Union. Reliable and comparable statistics are needed to underpin the planning and the monitoring of the implementation of Union policies.

The Commission (Eurostat) cannot work alone to provide the necessary statistics. The European Statistical System (ESS) is the partnership between the Union statistical authority, which is the Commission (Eurostat), and the national statistical institutes (NSIs) and other national authorities responsible in each Member State for the development, production and dissemination of European statistics. The ESS is established by Regulation (EC) No 223/2009 on European Statistics. The Member States ensure the actual compilation of statistical information at the national level. For the compilation of statistics at European level, a harmonised methodology and the definition of a common output to be delivered by Member States are indispensable in order to guarantee the required high quality statistics necessary to support the Union policies and robust decision-making. As the difference in methodology is not only caused by the diversity of statistical domains, but also by heterogeneous methodology in Member States, attaining harmonised methodology can be fully achieved only by way of EU action. The proposed Regulation will guarantee the required data quality e.g. in terms of coherence and comparability of the business statistics relevant for the activities of the Union.

Only the Commission (Eurostat) is in a position to propose the necessary measures for harmonisation of statistical information across Member States and produce harmonised business statistics at the European level on the basis of the data compilation carried out by Member States. Consequently, the European Union may adopt measures in accordance with the principle of subsidiarity as set out in Article 5 of the Treaty on European Union (TEU). Thus, the proposed EU action is fully justified.

Although this reasoning is also valid for all other domains where statistical information is needed for EU policy-making and monitoring, specific phenomena of business statistics are transnational or increasingly global. These cannot be adequately addressed by actions on the Member States level alone. Better monitoring of globalisation, based on a better knowledge of multinational enterprises and their activities, can only be done at European level. Also the structured and regular exchange of micro-data between Member States, which is needed to alleviate the burden on businesses, can only be established at the EU level.

Businesses undertake production, sales, research and development where it is the most advantageous and not necessarily within certain national borders. This has to be taken into account by statisticians. A correct statistical picture of the European economy can no longer be given by simply adding up national figures. This underpins the need for a uniform and coordinated EU action.

4.    OBJECTIVES

Effective tackling of the issues defined under Chapter 2 would necessitate a more robust and streamlined legislative framework for European business statistics that should ensure comparability and coherence of data in the future.

Three types of objectives are distinguished: general, specific and operational (or "measures"). Firstly, the general objectives indicate why a policy action is needed and are defined by taking into account the different aspects of the problem. Within this general framework, specific and operational objectives are then identified, illustrating respectively what should be done to reach these general objectives, and how this should be done in practice.

With a view of achieving the identified objectives and ultimately solving the problem, a number of different policy options are set out and analysed in the following chapters.

4.1    General objectives and links to horizontal EU objectives

In order to define the general objectives of the policy initiative, a broader view of the general long term EU policies is taken within which the initiative should be placed. More specifically, the monitoring of the goals set through the Juncker Commission 10 priorities requires harmonised and comparable European statistics to design policy initiatives to meet the priorities and to monitor the implementation of these initiatives.

Furthermore, the EU 2020 Strategy envisages a Europe that would "be turned into a smart, sustainable and inclusive economy delivering high levels of employment, productivity and social cohesion". Monitoring the objectives set by the strategy requires an information capacity that can deliver high quality, reliable and integrated European business statistics.

Moreover, ESS Vision 2020 strategy invigorates the modernisation activities on European statistics, already started in 2009 with the MEETS programme. The ESS Vision 2020 recognises the need to address emerging policy requirements for new statistical indicators and calls for the more integrated and deeper analysis of new phenomena (such as globalisation) and for better serving EU policies with high economic or social impact (e.g. trade or R&D policy).

The Commission’s initiative on Smart Regulation underlines that the evaluation of EU policies would lead to an increasing need for more relevant data (Commission Communication COM (2013) 686 "Strengthening the foundations of Smart Regulation – improving evaluation").

The FRIBS is one of the key initiatives of the Commission's Regulatory Fitness and Performance programme (REFIT). In that respect, it aims to increase the relevance of business statistics, streamline and simplify, facilitate the use of modern and efficient data production methods which would allow for the reduction of burden and costs in the long run.

The European Statistical Programme (ESP) 2013-2017 defines business statistics as one of the three pillars of the European Statistical System. The ESP underlined that business statistics need to support the decision-making process and help European citizens and businesses understand the impact of EU policies, focused on European businesses with an increasing need for detailed and harmonised statistics. Addressing globalisation, the ESP also envisaged among its objectives to enhance the indicators and statistical information available for better measuring economic globalisation and global value chains. This should be done by better analysing the global value chains through foreign trade and business statistics, in also linking the statistical micro-data.

The FRIBS objectives set out below are fully in line and coherent with the objectives of the ESP relating to European Business Statistics. In particular, the ESP emphasises the need for more efficiency and effectiveness of statistical production processes; the provision of high-quality statistics; the requirements to reduce the administrative and reporting burden.

The European Statistical System Committee's Riga Memorandum recognises the strategic challenges that the business statistics is confronted with and the need for business statistics to be reviewed regularly so that comparability and relevance of data is guaranteed. The Riga Memorandum confirms that measuring of economic globalisation should be anchored on consistent and aligned primary statistics on international transactions and businesses.

Given this context, the general objective can be formulated as follows:

streamline and rationalise the European business statistics in order to better respond to the changing user needs; simplify the respective European legislation; modernise and increase the efficiency of the production and dissemination of high quality European business statistics to make it fit for the future, while at the same time adhering to impartiality, reliability, objectivity, scientific independence, cost-effectiveness and statistical confidentiality. 11

4.2    Specific objectives

From the general objective defined above, five specific objectives can be derived:

Improve consistency;

Enhance flexibility;

Facilitate the use of innovative methods and sources;

Ensure quality of data production;

Reduce unnecessary burden on data providers.

The specific objectives listed above can be further translated into a number of operational objectives. The operational objectives would be defined on the basis of the policy options that will be set out in Chapter 5 of this report.

5.    POLICY OPTIONS

In order to solve the problem described in Chapter 2 Eurostat has considered several policy options to achieve the objectives set out in Chapter 4.

5.1    List of policy options

As a result of the consultations of stakeholders that took place in the context of the impact assessment and the extensive discussions with Member States, the following list of policy options, on which the detailed qualitative and quantitative assessment will focus, has been defined:

Option A - Baseline scenario – no EU policy change;

Option B - Implement legislative actions limited to certain business statistics domains, including alternative options for the modernisation of Intrastat.

Option C - Modernisation of business statistics in a single framework (FRIBS) using a mix of measures, including alternative options for the modernisation of Intrastat.

5.1.1    Baseline scenario – no EU policy change (Option A)

No action is taken and therefore the current legal framework of business statistics divided into 10 heterogeneous legislative acts with the shortcomings described previously would continue to exist.

This option A refers to the baseline scenario without any EU policy change and does not require any particular measure or action.

5.1.2    Implement legislative actions limited to certain business statistics domains, including alternative options for the modernisation of Intrastat. (Option B)

This option would implement the inclusion of additional data requirements, the identified simplifications of the data requirements and certain modernisation of the statistical production facilitating reduction of burden on data providers and minimisation of costs for data compilers through separate amendments to the 10 legislative acts currently existing in the area of business statistics. This option requires the use of different legal instruments to modify existing legislation on certain business statistics domains. Depending on the scope of the changes and the procedures foreseen in each individual legal act, an ordinary legislative procedure or/and the adoption by the Commission of delegated or/and implementing acts would be necessary.

For the collection of statistics on intra-EU trade (Intrastat), this option B includes four possible sub-options - SIMSTAT, Revised Intrastat, Single Flow and a combination between SIMSTAT and Revised Intrastat (explained below in 5.1.3)).

5.1.3    Modernisation of business statistics in a single framework (FRIBS) using a mix of measures, including alternative options for the modernisation of Intrastat (Option C)

The structure of the framework integrating business statistics (FRIBS) project is designed in a modular way, with the different components being discussed at different levels and adopted by different means:

-    On the one hand, all infrastructural elements would be part of a framework Parliament and Council Regulation. This would cover the business registers (including the EuroGroups Register), the basic data topics and their periodicity, the data sources, the exchange of micro-data for Intrastat, data transmission, data quality, data confidentiality, etc. This framework legal text would repeal the existing regulations for the various business statistics domains which will then be integrated into FRIBS.

-    On the other hand, non-essential elements can be supplemented or amended by delegated acts (e.g. detailed topics for all business statistics and for business registers as well as subjects and characteristics of the dynamic statistics- ICT, CIS, GVC 12 ), while the technical provisions (e.g. detailed data requirements in terms of for instance, breakdowns, statistical populations etc. , data quality reports and the technical provisions for the exchange of data and micro-data) will be part of implementing acts. The procedure for the adoption of these provisions will be included in the framework Parliament and Council Regulation..

Similarly to Option B, for the collection of Intrastat statistics four sub-options to lower the burden on data providers are considered (SIMSTAT, Revised Intrastat, Single Flow and a combination between SIMSTAT and Revised Intrastat).

Thus, for Intrastat, option C (like option B) is sub-divided into the following four sub-options:

Option C1: SIMSTAT –exchange of micro-data

Under the SIMSTAT sub-option, Member States remain responsible for compilation and transmission of statistics on both intra-EU exports and imports. The novelty of the SIMSTAT approach in comparison to current Intrastat is the creation of an additional data source. Each Member State collects data on intra-EU exports and makes it available to other Member States via a hub. Each transaction reported in one Member State may thus serve as a data source for two Member States: for compiling the intra-EU exports of the exporting Member State, and for compiling and/or verifying the intra-EU imports of the partner Member State. To enhance the reusability of exports data, it is foreseen that it will need to have a quality requirement on the collected trade (97 % like currently) and also information on the importing company (VAT number of partner operator) and country of origin of goods. For import statistics quality requirements will be for the statistical output rather than for the input (collection), making it possible to use either national data collection on imports or exchanged exports data or a combination of both.

Option C2: Revised Intrastat

Revised Intrastat means continuation of the current Intrastat system in a simplified form, but adjusting the coverage requirements. In this option Member States will need to compile and transmit to Eurostat statistics on both intra-EU exports and imports like they do currently. Less data will be collected from both companies' flows or for either of the two flows, depending on the final parameters of this option. The quality requirements will be defined either through the amount of collected data or through the statistical output. This option foresees neither mandatory exchange of micro-data nor collection of new data from exporters.

Option C3: Single flow

Single flow differs from the other two options in the sense that Member States will need to compile and transmit to Eurostat statistics only on intra-EU exports. The corresponding imports statistics are created by mirroring. Consequently, this option foresees no data collection on intra-EU imports. A quality requirement based on the amount of collected data is maintained for the exports flow (97 %). This option foresees neither mandatory exchange of micro-data nor collection of new data from exporters.

Option C4: Combination of SIMSTAT and Revised Intrastat

At the beginning of the so-called REDESIGN of Intrastat project these three different approaches for Intrastat have been considered separately in order to assess their impacts. The results from the extensive consultation exercise on Intrastat (and more specifically the analysis of costs and benefits for NSIs) revealed that Single Flow is not a viable option for MSs due to serious quality problems. On the basis of these assessments the preferred option for Intrastat is going to be the combination of sub-options C1- SIMSTAT and C2- Revised Intrastat, as decided by the ESSC in May 2016, in order to maximise the benefits while mitigating the risks and costs. The ESSC decision was to combine elements from the SIMSTAT option (mandatory exchange of micro-data) and from the Revised Intrastat option (more flexibility in the choice of data sources used).

5.2    Discarded options 

Taking into account the broad list of existing preliminary policy options, and against the background of the results from the stakeholder consultations, an initial analysis was firstly carried out regarding the overall practicability of all options. The goal was to identify any options that are considered unsuitable, so as to discard them from the detailed assessment. This analysis has led to the conclusion that the three options set out below are not suitable for further assessment and were discarded. Additionally, it should be clarified that a horizontal recast of the ten existing legislative acts governing European Business Statistics (as an alternative to a framework regulation) has not been suggested as an option during the stakeholders' consultation. The reason for that was the guidance provided by the Interinstitutional agreement ((2002/C 77/01)) setting up the rules for the use of recasts and stipulating that:

"A new legal act shall not constitute a recast act if, …, it makes substantive amendments to all the provisions of the earlier act, which it replaces and repeals."

The undertaken initiative to modernise and streamline European Business statistics necessitates substantive amendments to the provisions of the legislative acts currently in force. That is why a framework regulation has been suggested instead of a recast.

5.2.1    Better enforcement and implementation of existing legislation (the Statistical Unit Regulation- "Council Regulation 696/93") (Option B in the IAA)

In line with the Impact Assessment Guidelines, an option on the better enforcement and implementation of existing legislation, possibly with improved guidelines should always be considered. Current compliance by Member States, with existing legislation in the field of business statistics, is generally very good. This however is less the case for application of the Statistical Unit Regulation (Council Regulation 696/93) in the different business statistics domains for which implementation problems can be observed in many Member States. Currently, most of the Member States assume that legal units are identical to enterprises and thus take a juridical/administrative view instead of an economic view. This is not in line with the organisational development of the economy. Many enterprises consist of more than one legal unit. Moreover, several Member States have indeed implemented the Statistical Unit Regulation, but not in a uniform way. Inconsistencies resulting from different applications of the Statistical Unit Regulation have a considerable impact. The enterprises stored in the national business registers are not harmonised across countries and the resulting business statistics could be distorted. As mainly large entities are concerned, changes to the business statistics due to the harmonised implementation of the Statistical Unit Regulation are expected to be significant.

This option therefore focusses mainly on improving the application of the Statistical Unit Regulation (Regulation No 696/93).

However, given the serious impacts of the non-compliance or non-uniform compliance with the Statistical Unit Regulation and the urgent need to tackle this issue, a decision was taken at the end of 2014 to immediately address the issue and support further Member States' efforts in better complying with the existing Statistical Unit Regulation through national action plans supported by EU grants and by clarifying existing definitions through developing operational rules. These action plans have already been put in place and are supposed to lead to full compliance with the Statistical Unit Regulation and to quality improvements of European business statistics. Therefore, for the time being no further non-legislative measures could be implemented to improve the business statistics. All in all, due to the measures already taken to improve the application of the Statistical Unit Regulation, this option has become obsolete.

5.2.2    Integration of business statistics in a single framework (FRIBS) using a mix of measures and including one option (SIMSTAT) for the modernisation of Intrastat (Option D in the IAA)

This option would implement the simplification (burden and cost reduction) and modernisation as well as respond to additional user needs. In addition, it would allow for the alignment and harmonisation of the collection, compilation, transmission and dissemination of business statistics, with a much more integrated and cross-cutting output for users.

This option also envisages that clearer and better implementable definitions of the statistical units will be included in FRIBS. Regarding the modernisation of intra-EU trade in goods statistics this option only encompasses SIMSTAT.

Thus, this option is characterised by two major shortcomings. Firstly, (as also explained under 5.2.1) due to the serious impacts of the non-compliance with the Statistical Unit Regulation and the urgent need to address the problem, measures have already been taken to address the non-compliance and support further Member States' efforts in better complying with the existing Statistical Unit Regulation. Therefore this option has become obsolete in terms of its objective to include in FRIBS better implementable definitions of the statistical units. Secondly, limiting the modernisation of Intrastat only to the SIMSTAT sub-option is too restrictive. The NSA consultations have demonstrated that different alternatives for the modernisation of Intrastat should be considered, (SIMSTAT, Revised Intrastat, Single Flow, or a combination of them). Consequently, it is better to allow for an analysis covering all three sub-options for the modernisation of Intrastat. Against this background, this option has been discarded.

5.2.3    Implementation of a non-legislative strategy for the rationalisation of the statistical production of business statistics and the reduction of the burden on businesses

This option would focus on the better management of the production of business statistics from the procedural point of view, looking for synergies in the collection and compilation processes, enhancing the access and exploitation of existing administrative sources and promoting the use of modern tools. This option would require the use of agreements between Eurostat and the NSA or any other kind of bilateral or multilateral instruments outside of legislation.

However, given the various pieces of legislation underpinning the European system of business statistics, not many corrective actions could be undertaken without revising one or more elements of the legal texts. For instance, definitions are mainly fixed by legislation, which means that reducing the existing inconsistencies in definitions of certain variables would imply reviewing legislation. This would be contradictory with the non-legislative nature of the policy option under consideration and this is why this option has been discarded. Additionally, the exchange of micro-data for intra-EU trade in goods statistics (Intrastat), which has a considerable potential for burden reduction on EU businesses, could not be undertaken unless a legal framework is put in place. Furthermore, the budgetary provisions for the statistical production of the NSAs are determined on the basis of legislative acts, hence any initiatives which have cost impacts would be difficult to implement without a legal frame.

 6.    IMPACTS OF THE POLICY OPTIONS

Each of the shortlisted policy options presented in Chapter 5 has been assessed on impact of the actions on specific stakeholders (data users, data providers – businesses -, and data compilers). These assessments use on the one hand detailed qualitative information collected from the stakeholders and on the other hand quantitative information collected from data providers and data compilers. Given the fact that all expected benefits e.g. from the improved availability of data could not be monetised, an overall 'net benefit' in monetary terms cannot be obtained for each of the policy options.

Impacts Option A: Baseline scenario: no EU policy change

The baseline scenario foresees a continuation of the current system in terms of legislation, as well as the structure and practices of the collection, compilation, transmission and dissemination of statistical information on Member States' economic activity of businesses. No action is undertaken to improve the degree of consistency of business statistics across the EU and their quality, nor are attempts made to alleviate administrative burden on data providers. The problems encountered in the current situation continue to persist, as the current stovepipe approach is not adapted anymore to the changes in the business environment and changing user needs. Flexibility to more adequately and timely respond to these needs is not improved and it is not possible to use new methods or procedure to fulfil these new user needs.

Based on the results of the stakeholders' consultations, the total cost of the production of the business statistics by the NSAs is estimated to be EUR 290 million annually and the total burden by data providers at EUR 689 million.

Effectiveness of fulfilling the objectives of this policy initiative and impact on stakeholders

This option does not fulfil any of the objectives set for this policy initiative. The problems identified under chapter 2 e.g. the lack of responsiveness of business statistics would remain unresolved.

Efficiency

The combination of domain-specific Regulations and non-integrated production processes contributes to inefficiencies of the business statistics system. The current system maintains overlaps and redundancies in data compilation. The non-integrated system limits the potential for NSI or Eurostat to reduce such overlaps. The current system cannot really eliminate inefficiencies. As there will not be any change to the statistical system, no additional costs (for redevelopment and adaptation of the statistical system) for NSI are expected in the short term on top of the current running costs for maintaining the system. However, possibilities to address current inefficiencies would be limited.

Changes to the sources used for compiling the statistics will be limited as access to administrative sources will remain at its current level and the use of alternative sources such as big data and new procedures such as data linking will not be possible. The burden on the data providers will remain unchanged. The relevance of statistical information will become increasingly obsolete.

Coherence

The baseline scenario is not coherent with the European Statistics Programme 2013-2017 (and its ongoing extension) since the ESP already envisaged a number of modernisations and extensions of data collection in business statistics which would not be able to be put in place without a policy change - under this Option A.

Option B: Implement legislative actions limited to certain business statistics domains

The second policy alternative Option B foresees the amendment of some or all of the current legal acts in force. Hence, current data requirements could be simplified with the aim of reducing the cost of the NSIs and the burden on data providers. Also additional new data requirements can be added for better satisfying the user needs. All this would be organised through updates of the separate legislative acts currently in use for business statistics. Consequently, the production processes at Member State level may continue to be organised by different statistical domains.

However, the streamlining and harmonisation across all business statistics would hardly be achieved. Separate legal texts and separate statistical production processes would provide neither the necessary consistency across statistical domains nor allow for an integration of the collection, compilation, transmission and dissemination of business statistics. In addition, this option would not permit the users to use the statistical output in an integrated way.

Effectiveness of fulfilling the objectives of this policy initiative and impact on stakeholders

This option may fulfil most of the pre-identified objectives. In fact, a series of distinct legislative changes enables streamlining and rationalising the production of business statistics, somewhat reducing unnecessary burden on data providers. These changes could also incorporate the provision of EU-wide common infrastructure tools (e.g. business registers) to increase overall quality, accuracy and comparability of business statistics..

The legislative changes foreseen under this option significantly impact all categories of stakeholders. This option opens up some possibilities for data compilers to use and combine new/innovative methods for data compilation. It also allows for improving the production processes and statistical output quality to a certain extent. The maintenance of the various domain-specific Regulations in fact hampers the possibilities to introduce and fully apply innovative methods, and as such, also has an impact on the efficiency of the underlying statistical processes. The maintenance of fragmented production processes and data collections at the level of Member States counterweights the increased consistency which eventually could be achieved through this option. If Eurostat continues to request data from Member States using the traditional fragmented stovepipes, NSIs will continue to have insufficient resources for modernising their systems.

This option does not reduce the fragmentation of the business statistics system and it will not fully serve the user looking for consistent business statistics. The existing domain-specific Regulations require Eurostat to apply similar or identical concepts, methods and standards across domains for ensuring the quality of the output across the various domains and for increasing its comparability. There is a certain degree of hampering of desired results in view of the fact that coordination of implementation between statistical domains and with other operational objectives is not optimised (which e.g. impacts on opportunities for re-use and linking of data). This is not only a temporary effect but rather a continuous shortcoming. For instance, the removal of inconsistencies in the definitions of variables would create additional possibilities for improved re-use of data. In a non-integrated approach these inconsistencies are not necessarily solved in a sustainable way (e.g. there is a risk they will diverge again after future adaptations of the separate legal acts), so that the re-use of data in the long run is not sufficiently ensured. Unnecessary overlaps and redundancies in the data collections would remain. The continuing legal fragmentation would therefore still hamper Eurostat’s possibilities to ensure the quality of produced data in terms of comparability and coherence. Due to the need to amend domain specific regulations, also this option has considerable limits in responsiveness to changing user needs arising from societal changes.

In addition, any European level future changes to the contents of the data compilation and data output to better satisfy users’ needs will be required to be legally incorporated across the 10 domain-specific legislative acts, through separate legal procedures. This multiplies the efforts and resources needed for preparing and managing these legal processes. For this reason, considerable limits in responsiveness to business and societal needs at the EU level continue to be in place. This policy option therefore does not improve the responsiveness of business statistics to user needs. The national production processes would only be integrated across statistical domains as far the data requirements are kept consistent over possible changes to individual legal acts. All in all, this option might achieve some degree of harmonisation across the 10 different business acts, however this would be a cumbersome and partial solution only.

Efficiency

Assuming that all of the existing legal acts would be amended, current user needs could be addressed in a satisfactory manner. Given that new data requirements would be added, investment costs to adapt the current systems and increases in the initial operational costs would be incurred.. Bringing non-harmonised solutions, this option B would reduce the existing inconsistencies (present under the baseline scenario), but only to a limited extent.

The additional implementation costs for NSAs will be partly offset by simplifications to the data requirements and from the modernisation of Intrastat. The additional costs, incurred by NSAs, for implementing this policy option are estimated to represent an annual increase on average per Member State by:

6.8 % (or EUR 700 000) of the annual operating costs for sub-option B1 (SIMSTAT);

6.7 % (or EUR 691 000) of the annual operating costs for sub-option B2 (Revised Intrastat);

4.3 % (or EUR 436 000) of the annual operating costs for sub-option B3 (Single Flow);

Between 6.7 % (or EUR 691 000) and 6.8 % (or EUR 700 000) of the annual operating costs for sub-option B4 (combination of SIMSTAT and Revised Intrastat)

The estimated one-off investment costs per Member State are:

EUR 1.9 million for sub-option B1 (SIMSTAT);

EUR 1.25 million for sub-option B2 (Revised Intrastat);

EUR 1.16 million for sub-option B3 (Single Flow);

Between EUR 1.25 million and EUR 1.9 million for sub-option B4 (combination of SIMSTAT and Revised Intrastat)

Applying a cost model (See Annex IV) for calculating the net present value of the total EU-28 costs (NSIs' one-off investment costs + NSIs' annual implementation costs) over a 10-year implementation period, the following results have been obtained:

sub-option B1: net cost increase of EUR (+) 103 million,

sub-option B2: net cost increase of EUR (+) 83 million;

sub-option B3: net cost increase of EUR (+) 29 million;

sub-option B4: between minimum net cost increase of EUR (+) 83 million and maximum net cost increase of EUR (+) 103 million.

The new data requirements foreseen would increase the burden on the respondents (data providers). In order to off-set this increase in burden, simplifications would be introduced in the existing requirements of several business statistics regulations. Especially the simplifications or collaborative ways of data production foreseen in statistics on intra-EU trade (Intrastat) would enable the Member States to materialise the untapped potential for burden reduction on data providers. The extent of the burden reductions will depend on the way the final combination of sub-options chosen for the modernisation of Intrastat is implemented by each Member State following the subsidiarity principle. However, under each of those sub-options, the burden reduction would be more than off-setting the burden increases associated with the new data requirements. The burden reductions that may be introduced are the same for Option C, as further elaborated below.

Coherence

This Option B which foresees the implementation of legislative actions limited to certain business statistics domains is in line with the ESP 2013-2017 (and its extension). Namely, the measures planned in the ESP 2013-2017, regarding business statistics, could indeed be introduced by targeted legislative actions in the respective specific business statistics domains.

Option C: Modernisation of business statistics in a single framework (FRIBS) using a mix of measures

The third policy alternative proposes replacing the 10 existing statistical EU legislative acts for business statistics by a common coherent framework, by repealing the existing domain-specific Regulations, simplifying and replacing these with a single legislative package. This package covers a common statistical infrastructure (i.e. the business registers, the EuroGroups Register, the basic data topics, the data sources, the exchange of micro-data for Intrastat, data transmission, data quality, data confidentiality, etc.) which will then create the possibility to integrate the production processes at national and international level. The introduction of measures that remove the stovepipes in the production processes of NSIs in combination with coherent business statistics legislation, will lead to the most comprehensive harmonisation of European business statistics. Similarly to Option B, for the collection of statistics on intra-EU trade (Intrastat), this option C also includes the three alternative sub-options for the modernisation of Intrastat.

This option would implement the simplification and modernisation and create a system which facilitates increased cooperation in statistical production to respond to cross-cutting statistical needs, the reduction of burden for data providers and the minimisation of cost impacts for data compilers. In addition it would allow for better inherent consistency, the integration and harmonisation of the collection, compilation, transmission and dissemination of business statistics, with a much more harmonised and cross-cutting output for users.

Effectiveness of fulfilling the objectives of this policy initiative and impact on stakeholders

This is the only option that enables to fulfil all objectives identified for the policy initiative. The integration of EU legislation on business statistics and measures to enable the better integration of national production processes contribute to a higher responsiveness of the business statistics to changing data needs due to economic and societal changes. When new (cross-domain) issues arise, Eurostat can effectively use the overarching framework legislation instead of changing various domain-specific legislations. At the same time, the production processes for business statistics get more agile for accommodating new cross-domain issues, which will become increasingly important. As the framework Regulation establishes a European Network of Business Registers the central role of the European business registers in business statistics is enhanced making it easier to produce additional statistics by means of data linking. All this responds well to the needs of users and policymakers asking for more flexibility of the business statistics system. In addition, this option allows for the creation of a system which is more forward looking, with best potential for modernisation and collaborative approaches to statistical production and more openness for accommodating new and innovative production methods for statistics.

A reduction of fragmentation in data production processes in Member States creates incentives to use and combine alternative data sources and methods across various domains. The framework legislation will thus contribute to an increasing use of administrative sources and registers as input for business statistics. Such a focus not only reduces the cost of data production but also the response burden of data providers. Many Member States already allow their NSIs to access these sources, and a strengthened EU support could provide further leverage to NSIs to convince their governments to expand the use of these sources. Additional methodological work could be done to assess the quality of administrative or big data sources, to harmonise the concepts used with those used for the production of business statistics and then overall to decrease the production costs.

Option C provides in addition more benefits for users in terms of data quality, transparency, comparability and additional statistics (e.g. through linking). In addition, the effectiveness to ensure the quality of data production and output is maximised. The framework legislation inherently helps Eurostat in increasing the consistency of statistical output across different domains for instance by using the same definition for the same concept. Integrating business statistics in a harmonised data structure also makes the existing links between these statistics more apparent and NSIs will then adapt their production processes to take these links into account and profit from the harmonisation. This framework also facilitates to deal effectively with possible newly arising quality concerns that could come from innovative methods, like the use of new or big data sources. At the same time, the reduction of fragmentation of the production processes in Member States will lead to better data quality as more harmonised standards are used. For new cross-domain data production the quality can be better ensured through the effective integration of EU legislation. 85% of respondents in the first public consultation of data users have indicated that a single harmonised regulation and quality requirements would further increase their already overall high level of confidence of quality in European business statistics. Although improvements in quality of statistical data are highly appreciated by users, they also bring about breaks in time series resulting in problems to use the data in the short term. However, over time these problems will gradually disappear. In addition, some existing data requirements were cut to make room for new data requirements. For users who need these data for their analyses this could pose problems. However, the main stakeholders were consulted on the proposed data requirement simplifications and understood the need to balance the cost and burden as much as possible.

This policy option foresees the introduction of new data requirements. The additional data requirements to be introduced by FRIBS are proposed on the basis of extensive consultation of the main stakeholders and based on the outcome of pilot studies carried out to ascertain the feasibility of the data collection assessing the benefits and the costs. The usefulness of the additional data requirements has been confirmed by the outcome of a second public consultation of data users.

For the different sub-options the impacts on the stakeholders are included in the paragraphs below. The qualitative assessment relies on the SWOT (Strengths/Weaknesses Opportunities/Threats) tool and is based on consultation of the National Statistical Authorities for identifying positive and negative impacts which are expected to follow from the Member States' implementation of the sub-options.

Sub-option C1: SIMSTAT

Using the SIMSTAT sub-option for Intrastat has a substantial positive impact on the quality of the data concerned as the comparability of statistical data across Member States will be enhanced as the exchange of micro-data allows for identifying asymmetries at the micro-data level. In addition, it encourages the harmonisation of production methods across Member States. SIMSTAT implementation will increase the flexibility in the data collection and the relevance of the data thanks to the use of multiple data sources. The availability of the identification number of the importer improves the re-usability of the data. The results of a SIMSTAT pilot study indicate that smaller economies would gain information when using exchanged instead of nationally collected data. Data providers are sensitive to data confidentiality so the security of micro-data exchange between National (Statistical and Other) Authorities must be ensured to address their concerns.

Sub-option C2: Revised Intrastat

Under the Revised Intrastat sub-option Member States would continue compiling their own statistics independently. This option implies flexibility in both collection and compilation methods: if an exports-driven approach is chosen (Member States choose to maintain the coverage rate for intra-EU exports), the impact is expected to have similarities to SIMSTAT because there is dependency on other NSAs. However, as this option does not foresee micro-data exchange the impacts are assessed to be less pronounced than in SIMSTAT. As the share of the estimations increases, quality of data is lower. The current level of trust of data providers over data security will not be affected.

Sub-option C3: Single Flow

The Single Flow sub-option is considered to cause significant quality problems. As only one flow is collected and import statistics are derived from export statistics, it will no longer be possible to check the quality of the import data. Comparability over time will be seriously hampered as there will be significant breaks in series of import data. This could have a serious impact on trade balance and other relevant macro-economic statistics. The reliability and accuracy of imports data will be negatively affected because Member States are not able to control the data sources of the partner countries. It will not be possible to compile statistics on characteristics of businesses engaged in import activities anymore (Trade by enterprise characteristics (TEC)) as imports could not be allocated anymore to individual companies. Negative effects on timeliness of data availability are expected since data on intra-EU imports will exclusively depend on other countries. The current flexibility of the system of collection and compilation will be reduced because all Member States have to collect the same data elements and to apply the same methodology for estimation of imports, without the possibility to explore different data sources. NSAs will be fully dependent on NSAs of partner Member States for the compilation of statistics on intra-EU imports.

Sub-option C4: combination of SIMSTAT and Revised Intrastat

This sub-option would combine elements from both SIMSTAT and from Revised Intrastat. From SIMSTAT, the mandatory exchange of micro-data feature would be taken. While from Revised Intrastat, the key feature taken would be more flexibility in the choice of data sources used, meaning that less data will be collected from companies' imports and exports flows or for either of the two flows. The impact of this combined sub-option should be a reduction of the response burden in Intrastat of at least 25 % which would hence lead to a reduction of at least 13.5% of the overall burden of business statistics (for FRIBS as a whole) with a potential for an even larger reduction of burden. Sub-option C4 would provide a sound level of quality in response to stakeholders’ (i.e. users and respondents) needs. It would also allow for flexibility in data sources and compilation methods used by Member States and would prepare international trade statistics for the challenges brought by globalisation.

Efficiency

The efficiency gains to be made by including all existing business statistical legislation in an overarching framework legislation replacing the currently existing domain-specific Regulations will be substantial. The lengthy procedures currently associated with changing any element in business statistics legislations are inefficient. Streamlining and overall reducing these procedures under one framework legislation will lead to considerable efficiency gains for NSIs and also Eurostat and to fulfilling emerging user needs in a timelier manner. It creates the possibilities to develop more integrated data production processes and to use more innovative and alternative data sources, efficiency gains can also be expected on the response burden on businesses acting as respondents.

Efficiency gains for the business statistics system are possible due to the combination of legal integration and the better integration and modernisation of national production processes. This policy option explicitly opens the possibility for NSIs to modernise their systems and to reduce costs. The integration of data production processes in a more coherent manner is supported by the integration of legal instruments at the EU level. NSIs can use innovative and more cost-efficient methods of data production across the better integrated domains. This will further contribute to an expected reduction of costs for NSIs in the long run and to the reduction of burden on businesses, but it is not possible at this stage to quantify these reductions.

Despite these efficiency gains for the business statistics system for data production and dissemination, NSAs point out that there are cost increases in the short-term due to the necessary investments needed. This is mainly due to changes from heterogeneous processes to more integrated ones using more harmonised standards and methods. The costs for changes towards harmonisation and consistency are to be carried by NSAs, who will have to revise and modernise their statistical processes to accommodate the more consistent requirements. In addition, the new data requirements introduced by the framework Regulation and its implementing/delegated acts to fulfil longstanding user needs will also increase the costs for the data compilers.

In general terms, all Member States will face one-off implementation costs for adapting their national statistical systems to the new data requirements in the short run. The implementation of the modernisation and rationalisation possibilities offered by FRIBS will also entail initial investments before benefits, increased productivity and improved cost efficiency can be reached.

The countries most affected by the changes are those that either do not yet collect the additional data requested by FRIBS for national policy purposes or that have not yet participated in (pilot) actions for preparing the new data requirements financed by the Commission by means of grants. In the short term those countries – representing a minority of the Member States - would indeed face increases in the costs of compiling business statistics.

The FRIBS Regulation is an output-oriented Regulation, meaning that it does not prescribe the Member States which sources (input) need to be used for compiling business statistics, but it enforces the correct implementation of agreed definitions and data requirements for ensuring the comparability of the statistical output. Thus, Member States are free to choose the most cost and burden efficient data sources and production methods to provide the data requested. Due to the efficiency gains from other provisions of FRIBS which do not result in quick cost reductions, but will take more time, these initial cost increases are expected to be compensated in the medium and long run.

A number of provisions are foreseen to alleviate the possible initial cost and burden impacts of FRIBS. For example, the current simplifications exempting smaller countries from delivering all the requested data will be extended to the domains of PRODCOM, business services and global value chain statistics. Additionally, financing is foreseen (within the limits of budget availability) for actions to develop new data collections or to develop modernised data production processes. Furthermore, the provisions for improved access to administrative sources, a new data source for compiling the intra-EU trade in goods statistics at the national level (exchanged micro-data) and consistent definitions allowing for integrated surveys are examples of the rationalisation and modernisation potential offered by FRIBS. This is creating cost reduction possibilities after the initial investment phase.

The calculations for the long term development of costs (see below) were done on the basis of estimates at the European Statistical System level. Due to the lack of factual evidence on how the FRIBS implementation will be done in each Member State, it is impossible to estimate whether the countries with the highest increase in initial costs will be fully compensated by efficiency gains induced by FRIBS in the long term.

The additional implementation costs for NSAs will be partly offset by simplifications to the data requirements and from modernisation of Intrastat. The additional costs, incurred by NSIs, for implementing this policy option (similarly to Option B) are estimated to represent an annual increase on average per Member State by:

6.8 % (or EUR 700 000) of the annual operating costs for sub-option C1 (SIMSTAT);

6.7 % (or EUR 691 000) of the annual operating costs for sub-option C2 (Revised Intrastat);

4.3 % (or EUR 436 000) of the annual operating costs for sub-option C3 (Single Flow);

Between 6.7 % (or EUR 691 000) and 6.8 % (or EUR 700 000) of the annual operating costs for sub-option C4 (combination of SIMSTAT and Revised Intrastat).

The estimated one-off investment costs per Member State are:

EUR 1.9 million for sub-option C1 (SIMSTAT);

EUR 1.25 million for sub-option C2 (Revised Intrastat);

EUR 1.16 million for sub-option C3 (Single Flow);

Between EUR 1.25 million and EUR 1.9 million for sub-option C4 (combination of SIMSTAT and Revised Intrastat)

The results derived from the targeted NSAs' consultation on the additional data requirements show that the changes in annual operating costs do however vary significantly across Member States. For instance, the reported changes in annual operating costs, as forecasted by NSIs during the targeted consultation on the changes to the data requirements, due to the introduction of the new data requirements and deletion of obsolete data requirements range from -1 % to +20 % across Member States. After the initial investment costs and the increases in annual operating costs, considerable efficiency gains should however be reached by NSAs using more integrated processes. These efficiency gains are expected to significantly compensate for the initial investment and annual operating costs incurred in NSAs. A model for estimating the impacts on costs is presented in the box below. The Commission has already contributed around EUR 13 million in the form of grants in order to fund the additional costs that NSIs face in order to prepare and adapt their production systems to the changes introduced by the framework Regulation and will continue to make available financial resources in order to facilitate the implementation of the framework Regulation. It should moreover be noted that part of the increased costs for data compilers is related to actions reducing the burden on data providers.

To summarise, the Option C impacts both the annual operating costs of the production of business statistics and results in initial investments costs (one-off implementation costs) for the NSA. The simplifications and modernisation possibilities included in the Option will allow for a gradual reduction of the running cost and the initial investment costs will be partly borne by the Commission (grants to be given within the constraints of the Commission budget).

Applying a cost model (See Annex IV) for calculating the net present value of the total EU-28 costs (NSIs' one-off investment costs + NSIs' annual implementation costs) over a 10-year implementation period, the following results have been obtained:

sub-option C1: net cost increase of EUR (+) 9 million,

sub-option C2: net cost savings of EUR (-) 10 million;

sub-option C3: net cost savings of EUR (-) 62 million;

sub-option C4: between maximum net cost savings of EUR (–) 10 million and maximum net cost increase of EUR (+) 9 million.

The new data requirements introduced to fulfil longstanding user needs will increase the burden on data providers in the same way as for option B. The introduction of the new data requirements would amount to a burden increase of about EUR 24 million. This increase will be compensated by the simplifications and rationalisations foreseen in the different domains and by the modernisation of Intrastat.

Based on information collected from the respondents of the surveys regarding the administrative burden, the burden reduction which could be realised would depend on the approach chosen for the intra-EU trade data. Notwithstanding that the final approach chosen for the intra-EU trade will try and maximise the burden reduction while guaranteeing the quality and especially the timely availability of the statistics on intra-EU trade, the total potential burden reduction for FRIBS (for the EU on total) is estimated to be:

26.8 %     (or EUR 184 million ) for sub-option C1 (SIMSTAT);

12.1 %     (or EUR 83 million) for sub-option C2 (Revised Intrastat);

33.6 % (or 231 million) for sub-option C3 (Single Flow);

13.5 % (or EUR 93 million), at least, for sub-option C4 (combination of SIMSTAT and Revised Intrastat)

Some of the additional data requirements covered by the FRIBS Regulation would increase the burden on SMEs especially with regards to the extension of business statistics to cover more services than were previously not covered. This is, however, to respond to a long-standing specific user need to enable the monitoring of European and national policy actions with regards to SMEs for which sufficient information is currently lacking. The data compilers are moreover encouraged to use as much as possible existing administrative data (e.g. from tax authorities) to respond to this data request and to minimise the use of surveys and therefore limit the burden on SMEs.

The simplifications foreseen by FRIBS in the area of intra-EU trade statistics, on the other hand, could reduce the burden for SMEs. However, for intra-EU trade statistics the determining factor whether to be obliged to report or not is not the size of the company but the trade volume. Though usually one can expect to have some sort of correlation between the two, small businesses can have large trade volumes while large companies have very little or no trade at all. The intra-EU trade statistics system however benefits from the availability of VAT data on intra-EU sales and purchases which allows exempting the smallest traders from intra-EU trade statistics reporting.

The ESSC agreed on 18 May 2016 on the principle of burden neutrality on exporters. This could mean in practice that additional burden which exporters would face due to reporting of the new data elements needed for exchanging the micro-data could be compensated by lowering the overall minimum collection rate (threshold) of exports. Based on estimations, the burden neutrality could be ensured by lowering the collection rate from 97% to 95% of the total exports. In this case the exporters who would be exempted from reporting would be the smallest ones of which the majority can be expected to be SMEs. In addition, any reduction of the imports data collection would mostly be in favour for the SMEs so the overall impact on the SMEs should be very positive. However, the system will be all-encompassing; there might always be individual SMEs who may not benefit from burden reductions and may even face more burden.

Nevertheless, any possible burden increases on SMEs would be kept limited also because many countries will use administrative data for fulfilling new data requirements, in some cases combined with model-based estimation methods. In addition, some of the new data requirements (e.g. regarding Global Value Chains) are specifically directed to large enterprises (there is a threshold in terms of employment) and thus the SMEs are protected from excessive burden. Additionally, for the new data requirements for which surveys have to be organised, the practice to cover only a sample of small enterprises will also be applied (whereas large enterprises are covered exhaustively). What is more, the replies from the targeted consultation of the data compilers on the additional data requirements introduced by FRIBS indicated that the data compilers would make particular efforts to minimise the burden on SMEs for those new data requirements that are more burdensome on respondents.

Coherence

Option C is fully coherent with the ESP 2013-2017 and its extension to 2020. Modernisation of business statistics in a single framework (FRIBS) would fully and consistently address the measures and novelties envisaged for European business statistics by the ESP and its extension.

7.    COMPARISON OF OPTIONS

In this chapter the proposed three policy options are compared. This will be done using the thorough impact analysis as outlined in the previous chapter. The following table gives a summary overview of the impacts of the different policy options on the different stakeholders.

-- negative impact                - slightly negative impact            0 neutral impact            + slightly positive impact            ++ positive impact

Option A is keeping the status quo. This option does not look acceptable as the need for modernising European Business statistics was already recognised in 2008, mainly as the current business statistics serve the policy and other user needs only partially. Since then investments have been made to accelerate the modernisation of the system. These investments would be lost if the current system of European Business Statistics is maintained. Policy users would be increasingly dissatisfied with the data disseminated now and turn to other data sources. Over time, this would deteriorate the quality of European or national policy decisions. What is more, this option is not coherent with the ESP 2013-2017 and its extension (under preparation) because the modernisation and modifications which the ESP envisages would not be possible without undertaking legislative measures in the area of business statistics. That is why keeping the status quo, would even negatively affect coherence with other legal initiatives (see comparison table below).

Option B tackles the modernisation of the current system of European Business Statistics to a certain extent. This is in particular true for the upgrading of the data production and data output for policy and other users which will get more relevant. It is also true for the modernisation of the Intrastat system which should lower the burden on data provider. However this option leaves a series of current deficiencies unsolved. These are:

Keeping 10 separate legislative acts for Business statistics means that more efforts are needed to reduce inconsistencies and to preserve the coherence of data and indicators produced for policy users in the future. In view of the risks of non-optimisation of the coordination of implementation of the different legal acts, indicators produced on the basis of such a dispersed system would not be relevant enough and misguide policy users.

Secondly keeping 10 separate legislative acts also creates a high work load for the management and updates of these pieces of legislation; this affects NSAs, but also data providers (enterprises) who might need to answer to fast changing or sometimes inconsistent questionnaires.

In addition, the improvement actions proposed in this option do not really create more agility and responsiveness to changing policy and other user needs. Heavy procedures and long-lasting discussions do prevent more agility of the system and therefore hamper the relevance of the output for users.

In terms of coherence Option B is an improvement to the baseline scenario, since the measures planned in the ESP 2013-2017 and its extension could indeed be introduced by targeted legislative actions in the corresponding specific business statistics domains.

Option C as outlined above is the most advanced and most forward-looking one as it modernises the European business statistics system. This option covers a long-term investment in European business statistics in making them fit for the future. All European business statistics will be put into one single legal framework FRIBS. This has – compared to the previous options – a series of clear advantages. These are:

The assembling of 10 separate legislative acts including the Business Register Regulation into one single legal framework will implicitly guarantee much higher consistency of business statistics (e.g. in terms of timing of changes and harmonisation of definitions etc.). This allows for the drawing of higher benefits from the ESS system as a whole and maximises the value added of the EU at the same time. The frontiers between the various statistical domains will diminish or even completely disappear. This alignment will make it possible to serve the policy needs much better as more streamlined indicators and combinations of indicators can be disseminated. As an example, - contrary to the current situation - there will only be one single figure for the variable "number of active enterprises" for a certain industry which will be used all across the system.

The streamlining of all business statistics in FRIBS and the underlying FRIBS legal architecture will increase the agility of the system. New policy needs can be accommodated with rather short delays and be embedded in a functioning and well-designed system. This assures a level of agility which is not really possible under the two previous options.

In contrast with Option B the costs for adaptations of the statistics disseminated and of the underlying legal framework are minimised under this preferred Option C as all necessary revisions are introduced more easily in one go. The framework Regulation contains only the essential statistical elements. Non-essential elements can be supplemented or amended by delegated acts, while the technical provisions will be part of implementing acts. This brings much more agility and responsiveness to the system.

Most importantly Option C has – by far - the highest potential for reducing the burden on enterprises. Better harmonisation of business statistics will also streamline the data collection and production methods used by NSAs or also by Eurostat. For instance the "employment" figures will be asked from enterprises only once and then be reused for many different indicators made available to policy users. In many Member States the number and coverage of questions to which enterprises need to respond should considerably decrease over time. FRIBS encourages that more harmonised methods and processes are used by NSAs with the effect of a further decrease of burden on enterprises. Thus, the increased coordination and use of multipurpose methods (administrative data sources, combining of survey questionnaires, use of "big data" data sources etc.) will create better opportunities for the reduction of data production costs. FRIBS also allows for enhanced use of micro-data exchange in various domains in the future and better and more aligned output facilitating e.g. new indicator production.

Regarding coherence, Option C is fully in line with the ESP 2013-2017 and its extension and is therefore clearly superior to Option A. Only the integrated approach of Option C (FRIBS) could completely and coherently implement the actions and modernisations envisaged for business statistics by the ESP and its extension.

The table below summarises the scoring of each of the policy options regarding:

- the extent to which different options would achieve the objectives (effectiveness);

- the relation between costs and benefits;

- the coherence with the overarching objectives of EU policies.

-- negative impact     - slightly negative impact    0 neutral impact     + slightly positive impact ++ positive impact

To conclude, all these advantages clearly favour Option C. It responds best to the objectives of the REFIT programme by simplifying and streamlining the heterogeneous and inconsistent legal texts currently governing business statistics into one coherent legal framework. It is based on extensive user consultations and has the highest potential for reducing regulatory burden. This option makes the European business statistics system fit for the future in meeting much better the policy user needs compared to the rather inflexible system of today, while being also best placed to introduce the innovations envisaged for business statistics in the ESP and its extension. This option provides more possibilities for burden reduction as explained above. Option C is the only option which takes fully advantage of the EU in maximising its value added; this option creates the platform for a more collaborative and agile system allowing for further modernisation in the future. The increased cost for data compilers in the beginning of the implementation would be counterbalanced by potential benefits and efficiency gains that would be realised once the FRIBS measures start to take effect (see cost model in Annex IV). What is more, these increase costs would be completely offset by the advantages for data users and the decrease in the burden for the data providers which Option C would allow for. Moreover, the results of the first consultation of the NSAs clearly showed that Option C - FRIBS was their preferred option. Namely, around 55 % of the NSIs indicated FRIBS as their preferred option, while only around 10 % preferred Option B - the non-integrated approach of separate revisions of the individual acts and only 7 % of the respondents preferred Option A - keeping the status quo (no policy change ) 13 .

8.    MONITORING AND EVALUATION: HOW WOULD ACTUAL IMPACTS BE MONITORED AND EVALUATED

The European Commission REFIT agenda makes evaluation of all new legislative measures a priority for the Commission. The proposed FRIBS legislation will also be subject to a complete evaluation in order to assess, amongst other things, how effective and efficient it has been in terms of achieving the objectives presented in this report and to decide on whether new measures or amendments are needed.

It is not expected that the implementation of the FRIBS framework Regulation and its accompanying delegated and implementing acts would be potentially more difficult in certain Member States.

In general the existing monitoring and evaluation tools, in place and valid for the statistical production and dissemination of European statistics, will be used. These should enable a profound analysis of the effectiveness and efficiency of the new statistical initiative and of the quality of the data produced. These tools are:

The European Statistical Programmes (currently based on Regulation (EU) 99/2013) foresee systematically mid-term and final evaluations. Business statistics are part of this reporting mechanism 14 .

The Eurostat Management Plan foresees the follow up of key performance indicators, which also apply to business statistics 15 .

User satisfaction surveys are carried out on an annual basis 16 .

FRIBS also requires the production of standard quality reports, regularly produced by Member States and Eurostat, as part of the Statistical Quality Assurance Framework.

Eurostat and the NSAs will further improve the standard metadata and quality reporting system for business statistics. This will allow a more sophisticated monitoring and evaluation of the statistical processes used in Member states and of the output disseminated. For example, more detailed information will be available on the use of administrative data sources (leading to burden saving) by Member States or on the use of shared services or IT tools (leading to cost savings). The results and the effect of the FRIBS legislative initiative would regularly be monitored via the annual compliance and quality reports.

Measuring the progress towards achieving the objectives of the initiative the following list of monitoring indicators has been defined. Those indicators are juxtaposed against a list of operational objectives which have been derived from the specific objectives (presented in Chapter 4). The indicators would be measured against the benchmark targets indicated in the last column of the table below.

Table 5. Monitoring and evaluation indicators

(*) FRIBS does not impose a redesign of the national production systems to reduce the burden on data providers but makes it possible for the Member States to reduce their number of surveys, to use innovative or additional administrative sources or to reduce the burden on the data providers.

ANNEXES

ANNEX I: PROCEDURAL INFORMATION 

Basic information:

Title of the initiative: Framework Regulation Integrating Business Statistics (FRIBS)

Lead DG: Eurostat

Agenda Planning Number: 2012/ESTAT/011

In its Communication on EU Regulatory Fitness of December 2012 the Commission committed to strengthen its various smart regulation tools (impact assessment, evaluation, stakeholder consultation) and launched the REFIT programme. Through REFIT, the Commission services have mapped the entire EU legislative stock looking to identify burdens, gaps and inefficient or ineffective measures including possibilities for simplification or repeal. Thus, the Framework regulation integrating business statistics (FRIBS) falls under the scope of the REFIT initiative. Under this context, Eurostat started the preparation of this impact assessment in 2013 with a view to comparing the impacts of different possible options for integrating and rationalising business statistics. As this was prior to the adoption of the Better Regulation guidelines in 2015, no specific full evaluation of the current situation was done to support this initiative. However, the regular evaluation mechanism in place at Eurostat (e.g. quality reports, rolling reviews, evaluation of the European statistical programme, management plans) as well as an ex-post analysis of the current situation (through discussions with NSIs, interviews with main users, public consultation collecting feedback from general users and thorough desk research) provided a broad basis for the Impact Assessment.

A FRIBS roadmap has been agreed and published in January 2013. A revised roadmap has been prepared and published in January 2015 in order to reflect the latest state of play of the FRIBS preparation. Following the introduction of the Better Regulation guidelines in 2015, the roadmap has once again been revised under the form of an inception impact assessment that was published in January 2016.

Consultation of Impact Assessment Steering Group (IASG)

An inter-service steering group was set up in 2013 involving the following DGs: Secretariat-General (SG), Competition (COMP), Economic and Financial Affairs (ECFIN), Employment, Social Affairs and Inclusion (EMPL), Internal Market, Industry, Entrepreneurship and SMEs (GROW), Communications Networks, Content & Technology (CNECT), Trade (TRADE), Taxation and Customs Union (TAXUD), Research and Innovation (RTD) and Legal Service (SJ). The inter-service steering group met 5 times (the meetings took place respectively in October 2013, May 2014, the June 2015, and twice in April 2016). Following the results of the latest meeting of the IASG, a revised version of the Impact Assessment Report was prepared and sent to the Regulatory Scrutiny Board.

Consultation of Regulatory Scrutiny Board (RSB)

The RSB examined the Impact Assessment Report on 8 June 2016. The RSB gave a positive opinion and provided recommendations for further improvements. The report was further adjusted in order to integrate the Board's recommendations.

Evidence used

In reference to the evidence that constitutes the basis for the Impact assessment, it is first pertinent to mention the evaluation tools currently in place that already allow for good analysis of the effectiveness and the efficiency of new statistical initiatives and the quality of the data produced. These tools include: the European Statistical programmes (currently regulation 99/2013), which foresee systematic mid-term and final evaluation of the programme, of which business statistics represent an integral part 18 ; the follow-up of a series of performance indicators, as planned by the Eurostat management, which also applies to business statistics 19 ; and the Users satisfaction surveys carried out on a regular basis 20 .In addition, the ongoing assessment of the quality of the statistics undertaken by Eurostat, presented in the form of Quality reports 21 ,represent a source of extensive and reliable evidence for the preparation of the impact assessment and at the same time a tool for monitoring and evaluating the impacts of the new legislative proposal.

Systematic Rolling Reviews, in accordance with the binding standards approved by the Commission, were also carried out. They formed part of the Quality Assurance Framework developed by Eurostat in 2007 22 .

Further evidence for this impact assessment was collected from the consultations of the relevant stakeholders –Directorate Generals of the European Commission, National Statistics Institutes (NSIs) and National Central Banks (where relevant), and businesses and their associations and the public (i.e. academics/researches, interest groups, media and non-governmental organisation, and individuals) –conducted as part of the preparation of this impact assessment. The combination of the information provided from the data users –from within (DGs and EU agencies) and outside of the EU administration (e.g. external micro-data users); from data providers (businesses); and from the data compilers (National Statistics Institutes) and extensiveness of the consultations ensures the robustness and the relevance of the information collected. An additional aspect that contributes to the robustness and comprehensiveness of the evidence gathered thorough the stakeholders' consultation is the use of several modes of data collections (i.e. the use online questionnaires addressed to the open public, the data users and data providers, the targeted consultation of data compilers, in combination with in-depth discussions to data users and data producers), which broadened the scope of the exercise, and allowed to undergo a more detailed analysis of the problems at hand, and a thorough assessment of the possible solutions from the perspectives of the different stakeholders.

External expertise used

The inter-service group (ISG) created in the context of this impact assessment, in which the above-mentioned DGs were represented, constituted –besides its already established function of steering the IA process and contributing to the collective preparation of the IA report, and in relation to the fact that the DGs are main data users of the business statistics provided by Eurostat–, a most relevant source of external expertise for the preparation the Impact assessment of new legislation.

Besides the expertise and the advice sought from the inter-service steering group of the DGs, also the assistance of two external contractors has been used. Namely, the contractors assisted in executing the stakeholders' consultations, summarising the main results and analysing the evidence gathered.

The expertise of the NSIs has also been profoundly relied upon. Namely, there have been extensive discussions and exchange of views as part of regular domain-specific working groups, task-forces, directors groups and also informally through various bilateral consultations.

ANNEX II: STAKEHOLDER CONSULTATIONS

1. Introduction

Besides the regular and extensive consultations of the concerned parties (more than 100 meetings in total with Member states, EFTA countries and others) during the development of the FRIBS proposal (including the problem definition and the need for a EU rather than a national solution (subsidiarity)), three rounds of specific stakeholder consultations were organised for the preparation of the FRIBS Impact Assessment. The first round covered the FRIBS infrastructural elements (such as the Business Registers, micro-data exchange, quality issues and confidentiality). The second round focussed on the changes to the data requirements to be introduced by FRIBS and the third round collected stakeholders' opinions on the modernisation of Intrastat. Each round consisted of a targeted consultation of the data compilers (NSAs) 23 and an open public consultation also aiming at collecting feedback from the data providers (businesses) 24 and data users 25 . . The consultations meet the minimum standards of the European Commission for the consultation of interested parties. This Annex will first give an overview of the different consultations undertaken. In addition, the outcomes of the respective consultations are summarised.

1. 1. The first round of consultations on the FRIBS infrastructural elements 26

The consultation on the FRIBS infrastructural elements took place in 2014 In the targeted consultation of the NSAs, the data compilers were in particular consulted on the expected qualitative impacts of FRIBS, the qualitative comparison of FRIBS and alternative policy options, the expected quantitative impacts of FRIBS and alternative policy options and the expected quantitative impacts of FRIBS and alternative policy options for data providers.

29 of the 32 NSAs which were invited to participate in the targeted consultation of data compilers provided their contribution. 27 EU Member States and 2 EFTA countries finally responded.

In total, 54 respondents answered to the public consultation of data users and providers on the expected qualitative impacts of FRIBS and the qualitative comparison of FRIBS and alternative policy options. Professional associations were responding most, followed by academic and research institutes and by business themselves. Most respondents were users of European business statistics, only a few were providers of statistical data (i.e. enterprises). The remaining responses came from organisations linked to national and/or European institutions.

1.2. The second round of consultations on the FRIBS additional data requirements

During the Business Statistics Directors Group (BSDG) meeting in 12/2014, Member States asked for an additional consultation on the impacts of the additional data requirements in FRIBS. Therefore, this second consultation was organised in 2015 assessing the impact of the data requirements as changed by FRIBS (both for increases and reductions). This second round of consultations complements the first cost-benefit analysis of the infrastructural elements as described under section 1.

This second consultation phase also aimed at receiving input data compilers, data providers and data users. The data users were asked to describe the qualitative impacts of the proposed changes (quality, flexibility, timeliness),

The data compilers were asked to assess the one-off implementation costs and changes in yearly operating costs as well as the estimated changes in administrative burden for data providers caused by the changes in the data requirements of FRIBS.

All 28 NSIs from EU Member States and 1 NSI from EFTA countries answered to this consultation. In addition 17 replies came from NCBs (9 of those were joint replies from the NCB and the NSI).

For this round of public consultation separate questionnaires were addressed to data providers and data users respectively. The data users were asked to describe the qualitative impacts of the proposed changes (quality, flexibility, timeliness) and the data providers to assess the impact on administrative burden ( qualitative assessment) caused by the changes in the proposed data requirements of FRIBS.

52 data users answered; many replies came from National Accounts departments and business associations. In addition EU institutions, research institutes, public authorities' individual businesses and central banks replied. In spite of many extensive efforts from Eurostat, only a limited number of replies from data provides were received.

1.3. The third round of stakeholder consultations related to Intrastat

Separate targeted and open public stakeholder consultations have been carried out on Intrastat feeding into the impact assessment as well as to the modernisation of Intrastat. The overall aim of the modernisation project is to assess the impacts of the three options proposed – SIMSTAT, Revised Intrastat and Single Flow - in terms of their costs and benefits.

The NSAs have been addressed in a targeted consultation for collecting data for a cost-benefit analysis. This cost-benefit analysis aims at providing input to the ESSC to take a strategic orientation on the modernisation of Intrastat. At the same time the cost-benefit analysis is contributing to the more encompassing cost-benefit analysis of FRIBS.

The methodology followed for the cost-benefit analysis consists of two actions: quantitative assessment (cost analysis) and qualitative assessment (SWOT analysis).

The aim of the cost analysis was to estimate the costs (current costs, development and adaptation costs, future costs) incurred by the NSAs caused by the implementation of modernisation options for Intrastat. 24 Member States provided data for all three types of costs. Eurostat finally estimated the complete EU costs for the NSAs.

The SWOT assessment for the Member States was carried out in autumn 2015 based on a standard methodology. 26 Member States provided their SWOT assessments, with some updates in 2016.

For collecting data on the administrative burden, an extensive open public consultation of the data providers has been organised during the first quarter of 2016. More than 20.000 replies have been collected.

2. Outcome of the consultations

The three rounds of consultations have been summarised in this section by category of stakeholder in order to provide a comprehensive picture, given that the consultations have been complementary.

2.1.    Data users

In the first round, 48 users responded to the consultation, in particular reporting the following issues:

1.Combining data from different business statistical domains: many problems with inconsistencies between statistical domains exist.

2.Consistency of statistical data across Member States: many respondents believe that inconsistencies exist and that those have a negative impact on both the quality of their work and the time required for exploiting the data. Around 80% of respondents consider that an increased coherence and consistency of data would significantly improve both, the quality of their work and the time needed for exploiting the data.

3.Additional data needs: The majority of users judge the extension of the FRIBS data requirements positive.

4.Confidence of users in the European business statistics system: nearly 80% of the respondents are very/rather confident in the quality of the current European business statistics;

5.Nearly 85% of the respondents have indicated that a single harmonised regulation would further increase their confidence in European business statistics.

The second consultation focused on the FRIBS additional data requirements (i.e. new data requirements as well as deletion of some existing requirements). 52 data users responded to the consultation, covering EU institutions and research institutes (28%), National Accounts departments (25%), business associations (23%) and other public authorities, individual businesses or central banks (22%).

Around 50% of the data users emphasised that the FRIBS additional data requirements are crucial or highly important for them. Crucial new requirements are in particular indicators on globalisation and global value chains of enterprises in Structural Business Statistics and in FATS statistics; and the improved coverage of the services sector, especially the introduction of a monthly production index for services in Short-Term Statistics and the level of detail of Structural Business Statistics available for National Accounts.

2.2. The data compilers (NSAs)

The results presented here summarize the information collected on costs using the three targeted consultations of the data compilers. The NSAs that provided details regarding the current costs, caused by European business statistics, report total yearly costs of about EUR 290 million. This total translates into an average amount of EUR 10.5 million per Member State ( 27 ).

The total yearly costs caused by European business statistics can – as an average - be distributed as follows over the different statistical domains and the Business Registers (BR) 28 .

On average, the current annual costs are EUR 0.57 per EU inhabitant. An increase in yearly operating costs of NSAs between 4.3 % (EUR 436 000) and 6.8 % (EUR 700 000) (depending on the different Intrastat sub-options consulted and analysed) is expected due to FRIBS implementation as a whole, encompassing the infrastructural changes and the upgrade in the FRIBS data requirements.

FRIBS would also cause one-off investment costs, on average per Member State between EUR 1.16 million and EUR 1.9 million (or between EUR 32 million and EUR 52 million for the EU on total) depending on the different Intrastat sub-options consulted and analysed.

These increases (operational cost increase and one-off implementation cost) represent additional costs between EUR 0.08 and 0.14 per EU inhabitant. Cost increases are caused by all statistical domains, except International Trade in Goods Statistics (ITGS). The new data requirements for STS and SBS statistics cause most of the additional costs. For the ITGS, a reduction of costs through a modernised Intrastat system can be expected (between EUR 8 million and EUR 18 million depending on the different Intrastat sub-options consulted and analysed).

During the preparatory phase of the FRIBS, the Commission (Eurostat) provided a total of EUR 37 million, mostly to NSIs, through the MEETS programme on the modernisation of enterprise and trade statistics in addition to other pilot studies financed and carried out to test the feasibility of new data requirements. Eurostat aims to provide financing ( within the Commission budgetary constraints) to offset some of the investments costs needed for the further modernisation of the data production processes and systems as well as for capacity building and piloting of data production related to new data requirements. It can also be expected that once the initial investments have been made the yearly operating costs for the NSA would gradually decrease when multi-source and multi-purpose data collections and access to alternative data sources – administrative or collaboratively generated by the ESS (i.e. EGR and micro-data exchange) – will be increasingly used in many Member States.

Regarding the preferred policy option, the results of the first targeted consultation of the NSAs show that around 55 % of the NSIs indicated FRIBS as their preferred option, around 10 % preferred Option B - the non-integrated approach of separate revisions of the individual acts and 7 % of the respondents preferred Option A - keeping the status quo (no policy change). Same ranking of the policy options was indicated by the data users during the open public consultation. The results however depict even greater support for FRIBS - Option C with 74 % of the respondents supporting it, followed by Option B (non-integrated revision of separate legislative acts) with 13 % and finally Option A – no policy change was supported by 4 % of the respondents.

2.3.    Data providers (businesses)

In spite of considerable efforts from Eurostat to advertise and encourage the participation of data users in the open public consultation, only a limited number of replies by data providers were received on qualitative impacts in the first two rounds of the public consultations. Consequently, the responses could not be considered representative and not analysed further. Namely, during the first open public consultation only 6 partial responses (from 1 micro-enterprise, 1 small enterprise, 1 large company and 3 professional associations, all originating from different EU countries) were recorded. During the second consultation round (on the additional data requirements) only 5 replies were received. These came from: two individual businesses; a federation of businesses; a chamber of commerce, and a national statistical office.

From the targeted consultations on FRIBS infrastructural elements and changed data requirements (other than intra-EU trade statistics), information on the current burden and on changes in the burden for data providers (businesses) were estimated by the NSAs. For most statistical domains, between 26 and 28 NSAs provided estimates regarding the costs for businesses obliged to fill in business statistics surveys. Information on the administrative burden caused by the intra-EU trade statistics and the effects of its modernisation was collected by a public consultation of data providers (businesses). When the inputs from the consultation on businesses regarding the burden of the collection of Intrastat data is also taken into account the total administrative burden of business statistics is estimated to be around EUR 689 million.

Three statistical domains cause around 89% of this total burden:

67 % of the burden on businesses is caused by statistics on International Trade in Goods. Structural Business Statistics and Short Term Statistics are ranking second and third. The potential reduction of the total burden on businesses will depend on the level of national implementation of the simplifications and modernisation possibilities provided by FRIBS and is estimated to be:

26.8 %     (or EUR 184 million) for sub-option C1 (SIMSTAT);

12.1 %     (or EUR 83 million) for sub-option C2 (Revised Intrastat);

33.6 %    (or EUR 231 million) for sub-option C3 (Single Flow);

13.5 % (or EUR 93 million), at least, for sub-option C4 (combination of SIMSTAT and Revised Intrastat).

ANNEX III: WHO IS AFFECTED BY THE INITIATIVE AND HOW

The production and use of European business statistics involves 3 main categories of stakeholders:

•Data users: this includes institutional users –national administrations, other international organisations, or professional staff working within the Commission services or DGs and other EU institutions, and other external users, such as mass media and academics.

•Data compilers: this category comprises the authorities responsible for the collection and compilation of business statistics. It mainly includes the NSIs and Eurostat.

•Data providers: this category includes businesses which respond to surveys

The preferred policy option has a significant impact on each of the stakeholder categories.

1. Data users of European business statistics

Currently the possibilities for users to analyse data across domains of business statistics are hampered by differences in concepts, definitions and classifications. Data users will benefit of an increased coherence of the business statistics. The use of common concepts and definitions, classifications and breakdowns, will improve the quality perceived by the users of statistics.

The preferred policy option allows for greater flexibility in the statistical work programme by providing for improved mechanisms to accommodate emerging information needs. Under the preferred policy option, the system will increase flexibility as well as harmonise business statistics; this has the double advantage of covering new information needs by producing new data and by combining them with those already collected.

The producers of macro-economic accounts will benefit from the increased consistency as well as from the improved quality of the business statistics which are used as critical inputs for the calculation of main economic indicators such as the GDP.

To benefit of greater flexibility, users have to closely work with statistical authorities to achieve proportionality in the satisfaction of information needs and production efforts. This implies ensuring an effective mechanism for the identification of negative priorities. At the European Commission level, the institutional users (mainly (DGs) will continue working closely with Eurostat on defining their needs and will develop the mechanisms for budget delegations. This is normally formalised through Memorandum of Understanding between each user DG and Eurostat and the annual formal "hearing" with each DG that supports the definition of the annual and multi-annual statistical programmes.

2. Data compilers, the National Statistical Authorities

The production of European business statistics is mostly done at national level by NSIs and NCBs, which are public institutions and therefore with public budgets. Additional requirements of information lead to the need for higher budgets, unless measures to significantly increase efficiency are taken.

The cost of accommodating new information needs includes new methodological work, new data collection, new processing or dissemination methods. The bulk of costs go to data collection. The increased information requirements foreseen in the preferred option necessitate an assessment at the national level on the most cost-efficient way of data production, but an initial investment - irrespective of the choice on the production method- can be expected. As a consequence, the preferred option also foresees provisions which allow for the modernisation of production (e.g. using more administrative data, better technology or sharing methodological solutions among NSIs) and by data sharing ( as foreseen by the EGR and the modernisation of Intrastat) and collaborative approaches to data compilation.

National production processes would benefit from a network of integrated business registers, from the improved access to administrative and other data sources as well as from the possibility to use micro-data collected by other national statistical authorities. More integrated processes should also lead to increased cooperation between the national statistical authorities concerned and facilitate long term overall cost savings for them.

Carrying out new surveys or increasing the frequency of data collections has a significant impact on cost. Unless constraints on the rising cost are put in place, policy options increasing the flexibility may highly impact the budgetary needs of NSIs.

It has to be mentioned that the most expensive surveys are those carried out with large samples or with high frequency (e.g. STS). Policy options that consider mechanisms for (1) evaluating the relevance of additional information requirements and (2) simplifying the work programmes by excluding the data collections which are not deemed relevant anymore (negative priorities) would provide safeguards for increases in cost.

NSIs would however benefit of a simpler, integrated legislation that would allow for reducing the legal work to update the statistical regulations, that would consider the whole system of business statistics when planning and designing each data collection and that would promote the modernisation of processes (e.g. by facilitating access to administrative data and other sources) and increase the methodological coherence across surveys.

In practice, the joint work of Eurostat and the NSIs to integrate business statistics will continue with the elaboration of the necessary implementing measures. The role of the BSDG will thus be reinforced as the main technical forum of discussion on business statistics, advising the ESSC and the Commission. The balance of costs and benefits, in financial terms, is difficult to assess since it depends on the actual implementation of changes in existing data collections. In some Member States such collections existed or have already been implemented. Progress in the harmonised and comparable measurement of production costs in NSIs would not only improve the management at the national level, but also provide a sound basis for monitoring and evaluating the impact of any modernisation process – be it or not related to legal obligations.

3. Data providers, the businesses

While some of the new information needs proposed by the preferred policy option could be accommodated using administrative sources or data sharing, there will be impacts also on data providers. The increased coverage of the services sector and the increased frequency of some of the short-term statistics to be produced may create more burden on the respondents. However, this potential increase is expected to be more than offset by the possibility to re-use data collected by other NSAs for intra-EU trade in goods statistics and by modernising the trade in goods statistics. The SIMSTAT exchange of micro-data has the potential to reduce the overall burden created by business statistics by about 10 %. In addition, a reduction of the coverage rates (of imports and exports) for intra-EU trade in goods statistics , foreseen in the Revised Intrastat sub-option would also relieve some businesses from the burden to provide data for intra-EU trade in goods statistics.

Impact on SMEs:

The additional data requirements might result in additional burden for SMEs especially as regards the extended information on the services sector. In the qualitative consultation responses provided by the data compilers the NSAs have assured that the utmost care is taken to ensure that the burden on the SMEs is as limited as possible by e.g. the use of administrative data or advanced statistical methods. The modernisation of Intrastat as mentioned above would also reduce the burden of SMEs.

ANNEX IV: ANALYTICAL MODELS USED IN PREPARING THE IMPACT ASSESSMENT

The impact assessment has been based on a cost-benefit analyses of: on one hand the REDESIGN of Intrastat (i.e. the options for modernising the collection of Intrastat statistics) and on the other hand, all the envisaged modifications in all other statistical domains which FRIBS is covering. Those two cost-benefit analyses have been combined with the goal to evaluate and compare in detail the ability of the policy options to achieve the objectives, identified for solving the current problems. This cost-benefit analysis has utilised as much as possible the use of quantitative data, having in mind a principle of proportionality. For the assessment of some of the impacts quantitative data could however not be collected and for these elements qualitative elements have been collected. In summary, the following assessments are made for each of the shortlisted policy options:

• qualitative assessment of their impacts;

• quantitative assessment of:    expected costs for NSAs; expected costs for data providers (administrative burden);

The analysis thus implies that the outcome of the comparison of quantified costs (in Euros) and benefits is confronted with the qualitative assessment of other advantages and disadvantages.

Common to all approaches presented is the fact that in an impact assessment, impacts are typically to be expressed as an expected effect in comparison to what would happen under the no policy change option. As a consequence, the cost-benefit analysis will provide estimations of the additional costs and additional benefits compared to the baseline scenario. Indeed, the main objective of the impact assessment is to enable comparison between the options, which implies that expressing the correct relations between values and options (relative relationships) are of higher importance than determining the total cost/benefit of each option. An effort has however been made to quantify the baseline scenario. There was a lack of existing harmonised and comprehensible data. A targeted data collection was necessary.

The cost-benefit analysis brings together the comparative results of the different approaches, so as to allow an integrated comparison over all types of impacts of the different policy options. Namely:

- qualitative assessment of the impacts of each policy option;

- assessment of the costs for NSIs;

- assessment of the costs for data providers

Qualitative assessment of the impacts of each policy option

One of the cornerstones of a detailed comparative assessment of the policy options is a qualitative assessment of the way in which the options address the different objectives (i.e. the effectiveness of the policy options). In this way, it is possible to take account of impacts that cannot be quantified in Euros, such as the benefits.

For the qualitative assessment, a set of expected impacts relating to the identified objectives, are defined, and subsequently reviewed per policy option, in order to identify the most advantageous option. By linking the impacts directly to these objectives, the degree to which the options will reach the objectives will become apparent in a very straightforward way. Information on the impacts was collected via an open and a more targeted consultation.

The appropriate objectives for a suchlike analysis in this case are the operational objectives, as this level of detail allows a well-founded analysis and by consequence a true comparison of options.

Assessment of the costs for NSAs

A second part of the cost-benefit analysis consists of the assessment of the costs for NSAs for implementing the different policy options. This assessment is for the largest part based on information that was specifically collected from the individual NSAs during the consultation rounds (first one on FRIBS infrastructural elements and second one on introduction of new data requirements and on a separate track the consultation on the costs of different sub-options for modernising Intrastat). NSAs were asked to provide a baseline and to assess the cost of the changes brought about by the envisaged changes. NSAs have provided their estimations of costs (current and future) per statistical domain and an estimate for the one-off implementation costs of the changes which would also be incurred.

Assessment of the costs for data providers (administrative burden)

The third part of the cost-benefit analysis relates to the assessment of the costs for data providers (i.e. administrative burden) linked to the implementation of the different policy options. In order to give a complete overview of the impact on the administrative burden, the assessment is not limited to the costs directly linked to the completion of statistical surveys, but will consider the administrative burden of data provider (businesses) in a broad sense, including also qualitative elements , as well as non-staff costs (e.g. investments in IT-systems).

Regarding all the business statistics domains, except for the Intrastat part of International Trade in Goods (ITGS), the individual contributions of NSAs from the targeted consultations represent the main sources for the assessment of the costs for data providers 29 . Regarding Intrastat, the assessment of the costs for data providers is based on public consultations of the data providers themselves.

The Standard Cost Model (SCM) methodology was applied to the figures obtained through these consultations: by combining the total time spent on filling out surveys on a yearly basis and (cost) level of the employee completing the survey, the (current and future) administrative burden could be derived.

Overview of the costs of data production under different policy options:

In Chapter 6, the final results of a model based scenario were presented for the cost of data production under the different policy options. The cost model and its assumptions for Options B and C respectively are presented in the following two boxes:

It is important to note that the model is based on data collected from the data compilers and on assumptions summarised in the following table, together with the results aggregated for all the data collections.

The yearly details of the various sub-options 30 are presented in the following table.

ANNEX V: LIST OF THE CURRENT LEGAL ACTS GOVERNING EUROPEAN BUSINESS STATISTICS

1) Business Registers domain:

REGULATION (EC) No 177/2008 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 20 February 2008 establishing a common framework for business registers for statistical purposes and repealing Council Regulation (EEC) No 2186/93;

2) Structural Business Statistics (SBS) domain:

REGULATION (EC) No 295/2008 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 11 March 2008 concerning structural business statistics;

3) Short-Term Business Statistics (STS) domain:

COUNCIL REGULATION (EC) No 1165/98 of 19 May 1998 concerning short-term statistics;

4) Statistics on the production of manufactured goods (PRODCOM) domain:

COUNCIL REGULATION (EEC) No 3924/91 of 19 December 1991on the establishment of a Community survey of industrial production;

5) Foreign Affiliates Statistics (FATS) domain:

REGULATION (EC) No 716/2007 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 20 June 2007 on Community statistics on the structure and activity of foreign affiliates;

6) Statistics on trade with non-member countries (Extrastat) domain:

REGULATION (EC) No 471/2009 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 6 May 2009 on Community statistics relating to external trade with non-member countries and repealing Council Regulation (EC) No 1172/95;

7) Statistics on trade between EU Member States (Intrastat) domain:

REGULATION (EC) No 638/2004 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 31 March 2004 on Community statistics relating to the trading of goods between Member States and repealing Council Regulation (EEC) No 3330/91;

8) Research and Development Statistics and Innovation Statistics (R&D and CIS) domain:

DECISION No 1608/2003/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 22 July 2003 concerning the production and development of Community statistics on science and technology;

9) International Trade in Services Statistics (ITSS) and Foreign Direct Investment Statistics (FDI) domain:

REGULATION (EC) No 184/2005 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 12 January 2005 on Community statistics concerning balance of payments, international trade in services and foreign direct investment;

10) Information and Communication technologies statistics (ICT) domain: REGULATION (EC) No 808/2004 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 21 April 2004 concerning Community statistics on the information society.

ANNEX VI: LIST OF ABBREVIATIONS

BSDG - Business Statistics Directors Group

CIS - Community Innovation Survey (innovation statistics)

EGRs – EuroGroups Register

ESS - European Statistical System

ESSC - European Statistical System Committee

Extrastat – Statistics on trade with non-member countries

FATS - Foreign Affiliates Statistics

FRIBS - Framework Regulation Integrating Business Statistics

GVC – Global Value Chains statistics

ICT- Information and Communication technologies statistics

Intrastat – Statistics on trade between EU Member States

ISS – Information Society Statistics

ITGS - International Trade in Goods Statistics

ITSS - International Trade in Services Statistics

MEETS - Modernisation of European Enterprise and Trade Statistics

NCB – National Central Bank

NSA – National Statistical Authority

NSI – National Statistical Institute

PRODCOM - "Production Communautaire" or statistics on the production of manufactured goods

SBS- Structural Business Statistics

STS- Short-term Business Statistics

(1) The ESS is the partnership between Eurostat and the national statistical institutes (NSIs) and other national authorities (ONAs) responsible in Member State for the development, production and dissemination of European statistics.

(2)

 Decision No 1297/2008/EC of the European Parliament and of the Council of 16 December 2008 on a Programme for the Modernisation of European Enterprise and Trade Statistics (MEETS)

(3) Currently, work is under way for extending the European Statistical Programme (2013-2017) to 2018-2020 and thus aligning it to the cycle of the Multi-annual Financial Framework (MFF)

(4) Commission Communication COM (2013) 686 "Strengthening the foundations of Smart Regulation – improving evaluation"

(5) Please see more in detail in Chapter 5- Policy Options

(6) SBS- structural business statistics; FATS- foreign affiliates statistics; STS- short-term business statistics; CIS- community innovation survey (innovation statistics); ICT- information and communication technologies statistics

(7) "Persons employed" comprises both the "employees" and the "self-employed"

(8) In the context of the European Statistical Law (Regulation (EC) No 223/2009), the criteria defining statistical quality are: relevance; accuracy, timeliness; punctuality; accessibility, comparability, coherence.

(9) Exchange of the importer’s VAT number i.e. the exchange of the identification number of the partner in the Member State of imports. The current legislation does not require the collection of such an ID number.

(10) There is a whole range of EU policy areas whose development and monitoring of implementation relies on data covered by the business statistics domain: e.g.- employment, competitiveness, innovation, trade, industry, regional, economic and monetary policy.

(11) As provided for by Art 338 TFEU

(12)  GVC – Global Value Chains statistics

(13) Another 25 % of the respondents to the consultation did not have a clear preference

(14) http://ec.europa.eu/eurostat/web/quality/general-evaluation-results See

(15) The five key performance indicators are the following ones: Number of data extractions made by external users from Eurostat reference databases (EuroBase and Comext) via the Eurostat website, Percentage of users that rate as "Very good" or "Good" the overall quality of European statistics, Percentage of users that rate as "Very good" or "Good" the timeliness of European statistics for their purposes, Percentage of users that rate as "Very good" or "Good" the comparability of European statistics among regions and countries, Residual error rate (RER)

(16) http://ec.europa.eu/eurostat/web/quality/general-evaluation-results See

(17) For KPIs 3, 6, 7, 12, 13: no exact benchmarks can be specified as FRIBS does not impose any obligation to the Member States but only enables use of new sources and simplifications regarding those indicators.

(18) http://ec.europa.eu/eurostat/web/quality/general-evaluation-results See  

(19) The five key performance indicators are the following: Number of data extractions made by external users from Eurostat reference databases (EuroBase and Comext) via the Eurostat website, Percentage of users that rate as "Very good" or "Good" the overall quality of European statistics, Percentage of users that rate as "Very good" or "Good" the timeliness of European statistics for their purposes, Percentage of users that rate as "Very good" or "Good" the comparability of European statistics among regions and countries, Residual error rate (RER)

(20) http://ec.europa.eu/eurostat/web/quality/general-evaluation-results See

(21) http://ec.europa.eu/eurostat/web/quality/quality-reporting See:

(22) http://ec.europa.eu/eurostat/web/quality/evaluation More detail can be found under

(23) National statistical authorities (NSAs) of the 28 EU Member States and 4 EFTA countries responsible for the collection, compilation and dissemination of statistics at the national level; this includes the National Statistical Institutes (NSIs), but also other compilers such as National Central Banks.

(24) Businesses (including SMEs) responding to statistical surveys at national/regional level.

(25) All actors who are frequently using European business statistics, such as other Commission services, national policy makers, NSAs, National Central Banks and the European Central Bank, professional associations, businesses and researchers.

(26) The infrastructural elements include the Business Registers, micro-data exchange, quality issues and confidentiality

(27) Median value: EUR 4.9 million

(28) The business register is considered at the same level as the statistical domains.

(29) Through the consultation of stakeholders, it was also attempted to obtain direct inputs from data providers, but this has not resulted in usable information.

(30) Sub-option B4 and sub-option C4 are not explicitly inserted in the table since, as indicated in the boxes above, their values are the ranges between the values of sub-options B1 and B2; and C1 and C2 respectively.