Business Model for Integration of IPT into EGR

Transcription

1 EUROPEAN COMMISSION EUROSTAT ESSNET ON A EUROPEAN SYSTEM OF INTEROPERABLE STATISTICAL BUSINESS REGISTERS PHASE 1 ( ) MBGA N Deliverable No A4.1, WP3 Business Model for Integration of IPT into EGR Task: WP3, Block 3 Author: ESSnet ESBRs, WP3 Date: September 2016

2 The content of this deliverable is the result of the work of the ESBRs ESSnet. The reader should consider that Eurostat might not agree with some of its content nor is bound by it.

3 ESSNET ON A EUROPEAN SYSTEM OF INTEROPERABLE STATISTICAL BUSINESS REGISTERS (ESBRS) PHASE 1 WORK PACKAGE 3 PROCESS DEVELOPMENT AND DATA QUALITY PROGRAM Deliverable A4.1 Business Model for Integration of IPT into EGR Table of contents 1 Introduction Purpose Scope 2 2 As is state - integration of into EGR Business entity model Entity specifications Current state - Partial integration in a few countries 8 3 Toward To be state prerequisite statements Statements for validation/confirmation 9 Statement 1 - EGR users are to be served in the integration process 9 Statement 2 - ENT is the only enterprise unit to be stored in the EGR 9 Statement 3 - GEGs and GENs are candidates for integration into the EGR 9 Statement 4 - under European initiative is implemented on a voluntary basis 10 Statement 5 - information refers to a given reference year Options to be considered - for discussion and choice on preferred option Channel, responsibilities and priority rules 10 Choice 1 - Channel for integrating information on global groups and its legal unit structure 10 Choice 2 - GDC NSI responsibility in integrating information on global groups and its legal unit structure 11 Choice 4 - Priority rules for information on global groups and its legal unit structure from profiling Other choices 12 Choice 3 - Global Enterprise Group perimeter to be stored in the EGR CORE 12 Choice 5 - Information to be integrated - status of previous profiling results 13 Choice 6 - The GEN unit is candidate for integration into the EGR 14 4 Final choices made for the short term business model Preliminary note Statements confirmed Options validated 15 Annex 1 - under European initiative: process overview and shortcomings 16 Annex 2 - Integration of information from profiling into EGR - example from the UK 18 Annex 3 - Validation on GEG - reasons for divergences between profiling and EGR 20 A project funded by the European Union MBGA No.:

4 1 Introduction 1.1 Purpose This paper raises the major questions to be answered so that a business model for integration of IPT into EGR can be proposed and, subsequently, the business specification for the integration of IPT into EGR 2.0 can be drafted, i.e. ESSnet deliverable A4.2. The paper first describes the current state of the Business Model for integration of IPT into EGR (part 2); then it identifies the requisite statements and options that should be taken into account for setting up the business model for integration of information from manual profiling 1 activities conducted under European initiatives into the Euro Group Register (EGR) (part 3). Finally, the choices made, taking into account ESBRs SG advice, are presented in the last part (part 4). Parts 1 to 3 of this paper were presented and discussed at the ESBRs Steering Group s 4 th meeting held in December 2015 in Luxembourg. At the SG meeting the ESSnet asked Steering Group members for validation and advice on the options envisaged. Following the feedback received from the Steering Group Members and according to the requirements detailed in the Eurostat ESSnet ESBRs call for proposals some decisions were made on options to be followed when drafting the business model for integration of IPT into EGR and completing business specifications (ESSnet deliverable A4.2). Those decisions are presented in this paper (part 4). 1.2 Scope This paper deals with the integration of information from manual profiling activities conducted under European initiatives into the Euro Group Register (EGR). The profiling activities under European initiative target the large and complex Global Enterprise Groups (GEGs) within the top 600 that have at least one affiliate in a EU+EFTA country. 97 GEG have been profiled as part of the European activities already; this number should reach 160 by the end of It is planned that 300 will have been profiled at least once by The relation between profiling and EGR has many dimensions. The present paper does not cover them all. Thus, several aspects listed below are out of scope of the paper: Integration of information from manual profiling activities under national initiatives (i.e. with no European coordination). activities under national initiative can target Domestic Enterprise Groups and the truncated part of Global Enterprise Groups. It is expected that this information be integrated when each NSI provides to the EGR data input including the Enterprise statistical units that belong to the GEG. Each NSI is responsible for integrating the information from profiling collected under national initiative into the EGR. Integration of information from automatic procedures on small and medium Enterprise Groups. Information from EGR into : the paper discusses the profiling information as input to the EGR process. It should be noted that the EGR is also used as main input to the profiling work; flows going from the EGR to the process is out of scope of this paper. 1 is a methodology intended to delineate the enterprise in the groups by analysing their legal and operational and accounting structure. It is the preferred method supported by Eurostat to get relevant enterprise units complying to the 1993 regulation. Depending on the size of the GEGs Eurostat recommends implementing automatic or manual profiling; Manual profiling consists in a desk analysis (including the analysis of the documents made publicly available by the group (annual reports, for example)) and possibly a contact with the group, which will confirm the desk analysis and provide relevant information. Implementing a manual profiling is costly so it is done on a targeted number of Enterprise Groups when it has a significant impact on business statistics. Version September 2016 page 2

5 Sustainability of manual profiling under European initiatives: in the paper it is assumed that profiling under European initiatives is sustainable meaning that Eurostat will support the implementation through individual grants, Eurostat/Centre of Excellence organises the coordination; a minimum number of NSIs (GDC and partnering countries) are involved in implementation. Discussion on the requirement needs to settle a sustainable system is out of scope. methodology: The profiling method is not discussed in this paper. The paper deals with how to integrate the profiling information as currently collected under European initiatives. Important notes Warning - in this paper the term profiling with no other mention refers to manual profiling of large and complex GEG when conducted under European initiatives. Warning While drafting the business model deliverable, the ESSnet raised some issues about methodology on European profiling developed under the MEETS program and about potential mismatches between the delineation of the national enterprise (ENT) carried out at national level and the results from European profiling, i.e. the delineation of truncated enterprises (TEN) made up by the national part of a Global enterprise (GEN). The deliverable has been produced as an input for the ESBRs Business Architecture (BA) without any view of the final version of the BA itself; BA was delivered afterward. Wording clarification - The question about the integration of profiling information into the EGR is often referred as integration of IPT into EGR. This sentence requires clarification. The IPT (Interactive Tool) is a prototype that is currently under IT development. It has been designed for helping profilers to perform manual profiling conducted under European initiatives. Currently the work performed under European initiatives is done through a very burdensome manual process using Excel files exchanged through Edamis. The IPT tool aims at facilitating the work process and the access to the profiling results. It will allow profiling teams to smoothly exchange information during the profiling process and to centralize information on the group under profiling process. The tool itself will not change the work process and the current profiling model. In 2016 a prototype of the IPT tool will be made available to NSIs who will perform manual profiling activities under European initiatives, for testing purpose. The experiences with the IPT could impact the process as tested in The paper refers to the integration of profiling information (under the scope of the IPT) and does not refer to the integration of the IPT tool itself. This is because what matters for EGR users is that information from profiling when relevant is made available in the EGR whatever the tool and the channel used (either by a manual or interactive way). 2 As is state - integration of into EGR This part presents the current state (2015) of the integration of (manual profiling under European initiative) and EGR. Similar business model will run in The availability of the IPT as a prototype will ease the work but will not change the current Business Model. Version September 2016 page 3

6 2.1 Business entity model The Business Entity Model provides a picture of the main entities at stake in the profiling process and how they relate to EGR entities. The model is divided in two parts: the top part represents the EGR activities; the bottom part represents the profiling activities. contains object Business contains Entity sends and validates Data sets * EGR IM EGR FATS EGR Business Register Staff is part of Global Enterprise Group (GEG) Enterprise (ENT) EGR-CORE EGR IS runs Consults and retreives data from Exchange information and data is part of Relationship between Legal units is linked with Legal Unit Retrieves Data from EGR CORE and EGR IS oversees Eurostat supports Users controls generates analyses Profiler supports Process owner (GDC Profiler) creates Global Enterprise (GEN) process contains contains concerns Global Enterprise Group (GEG) result * is part of ESSnet updates Truncated Enterprise (TEN) is linked with ESSnet Partnering country updates Legal Unit *Notes - changes in 2016: results - There is currently no central profiling database containing the information that has been gathered from previous profiling cases. The profiling results stored at Eurostat are still in the Excel format as sent by each country. So profiling results are currently not easily accessible to profilers. In 2016 the IPT tool should for the first time generates a profiling database including data that are collected by manual profiling. The IPT will make data access easier. EGR-CORE and profiling process: Currently Eurostat is in charge of retrieving data from the EGR manually and to send it to profilers so that they can start the profiling process. In 2016, this function should be made automatic through the IPT. EGR FATS currently provides cross border frames for IFATS and OFATS. Data sets sent by Business Register staff refer to files such as LEU, REL, GEG, LEL, ENT etc. Version September 2016 page 4

7 2.2 Entity specifications The entities shown in the above Business entity model are shortly described below. Entity Specification Part Actors Eurostat Process owner Partnering country profiler ESSnet Business Register staff Eurostat is profiling administrator and as such oversees the profiling processes and check consistency with the signed contract. Eurostat is also in charge of analysing profiling results 2. Eurostat is EGR manager and runs the production cycle. Process owner is profiler (member of profiling team) who controls profiling process, creates group structure and approves updates proposed by partnering countries. Process owner is usually GDC NSI profiler, however in some cases it might be profiler outside GDC country (after Eurostat acceptance) Member of partnering country profiling team is responsible for proposing information updates on TENs and LEUs and confirming of TENs when no changes are to be requested. ESSnet is responsible for answering questions from testing countries on profiling activities, organizing training seminars, supporting Eurostat in specifying improvements for the profiling tools (Excel templates and interactive profiling tool). The ESSnet is also responsible for updating and improving profiling documentation (methodological report, guidelines). Business Register staff is primary actor in the EGR process. He/she sends the information to the EGR concerning legal units, relationship between the legal units (LEU), enterprises (ENT) and on global enterprise groups (GEG) to the system. He/she sends information through data files or using interactive tools either during the initialization phase or during the validation phase. User User is able to consult EGR results. For now it is FATS users. EGR EGR EGR Entity Specification Part Units Global Enterprise Group Global Enterprise * see note below Truncated Enterprise * see note below GEG is main object under profiling. The aim of profiling process is to create appropriate structure of the Global Enterprise Group including identifying GEN and TEN within the group. GEN is created by GDC NSI profiler and contains Truncated Enterprise. It is a technical unit supporting the delineation of ENT TEN is created by GDC NSI profiler as derived from GENs. TEN are national part of GENs. Partnering country is allowed to propose updates on TEN s data. In the current process however the process owner is responsible and EGR 2 Up to now Eurostat has been assessing the results partially only as mentioned in the assessment report available on CIRCABC. Version September 2016 page 5

8 Entity Specification Part for the final decision on TEN. TEN is not a statistical unit. It is a tool used by profilers to try to ensure as much as possible consistency in the delineation of the statistical unit ENT. It is expected that in most cases the TEN will coincide with the statistical unit ENT active in the national territory. It is a technical unit supporting the delineation of ENT. Legal Unit Enterprise LEUs are initially retrieved from EGR but it is possible to create new LEUs within the profiling process by GDC NSI profiler (also based on partnering country proposal). ENT is the statistical units used in the production of national business statistics. NSI is owner of the entity ENT. The national statistical business register of a given country is the authoritative source for the EGR concerning the ENT resident in the country. and EGR EGR Note: example of GEN, TEN and Legal Units (LEU) in a form of graph for a given GEG. It should be noted that the present version of the operational guidelines agreed by the BR Working Group regarding the delineation of statistical units doesn t mention GENs and TENs. According to the Task Force Statistical Units: «For national statistics the national part of the unit is to be considered an enterprise. Such a resident unit is regarded an enterprise unit in the economic territory where it is located. However in some specific cases this may lead to more than one national enterprise.» (Operational rule n 6). In other words, in a few cases TEN can differ from ENT. This can be due to the coexistence of different and relevant points of view (national and global). Example of l Oréal (France: GDC country): In the following example one GEN, so one TEN by country, were stated from EU profiling. However, from a national point of view there were two TENs. At global point of view, Cosmetic l Oréal group is made of a single GEN, although the group acquired in 2006 Body Shop a UK sub group specialized in retail sales of natural cosmetics as Body Shop turnover was less than 4 % of the total GEG turnover. The activity can be larger for the national economy. Thus, Body Shop accounted for 60 % of employees and 36 % of turnover of the UK part of the GEG. That s why UK creates two enterprises: one for l Oréal and the other for Body Shop. Version September 2016 page 6

9 Symmetrically, there are cases where a NSI may wish to distinguish just one national enterprise ENT while top down profiling would lead to several TENs. Indeed, we have to take into account size, activities and governance of the truncation of the group in that country. This may be illustrated by the following example of the US controlled group Procter and Gamble in France. The US group Procter and Gamble (P&G) has 118,000 employees worldwide (with local units in 70 countries)1. Under US gaap, its structure is organized into 5 reporting segments (Beauty, Grooming, Health Care, Fabric & Home Care, Baby, Feminine & Family Care) and 10 global business units. At the global level, a top down profiling conducted by US profilers would probably lead to at least 5 GENs. However, the truncation of the group in France is an intermediate sized enterprise (IZE) of 2600 employees and 11 legal units. The French profiling team estimates that there are several reasons to consider it as a single ENT. From a practical point of view, considering several TENs as ENTs would lead to split legal units. It is true that is a technical reason, which may be not enough to take the decision, even if splitting legal units bring a lot of complexity. Other economic and governance reasons lead to retain only one ENT. According to its report, P&G in France presents itself as one responsible enterprise, managed by one director. Its global head in France is a société par actions simplifiée (SAS), a legal form in which the management keep most of the power of decision. If we consider The European regulation 69/93, the level of decision, especially for the allocation of its current resources is at one level, not at several ones. The products of different segments are manufactured together in the same plants, as the Amiens one. In 2012, P&G France contacted INSEE to receive all the statistical survey at its French headquarters. Considering P&G France as several ENTs would lead to define it, not as an IZE, but as several SMEs. This would contradict the representation of enterprises in terms of size, which now has in France an official definition and an institutional recognition. According to the statutes of the union representing the enterprises, P&G France has to right to register to the union of IZEs, while its different parts would never have to right to register to the union of SMEs. For all these reasons, it is impossible in the French context, to consider P&G France as several ENTs. These two symmetric examples show the need for pragmatic approaches to establish the links between TENs and national ENTs. In most cases, for most countries, TENs and ENTs well coincide and one can manage, as far as possible, that this is the general case. Nevertheless, there may be also the two other following cases: Several ENTs fit together into a TEN, like L'Oreal in UK; Several TENs fit together into an ENT, like P&G in France Entity Specification Part Others process EGR-CORE result Cross border frame A profiling process is identified by a reference period and a GEG. process controls the flow of the events and contains all mandatory information to aid GDC and partnering countries profilers to create appropriate group structure, including identifying GEN and TEN within the group. The core of the EGR system receives data from NSIs and Commercial data providers; transform the data and then create the Global Enterprise Group with its statistical unit Enterprise (ENT) and legal unit structure. EGR-CORE provides the initial data to profiling process. The result is generated when GDC NSI profiler decides to complete (light/intensive) profiling process. The result can be consulted by all actors of the profiling process. For now the EGR provides cross border frames for IFATS and OFATS. EGR EGR Version September 2016 page 7

10 2.3 Current state - Partial integration in a few countries activities under the supervision of Eurostat follow three main objectives for a given reference year: To collect control/consolidated GEG perimeter. This stage includes the stage of identification/validation of GDC and main attributes (such as name and country) and of main attributes of controlled/consolidated affiliates (such as name, country, consolidation method and controlled rate); To define agreed GENs and collect basic attributes (name, main activity code, consolidated turnover, employees); To encourage the use of TENs as ENT in the national BRs and data collection for those units. The implementation is currently conducted under individual grant agreements between each NSI and Eurostat with a support from the ESSnet. As GDC profiler the NSI is responsible for choosing a GEG to be profiled for a given reference year. Once validated by Eurostat the profiling process starts. The GDC country retrieves data from the EGR for that group, checks GEG data and legal units in the group perimeter, and introduces information on GEN and TEN. The partnering countries notified check legal units in the group perimeter in their countries and the information on TEN. Exchange between GDC and partnering profilers may lead to revision. GDC NSI is responsible for completing the profiling process and send profiling results to Eurostat (for more information about the profiling process - see annex 1). During the profiling process the EGR data initially retrieved are modified (on the GEGs perimeter and attributes) and new information is made available (on GENs and TENs). The profiling actions conducted in actually show the importance of profiling to detect new legal units or legal units in the GEG perimeter (4248 legal units inserted, and 4247 legal units removed for a total of 97 GEG including legal units). When the process owner initiates a profiling process he/she (or Eurostat on behalf) retrieves EGR data on reference year N. During the profiling process profilers work on reference year N+1 that means that they collect information in advance that could be used for next EGR cycle. In the current process a few NSI make those corrections available to EGR FATS users already. Once the profiling process completed they indeed use the updated information as input to their national statistical business register. The Business Entity Model shows the channel through which the information from profiling is currently used to integrate information into the EGR (line between the profiler and the Business Register staff). An example of such integration was discussed during the ESSnet ESBRs WP3 Seminar held in Paris in June 2015 (for more information about the UK example see annex 2). The presentation from the UK showed what the situation is in the UK and showed that only partial integration can be achieved mainly due to restriction in the UK register. It is more than likely that every register in Europe is different in terms of perimeter and timeliness. It is difficult then to define what can be integrated into the EGR through national statistical business register (NSBR). So in the short-term the ESSnet should work in finding another way for the integration. Considering updating the EGR directly at the central level without link to the NSBR may be an option. Version September 2016 page 8

11 3 Toward To be state prerequisite statements This part identifies several points that directly impact integration and that need to be validated to make sure that the detailed work that will be on-going on integration is going in the right direction. In the first paragraph (3.1.) statements are proposed for validation. In the second paragraph (3.2) several options are listed and proposed for discussion and for making decision on the preferred option. 3.1 Statements for validation/confirmation Statement 1 - EGR users are to be served in the integration process In the short term main EGR users are FATS statisticians 3 who will retrieve cross border frames from the EGR to produce business statistics for a given reference year. Those users want accurate and timely information available on Global enterprise Groups (GEG) and its constituent units namely the statistical unit Enterprises (ENT) and legal units (LEU). The purpose of the integration is to make sure that the information collected by profilers through the activities conducted under European profiling will be taken into account in the EGR cross border frames whenever relevant. Statement 2 - ENT is the only enterprise unit to be stored in the EGR Only one version of the enterprise units should be stored and maintained in the EGR. The statistical units ENT should be the Enterprise unit reflected in the EGR-CORE and as a consequence in the EGR-FATS interface. The unit TEN derived from profiling under European activities will be reflected in the EGR only if NSIs decide that the TEN is relevant to be used as their ENT (in which cases ENT=TEN). TEN should not be referred as such to in the integration process. Statement 3 - GEGs and GENs are candidates for integration into the EGR The integration of information from manual profiling under European initiative into the EGR could target the information collected by profilers on GEGs and its legal unit structure. This could possibly cover information on: - Global enterprise group (GEG) and its attributes - GEN and its attributes - The relationship between a legal unit and the GEG /GEN perimeters The profiling information on the unit enterprise (TEN), legal unit and direct control relationship between legal units are out of scope of the option list. The purpose of such integration would be to make use of the information from the group on the legal unit structure of the GEG together with the GEG Attributes. 3 Note: the second user target for the EGR could be FDI statisticians. However, for now neither the EGR nor profiling as conducted under European profiling serves the FDI requirements. An analysis of the FDI requirements and what this means for the to be state of the EGR is still to be performed by the ESSnet. Version September 2016 page 9

12 Statement 4 - under European initiative is implemented on a voluntary basis activities under European initiatives (with European coordination) are conducted on a voluntary basis. Only NSIs that on a voluntary basis decide to take part in those activities will do it under a bi-lateral contract with Eurostat. Warning: it is assumed here that Eurostat will support implementation of profiling coordinated under European initiative from 2017 onwards and that a certain number of NSIs will commit to take part in those activities from 2017 onwards. Statement 5 - information refers to a given reference year The profiling process gives information for a given reference year and should be processed by the EGR for this given reference year Options to be considered - for discussion and choice on preferred option This paragraph presents different options that can be envisaged for six main strategic choices impacting the integration. In each case the question is: What is the preferred option? First three choices on channel, responsibilities and priority rules are presented ( 3.2.1) all together as choices on responsibilities and priority rules depend on the option chosen on the channel. Then other choices that do not depend on the first ones are listed ( 3.2.2) Channel, responsibilities and priority rules Choice 1 - Channel for integrating information on global groups and its legal unit structure For integrating information from profiling (on global groups and its legal unit structure collected under European initiative) into the EGR three channels can be envisaged: Option 1 - Information from the profiling platform (IPT) on global groups and its legal unit structures 4 go to National Statistical Business Register and is then sent to the EGR by Business Register staff; Option 2 - Information from the profiling platform on global groups and its legal unit structures go to the EGR directly and then it is the responsibility of NSI to update their Business Register; in such a case it should be decided what the roles and responsibilities are between GDC country and partnering country (see Choice 2 below). Option 3 - Profilers are requested to enter the information from profiling directly in the EGR (by using the EGR Interactive Module). 4 This could possibly cover information on: i) Global enterprise group and its attributes; and ii) the relationship between a legal unit and the GEG perimeter. The profiling information on the unit enterprise (TEN), legal unit and direct control relationship between legal units are out of scope of the option list. Version September 2016 page 10

13 Pros and cons of each option are listed in the table below: Option 1 - via NSBRs (single flow model) Option 2 - via IPT Option 3 - via IM PROS -Make sure NSBR and EGR consistent - EGR input data flow does not change -Make sure EGR is up to date on time when EGR production cycle runs -Make sure EGR is up to date on time when EGR production cycle runs - EGR input data flow does not change CONS -Will delay the actual integration as not all BR can currently process and store information from profiling -NSBR may diverge from EGR -NSIs may lose autonomy for part of their NSBR (see choice 2) - EGR needs to be able to receive Input data flow from IPT -Double work for profilers (IPT and IM) -NSBR may diverge from EGR Choice 2 - GDC NSI responsibility in integrating information on global groups and its legal unit structure Choice 2 is discussed here in the context where information from profiling platform (IPT) on global groups and its legal units structures go to the EGR directly (Choice 1 - option 2). Also it assumes that the GDC country is the process owner for the process of profiling in the IPT, as it is the case today. In such case the question about the responsibility in integrating information is about who has the responsibility to push the button of the IPT allowing a data flow going from the IPT into the EGR. Two scenarios can be envisaged for GDC NSI responsibility: Option 1 - The GDC country is responsible for the final decision (after consultation with partners and with the help of Eurostat in case of conflict). The GDC country (together with Eurostat) is the only one with right to send a data flow from the profiling platform to the EGR. It should be underlined that this option is in line with profiling as a shared process, in which NSIs as partner s countries that have not validated the information could lose their autonomy of part of the national frames. Option 2 - The GDC country is responsible for launching a profiling case, asking partnering countries to validate; each NSI is eventually responsible for sending data flows from the profiling platform to the EGR. Pros and cons of each option are listed in the table below: Option 1 - GDC country has the final decision (shared process) PROS -Make sure that the overall picture makes sense CONS -NSIs as partner s countries that have not validated the information could lose their autonomy of part of the national frames -Need to adapt the EGR rules as it would not be aligned with the EGR 2.0 Full authentic source principle Version September 2016 page 11

14 Option 2 - Each NSI eventually responsible for sending data PROS -Consistent with the EGR 2.0. Full authentic source principle (NSIs can keep full autonomy on their resident data) -No need to adapt the EGR principles CONS -Could allow inconsistencies in the overall picture of the GEG -No use of information collected from the GEG Choice 4 - Priority rules for information on global groups and its legal unit structure from profiling Choice 4 is discussed here in the context where information from profiling platform (IPT) on global groups and its legal units structures go to the EGR directly (Choice 1 - option 2). In such case a new kind of source will have to be taken into account in the EGR consolidation process. The question once the data enters the EGR system how should it handle the information. For a given reference year EGR cycle priority rules will have to be carefully thought. Two scenarios can be envisaged: Option 1 - Considering by default that information from the profiling platform (IPT) is better than any other source of information (such as NSBRs); Option 2 - Considering that the information from the profiling platform should compete with other sources of information available. Such rule could be that the profiling platform is considered as source with higher priority in case no authoritative source has been defined. Pros and cons of each option are listed in the table below: Option 1 - information is higher quality Option 2 - information compete with other sources PROS -The information from the IPT is the one reflected in the final EGR frame -Priority rules could consider that information sent by BR NSI may overwrite the information from the GEG in specific cases (e.g. in case there is an authentic source for the entity - No loss of autonomy CONS -NSIs as partner s countries that have not validated the information could lose their autonomy of part of the national frames -The information from profilers may not be reflected in the EGR Other choices Choice 3 - Global Enterprise Group perimeter to be stored in the EGR CORE On the one hand the EGR CORE derives the GEG control perimeter. On the other hand profilers collect information on the consolidated GEG perimeter. Both perimeters can diverge. Two scenarios can be envisaged for the integration: Option 1 - Allowing EGR users to generate both the control GEG perimeter and the consolidated GEG perimeter on demand. Option 2 - Providing EGR users with one single perimeter, the control GEG perimeter. Version September 2016 page 12

15 Pros and cons of each option are listed in the table below: Option 1 - Control and consolidated GEG perimeter Option 2 a - One single perimeter - control GEG perimeter Option 2 b - One single perimeter - consolidated perimeter or control perimeter PROS -Users could access to both information -Consistency between economic variables and the perimeter of the GEG -Consistent approach for all GEGs -Consistency between economic variables and the perimeter of the GEG CONS -Economic variables (turnover, employment, NACE) would have to be provided for both perimeters -Changes to the EGR system would have to be implemented -Burden in maintaining 2 GEGs perimeter -Possible inconsistencies between the economic variables collected through profiling (consolidated turnover, employment) and the GEG unit perimeter -Two different perimeter concepts combined that could confuse users Choice 5 - Information to be integrated - status of previous profiling results Up to now the results of manual profiling large and complex GEG under European initiative have been stored in Excel files at Eurostat. From 2016 onwards, profiling countries will use the IPT prototype that will feed in IPT database on GEGs with data on references years possibly from 2015 onwards. This information will be in a good format to be candidate for integration once the IT tool will be ready for integration. However, the results of previous profiling cases conducted prior to 2016 recorded in Excel spreadsheets are not suitable for integration. The question raised: will we use the data from profiling conducted prior to 2016 in the EGR? If yes, how and where these data will be stored and used? Three scenarios can be envisaged: Option 1 - Result from previous profiling cases will neither feed in the IPT profiling platform nor the EGR; Option 2 - Results from previous profiling cases will feed in the IPT profiling platform but not the EGR; Option 3 - Result from previous profiling cases will feed in the IPT profiling platform (for profilers consultation) and a study will be conducted on whether there is room for integrating manually information that could still be relevant. Pros and cons of each option are listed in the table below: Option 1 - No use PROS -No additional burden on NSIs and Eurostat that could focus on current cases CONS -The 300 target will never be met in the IPT - Lose the opportunity for some NSIs to use the information collected to identify their enterprise units (lose opportunity for harmonization) Option 2 - Use in IPT -All NSIs (not only partnering -Huge workload as the files have Version September 2016 page 13

16 PROS CONS not in EGR Option 3 - Use in IPT and in EGR countries) could access to the results and use it to identify the enterprise unit perimeter -Make full use of the work done and the information collected changed format; -Huge workload as the files have changed format; -Need to decide whether cases were agreed enough for integration in EGR Choice 6 - The GEN unit is candidate for integration into the EGR The GEN unit is currently not included in the data model of EGR version 2.0 so the information cannot be integrated. Work on integration should aim at including the GEN units in EGR-CORE. The attributes to be included in the EGR for the unit GEN will have to be clarified with users. Pros and cons of each option are listed in the table below: Option 1 - GEN unit integrated Option 2 - GEN unit in the IPT only PROS -Statistics on GEN could be produced (IF GEN as some stage (If delineated and maintained on the full scope of MNEs) -Can be used by profilers to identify the enterprises units -No need to be maintained as a statistical unit CONS -Need adaptation of the EGR data model -Need to be maintained as a statistical unit (workload) -No statistics can be produced on GENs 4 Final choices made for the short term business model 4.1 Preliminary note At the ESBRs 4 th Steering Group meeting the point of view varied between ESSnet Members, Steering Group Members and Eurostat (see minutes). Also at the ESSnet ESBRs-Eurostat coordination meeting that follows the Steering Group meeting Eurostat reminded the ESSnet partners about the requirements detailed in the ESSnet ESBRs call for proposals that could contradict the point of view expressed by some ESSnet Members and Steering Group Members (see minutes). The content of part 4 then presents the main business model principles that were chosen and followed in the business specifications for integration of IPT into EGR 2.0 (deliverable A4.2) in accordance with the requirements detailed in the ESSnet ESBRs call for proposals. 4.2 Statements confirmed The following statements were validated: Statement 1 - EGR users are to be served in the integration process Statement 2 - ENT is the only enterprise unit to be stored in the EGR Version September 2016 page 14

17 - No matter which input source feed in the EGR (either NSI Business Register Staff or NSI Profilers) only one kind of enterprise unit is stored in the EGR, namely the ENT, the unit that is used to produce business statistics - In the EGR NSI are authentic source for the ENT. This principle should remain. NSIs will not loose autonomy on the statistical unit Enterprise. Statement 3 - GEGs and GENs are candidates for integration into the EGR Statement 4 - In the short-term profiling under European initiative is implemented on a voluntary basis Statement 5 - information refers to a given reference year. 4.3 Options validated The following options were validated: Choice 1 - channel for integrating information on global groups and its legal unit structure - Option 2 is chosen: a data flow going directly from the IPT to the EGR will be generated automatically when the profiling process is completed in the IPT - It is the responsibility of the NSI to update its BR accordingly if needed (missing information on relationship) Choice 2 - GDC NSI responsibility in integrating information on global groups and its legal unit structure from profiling - Option 1 is chosen: the GDC country when completing the profiling process is responsible for the final decision on what is sent to the EGR regarding information on global groups and its legal unit structure. - The GDC Profiler NSI is considered as the source with the highest priority in the EGR in case no authoritative source for the relationship has been defined - NSIs will not loose autonomy on the statistical unit Enterprise, nor on resident legal units (and its attributes), nor on the direct control relationship between resident legal units. Choice 3 - Global enterprise group perimeter to be stored in the EGR CORE - Option 2 is chosen: the EGR frame will only provide the control GEG perimeter. Choice 4 - Priority rules for information on global groups and its legal unit structure from profiling - Option 2 is chosen: information from profiling is an input to the EGR that will then follow the EGR priority rules. The profiling source will compete with other source. Choice 5 - Information to be integrated - status of previous profiling results - Out of scope of the business specification paper. The choice depends on Eurostat. No position taken from the ESSnet. Version September 2016 page 15

18 Annex 1 - under European initiative: process overview and shortcomings activities under the supervision of Eurostat follows 3 main objectives: for a given reference year. To collect control/consolidated GEG perimeter. This stage includes the stage of identification/validation of GDC and main attributes (name, country, town) and of main attributes of controlled/consolidated affiliates (name, country, town, consolidation method, controlled rate). To define agreed GENs and collect basic attributes (name, NACE, consolidated Turnover, employees) To encourage the use of TENS as ENT in the national BRs and data collection for those units. process described in the figure below represents the general approach to the profiling followed by countries that are currently manually profiling GEG under European initiatives. Figure High level profiling process 5 Before starting of the profiling, GDC NSI profiler has to be sure that the GDC is in line with his own perception. This checking should be made directly in available data sources (i.e. in EGR). The GDC NSI 5 This diagram is extracted from the document EUROSTAT/ Business Use Case Specification Version 3.0. To be noted - UCI refers indeed to the GDC (Global Decision Center). Version September 2016 page 16

19 profiler must extract information from available source (mainly the EGR) to prepare input for profiling. When all information is in place it is possible to start profiling by GDC NSI profiler. The following operations are performed on initial data: check/update of Group information, compare with a list of LEU displayed by the group, check the legal units in the group perimeter in the country of the GDC, check the global information of the group, introduce information on the GENs, and introduce information on the TENs. When the profiling by GDC NSI profiler is completed it should be sent to partnering countries profiling teams to carry out the next step of profiling with the following operations: check the legal units in the group perimeter in the partnering country, check information on the TEN provided by the GDC NSI, and include national information on the TEN. The updated version is sent back to GDC NSI profiler for validation and verification. In case when GDC NSI profiler agrees with all updates from partnering countries the profiling is completed. On the other hand, it is possible that GDC NSI profiler does not agree with some parts of updated data. In that case he prepares counter proposal and starts next iteration of partnering country profiling by sending the data again. The process of information exchanging lasts until GDC and the partnering countries agree with the delineation of the TEN and GEN. The economic data on TEN (NACE, Employment, turnover) must be reported, and exchanged if easily available for the GDC country (through OFATS data or easy to get when visiting the GEG). If not, the work of the GDC and partner countries should focus on the GEG structure and the GEN data. This is because Eurostat has put central collection of data as a lower priority. There are well-known shortcomings regarding the high level profiling process: A) There is no well-known guidance on how long a profiling process can last and what makes it complete; in the meantime the individual grants organization clearly assume that the profiling should be completed within a year; also Eurostat set up a list of criteria to meet to have a profile considered as completed; B) Nothing is envisaged in case of remaining conflict between the GDC and partner country; C) There is no procedure specified to make use of the information from profiling into the EGR. NSI Business Register staff has possibility to update information stored and processed in the EGR Core system. As the profiling process results in creation of a Global Enterprise Group structure basing on EGR Core information, there is a need to align data in the EGR Core with the results of the profiling process. Profiler may exchange information and request updates in EGR Core from the NSI Business Register Staff at any point of time during profiling process. That data alignment needs to be done at least once when profiling process of a given GEG is finished. There is a need to have this alignment realized on a procedural level. D) Duration - The profiling of the GEG is considered as completed within a year by the individual grant while it occurs that one-year might not be enough to complete it. E) Reference year - To allow partners to validate data profilers may have to work on data with one year lag compared to what was asked in the individual grants -> there is a need to know whether this practice is allowed or not when building an integration model. F) Non-mandatory activities - Currently profiling activities under European initiatives are conducted on a voluntary basis. Only NSIs willing to do so participate while the process proposal assumes that: 1. It is possible to coordinate resources and work planning of 32 NSI s, 2. It is possible to implement simultaneously profiles in national business registers and national statistics and 3. It requires a strong centralized orchestration process. Version September 2016 page 17

20 Annex 2 - Integration of information from profiling into EGR - example from the UK The integration of information from profiling with information from the EGR was discussed at the Paris June EGR and ESSnet ESBRs Workshop. An example of partial integration through National Statistical Business Register was given by the UK and is detailed below. The UK updates the NSBR with profiling information when the register allows it and then it fed through the EGR following the current EGR frame methodology. At national level the information collected through profiling is restricted to what can be integrated in the UK register. As a consequence there is a full integration of the information collected from UK profiling activities in the UK register. When coordinated at the European level, the information collected from UK profiling is broader. Big differences between the information collected through profiling coordinated at EU level, the UK register and the EGR mean that the UK can only integrate some of the information from profiling under European initiatives to the EGR. The three main reasons explaining why the information from profiling coordinated at EU level is not fully integrated are due to: Limitations in the UK register 6 : o o o Cluster of Control - only legal units that are 50% or more controlled are stored in the UK register; Identification number only legal units with Dun and Bradstreet number are stored in the UK register; Threshold only legal units that fall above a Tax and employment Thresholds are stored in the UK register (very small businesses are not captured); The profiling scope covers all legal units that are consolidated in the annual report and balance sheets it may include legal unit that are less than 50% controlled, legal unit not recognized by Dun and Bradstreet, or very small businesses. If information on such legal units and relationships are collected by profilers it will not be integrated in the UK register and will not be sent to EGR. Limitation in the EGR frame methodology: o o EGR Timing - a snapshot of the live UK register is extracted once a year (e.g. December T) and sent to the EGR (November T+1) to be used for the frame released in March T+2 (with year T-data). Any profiling information collected after December T that would be of interest for data on year T will not be used in the EGR Frame produced in March T+2. Disagreement on the TEN proposal: if the National Enterprise is different to the TEN proposed by another GDC, then the UK register does not reflect the TEN data. 6 Note scope differences - only British legal units and their links to foreign parents and/or subsidiaries are stored in the UK register (not all foreign legal units in GEGs are stored). So any information on legal units that are not in direct relation to a British legal unit part of the UK register cannot be updated through the UK register. Version September 2016 page 18

21 Evidence from the UK view shows that two major improvements would be needed to allow for a better integration of profiling information to the EGR through NSBR: Getting consistency between NSBR and the EGR - there is a need to be consistent in methodology and approaches in all the NSBR and the EGR Aligning at methodological and process level the profiling and the EGR activities. Version September 2016 page 19

22 Annex 3 - Validation on GEG - reasons for divergences between profiling and EGR This annex discusses what the requirements are for a relevant integration of the information from profiling on GEG perimeter 7. It is considered that the question of the integration starts when the GEG profiled has a status completed. The question of when and by whom the integration process is triggered (it could be at any time or at specific point in time; by profilers themselves on a case by case basis or by the system) is not discussed yet. The final aim in this particular case is to use information from profiling to make sure that the right LEU structure is stored in the EGR for a given reference year. This requirement is based on the assumption that information coming from profiling (desk analysis of annual report and face-to-face visit) is better than the information stored in business registers. The profiling actions conducted in actually have shown the importance of profiling to detect new legal units or legal units that should not be in the GEG perimeter (4248 legal units inserted, and 4247 legal units removed for a total of 97 GEG including legal units). The reasons why the GEG perimeter can be different in the profiling and the EGR databases are listed (based on previous experience). Difference in reference year: For most of the GEGs profiled in the differences observed between the EGR values and the profiling values were determined by more up-todate information meaning that the reference year was not the same. Indeed for the GEGs profiled in 2013 the GEG information reported referred to the year 2012, while EGR information generally refers to Same reference year: In case of same reference year there are three possible situations. o o o Legal unit in the GEG perimeter IN the database and NOT IN the GEG perimeter in the EGR CORE database; Legal unit in the GEG perimeter NOT IN the database and IN the GEG perimeter in the EGR CORE database; Legal unit in the GEG perimeter IN BOTH databases but with different legal units attributes (impacting the GEG perimeter); In every of those three cases two main reasons explain divergences: o o Difference in concepts - First it should be specified what is meant by the right LEU structure. Indeed there are three possible GEG perimeter namely: Perimeter of control; Perimeter of detention; Perimeter of consolidation. Mistake in the business register - If not a difference in concept then it should be considered that the information from the profiler is better and the update is needed. Indeed while conducting the profiling cases, and before completing it the GDC profiler should have analysed carefully the differences between the EGR and the information from the GEG or from the annual report. The differences should have been discussed with partnering countries. So it is expected that the differences be worth integrating by default. 7 It does not cover the GEG attributes (NACE code, name, address ). Version September 2016 page 20