EUROPEAN COMMISSION

Brussels, 30.7.2025

SWD(2025) 232 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for a
REGULATION OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

on European fisheries and aquaculture statistics and repealing Regulations (EC) No 1921/2006, (EC) No 762/2008, (EC) No 216/2009, (EC) No 217/2009 and (EC) No 218/2009

{COM(2025) 435 final} - {SEC(2025) 224 final} - {SWD(2025) 233 final}

Table of contents

1.Introduction: Political and legal context

2.Problem definition

2.1.What is/are the problems?

2.2.What are the problem drivers?

2.3.How will the problem evolve?

3.Why should the EU act?

3.1.Legal basis

3.2.Subsidiarity: Necessity of EU action

3.3.Subsidiarity: Added value of EU action

4.Objectives: What is to be achieved?

4.1.General objectives

4.2.Specific objectives

5.What are the available policy options?

5.1.What is the baseline from which options are assessed?

5.2.Description of the policy options

5.2.1.    Option 1: Baseline scenario – continuation of the current policy    

5.2.2.    Option 2: Discontinuation of EFS    

5.2.3.    Option 3: New streamlined legal framework for EFS    

5.2.4.    Option 4: New legal basis for aquaculture and compilation of other EFS from EU level administrative sources    

5.3.Options discarded at an early stage

6.What are the impacts of the policy options?

6.1.Option 1: Baseline scenario – continuation of the current policy

6.2.Option 2: Discontinuation of EFS

6.3.Option 3: New streamlined legal framework for EFS

6.4.Option 4: New legal basis for aquaculture and compilation of other EFS from EU level administrative sources

7.How do the options compare?

8.Preferred option

8.1.REFIT (simplification and improved efficiency)

9.How will actual impacts be monitored + evaluated?

Annex 1: Procedural information

Annex 2: Stakeholder consultation

Annex 3: Who is affected and how?

Annex 4: Analytical methods

Glossary

1.Introduction: Political and legal context

Eurostat and the European Statistical System

Eurostat, the statistical office of the European Union (EU), ensures the production of high-quality, comparable European statistics 1 according to established rules and statistical principles, notably those laid down in the European statistics Code of Practice 2 and in the legal framework for European statistics known as the “Statistical Law” 3 . These rules and principles aim to ensure, among others, the independence, impartiality, objectivity and reliability of European statistics, and through those objectives public trust in the statistics. The main uses of European statistics are to serve EU policy design, implementation and monitoring, and their main users are EU Institutions.

The European Statistical System (ESS) is the partnership between Eurostat and the national statistical institutes (NSIs), as well as other national authorities which are responsible for the development, production and dissemination of European statistics in each EU Member State. This partnership also includes the European Economic Area (EEA) and European Free Trade Association (EFTA) countries, i.e. Iceland, Liechtenstein, Norway and Switzerland. The ESS functions as a network in which Eurostat's role is to lead the way in the harmonisation of statistics in close cooperation with national statistical authorities, which collect data and compile statistics for national and EU purposes. In order to fulfil this role, Eurostat issues statistical regulations and methodological guides, organises expert groups, and assesses the quality of statistics and countries’ legislative compliance. In accordance with the EU principles of subsidiarity and proportionality, each ESS member develops a statistical system suitable to their individual institutional context, while still following the common rules. The ESS also coordinates its work with candidate countries and, at EU level, with other European Commission services as well as with international organisations such as the Organisation for Economic Co-operation and Development (OECD), an intergovernmental economic organisation, and United Nations bodies such as the Food and Agriculture Organization (FAO), an agency tasked to defeat hunger and improve nutrition and food security.

Eurostat’s activities are further influenced by overarching policies such as the EU’s Better Regulation agenda 4 , which promotes open and transparent EU decision-making and evidence-based decisions, and European Commission President Ursula von der Leyen’s six political priorities, among them “A European Green Deal” and “An economy that works for people” 5 . Implementing, monitoring and assessing these priorities requires impartial and objective data – that is, official statistics.

European Fisheries and the Common Fisheries Policy

The EU’s Common Fisheries Policy (CFP) 6 , a major common EU policy next to the Common Agricultural Policy, regulates the fisheries sector in the EU by creating a framework that gives direction to the fisheries sector, lays out the rules for managing the EU fishing fleet and strives to ensure the long-term economic, environmental and social sustainability of European fishing. The CFP was last updated in 2014 and currently rests on four pillars:

1)fisheries management to balance fish stock sustainability with maintaining livelihoods for EU fishers, including setting rules and regulations such as annual quotas;

2)international policy to synchronise EU practices with practices and requirements of international organisations and regional fisheries management organisations (RFMOs), as one in five EU fishing vessels fish outside of Union waters;

3)market and trade policy to ensure that the fisheries sector complies with and fosters the common market policy of the EU and enables equal access to all Union waters to vessels from all Member States; and

4)funding to support the previous three pillars.    

The fisheries authorities of EU Member States collect data related to fisheries control covering the entire production and distribution chain (e.g. catch, landing, transport and first sales, as well as data relating to fishing effort, vessel characteristics, fishing licences etc.) under the Control Regulation and reporting obligations stemming from specific regulations and in particular from RFMO recommendations and international agreements 7 . The main data sources include logbooks, landing declarations, transport documents, sales notes, inspection reports, and vessel monitoring system records. Based on this information, Member States must transmit aggregated catch data and fishing effort data to the European Commission’s Directorate-General for Maritime Affairs and Fisheries (DG MARE) on a monthly basis for species subject to quotas and on a quarterly basis for all other species. Aggregated data is stored in the Aggregated Catch Data Reporting (ACDR) database, a Commission repository which is not publicly accessible. In addition, DG MARE manages data collection under the Data Collection Framework (DCF) 8 , with the aim of providing scientific advice for the CFP. Under the DCF, Member States collect biological, environmental, technical, economic and social data to help assess the state of exploited marine biological resources, the level and impact of fishing, and the socio-economic performance of the fisheries and aquaculture sector. Some assessments used for policy-making are performed by the Scientific, Technical and Economic Committee for Fisheries with technical assistance by the Commission‘s Joint Research Centre (JRC), whereas others are performed by the International Council for the Exploration of the Sea (ICES), the world’s oldest intergovernmental science organisation. The data collection is coordinated between Member States, end users and the Commission in regional coordination groups organised by sea basin, or by fisheries sector (e.g. long-distance or tuna fisheries). A dedicated group coordinates the socioeconomic data collection.

The EU regulates the fisheries sector because it is inherently international. Fish move on their own, often in international waters, and when exploited sustainably, are a renewable natural resource providing important nutrition for millions of people, as well as economic opportunities for fishers and coastal communities. Fish also play key roles in marine ecosystems. However, they are vulnerable to a collective action problem, as each individual fishing actor has a rational self-interest to fish as much as possible, with no initial possibility of excluding any actor from doing so. Mismanaged fisheries could lead to a rapid depletion of fish stocks, which indeed has happened in many cases in the past. For example, scientists of the JRC estimate that the Mediterranean Sea lost 34% of its total fish population between 1950 and 2011 9 . To prevent this collective action problem harming fisheries, the economy, the environment and society at large, international fisheries governance is therefore necessary (see also below).

EU-27 fishery production had a volume of 5.7 million tonnes live weight in 2018 and consisted to about 4/5 of catches of wild fish, and to about 1/5 of aquaculture production 10 . Globally, China accounted for about 35% of total fisheries production in 2018, followed by Indonesia with 13%, India with 6.1%, Vietnam with 3.6% and the EU-27 in fifth place with about 3% 11 . The production value of EU fisheries and aquaculture was estimated at 10.2 billion € in 2018 (6.5 billion € for landings and 3.7 billion € for aquaculture production), or 0.08% of the EU-27 GDP of 13.5 trillion € in that year.

Figure 1: EU Fisheries infographic, 2020. Source: Eurostat 12 .

In contrast, CFP support via the European Maritime and Fisheries Fund to the EU fisheries sector amounted to 6.4 billion € for the 2014-2020 programming period, or 0.6% of the EU budget for that period. Almost 80% of that amount is used to promote sustainable fisheries, around 10% is for control and enforcement, and about 9% is for Control Regulation and DCF data collection (EFS production costs are in addition to this cost, see below). 163.000 people, or 0.08% of all EU workers, were employed in the primary fisheries industry in the EU-27 in 2018, with Spain, Italy, Greece, France and Portugal making up about three quarters of that employment. There were a bit more than 76.000 fishing vessels in the EU-27 in 2018, with the highest numbers in Greece (almost 20% of the EU total), Italy (16%) and Spain (almost 12%), yet the largest gross tonnages were registered in Spain (about 25%), France (about 13%) and Italy (11%). The vast majority of EU fishing vessels is less than 10 metres long, but the larger vessels catch three quarters of the fish. Most companies in the fisheries and aquaculture sectors are small. In the fishing sector, 90% of companies operate only one vessel (see Figure 2), and it is estimated that 90-95% of these companies employ a maximum of 10 people. In the aquaculture sector, it is estimated that 92% of the enterprises employ a maximum of ten people 13 .

Figure 2. Vessel distribution by company size in 2018. Source: Scientific, Technical and Economic Committee for Fisheries  14 . Note: Data for Germany are missing.

The EU imports more than five times more fish by value than it exports 15 . Despite this comparatively low economic impact, fisheries are politically important, and significant for some coastal regions.

European Fishery Statistics

European Fishery Statistics (EFS) are official Eurostat statistics on the production, volume and value of fisheries products caught from the sea or cultivated in aquaculture facilities in the EU, Norway and Iceland (Liechtenstein and Switzerland are landlocked and do not report either sea fishery or aquaculture data to Eurostat). They are originally intended to contribute to the development, implementation and monitoring of the CFP, as well as to support EU trade policy, economic analysis, and environmental policy with the aims of long-term environmental, economic and social sustainability of European fisheries.

Eurostat has produced EFS since the inception of the European Economic Community in the 1950s. Currently, they are based on five statistical regulations 16  covering catches 17 , landings 18 and aquaculture 19 data, which are binding for EEA countries. The regulations lay down, among others, the fishing areas covered 20 , the statistical variables, data and metadata, transmission deadlines and statistical quality criteria. In addition, Eurostat compiles fishing fleet statistics directly from the EU fishing fleet register managed by DG MARE under separate regulations for the purpose of proper registration of all vessels allowed to exploit EU fisheries resources 21  (Iceland and Norway submit their fleet data directly to Eurostat). Most catch and landings statistics are now produced from administrative sources 22 developed under the CFP such as logbooks, landing declarations and sales notes, but national authorities also carry out statistical surveys, especially when administrative sources do not cover the complete sector. In the aquaculture sector, surveys and, most often, censuses are used to cover the largely fragmented sector and the detailed data requirements.

Eurostat processes, validates and finalises the data originally collected and first treated by EU Member States and then publishes them on the European statistics database on Eurostat’s website free of charge. The data is further published in statistical books, feeds diverse web applications, and is sent to the EU Institutions needing them and to various organisations (see also below). For example, EFS aquaculture statistics are relevant as the basis for other data collections 23 , in particular for freshwater aquaculture, where no other EU-level dataset is collected and published. They are also relevant for sustainable food production, especially in the context of the European Green Deal, whose “Farm to Fork” strategy 24 includes the share of organic aquaculture production in the EU as an indicator.

A recent evaluation of EFS 25 , conducted in 2018-2019, included an online survey on the purposes data users access EFS for. The most-mentioned purposes were (multiple mentions possible): CFP design, implementation and monitoring on national, EU or international level (62%), research (28%), commercial purposes, for example market intelligence (25%), environmental policy purposes (22%), and other policy purposes (13%). However, fewer than half of the users who identified the CFP purpose represented national authorities with responsibility for development of fisheries policy, with the majority being either research institutes supplying data for the CFP, or individuals using the data for independent research or analysis not necessarily directly related to official CFP purposes. As another data point, downloads of Eurostat fisheries datasets (catches by fishing area, landings of fishery products, aquaculture production by species and fishing fleet) from 2013 to 2018 and page views for the same time period have tripled from 8.565 to 27.647 26 , likely reflecting an increasing general interest in EFS data.

The estimated total cost of producing EFS is about 5.6 million € per year 27  for the 27 EU Member States and the European Commission, of which approximately 5% are incurred by the European Commission. The costs are higher for Mediterranean countries that have large small-scale fishing fleets whose fishing data is not recorded automatically for the CFP (this is because EU fishing vessels below 12 metres are not obliged by the Control Regulation to have electronic recording systems on board, so data is mainly recorded based on paper logbooks and must be fed into the electronic database). These costs refer to direct and indirect costs for data producers (NSIs and other national authorities), but do not include costs for data providers as a separate statistical data collection is only organised exceptionally for catch and landings data, and filling in the aquaculture questionnaire only takes a few hours per year (see below).

EFS costs thus represent only 0.05% of the annual EU fisheries and aquaculture production value. This is very low if compared, for example, to the cost of the decennial agricultural census alone, which represents approximately 0.6% of the production value of the far larger EU agricultural sector. The reason for this low cost is the wide reuse of available administrative data as source data for catch, landings and fleet statistics. These “raw data” are not specifically collected for EFS, but have to be collected by Member States under the Control Regulation for catches and landings, and under the fleet regulation for fishing fleet data. Additional costs for EFS over the CFP catch and landings data collection mostly arise from collecting data for vessels under the CFP’s recording thresholds, specific database queries, and the EFS validation procedure and quality reports. For aquaculture statistics, most EU Member States carry out joint collections for statistical and DCF purposes, which reduces costs. For example, the 2017 report from the Commission to the European Parliament and the Council on the submission by Member States of statistics on aquaculture 28 indicated that operators usually spend only around three hours to fill in the joint questionnaire per year. Aquaculture data costs related specifically to EFS mostly arise from adapting the questionnaires to the EFS requirements, the production of certain aggregates, and the EFS validation procedure and quality reports.

EU and global fishery data and governance environment

EFS are required by several EU Institutions and international organisations and exist in a complex environment of regional, national, international, supranational and global fisheries data flows, organisations, responsibilities and governance agreements.

Within Eurostat and the EU, statistics related to fisheries are collected in many places:

·statistics on fish processing are included in the statistics on the production of manufactured goods;

·statistics on the businesses working with fisheries and aquaculture are found in Structural Business Statistics, which focus on describing the structure, conduct and performance of businesses;

·statistics on the international trade in fisheries products are included in international trade statistics;

·statistics on the employment in fishing, aquaculture and related service activities are covered in labour force statistics;

·statistics on the organic production of aquaculture products are collected under organic farming statistics; and the European Food Price Monitoring Tool contains information about fish and seafood prices and their annual rates of change.

Eurostat also provides EFS to the European Market Observatory for Fisheries and Aquaculture Products (EUMOFA), a service established by DG MARE to provide weekly, monthly and yearly market information to the fisheries sector to enable better production planning and increased production 29 . EUMOFA uses various EFS data and collects specific additional data from auctions and markets in EU Member States.

Internationally, FAO compiles national catch and aquaculture statistics on a global scale, for which data are collected directly from countries. In the EU’s case, FAO uses Eurostat data as much as possible. The OECD also compiles and disseminates statistics on landings and aquaculture production for its members, drawing mainly from annual questionnaires, but also from EFS and other data. ICES uses DCF data and receives catch data from its 20 members, most of which are EU members for whom Eurostat provides most of the statistics.

Regional fisheries management organisations were established following the adoption of the United Nations Convention on the Law of the Sea in 1982, which called on participating countries to cooperate at global, regional and sub-regional level to manage and conserve fisheries on the high seas. They assess the status and commercial value of certain fish stocks they are responsible for, set catch limits or conduct inspections. Through the CFP, the EU has the mandate on fisheries management in EU waters, and the EU is a member of 17 RFMOs or other regional fisheries bodies. The reporting obligations of the EU towards RFMOs are mostly covered through the data collected under the DCF, and DG MARE reports fisheries activity aggregates to RFMOs on behalf of the EU.

FAO’s Coordinating Working Party on Fishery Statistics (CWP) provides a mechanism to coordinate the fishery statistical programmes conducted by regional fishery bodies and other intergovernmental organisations. It has served as the most important international and inter-organisational forum for agreeing common definitions, classifications and standards for the collection of fishery statistics since 1960. The group has 19 members, such as RFMOs, ICES, OECD, and FAO. Eurostat participates in the CWP on behalf of the European Commission in close cooperation with DG MARE and EU Member States.

Issues with European Fishery Statistics

The current EFS catch regulations recast older regulations adopted in the 1990s. The landings regulation repealed an older regulation in order to establish a common legal framework for data production based on principles of impartiality and reliability, ensure a better data transmission frequency and improve the data submission system. The aquaculture regulation also repealed an older regulation and aimed to organise the provision of statistics supporting the development and management of the aquaculture sector in the EU within the CFP, as the sector’s contribution to total EU fisheries production had increased over the years. All five regulations were adopted between 2006 and 2009, and it was expected that most source data would be collected by statistical sample surveys.

However, the fisheries data landscape has changed since then, following subsequent reforms of the CFP with the most recent one taking effect from 2014 on, the DCF recast in 2017 to provide scientific data, recent EU initiatives such as the European Green Deal with its focus on data for implementation and monitoring, and technological, economic and environmental changes in the fisheries sector. Consequently, fisheries statistics needs and requirements have changed too. The wealth of CFP administrative data now used as the main source for EFS was not foreseen when adopting the current regulations, and EFS are therefore ill-adapted to this data flow, generally inflexible and do not offer easy ways to include or accommodate new data needs, sources or methods.

Thus, there have been mounting indications from within the European Commission and from data users that the EFS regulations do not fully meet user needs anymore and place comparatively high administrative burdens and costs on data respondents and data producers. Moreover, since 2007, the European Court of Auditors has released three audit reports 30  highlighting redundancies and inconsistencies between the data collected and disseminated by Eurostat and DG MARE, as well as inherent problems within EFS.

For these reasons, the Commission started a project to streamline and simplify EFS in order to analyse and then improve this situation. The project is part of the European Commission’s Better Regulation Agenda’s programme on Regulatory Fitness and Performance (REFIT) because it aims to improve the performance of the underlying legislation while reducing burdens and costs.

The first step of the project was the EFS evaluation. It focused on the EFS regulations and the current CFP period starting in 2014 and covered all then 28 EU Member States, Norway and Iceland. Related information and fisheries data collections managed by other parts of the European Commission than Eurostat, mainly under the responsibility of DG MARE, were not subject to the evaluation, but were taken into account as contextual factors. The same applied to European socio-economic statistics on the fisheries sector such as production, trade, business and employment statistics, and to organic production statistics.

The evaluation found that EFS are an important and independent source of information for a wide range of users, but that their added value to manage the CFP is decreasing; in comparison, the Control Regulation and the DCF better meet fisheries policy management and analysis needs. The evaluation also found, among other issues, that EFS only partially cover EU catches, that landings of fishery products are not precisely and comprehensively defined, and that aquaculture statistics remain to a large extent confidential due to very detailed breakdowns of multiple statistical dimensions potentially allowing the identification of individual enterprises. Moreover, as EFS are part of a complex system of fisheries data in the EU and globally as described above, the evaluation found overlaps and discrepancies in parallel data flows, instead of them complementing and supporting each other. The evaluation thus confirmed the European Court of Auditors’ findings.

Eurostat therefore deemed it appropriate to conduct an impact assessment as the next step of the project to streamline and simplify EFS. Given the issues found by the evaluation, the impact assessment is to present objectives for improvement and systematically analyse and compare possible options to achieve them, as well as to analyse and compare the impacts of the options.

Scope of the impact assessment report

The scope of this report is accordingly the five regulations governing EFS, as these are under Eurostat’s full control to change. Related EU and other policies such as the CFP as well as the wider fisheries governance environment are taken into account as important context. However, as Eurostat’s influence on these is limited to cooperation agreements and/or being one of many members of groups such as the CWP, the related policies and governance venues are only included in the scope insofar as Eurostat can possibly influence them.

To understand this limitation of scope and of the project to streamline and simplify EFS altogether, it is important to understand the difference between European statistics and other types of data, such as administrative data or big data. Statistics can be based on these and other types of data, but they go through more validation and verification according to established, public principles, standardised definitions and so on in order to create an as accurate and reliable picture of reality as possible. Statistical authorities also take care to be independent in their work to prevent undue influence either via low quality data (hence the validations) or irregular processes. Statistics are often the basis for political decisions and scientific research, so it matters that they are as objective and independent as possible.

For these reasons, it is important and established institutional policy that European statistics are regulated by independent legal acts, and therefore EFS need to have a legal basis separate from the Control Regulation and the DCF. While this impact assessment report is accompanying the modernisation of the EFS legal basis, the Control Regulation and the multiannual programmes under the DCF are currently also being reformed by DG MARE in close cooperation with Eurostat (before entering their respective legislative procedures) in order to make the data optimally reusable for fishery statistics. However, Control Regulation and DCF data fundamentally serve different purposes – e.g. mandatory control, monitoring and research – than statistics, so it is justifiable from this angle that each domain maintains its own legal basis, optimally tailored to its own characteristics and needs.

The Commission proposal to revise the Control Regulation was tabled in 2018, and it is currently (September 2021) entering the trilogue phase of negotiations in the European Parliament and the Council of the European Union 31 . This revision aims at improving the collection of fisheries activity data. Reporting obligations are also not only established via the Control Regulation, but by other regulations as well, in particular one transposing international agreements into EU law. As regards the revision of data requirements under the DCF, the multiannual programme was adopted on 27 April 2021 32 after thorough stakeholder consultation, including Eurostat. The data collected for aquaculture under this programme concerns mainly socioeconomic variables; DCF thresholds below which EU Member States do not have to collect the data are based on Eurostat data on the weight and value of aquaculture production.

Eurostat and DG MARE are cooperating intensively to align their data collection parameters such as definitions and classifications in order to facilitate the work of data collectors and contribute to coherent data sets. All Member States have welcomed this work. Many EU countries choose to use Control Regulation data for EFS because they deem the data quality adequate, and according to the European statistics Code of Practice, already existing data sources should be used to decrease the burden on respondents. alternatively, Member States could also choose to use for example surveys, which they are free to do following the EU principle of subsidiarity. The importance of Control Regulation data as an EFS data source is thus not a de jure obligation, but a de facto choice. Or put another way, source data, whatever the source, are a prerequisite for statistics, but not statistics themselves, and this report focuses on statistics.

2.Problem definition

2.1.What is/are the problems?

The evaluation of EFS, the three European Court of Auditors reports on aspects of the EU fisheries data system and consultations with expert groups such as the Eurostat Working Group on Fishery Statistics and other stakeholders have resulted in concordant lists of problems, but also strengths of EFS.

As to the strengths of EFS, the evaluation found that the statistics are in general an important and independent source of fisheries information for a wide range of users in fisheries management, market monitoring and research and media. The main user groups of EFS are national authorities, non-governmental organisations (NGOs), national fisheries organisations, international organisations and research institutions. The most common uses have been linked to fisheries policies either at national, EU or international level. Research and commercial purposes were the second and third most common uses, followed by environmental policy and other policies. EFS thus appear to provide relevant statistics to monitor and evaluate European fisheries policy and to inform policy making in a wider context.

Moreover, stakeholders see the quality and independence of the data as high, and the EFS datasets are generally internally coherent. This means that the five legal acts framing EFS complement and do not contradict, overlap or duplicate each other. The internal coherence has improved in recent years because Eurostat coordinates closely with Member States in the ESS to guide and support countries and share best practices for the statistical process. EFS are also complementary to and not overlapping with or duplicating other European statistics such as trade, labour or organic farming statistics. In addition, data users consider EFS to be more accurate, reliable and complete than comparable collections, and they are seen as adding EU value because they offer long, independent, comparable, consistent, validated and publicly available data time series, which individual countries acting on their own could not easily compile. Respondents to the online survey in the EFS evaluation concurred and named as further benefits of the data a general European overview among countries and sectors which allows comparative analysis and benchmarking, and support for stakeholders’ advisory with official facts and data for modelling. The respondents also had general confidence in the statistics and appreciated the easy and free access, the added value in data quality verification and compliance monitoring, the timeliness and reliability of the data, and Eurostat as a one-stop shop for data instead of having to access up to 27 NSI pages with various language or technical barriers. These benefits were also confirmed by the case studies and stakeholder interviews undertaken for the evaluation. As stated, the statistics are also in high demand on Eurostat’s online database, more than comparable data collections in other domains, which indicates that the EFS target of reaching the general public is attained. The EFS regulations are crucial for this availability and quality of fishery statistics by stipulating, among others, lists of variables to be provided, main definitions, reporting and quality control procedures, and deadlines, with legal force.

Lastly, the evaluation also considered statistical quality as an additional, overarching criterion with metrics like accuracy and reliability, timeliness and punctuality, coherence and accessibility as well as process control, cost-efficiency and other factors. It found that EFS have high quality standards valued by data users, and are also in general published in a timely, punctual, accessible and clear fashion. The regulatory compliance monitoring system of Eurostat and the reaction speed of countries to compliance issues are also rated as good. As stated previously, the cost and additional administrative burden of producing EFS is very low (about 0.05% of the annual EU fisheries and aquaculture production value) in comparison to their possible benefits because they are mostly produced from administrative data already collected for CFP purposes.

However, this means there are few opportunities to reduce EFS costs and burdens further, either in absolute or relative terms. For example, the questionnaires sent to aquaculture facilities create a burden on respondents, but it is estimated (in the national metadata reports for aquaculture statistics sent to Eurostat) that it takes approximately only three hours to complete the joint questionnaire for EFS and DCF data collection. Given the size of the aquaculture sector and the nature of the data collection, the total EU-wide cost of the time spent to fill in the questionnaire is thus estimated as less than 200.000 €, i.e. small if compared to the economic value of the sector of around 3.7 billion € in 2018. In order to reduce the burden, countries have made the questionnaires more user-friendly and offered possibilities to report the data online, but any such efforts only have a very small financial effect as long as the statistics are being collected.

In all, EFS can thus be seen as relevant to the needs of professional users such as research institutions, national and international fisheries organisations and data redistributors (such as FAO, OECD and ICES) who use EFS as a reference or a validation source for their own statistics, and for purposes like market monitoring and analysis in the frame of EUMOFA. And while DG MARE fulfils EU reporting obligations towards RFMOs, some of the organisations use EFS for validation and country disaggregation purposes. In this manner, EFS fulfil stakeholders’ needs and provide timely and official information that is comparable across EU Member States.

However, despite these strengths and as referred to above, the evaluation also showed the statistics’ decreasing added value with regard to the purpose of supporting CFP policy-making and management, and diagnosed several other problems of EFS. In detail, these are:

·Sinking relevance: Subsequent and continuing reforms of the CFP have changed the policy’s data needs and data availabilities, to which EFS have structurally been unable to react adequately. EFS were originally supposed to support CFP development, monitoring and implementation. But data collection under the Control Regulation and the DCF, adopted after the current EFS regulations had been, now generates more detailed, disaggregated, frequent and timely catch and landings data than EFS, so DG MARE uses those data more frequently than statistics. For example, the ACDR dataset provides monthly figures to measure catches against the total allowable catches, whereas EFS only provide annual figures. EFS can also not easily meet the data needs resulting from the new CFP landing obligation, which mandates that all catches of certain marine species must be landed instead of being partly discarded. This is because the “unwanted catch” may not be used for human consumption, but EFS only focus on catches and landings for human consumption. In general, advising and reporting on progress in achieving CFP and European Green Deal objectives requires biological, economic, social, and environmental data beyond the data available in the EFS.

EFS are also in principle relevant for aquaculture to define thresholds for the DCF, as a reference for European Maritime and Fisheries Fund actions, and as an information source for DG MARE on fisheries in Norway and Iceland, especially regarding structural information, but the current EFS do not deliver the needed data, nor deliver them early enough. For example, the dataset on the structure of the aquaculture sector is currently collected, but not published due to persistent quality problems. Moreover, FAO has indicated that it would be interested in using EFS aquaculture statistics if they were published earlier and had fewer confidential values, and OECD has expressed a preference to use EFS instead of collecting fisheries data from EU Member States if they would fit their specific needs better.

More generally, the current EFS regulations were designed with the expectation that the statistics would be collected mainly by sample surveys, and for minor species by expert estimates. The availability of detailed CFP administrative fisheries data has upended this expectation, so the EFS regulations are not well-tailored to the predominant mode of EU fisheries data collection anymore. The aggregation level of EFS is generally considered as too high for almost all dimensions (species, geographical units and time), and some users would prefer them to be more timely, even if that implies a trade-off with quality and reliability – the earlier data is collected, the more one has to work with preliminary, not yet fully verified data. Users also ask for better data on by-catches, landings outside of the EU, catches in more than the seven FAO fishing regions currently included, and sometimes for easier usability of Eurostat’s on-line database. For these reasons, users currently seek the data they need from other sources, even if they are aware of the differences e.g. in quality between official statistics and other types of data. For example, fisheries data users at national level often use national fisheries statistics or mixtures of e.g. national, DCF and FAO data due to them having more detail or a broader scope.

Some of these data, for example those of FAO, are either based on EFS via transmission by Eurostat or structured by the requirements of the EFS regulations, possibly without users’ knowledge. This means that the relevance of EFS is higher than some users are aware of and than e.g. the number of direct downloads from Eurostat’s online database suggests (a “Brussels effect” 33 for fisheries data). Nevertheless, improvements in coverage, aggregation levels and dissemination aspects would increase the relevance of EFS, and EFS should be adapted to the changed information ecosystem in which they now operate by redetermining the desired relevance level and importance of fishery statistics (see also “fisheries data landscape” point below).     

·The EFS legislation is inflexible and data needs go unfulfilled: As stated in section 1, the EFS regulations are either recasts or moderate updates of older regulations originating in the 1990s. They were aligned with the treaty of Lisbon in 2013 by Regulation (EU) No 1350/2013 in order to increase their flexibility somewhat 34 , but much more legislative flexibility would be beneficial to enable quicker reactions to new or changing data needs, for example to more easily change lists of fish species to be collected, or to react to political developments, e.g. those following the EU exit of the major fisheries stakeholder of the United Kingdom, where important fishing waters lie.

As also stated previously, EFS currently only partially cover catches by EU fishing vessels, landings of fishery products are not precisely and comprehensively defined, emerging data needs e.g. on molluscs and seaweeds are not fulfilled or not fulfilled in a timely enough fashion, the aquaculture economic value at first sale is ill-defined and often estimated which can lead to large fluctuations between years and countries, and fisheries effort data is not well-developed and seems not to be used much. Solving these problems is not possible only via delegated and implementing acts because that would necessitate a wider restructuring of the EFS regulations. Moreover, the regulations are at the moment not open enough for technological and statistical innovation, for example with regard to new data collection methods.

·Too much confidential data: Some of the data collected for EFS are very detailed, leading them to be confidential and hence not publishable. This issue concerns half the countries and is particularly acute in aquaculture, where 13% of values are confidential. The detailed data needs of the EFS aquaculture regulation necessitate large samples, in many cases censuses, and long and complex questionnaires which place administrative burdens and costs on data producers and respondents and then result in confidential and publicly unusable data for about half of EU countries, and correspondingly a lack of EFS aggregates for aquaculture. The confidentiality is necessary because detailed public breakdowns could allow the identification of individual companies’ production or even of individual stocks they grow. This is because the EU aquaculture sector is highly specialised and dominated by only few companies in many countries, and because the current Eurostat statistical confidentiality rules make aggregates confidential even if only a small part of them originally was and even if the confidential information is of little economic importance in order to prevent the identification of other individual cells by subtraction. So far, Eurostat and the EU Member States have only made little progress in agreeing on new rules to make more EU aquaculture statistics aggregates available while preserving data confidentiality.

A similar problem is growing in landings statistics where four Member States have lately flagged a part of the data as confidential, yet at the same time, some EFS users request even more detailed statistics which would exacerbate the confidentiality issue. Therefore, this problem should be resolved with a coherent and standardised solution that balances confidentiality protection with data access. 

·Quality problems of EFS source data: As explained, EFS base data today come largely from administrative data sources such as data collected under control requirements (about two thirds) or statistical collections (about three tenths), so their accuracy and reliability depend in large part on the quality of this third-party source. But data collected under control requirements sometimes contain inaccuracies and mistakes such as wrongly reported or unreported species, particularly for smaller vessels without an electronic recording and reporting system, the fishing areas used are partly unaligned with those used for EFS, and there can be different deadlines. As the European Court of Auditors found, this is in part because EU fisheries control is not effective enough: some Member States are not rigorously following up on missing, incomparable or erroneous data or do not apply strong penalties. However, these source data problems are hard to resolve by legislative measures under Eurostat remit.

In general, third-party data sources can present multiple problems for statistical bodies: technical problems such as creating interfaces to access administrative databases, organisational problems such as establishing durable, if possible contractual relationships with the holders or managers of such data sources, and substantive problems if, for example, those sources change their classifications and definitions in justified response to their use cases changing.

These problems have also been confirmed within the EU Institutions by a 2017 audit of the European Commission’s internal audit service on “other statistics” 35 , i.e. statistical activities of the institutions and bodies of the EU which are coordinated and steered by Eurostat to ensure data quality and consistency, with a mandate stemming from the “Statistical Law”. The audit found that there are many issues with the independence, robustness, sustainability, reliability and quality of these data and the process of disclosing, managing and producing them, and some relevant data are not even declared as “other statistics” yet. As statistical services in the EU and around the world keep facing budget cuts while the need for official statistics keeps growing, the services turn more and more towards already existing third-party data sources such as administrative data, which is also recommended by the European Statistics Code of Practice. Solutions for the problems encountered with these sources would therefore benefit not only EFS, but other statistical data collections too.

·Gaps, overlaps and discrepancies in the EU and global system of fisheries data: Ideally, the parallel fisheries data flows within and outside of the EU should complement and support each other, especially if they mostly originate from the same source data, as e.g. data collected under control requirements and EFS data do. But instead, there are still significant differences between EFS and data collected under control requirements in Mediterranean countries even though the difference is small in the EU-wide aggregate and has decreased over time due to stakeholders’ efforts. This is partly due to the large number of small vessels in these fleets, and the fact that these small vessels can benefit from derogations to some of the Control Regulation and therefore need to be covered by surveys. In 2016, EFS showed 2.9% higher total catches than ACDR data, with most of the difference stemming from Greek (+34%) and Italian (+42%) data. The evaluation of EFS also found that there are discrepancies of more than 100 tonnes in 15% of records between Eurostat, FAO and ACDR data in 2016. The main reasons for the differences are different methods in catches estimation (weighted landings and sales notes in EFS versus logbooks and landings in the control framework), different species reporting and classifications, differences in population and coverage, different organisations submitting the data and different deadlines and revisions. For example, ACDR includes data for 2289 species, the DCF for 2014, EFS for 1273, and FAO “only” for 847. The Commission is currently tackling these differences bilaterally with the concerned countries 36 .

This lack of harmonisation and coherence makes the data less comparable than it could be and is a source of additional burden on fisheries data producers. This is because while data is originally mostly collected through a single process, the need to recollect, reprocess and submit it in different formats with different definitions, classifications, aggregations and validations and at different times for different organisations imposes time and resource costs and a potential for errors that could be reduced through greater harmonisation between EFS, Control Regulation, DCF and international organisations’ data requirements. For example, both Member States and Eurostat report certain data to OECD, sometimes resulting in discrepancies. These could be eliminated or reduced by having single transmissions according to the principle “collect once, use multiple times”. This has already been implemented with FAO catch data which are now only collected from Eurostat and no longer from EU Member States, at least for those FAO fishing regions covered by the EFS regulations.

Within Eurostat, the other European statistics linked to fisheries are statistics on international trade, production of fisheries products, businesses and labour force working in the sector, and organic aquaculture statistics. The level of detail of these data is not always compatible with EFS. For example, trade and processing statistics are not broken down by catches and aquaculture. As stated above, these statistical domains complement each other, but they do not offer an optimal set-up for analysing e.g. “blue growth” 37 potential in detail. This could also be improved.

As stated, the costs of producing EFS are low in absolute terms and compared to the production value of the fisheries sector, but they are high compared to an ideal state without double reporting and discrepancies. National statistical authorities are under severe budget pressure and therefore constantly seeking to reduce costs, for example by asking the Commission to modernise European statistics data collections in order to enable savings. From their perspective, the costs of EFS are therefore higher than they could be.

Finally, these inconsistencies in EU, Member State and international organisations’ fishery data sets are confusing and difficult to explain to data users and thus pose risks to the credibility and reputation of the EU.

·The place of EFS in the EU and global fisheries data landscape is unclear: This is in part a consequence of the other problems. Each of the EU and international fisheries data sets, whether statistical, scientific or administrative, has a certain purpose and should be relevant for that purpose. Not all data have to or can be relevant for every purpose; for example, scientific data can be more relevant for stock assessments than EFS data, even though e.g. EFS time series on catches can serve as important contextual information for the setting of fishing quotas. However, as described above, EFS are currently not used much for their intended purpose of serving the CFP. Given the changes in the EU and global fisheries data landscape following CFP reforms and international organisations’ evolving data needs, the overall role and purpose of EFS within it should therefore be redefined so they can be optimally produced and used. EFS should also be flexible enough to be able to react to further changes in the landscape, as new policies, new technologies and other developments may make the most suitable data sources for a task change over time. For example, research on fishing effort in an area was previously based on statistical data with a low spatial resolution and a long time lag. Today, it can be conducted based on vessel monitoring systems and with electronic logbook data in close to real time, so fishing effort statistics play less of a role and could conceivably be flexibly de-emphasised or even taken out in a new EFS setup.

EFS are official statistics which means their production follows certain principles such as objectivity, impartiality and reliability. They are therefore in principle well-tailored to provide validated and reliable data on unit or aggregated level which show major trends and serve as an official one-stop reference and comparability benchmark for other data collections. Official statistics should also in general be the main evidence at the heart of evidence-based policy-making because they are more independent, validated and trusted, i.e. of higher quality, than other sources of data. The better the data, the better decisions are likely to be. Improvements to EFS that resolved the problems described above would strengthen these characteristics and enhance the EU’s reputation through better policy design and delivery.

On the other hand, official EFS cannot and should not serve short-term administrative needs, policy enforcement needs or scientific biological needs, to take some examples. The validation processes of official statistics take time, and they are thus unsuited for very quick outputs. Enforcement has political and economic consequences which puts data used for enforcement under political pressure, an incompatible state of affairs for official statistics which need to be independent. Lastly, statistics cannot cover complex and changing biological phenomena with enough detail and speed. Data users should be supported in discerning which source is optimal for which need and in which context.

EFS should thus respond to the changed fisheries data landscape by redefining their own place within it, but also by helping to clarify the roles and purposes of other fisheries data collections, not least to reduce the gaps, overlaps and discrepancies between them, as described above. A well-organised EU and international fisheries data landscape would allow each data collection to be used as optimally as possible and improve comparability. Eurostat as the statistical office of the EU is well placed to play a leading role here due to its long experience with data collection, validation and transmission and its cooperation with services both within and outside of the European Commission. As noted, the Commission is represented by Eurostat (in co-operation by DG MARE) in the CWP, which has developed common procedures to streamline processes, provided technical advice, and published methodological and reference documents. Some non-statistical fisheries data sources seek to align with CWP definitions and recommendations, but not all do, in particular when these definitions differ from FAO’s generic ones. This could be improved in a concerted effort. For example, an opportunity could be used to further harmonise EFS with the currently ongoing update of the Control Regulation aiming to improve data quality and availability, as well as with the revision of the DCF’s multiannual programmes, where Eurostat’s suggestions on coverage and exemptions have already been taken on board.

Harmonisation is especially beneficial in this field where clear responsibilities, accountabilities, procedures and definitions are necessary to solve the inherent collective action problem of fisheries. An analysis of existing agreements of Eurostat with various international organisations shows that some are outdated, incomplete or allocate responsibilities unclearly. Eurostat recently signed an administrative arrangement with FAO to improve data exchange based on domain-specific technical arrangements specifying contents and conditions of data sharing, but this is only a beginning. EFS are used as inputs by ICES (catches), OECD (fishing fleet, in principle landings but for EFS incompleteness), FAO (catches, fishing fleet, in principle aquaculture but for confidentiality and timeliness) and RFMOs (catches). DG MARE represents and provides data to six tuna RFMOs and nine non-tuna RFMOs 38 . 

Lastly, the objectives and purposes of EFS are currently not clearly stated in one legal document, but sometimes only mentioned implicitly, spread across several policy documents and regulations, or inferred from wider legislation relating to Eurostat’s mandate and the European fisheries sector. A clear, single definition could serve as a polestar that EFS could be oriented towards and that could be used to resolve uncertainties as to what EFS are for.

As to the coronavirus crisis, it has had little effect on the EU fisheries sector as an important part of EU food production, and most fishery data work can be done remotely, particularly if data collection is digital as with fishing vessel electronic recording and reporting systems. The problems outlined in this section are also unlikely to be affected by the crisis because they are general issues irrespective of possible short-term data collection problems; if anything, the coronavirus crisis will contribute to a further push to digitise and harmonise data collection, which EFS should take advantage of. But not every possible crisis is likely to leave the sector mostly unscathed, so EFS should be set up in a way that allows them to react flexibly to rapidly changing data needs or availabilities and other important external developments.    

2.2.What are the problem drivers?

The problems described above are mainly driven by the following factors or developments:

·The EU and global fisheries sector keeps changing e.g. due to

oThe climate crisis, which has effects, among others, on fish stocks and marine ecosystems, which can have effects, among others, on the fishing areas frequented, the most important species fished or the gear used

oPersistent overfishing and other sustainability challenges which have effects, among others, on fish stocks and marine ecosystems

oSocioeconomic developments, such as more expensive fuel for fishing vessels, the shrinking number of European fishers, the difficulty to attract newcomers to the sector and changing customer tastes and sensibilities such as a higher demand for sustainably caught fish, which have effects on the fisheries sector’s behaviour

oTechnological developments, such as more effective means to track shoals or innovations in aquaculture feed and rearing

oPolitical developments such as Brexit, which could affect EU access to the important fishing waters around Great Britain.    

These changes are outside of Eurostat’s power to influence because they are based on large-scale environmental, social, economic and political forces, but through their effect on the fisheries sector, they determine which fishery statistics become available, relevant, important, and needed.

·Partly in response to these changes in the fisheries sector, but also to proactively shape fisheries and due to political pressures, the EU keeps regularly changing the CFP in order to safeguard the sector’s economic, environmental and social sustainability. This important regulatory and financial environment in turn shapes the fisheries sector’s behaviour and incentives, which again has effects on data relevance and needs. As described above, the CFP also heavily influences basic data availability for Eurostat as well as the EU and global fisheries data landscape through large administrative data collections such as the ACDR database. Eurostat is part of the European Commission and cooperates with DG MARE on the data side of the CFP, but has little input into the main drivers shaping each reform.

·Other EU policies such as the European Green Deal and other organisations’ actions to regulate the fisheries sector and/or to collect, treat or redistribute fishery data further contribute to a dynamic EU and global fisheries and fisheries data landscape.

·Lastly, the EFS legislation being inflexible is both a problem of EFS – in that it is an issue with negative effects that can and should be solved to improve EFS – and a problem driver – in that this inflexibility contributes to other problems such as the sinking relevance of EFS, gaps, overlaps and discrepancies with other fisheries data collections, and an unclear place of EFS in the fisheries data landscape. As long as it remains unaddressed, this issue will drive EFS further losing relevance.

2.3.How will the problem evolve?

In the last decade, the information needs for the CFP have grown a lot, changing the framework in which EFS operate and their role in the fisheries data ecosystem. The reporting obligations of the reformed Control Regulation have made a wealth of administrative information available, not only e.g. for monitoring quota uptakes, but also as a building block of many other datasets, including EFS on catches and landings. However, the EFS regulations were not specifically designed to use mainly administrative data as source data, but sample survey data. This original premise had a strong impact on the EFS regulations, for example as regards available aggregation levels. DG MARE and other users have consequently been using CFP and other data to fulfil their data needs because EFS cannot match the wealth of information offered.

During the last decade, DG MARE also recast the DCF which collects not only marine biological data for scientific purposes, but also data for other policy needs. For fisheries, it entails variables not collected by EFS or other European statistics in detail, such as socioeconomic data. Thus, the relevance of EFS has declined both in the European Commission and among external users as compared to their goal of supporting CFP policy development, implementation and monitoring.

Without changes to EFS and considering the problems and problem drivers outlined above, EFS would likely decline further in relevance for global fisheries stakeholders and for CFP purposes, as fisheries data users would more and more use other data sources that fulfilled their needs better, responded more flexibly to evolving requirements, and fit better into the global fisheries data landscape. EFS would thus slide towards obsolescence, and fishery stakeholders would progressively be deprived of a high-quality, relevant and reliable data source, with consequent negative impacts on EU and global fisheries policy. As stated above, the better the data, the better decisions are likely to be. The converse is also true.

3.Why should the EU act?

3.1.Legal basis

Article 338 of the Treaty on the Functioning of the European Union 39 outlines the EU competence to adopt measures for the production of statistics where necessary for the performance of the activities of the Union. In order to allow an accurate and independent follow up of the CFP, an exclusive competence of the Union as regards the conservation of marine biological resources (Article 3 of the Treaty), high quality, comparable and harmonised statistics across the EU and internationally are needed. This is because common policies in and of themselves require such data to allocate resources fairly, efficiently and effectively and help make the best possible decisions across Member States.    

3.2.Subsidiarity: Necessity of EU action

Fisheries are a natural, renewable and movable resource for food and other uses seen as part of humanity’s common heritage. In the EU, they are governed by a common policy, the CFP, with common rules adopted at EU level and applied in all Member States. The objectives of the CFP are to ensure the environmental, economic and social sustainability of fisheries and aquaculture and to provide a source of healthy food for EU citizens. Without a common policy at supranational level to ensure a fair and level playing field, a collective action problem would ensue where each individual fisheries actor would have a rational self-interest to fish as much as possible, with no initial possibility of excluding any actor from doing so, as it is impossible to monitor the entirety of the wide seas even today. As referred to above, many fish stocks have historically been depleted or put into a precarious state in the absence of strong common rules.

As an exclusive competence of the EU, the CFP is inherently subject to EU action. Producing EFS can be better achieved at EU level because only the Commission can coordinate the necessary harmonisation of the statistical information at Community level to serve the needs of the CFP managed at the EU level, and those of other EU policies requiring fisheries statistics.

3.3.Subsidiarity: Added value of EU action

EU action in fisheries statistics, i.e. collecting, harmonising and publishing fisheries data based on common regulations, facilitates the collection of data at the same level of detail in all Member States. It imposes the application of common standards, definitions and methodologies which, in addition to producing comparable results at the service of management and analysis needs for the CFP and other EU policies, improve efficiency, timeliness, reliability and harmonisation, or in short, increase data quality. A common fisheries policy must inherently be based on comparable, harmonised and high-quality data which can only be ensured by action at EU level.

The regulations framing EFS are crucial for the availability and quality of the statistics. They stipulate the lists of variables to be provided, main definitions to be used, the reporting system, submission and assessment procedures, deadlines and geographic coverage. In addition, the regulations ensure quality control mechanisms and the availability of metadata. For these reasons, national EFS providers and institutional stakeholders such as Commission services, FAO or ICES have stressed the importance of a legal basis for EFS.

The added value generated by the obligation to comply with the EFS acts stems from the complete chain starting from user needs, definitions and requirements, continuing with the co-ordinated compilation and transmission of statistics and with the shared quality criteria and validation framework, and finishing with the monitoring ensuring compliance with the legal frame. This complete process ensures the availability of EFS covering the EU and EFTA countries to the wide user community. The availability of long, comparable time series of EFS in a consolidated publicly available, free of charge database accompanied by harmonised metadata could not be achieved by individual actions of countries, yet is advantageous to track trends and monitor and evaluate policies. The use of separate national sources, even if they were harmonised – which they would have little incentive to be without a common rationale and coordinating body –, would be much more cumbersome and hindered by technical and linguistic barriers.

EFS also create added value when compared to other types of fisheries data; institutional users consider the independence of official statistical information to be fundamental. Policies cannot be evaluated and analysed credibly without independent, objective data sources such as European statistics. Moreover, the economic playing field is levelled by reducing information discrepancy between larger and smaller actors, leading to healthier competition.

The ESS with Eurostat as the coordinating body and guardian of statistical principles ensures the application of the European Statistics Code of Practice, the harmonisation of definitions and statistics, the maintenance of updated statistical laws, and alignment with international statistical standards such as those of the CWP. These statistical principles are enforced by Eurostat validation and quality reporting and highly valued by national and international users because they meet their needs. They are also taken as an example by other institutions working with fisheries data, and due to the complex web of EFS data transmissions to international organisations mentioned above, users of e.g. FAO fisheries data based on EFS also benefit from European statistical principles, sometimes without being aware of it. Similarly, national aquaculture data, which some users prefer over EFS aquaculture data, have often been collected under the rules of the EFS aquaculture regulation. This wide relevance of EFS beyond the data in Eurostat’s online database has been confirmed by case studies and interviews conducted for the evaluation.

In short, the EU added value of EFS lies in the consistency, quality and internal coherence of the statistics, as well as their accessibility and clarity. EFS provide an EU-wide framework for data collection and supply of fisheries and aquaculture statistics using concepts and definitions that are harmonised across all Member States. This could not be achieved by Member States acting in isolation, but only by a common and coordinated approach in the ESS.

4.Objectives: What is to be achieved?

4.1.General objectives

The general objectives of EFS are currently outlined in Article 2 of the “Statistical Law”, Regulation (EC) No 223/2009, which lists the statistical principles governing the development, production and dissemination of European statistics; intermittently and implicitly in the five EFS regulations; and in objective 8.4. of the European Statistical Programme (ESP; the framework in which all statistical activities of Eurostat are embedded), which states as the goals of EFS to process and publish independent, accurate, timely and comparable fisheries statistics at EU level to support the development and monitoring of the CFP, the management of EU fishery resources and of the fishing and aquaculture sectors and markets, and meeting the EU’s international obligations to provide fishery statistics to international organisations and RFMOs. These aims are also tied to the six main priorities of the von der Leyen Commission, in particular “A European Green Deal” and “An economy that works for people”.

In order to optimally fulfil these objectives, the project aims to streamline and simplify EFS by improving the system of collecting, producing and disseminating European fishery statistics so that they can better and more flexibly meet user needs, while at the same time reducing burdens and costs on data respondents and producers, eliminating gaps, overlaps and discrepancies with other fisheries data collections, and finding the appropriate place for EFS in the global fisheries data landscape.    

4.2.Specific objectives

·Increase the relevance of EFS. For example, the EFS regulations should be aligned with the latest CWP concepts and classifications. EFS catch data usually do not include discards or other catch that is not landed because they are mostly calculated from landings using different conversion factors by species and presentation (e.g. gutted mackerel or plaice fillet) as “nominal catches”, which makes them less usable for the CFP landing obligation. This could be changed. And EFS aquaculture data should be improved to contain fewer confidential values and more timely data publication.

The attainment of this objective could be measured by an increase in the number of data downloads from Eurostat’s database and by making an inventory of new data needs to check when they have been answered. These metrics would point to an increased interest in EFS, suggesting a higher relevance ceteris paribus, and show EFS answering to evolving data requirements.

·Meet EFS user needs better and more flexibly. As stated, the current legal framework of EFS is successful in producing quality statistics, but user needs have changed and continue to change, and the evolution in the nature and availability of alternative data sources merits reexamining and updating the legal acts to ensure that they are flexibly addressing those user needs that cannot or should not be met by other data sources. For example, users have requested that EFS should cover more FAO fishing areas than the current seven. In order to make better use of delegated and implementing acts that offer the possibility for comparatively rapid changes e.g. in variable lists or allowed methodological approaches, the regulations could be reconstructed in a modular or framework fashion. In order to resolve definitional issues and reduce complexity, the rules that are currently spread across five regulations, but also across parts of the control framework and the DCF, could be consolidated into a single legal act. Legislative flexibility is also needed to take account of innovations and emerging phenomena in the aquaculture sector, where e.g. more seaweed and algae data and better definitions of “first sale” and “juveniles” could improve the economic picture.

Lastly, while EFS provide high-quality statistics overall, there are some weaknesses in the level of quality of data collection at Member State level that should be addressed to further improve the overall quality of the statistics. The evaluation found different levels of quality checks undertaken and different data collection processes at national level. Additional training and support for Member States as well as clearer quality guidelines in legislation would improve the quality of EFS overall in future years, particularly with respect to the design, implementation and quality assurance of the data collection.

This objective could be measured by indicators such as the share of FAO fishing areas covered by catch statistics and the shares of EFS users rating their quality, timeliness and comparability as good or very good, in order to create an external quantitative assessment in addition to the internal, more process-focused quality assessment.

·Resolve the EFS confidentiality issues. In order to reduce the high share of confidential data in EFS aquaculture and increasingly in EFS catch statistics and thus to increase their usefulness, fewer breakdowns for less important aquaculture and catch species should be requested, and Eurostat should also adopt more advanced rules of protecting confidentiality while ensuring as much data access as possible.

This objective could be measured by counting the number of confidential data cells and the number of published EU aggregates in aquaculture statistics.

·Solve the quality problems of EFS source data. Eurostat should better clarify and codify the responsibilities and rules for the access, use, amendment and maintenance of administrative data sources with the holders of those data sources, for example as regards interfaces, validation, quality and adherence e.g. to the European Statistics Code of Practice. These cooperations should be tailored to the respective relationships which can differ for example with regard to the frequency and importance of data extraction or whether public or private entities are holding the data. The instruments used to formalise the cooperations could accordingly range from Memoranda of Understanding to legislative powers such as those already included in the “Statistical Law” or those recently included in Regulation (EU) 2019/2152 on European business statistics (Article 5) 40 .  These instruments would help resolve issues, minimise transaction costs, stabilise the relationships and also serve as a model for similar relations in other domains of statistics. In the future, statistical offices and administrative data holders should cooperate from the beginning in order to set up administrative data collections from the ground up so that more synergies are achieved, for example by aligning definitions from the start and periodically realigning them, as different implementations and uses will over time inevitably lead to discrepancies. These actions would help overcome otherwise systemic difficulties for statistical offices to access, influence and validate administrative data sources.

Measuring this objective would be more challenging than for other objectives, but could be achieved, for example, by checking whether the quality assessment of administrative source data is better aligned with statistical quality criteria.

·Reduce gaps, overlaps and discrepancies in the EU and global system of fisheries data. Persistent large overlaps and discrepancies as evidenced in the EFS evaluation incur a reputational risk and confuse statisticians, policy-makers, media and the public. The principle of “collect once, reuse multiple times” should be followed as much as possible to reduce burdens and costs and increase clarity. For this, data definitions, reporting requirements, handbooks etc. should be aligned between Eurostat, DG MARE, international organisations and RFMOs as much as possible, and data flows should be optimised towards single data collection points and/or standardised information exchange platforms. For example, ideally none of the data collected by Eurostat should also be collected from EU Member States by international organisations. At a minimum, the EFS evaluation showed that optimally, only Eurostat should deliver EFS to FAO and OECD on behalf of EU and EFTA countries. Eurostat should therefore intensify its harmonisation efforts which have so far yielded some successes, e.g. OECD sourcing aquaculture data only from FAO instead of also from EU Member States, but still need to tackle important areas of gaps and overlaps such as between EFS and DG MARE’s fisheries data collections. Eurostat would be well-placed to take on a central coordinating role in this task by continuously engaging and consulting with Member States and international organisations, as it is already doing so. Revisiting the legal framework for EFS would also help in addressing these issues.

Measuring the attainment of this objective would be relatively straightforward by, for example, assessing the number of administrative sources used replacing survey data, and finding a reduction in the number of data flows from countries to the EU and international organisations.

·Clarify the place of EFS in the EU and global fisheries data landscape. As described in section 2, the objectives of EFS are currently spread across several policy documents and legal acts and not entirely clearly stated, the statistics are not used much for the purpose of serving the CFP, and the EU and international fisheries data landscape has changed a lot since the adoption of the EFS regulations. The objectives of EFS should therefore be redefined and stated in a clear and concise manner in a single document, preferably with legal force, and Eurostat should support further efforts to reorder the EU and international fisheries data landscape for the benefit of data users and producers. Areas of contradiction, overlap, duplication or complementarity as well as stakeholders’ relevant experiences and perceptions should be identified and then improved in cooperation between Eurostat, other Commission Directorates-General and Services, and international organisations.

In addition, given the continued precarious state of many marine basins and fish stocks particularly in the Mediterranean, an argument could be made that the current EU and global fisheries data landscape is not contributing enough to the environmental, social and economic sustainability of fisheries, and should be reordered accordingly.

This objective could be assessed as attained if there were a single reference document stating the objectives of EFS and if the wider fisheries data landscape were discussed in the Directors’ Group on Agricultural Statistics (DGAS), within the EU and in international venues such as the CWP.

These objectives are interlinked in a number of ways. For example, meeting EFS user needs better and more flexibly would also increase EFS’ relevance and help resolve confidentiality issues. Clarifying the place of EFS in the fisheries data landscape would go a long way towards reducing gaps, overlaps and discrepancies and give impulses for how best to meet user needs. And mproving EFS source data would contribute to reducing gaps and discrepancies as well as to increasing the relevance of EFS. Figure 3 illustrates the full relation of the problem drivers to the interconnected problems, the objectives and the options (see sections 5 and 6) of the modernisation of EFS. Solid lines indicate full influences, dotted lines partial influences. Except for the inflexibility of EFS, most problem drivers cannot be directly tackled because they are either not under the control of Eurostat or, as e.g. in the case of regular CFP reforms, a necessary reaction of the European Commission to changes in the fisheries sector. As the EU wants to keep regulating the sector, it needs to adapt its relevant legislation to changes in the sector.

Figure 3: Problem drivers, problems, objectives and options of streamlining and simplifying EFS. Source: Eurostat.

5.What are the available policy options?

5.1.What is the baseline from which options are assessed?

As described above, Eurostat produces EFS based on five statistical regulations covering catches, landings and aquaculture, which are binding for EEA countries. These regulations have no end date and could therefore continue applying indefinitely. EU Member States fulfil their EFS obligations for catches and landings mostly by reusing administrative data collected for the implementation of the CFP, which they supplement with further data e.g. on small fishing vessels which are not covered by the administrative data, but required by EFS. Aquaculture statistics, on the other hand, are mostly collected using surveys and censuses. In addition, Eurostat produces EU fishing fleet statistics directly from the EU fleet register maintained by DG MARE without receiving data from Member States itself, and Eurostat also delivers fishery statistics to various international organisations such as OECD and FAO.

Hence, the baseline or status quo option would mean that the EU Member States and Eurostat would continue producing EFS under the current legal framework of five regulations, and Eurostat would streamline and simplify the framework as far as possible through delegated acts. Such adaptations could simplify and align some definitions, the lists of species and fishing regions, and in the case of Regulation (EC) No 218/2009 also the permitted data aggregation level, yet they would by design only affect non-essential elements of the legislation, i.e. be far from a significant overhaul.

The groups mainly affected by EFS are data providers, i.e. fishers, aquaculture producers and others providing source fisheries data, and data producers, i.e. national statistical institutes and other national authorities collecting and treating fisheries statistics. Data users such as EU bodies, redistributors or scientists are also affected by the statistics as they use them for their work, but they do not have to do work themselves to create them. Very small enterprises, so-called “micro-SMEs” with fewer than 10 employees cannot practically be excluded from EFS data collection because most EU fishing vessels have crews smaller than that.

Producing EFS only costs about 5.6 million € annually for the entire EU, about 0.05% of the annual value of EU fisheries and aquaculture production. Whatever the option chosen, it can be expected that further digitalisation, the consolidation of the fisheries sector into fewer, but bigger operators, more efficient practices at statistical authorities and similar developments will over time reduce the costs of data collection and production. Any change to EFS has therefore only a small potential to reduce costs either in absolute or relative terms, but as stated, the costs are high compared to an ideal state, which national statistical authorities seek. However, several not directly monetary benefits such as more and better harmonised fisheries data, which could be used to improve the design, implementation and monitoring of EU fisheries and related policies, are sought with some options, as analysed below. In general, high-quality official statistics are a comparatively cheap public service that indirectly more than pays for itself by providing an important part of the necessary objective evidence for evidence-based policies.

The main benefits of EFS, as evidenced by the evaluation and the previous sections, can be summarised as follows:

·offering a one-stop shop overview of the European fishing and aquaculture sector which allows comparison, benchmarking, and analysis of long time series for national, EU and international policy making and research, and thus creates a high EU added value;

·supporting users with fisheries facts and figures to level the playing field and enable informed commercial and market decision making with reliable statistics and trends;

·providing all users free of charge with official, reliable, high quality fishery statistics that are subject to high levels of data quality verification, and are understood by users to be of high quality.

These benefits are hard to quantify in a meaningful way, but are broad-ranging and consistently mentioned by stakeholders. EFS are highly regarded among all types of users, especially for their comprehensively validated quality and reliability, and are used as a reference source for many users and as a validation tool by many important international organisations.

However, the problems of EFS as documented above affect assessment criteria such as those used for the evaluation:

·EFS have less relevance than they could have.

·EFS show effectiveness, but could be even more effective.

·EFS are produced with efficiency, but it could be improved.

·EFS show internal coherence, but their external coherence could be improved.

If the status quo of EFS continued without solving these problems, the cost-benefit ratio of EFS would likely shrink, as they would fulfil evolving requirements for the CFP and other EU policies less and less. The costs and administrative burden would remain about the same as now, with small improvements due to some streamlining of definitions and updating of classifications via delegated acts. On the other hand, data providers, data producers, Member States and users have a good understanding of the current policy and legal framework of EFS, so the data collection and validation systems would continue working and producing high-quality and comparable European fishery statistics, even if new needs were not fully satisfied. A more detailed impact analysis for this baseline option can be found in section 6.1.

The six assessment criteria indicated above can be mapped to the objectives described in section 4 as follows:

·Increase the relevance of EFS maps to the criterion of relevance.

·Meet EFS user needs better and more flexibly includes the criteria of effectiveness and efficiency.

·Resolve EFS confidentiality issues maps to effectiveness.

·Solve the quality problems of EFS source data includes effectiveness, efficiency and statistical quality.

·Reduce gaps, overlaps and discrepancies in the EU and global system of fisheries data maps to the criteria of efficiency, coherence, EU added value and statistical quality.

·Clarify the place of EFS in the EU and global fisheries data landscape includes efficiency, coherence and EU added value.

Going forward, these objectives will be used for the impact assessment in section 6. In addition, the opinions of the EFS stakeholders on each option are noted.

5.2.Description of the policy options

5.2.1.Option 1: Baseline scenario – continuation of the current policy

See section 5.1.

5.2.2.Option 2: Discontinuation of EFS

All EFS regulations would be repealed. EU fishery statistics would from then on be derived from EU level administrative data to the extent possible, similar to the current approach for fishing fleet statistics, which Eurostat derives directly from the EU fleet register administered by DG MARE and then publishes without asking Member States for data separately or having a separate legal basis for fleet statistics. However, these administrative data would not be supplemented by requirements specific to EFS anymore, and they would not be treated statistically (validation, quality reports, metadata provision etc.) anymore. EU Member States would continue collecting the fishery statistics they need for their national and international purposes, but not send them to Eurostat anymore.

5.2.3.Option 3: New streamlined legal framework for EFS

A new, single streamlined framework regulation for EFS covering catches, landings, fishing fleet and aquaculture statistics would be introduced, and the current five regulations on catches, landings and aquaculture would be repealed. This would be in line with the REFIT programme and satisfy the von der Leyen Commission’s “one in, one out” principle for new legislative proposals by reducing the burden of EU legislation.

The structure of the new EFS framework regulation would likely resemble that of other recent Eurostat framework regulations: a basic act fixing common statistical building blocks such as scope, definitions, coverage (e.g. of FAO fishing areas and fish species), allowed data sources (e.g. administrative data, surveys and novel data sources such as Big Data) and quality assurance procedures, as well as the names of the topics and detailed topics to be covered. These topics (for example catch, aquaculture and fleet data), their variables and further domain-specific requirements such as statistical observation units, precision requirements, specific deadlines and exemptions would be described in implementing acts, which are not adopted by the European Parliament and the Council of the European Union, but in the comitology process by the European Commission and the responsible comitology committee, in this case the European Statistical System Committee. Delegated acts would cover amendments to non-essential elements of the basic act, and those could be adopted by the Commission alone after consultation with Member State experts and the possibility of a European Parliament and Council veto. Such a structure would make the statistics more comparable and ensure that the most burden-relevant elements would be under the full control of the EU legislator, increase coherence across the different implementing acts, allow flexibility to address domain-specific differences, and enable comparatively rapid changes without having to change the entire framework in a long legislative process with an uncertain outcome (on average, it takes more than 18 months for a Commission proposal to become law). In order to give the EU legislator a full picture of all intended changes, it would be advisable to have draft implementing regulations ready at the time when the basic act is adopted.

EU Member States would adapt their national systems to the new framework regulation and then collect EFS data according to its requirements before sending them to Eurostat for compilation, validation and dissemination. Redetermining the place of EFS in the EU and international fisheries data architecture through this option could also mean that Eurostat would become the single contact point for FAO and OECD in the ESS, instead of both Eurostat and the Member States sending data to these organisations, as they currently do to some extent.

Eurostat, EU Member States and DG MARE could agree for Eurostat to directly access the ACDR system for data production and validation, a right in principle given by the “Statistical Law” but in this case not yet exercised in practice, among other reasons due to the discrepancies between Control Regulation and EFS data. Similar to Eurostat access to the EU fleet register, this could further simplify data flows in particular for catches, improve timeliness and coverage, and reduce transaction costs and burdens for Member States, since they could simultaneously ensure the quality of ACDR data and fishery statistics, as the validation would already take place at the administrative data level.

5.2.4.Option 4: New legal basis for aquaculture and compilation of other EFS from EU level administrative sources

A new legal basis for aquaculture would be developed to resolve the issues with the current legislation; the legal basis for catch and landing statistics would be discontinued, and European catch and landing statistics would from then on be derived from EU level administrative data to the extent possible. The discontinuation of the catches and landings regulations and modernisation of the aquaculture statistics regulation would be in line with the REFIT programme and satisfy the von der Leyen Commission’s “one in, one out” principle for new legislative proposals by reducing the burden of EU legislation.

This option would be a combination of option 3 for aquaculture statistics and option 2 for catch and landings statistics with their respective benefits and drawbacks, and with less cross-domain harmonisation than would be possible in option 3.

EU Member States would accordingly adapt their national systems to the new aquaculture regulation and collect data according to its requirements before sending it to Eurostat for compilation, validation and dissemination, whereas they would continue collecting the catch and landings data they need for their national purposes, international obligations and non-Eurostat EU requirements. This is because the legal obligations to provide fisheries data to various international organisations would have to be met by EU Member States and potentially DG MARE only instead of also by Eurostat.

5.3.Options discarded at an early stage

Option 4 was originally planned to include the integration of aquaculture statistics into the proposal for a new framework regulation on statistics on agricultural input and output (SAIO) that the Commission recently adopted 41 . The reason was that some organisations are of the view that aquaculture production has more similarities with animal agricultural production than with wild fisheries, for example controlled rearing, feeding and monitoring conditions. This is why e.g. FAO considers aquaculture to be a part of agriculture. However, more organisations group aquaculture in the same policy area as wild fisheries, and funding instruments are accordingly tied to the fisheries policy area. EFS stakeholders also felt that aquaculture should be grouped with wild fisheries. In addition, SAIO is already advanced and would have been slowed down by a year or more while waiting for the development of a new aquaculture regulation, which would have been risky particularly in light of the new Common Agricultural Policy reform and the European Green Deal, which require modernised agricultural statistics soon. Therefore, Eurostat decided to discard this option in the end.

Other possible options would have lain between options 3 and 4, i.e. with various combinations of modernised and possibly integrated legal bases for catches, landings and/or aquaculture and discontinuations of some data collections. For example, it would in principle be possible to update the catch and landings regulations, yet not integrate them, and not to update the aquaculture regulation. Or to update and integrate the catch and aquaculture regulations and discontinue the landings regulation. But ultimately, modernising the EFS regulations without integrating them would forgo obvious opportunities to harmonise common elements and increase efficiencies, for little apparent reason. For example, modernising either catches or landings statistics, but not the other would not make much sense as both depend on the same data sources, require each other to provide a fuller context, and nominal catches are in fact calculated from landings using a conversion factor. For these reasons, all possible “in between” options were crystallised in options 3 and 4, which both make the most sense prima facie and together incorporate the main benefits and drawbacks of full (option 3) or partial (option 4) modernisation, as well as those of partial discontinuation (in option 4).

It is also not possible to amend just some parts of the EFS regulations to create additional options between the baseline/status quo option 1 and options 3 and 4. This is because the regulations are internally an integral whole where the articles build on each other, so significant changes to one part would necessitate changes in all other parts as well. Moreover, the ordinary legislative procedure is controlled by the European Parliament and the Council of the European Union which are essentially free to substantially add to, change, or remove from European Commission proposals, so there would be no guarantee an intended partial change would not result in a more complete overhaul similar to options 3 or 4. The Commission can amend parts of the current EFS legislation by itself with delegated acts, but these are only allowed to change non-essential elements. So if a lengthy and complicated ordinary legislative procedure were undertaken to modernise EFS, it should be worth the effort by proposing substantial changes to the regulations, as the workload during the procedure would be similar either way. These are additional reasons for the crystallisation of possible options into options 3 and 4.

As a possible technical or practical option, data coordination centres were considered. As explained, data compiled under the Control Regulation are currently first provided to different national institutions which treat them to meet the statistical requirements of EFS, FAO and OECD and scientific requirements under the DCF. Differences between the data transmissions occur, among other reasons, because of different definitions, aggregations, deadlines and quality checks. It is however conceivable that the data could first be treated within one central node either at national level (option A: national data coordinator) or at EU level (option B: EU data coordinator), which would then transmit the data to all concerned inter- and supranational organisations. This could reduce the burden on Member States which would then not need to separately provide data to FAO, OECD and RFMOs, but would likely increase the costs at EU level. Both option A and option B assume that Eurostat would be responsible for the provision of statistics to FAO and OECD. This would be possible because the data definitions are similar, but effort would have to be expended to fully align them.

However, such centres would not obviously fall under the purview of EFS because they would be neither data nor data respondents or data producers, and their operation would in any case depend on better data harmonisation through one of options 1, 3 or 4, so they would not be an option on their own, but a part of different options. Therefore, Eurostat decided to limit the scope of this impact assessment report to changes to the EFS regulations proper, whose remit is comparably well defined, and which are under the control of Eurostat.

6.What are the impacts of the policy options?

The impacts of statistical regulations are more difficult to identify and quantify than those of many other EU initiatives. They mostly apply directly only to data providers, i.e. usually a narrow group that is usually not overall significantly affected by responding to a survey or census, and to data producers, i.e. a small group of organisations such as national statistical authorities whose job it is to create statistics efficiently, even if the creation of some statistics can be relatively burdensome. The direct impacts are here the resources needed to fulfil the legislative requirements of EFS, such as management and time/inputs required by data providers, data producers, and Eurostat. The legal acts are regulations and as such directly applicable in Member States, and the resulting activities triggered by their implementation, such as collection, processing and validation of statistics by the countries and Eurostat, constitute impacts which have a resource cost. The resulting statistical data do not have an impact on the world by themselves, only indirectly through their use e.g. for policy-making, research or media. Such indirect impacts are hard to measure because it is often difficult to know, for example, who uses statistics for which purposes, how large a part statistics play in the taking of certain decisions, and which effects those decisions have in the short and long run. Trying to assess such indirect impacts to the second or third degree would be too expansive and beyond the scope of this report, which is why it is not attempted here.

But in general, it is reasonable to assume that the provision of comparable, high-quality European statistics can lead to outcomes such as better development, implementation and monitoring of the CFP, improved market monitoring, and better assessments of economic impacts on fisheries and aquaculture. Fishery statistics are also considered to contribute to higher-level impacts such as supporting the sustainable exploitation of aquatic resources and better evidence-based decision-making overall. But as these outcomes are indirect, they cannot be readily quantified. In addition (as stated in section 1) the direct costs of producing EFS are very low (5.6 million € per year for the whole EU), so any change would only have a modest financial impact.

For the reasons described, the direct economic, social, environmental and fundamental rights impacts of the production of official statistics are low, and as stated, “micro-SMEs” with fewer than 10 employees cannot practically be excluded from EFS data collection (because most EU fishing vessels have crews smaller than that). The impacts on these areas and on SMEs are therefore not separately analysed below, except for the burden and costs for data providers and producers. Nevertheless, in order to have a proxy for indirect impacts of statistical regulations on other groups, the analysis notes the effect of each option on the statistical quality, i.e. the independence, comparability, availability of long time series on EU and country level, and accessibility in a public database of EFS. The reason for this is that these criteria are relatively tangible and that better statistics, i.e. produced through a better quality assurance process, can lead to better indirect impacts because they provide a better picture of reality to base decisions and policies on.

However, mainly “measuring the measurable” risks that potentially important impacts go undetected, and inferring that better statistics result in better decisions rests on the potentially wrong assumption that statistics and other data are used consistently and faithfully in decision-making. Regardless of the quality of EFS, policymakers could weight other factors higher in their decisions in an unsystematic fashion, for example if public attention and pressure were suddenly drawn to a fisheries sustainability crisis. The analysis below tries to note such uncertainties and potentially significant second order impacts where possible, but a complete assessment of every conceivable consequence of changing or not changing EFS is impossible (see also Annex 4).

As EFS are statistical regulations that are directly applicable in EU Member States, no special support or implementation plan is needed whose impacts would need to be analysed.

6.1.Option 1: Baseline scenario – continuation of the current policy

Continuing with the status quo of EFS would not solve the main problems detailed above of unfulfilled data needs, non-harmonised definitions, double reporting, inflexibility and an unclear place of EFS in the international fisheries data landscape. The cost-benefit ratio of EFS would likely shrink, as they would fulfil evolving requirements for the CFP and other EU policies less and less. The costs and administrative burden would remain about the same as now, with small improvements due to some streamlining of definitions and updating of classifications via delegated acts. But these delegated acts could not address the main problems as they can only change non-essential elements of the regulations, and changing for example the structure of the EFS regulations so that they could react to new developments and data needs more flexibly would be a major change.

On the other hand, data providers, Member States and users have a good understanding of the current policy and legal framework of EFS that has grown over years. The data collection, validation and treatment systems are in place and do not need to be changed, but would be maintained. And the long, useful time series of fishery statistics would also continue unbroken under option 1, as would high-quality and comparable statistics on catches, landings, fishing fleet compositions and aquaculture at EU level in general.

Data producers would continue collecting and submitting catch, landings and aquaculture statistics to Eurostat, DG MARE and other organisations requesting fisheries data, while continuing to send fishing fleet data to DG MARE only (Iceland and Norway send theirs to Eurostat because they are not EU members).

Correspondingly, Eurostat would continue compiling, validating and disseminating catch, landings and aquaculture statistics while taking fishing fleet data for EU Member States from the EU fleet register and for Iceland and Norway directly from them, compiling and disseminating the statistics. Eurostat would also continue submitting fishery statistics to OECD and FAO to the extent possible due to slightly different data collections.

Users would likely continue relying on EFS for comparable statistics on catches, landings, aquaculture and the fishing fleet at EU level, but among the other problems detailed above, their new and emerging needs would not be fully reflected.

Overall, the possible changes within this status quo option would mean that:

·The relevance of EFS would on the one hand slightly increase due to the possible improvements done by delegated acts, but on the other hand continue to decrease over time as the CFP and fishery data needs evolved without an adequate response by EFS.

·Meeting EFS user needs better and more flexibly would be slightly more possible due to the improvements done by delegated acts and because the strengths of the current legal framework would be maintained, i.e. high-quality, reliable and comparable official statistics that are accessible to the public. However, most weaknesses of the current legal framework would remain unresolved, such as a high amount of confidential data in aquaculture, mostly unmet needs, several inconsistencies and discrepancies, double reporting and inflexibility.

EFS would remain efficient due to the low cost of producing them. As Member States need to collect the data underpinning EFS for other purposes as well, the additional administrative burden related to the production of EFS would also remain limited. Improved IT tools and more use of administrative data could help increase the efficiency further. However, there would be no large efficiency gains, for example through systematic harmonisation and elimination of double reporting, and developing new delegated acts would incur a small cost. It is also doubtful whether the current legal framework of EFS with its five regulations and overlapping reporting requirements is still in line with the REFIT programme, which aims to keep EU law simple, remove unnecessary burdens and adapt existing legislation without compromising on policy objectives.

·The EFS confidentiality issues would not be resolved under this option because the improvements possible under this option could not change either the fragmentation of the aquaculture sector leading to easily identifiable data points, the Eurostat confidentiality procedures, or the general structure of the aquaculture regulation mandating very detailed data.

·Solving the quality problems of EFS source data: While the statistical quality of EFS themselves would remain high as the statistics would continue to conform with the principles of impartiality, reliability, relevance, cost-effectiveness, confidentiality and transparency, they would not be able to be modernised and aligned well to current and future CFP reforms nor be able to exert much of an influence on EFS source data due to a lack of formalised cooperation, common database access and validation rules etc.

·Reducing gaps, overlaps and discrepancies in the EU and global system of fisheries data: The coherence of EFS would remain as it is, i.e. high internally, but hard to maintain, and leaving overlaps and discrepancies with other fisheries data collections untouched, even if based on the same source data. However, EFS would still add EU value as the objectives of EFS can only be achieved on the basis of a common legal framework, such as the one currently in place, because only the EU can coordinate the necessary harmonisation between Member States.

·Clarifying the place of EFS in the EU and global fisheries data landscape. A lack of systematic harmonisation and a thorough revamping of the EFS regulations would mean that there would be little space to rethink the place of EFS, and little room for manoeuvre.

·The EFS stakeholders consulted for the evaluation are interested in streamlining the data collection and presentation of the statistics, but they are generally not interested in only amending the existing EFS legislation as this would not deliver some of the changes that they desire, such as the inclusion of more fishing areas. The respondents to the public consultation of the impact assessment considered the impacts of this option on all stakeholder groups to be rather neutral, which is expected for the baseline.

This option is also technically feasible as it would mainly be a continuation of the current EFS system, and likely politically feasible as it would only introduce small changes through the adoption of delegated acts which would be unlikely to face strong opposition – unless the European Parliament or the Council wanted more reform and vetoed the delegated acts, which is possible.

In sum, option 1 would mostly be a continuation of the current system with all its benefits and drawbacks, with some small improvements through the adoption of delegated acts. Their scope would however be limited as delegated acts cannot change essential elements of regulations.

6.2.Option 2: Discontinuation of EFS

Option 2 would mean repealing the five EFS regulations, and Eurostat would stop producing EFS. This would be a technically relatively straightforward administrative procedure including repealing the regulations themselves and amending any EU or other legislation referencing them. It could lead to some public opposition from data providers and users, but they would have little power to stop the procedure.

Discontinuing the EFS regulations and only creating European fishery statistics from such administrative data as are available, while Member States would continue collecting the data they need for national, EU and international purposes, would have several mostly negative impacts:

·Relevance: Repealing EFS would lead to a loss of some important statistics which are usually not sourced from administrative data, such as most aquaculture statistics. Users and redistributors of EFS such as OECD and EUMOFA would have to shift their data sourcing either to the DG MARE data networks or to national sources, which would be more time consuming and cumbersome for them and reduce the availability and quality of data. This is because of technical and language barriers and because these alternative sources are perceived as being of lower statistical quality, as they use less structured quality assurance frameworks than the one implemented by Eurostat.    

·Meet EFS user needs better and more flexibly: Repealing EFS would mean a loss of the legal structure that standardises the approach to producing fishery statistics across the EU resulting in reasonably complete, comparable, independent and validated high-quality catch, landings, fleet and aquaculture statistics with long time series that are publicly available to all users. As evidenced in the evaluation, this harmonised system represents a higher added value than the sum of each participant’s individual contribution because it establishes a trustworthy and comparable framework. Moreover, a lack of a regular, legally regulated statistical data collection and independent EU-level statistical data could lead to data only being collected in an unsystematic fashion to respond to short-term policy needs, which would result in quality, comparability and predictability issues.    

While EU Member States would still collect Control Regulation and DCF data in a standardised way based on EU legislation under this option, they would not collect fishery statistics in a standardised way based on the EFS legislation anymore. If this legislation fell away, countries would thus still have CFP administrative data but not a harmonised approach to make statistics out of them anymore.

As for efficiency, repealing EFS would mean that there would be no more costs and burdens explicitly for EFS such as producing data according to EFS requirements and transmitting it to Eurostat, and compiling, validating and disseminating it in Eurostat. However, Member States would have to produce “EFS-like” fishery statistics beyond the data available in administrative sources anyway for national purposes such as national accounts, EU purposes such as DG MARE data requests, and requirements of international organisations such as FAO. Therefore, the actual savings could be low or even negative both at Eurostat where officials currently tasked with EFS would likely still be tasked with creating and publishing such fisheries statistics as are possible to be made from administrative sources (yet without as much influence on these sources as under other options with a legislative basis), and in Member States, as the current single EU-wide EFS system would be discontinued and up to 27 new national systems would have to be built up first. These systems would also differ between countries, increasing aggregation and comparison costs, and the EU would have little influence over them because they would be based on national laws. As these data would then exist in any case, Member States and the EU would have forgone an opportunity to harmonise them and obtain the benefits of an EU-wide comparable system. Moreover, without a common EU source, the burden and cost of finding the statistics necessary for calculating EU aggregates would be shifted from one organisation to many users, leading to a large duplication of work. In short, there would be a loss of efficiency because Eurostat legislation and the ESS offer harmonisation, comparability and standardised procedures.

Repealing EFS would also result in lower data quality as the co-ordination, harmonisation and quality assurance provided by the EFS regulations would not be in place anymore.      

·Resolve EFS confidentiality issues: As the detailed EFS breakdowns and the Eurostat approach to confidentiality would fall away, it would be up to individual countries to deal with any confidentiality issues. These could even be exacerbated as producers which would be anonymous in EU aggregates could become identifiable in national aggregates.

·Solve the quality problems of EFS source data: Without EFS legislation, national fishery statistics could not be well aligned to current and future CFP reforms and systems nor be able to exert much of an influence on the source data due to a lack of same-level formalised cooperation, common database access and validation rules etc.

·Reduce gaps, overlaps and discrepancies in the EU and global system of fisheries data: Without common regulations imposing e.g. common definitions and quality assurance procedures, the “EFS-like” fishery statistics in EU Member States and the statistics created at Eurostat from such administrative sources as there are would have less internal coherence than they have now, and even less external coherence because every data producer would work on their own without EU-level coordination.

·Clarify the place of EFS in the EU and global fisheries data landscape: Repealing EFS would incur a reputational risk for the EU, as non-EFS data made available to users such as from EUMOFA and FAO could differ across sources, leading to confusion, uncertainty and a loss of trust. The convenient EFS one-stop shop for users of high-quality fishery statistics would be lost, and users would find it harder to access data with characteristics such as long, consistent time series from 1950 on elsewhere; for example, DCF data only go back to 2003 at the earliest, and  EFS are easily publicly accessible for data users, but other sources not always. Moreover, the value of EFS as independent, high-quality and comparable baseline statistics used to support CFP and EU policies work would also be lost, and e.g. fisheries market monitoring would be negatively affected. Without EFS, Eurostat would also likely not contribute anymore to the development of international fisheries statistics standards in the CWP. This would be a loss of expertise for the CWP, and for Eurostat a lost opportunity to develop international fishery statistical standards further.

   
It would also be hard to explain why one of the main common policies of the EU was no longer supported by independent EU statistics and invite questions for Eurostat, which is responsible for decisions on statistical methods, standards and procedures to be used for European statistics, in accordance with the “Statistical Law”. Moreover, the Commission Decision on Eurostat 42 provides that the Director-General of Eurostat act in an independent manner when carrying out statistical tasks and should neither seek nor take instructions from Union institutions or bodies, from any government of a Member State, or from any other institution, body, office or entity.    

In addition, those commitments of sending EU fisheries data to international organisations which Eurostat is currently fulfilling with EFS would have to be fulfilled with different quality data, which could in some cases be difficult and incur new burdens and costs. Repealing EFS would also go against Eurostat’s mission, as expressed in the ESP, that includes providing fishery statistics for the development and monitoring of the CFP. The “costs of non-Europe” would therefore be severe. Moreover, statistical institutes should in principle add value to data based on their expertise instead of just redistributing administrative data, lest their contribution is devalued.

·EFS stakeholders disliked this option the most according to the evaluation because it would not solve their problems. The respondents to the public consultation of the impact assessment considered the impacts on data providers and producers of this option as positive due to a smaller burden, but as negative or very negative on various user groups, as the availability of statistics would drastically decrease.

6.3.Option 3: New streamlined legal framework for EFS

This option means that the current five EFS regulations would be replaced by one new framework regulation for catch, landings, aquaculture and fleet statistics that would address the problems described in section 2 with new legislative provisions and data collection requirements in order to simplify the system and streamline and update definitions, classifications and deadlines. For example, the geographical coverage of EFS catches and landings could be extended as requested by users. Aquaculture statistics would be reworked to reduce the amount of confidential data and improve timeliness. Legal and organisational arrangements would be made for better Eurostat access to and use of administrative data sources such as the ACDR database, e.g. by pre-validating data already on that level. In addition, mechanisms for allowing more flexibility to accommodate new needs could be integrated into the new legal architecture. This improvement of the legal framework would be in line with the Commission’s REFIT programme.

The option is technically relatively straightforward, and the responsible unit in Eurostat tasked with creating the new framework regulation has recent experience in writing framework regulations to modernise agricultural statistics (Integrated Farm Statistics, adopted as Regulation (EU) 2018/1091 43 , and SAIO) and in accompanying them through the legislative process. Other Eurostat units have similar recent experience. Although there could be initial costs of adaptation to a new legal framework for the Member States due to various reasons, they have expressed support for modernising the legal basis for EFS. In the medium to long term, the new legal basis could lead to savings, e.g. by increasing the use of administrative data and estimating and aggregating data for species of minor economic importance.

In addition, NSIs, DG MARE and the JRC within the Commission, and international organisations such as FAO and OECD, should be involved in the EFS modernisation process, respectively via established venues such as the ESS, annual hearings between Commission services, and the CWP. This is important to ensure that their needs are addressed, to align definitions, data flows and deadlines, and to reorganise the EU and international fisheries data architecture so that user needs are met with a lower burden due to less double reporting and fewer gaps and overlaps, in line with the principle of “collect once, use multiple times”. The cooperation between institutions should accordingly be stabilised and regularised with formal agreements or legal provisions. For example, Eurostat producing fishing fleet statistics from the EU fleet register should be put under the legal framework, which non-EU members could then follow to deliver fleet data. Furthermore, all user needs should be reviewed to ensure that the new fishery statistics are useful now and in the future. However, this represents a higher workload than for the other options, and not all data may be able to be aligned as different organisations can have justified different data needs. For example, administrative control data can have a different scope than statistics intending to capture a full population. Moreover, negotiations with different organisations can result in compromises that do not fulfil all of the initial goals.

The main risks of this option are associated with a protracted legislative procedure due to divergent interests of the Council and the European Parliament, but the modernised legislation itself would not be too technically complicated, and as stated, Eurostat has recent experience in the legislative process for framework regulations. There could also be a risk in Eurostat and the Member States not being able to agree on a renewed confidentiality charter so that some confidentiality problems would persist even if the legislation would request fewer disaggregations.

The following impacts can be expected:

·Relevance: EFS would become more relevant as the new legal framework would be updated to better reflect current and future user needs in the context of the new CFP and of other evolving fishery policy data needs. Redetermining the place of EFS in the EU and international fisheries data architecture could for example mean that Eurostat would become the single contact point for FAO and OECD in the ESS, instead of both Eurostat and the Member States sending data to these organisations, as they currently do to some extent. The “Brussels effect” of EFS setting standards for other fisheries data collections, sometimes unbeknownst to their users, would likely increase.

·Meet EFS user needs better and more flexibly: EFS would become more effective as this option would keep their existing benefits, e.g. being a one-stop shop for high-quality, comparable fishery statistics with long time series accessible to all, while providing a simpler and more streamlined legal framework that can better integrate and serve new user needs and increase coverage. Reducing double reporting through aligning definitions and reorganising data flows would also reduce discrepancies between different data sources, thus increasing the reliability, accuracy and international comparability of EFS.

Regarding efficiency, EU Member States would continue submitting landings and aquaculture statistics to Eurostat, and Eurostat would continue compiling, validating and disseminating them. Eurostat, the Member States and DG MARE could agree to grant Eurostat direct access to the ACDR system for catch data production, similar to the status quo for the EU fleet register for fleet statistics. This could further simplify data flows, improve timeliness and coverage, and reduce transaction costs and burdens for Member States.

A lower burden and costs for data producers and Eurostat could be expected due to the new harmonised framework improving on the problems of the status quo, such as discrepant definitions and aggregations between administrative data and statistics, difficult to collect data, a lot of confidential data particularly in aquaculture, and double reporting. The potential for improvement is higher for catches and landings than for aquaculture and fleet data because the EFS aquaculture data collections are already integrated to serve both EFS and DCF needs in most countries, and fleet statistics are produced by a simple extraction from the EU fleet register. By comparison, EFS catches and landings data overlap and are not always consistent with ACDR data e.g. because of different definitions. In addition, countries with a large number of small vessels (e.g. Italy and Greece) currently have to conduct surveys on them to provide EFS, as the existing Control Regulation does not oblige them to install electronic recording systems on those vessels. However, the new Control Regulation could enlarge the reporting obligation to cover small vessels as well, which would allow reusing related administrative data for statistics and obviate the need to conduct additional surveys.

Thus, under the simplest case of option 3, Member States just deliver the necessary data under the Control Regulation, and Eurostat extracts the catch statistics needed at EU level from them. International organisations would download the data from Eurostat’s website, reducing the burden on data producers in Member States. Thus, EU countries would not need to deliver catch data to Eurostat or any data to international organisations anymore and could benefit from the reduction of the data flows.

   The burden reduction potential for data providers is limited. For aquaculture, the burden of EFS usually consists of some extra questions in a joint EFS/DCF questionnaire that plant managers fill out once a year; a few more or fewer questions would change the burden only slightly, even across all EU aquaculture plants. For catches, data providers would still have to provide CFP data, and after the Control Regulation reform, also more administrative data on small vessels instead of filling out statistical surveys on them.

Some adaptation costs would be incurred in the short term by national statistical authorities through for example having to change questionnaires and databases and by Eurostat. But this is a regular occurrence for those authorities and therefore unlikely to create implementation challenges. Burden and cost reduction through harmonisation, reduction of discrepancies etc. would apply in the mid to long term and therefore have a greater effect over time. The annual savings in direct costs linked to the catch data collection for national statistical authorities are estimated to be approximately 1.2 million € as compared to the baseline, mostly due to reduced survey and indirect costs, representing a total EFS cost reduction of about 20% (option 3 is expected to be cost-neutral for data providers and Eurostat) to 0.04% of the EU fisheries and aquaculture sector’s annual production value. This is estimated on the basis of the plan for Eurostat to collect catch statistics directly from administrative data collected by Member States and providing it to DG MARE, without Member States having to treat it first. The savings potential of option 3 can differ between countries because of different national implementations and the varied fisheries sector across the EU, e.g. vastly different numbers and sizes of vessels with different data collection and treatment challenges. Increasing and evolving demands for more fisheries data could over time also increase burdens and costs again, for example by requiring data on uses of aquaculture products beyond human consumption, as some users ask for. On the other hand, streamlining the data flows between Member States, the EU and international organisations would increase the long term efficiency of fishery statistics and the fisheries data architecture overall.

As an example of the improvements possible under option 3, currently the EFS catch regulations only cover seven FAO fishing areas near the EU: 21 – Northwest Atlantic, 27 – Northeast Atlantic, 34 – East-Central Atlantic, 37 – Mediterranean and Black Sea, 41 – Southwest Atlantic, 47 – Southeast Atlantic, and 51 – West Indian Ocean. But e.g. Spanish vessels already fish almost around the globe, and if for exampleUK-EU conflicts about the traditional fishing waters around Great Britain increased, it is conceivable that more EU fishing nations would explore other fishing areas with their bigger vessels, e.g. area 57 – the Eastern Indian Ocean. Incorporating statistics from this area would be hard to do in the current EFS regulations as they would need to be reopened in a lengthy legislative procedure. But if such a procedure is inevitable, it would be more beneficial to implement bigger changes such as those included in option 3. This is because legislative procedures require an in large part “fixed” amount of effort (e.g. meetings with the European Parliament and the Council of the EU, reconciling different versions of amendments etc.) and only some “variable” amount of effort (e.g. due to the length and complexity of a proposal).

·Resolve EFS confidentiality issues: Rewriting the EFS aquaculture regulation could go a long way towards fixing aggregates at a level that balances data needs and the need for confidentiality. Developing problems could be tackled by an easier possibility to amend the legislation as well as by agreeing to a confidentiality charter in the course of developing and negotiating the new legislation.

·Solve the quality problems of EFS source data: Cooperation between Eurostat and DG MARE in the CFP reform and adapting the new EFS legal basis to the characteristics of the changed source data can help remedy this issue by improving harmonisation between CFP and EFS definitions and classifications and establishing a more formalised cooperation in general, for example with regularly scheduled realignments. The improvements outlined in option 3 will also serve to improve the quality of EFS coming from other sources.

However, this problem is in general difficult to tackle. A series of data comparisons between EFS and CFP data has been made that show the extent of the problem (see section 2). The root causes lie in the quality of the data collected and/or in the post-collection treatment of the data by national authorities. In both cases, joining efforts across the data flows of the statistical network and of the fisheries data network is necessary to improve data quality. DG MARE has launched a series of action plans to support improvements of the catch reporting systems in Member States that will benefit not only the administrative data, but also the quality of statistical data.

·Reduce gaps, overlaps and discrepancies in the EU and global system of fisheries data: The internal and external coherence of EFS would increase, as developing and implementing the new framework regulation would lead to more alignment both within and across the EFS domains as well as with external fisheries data sources.

·Clarify the place of EFS in the EU and global fisheries data landscape: This option would strengthen the Eurostat objective of the ESP to provide fishery statistics for the development and monitoring of the CFP. Users could continue accessing reliable and comparable fishery and aquaculture statistics covering all EU Member States and have their needs better fulfilled. The new legal framework would be able to include future needs to ensure the continued relevance of EFS. The statistical quality of EFS would also at least remain the same by them continuing to follow the principles of impartiality, reliability, relevance, cost-effectiveness, statistical confidentiality and transparency, and would likely increase e.g. through better operational arrangements with holders of administrative data sources and better availability of the data due to fewer confidential values.

·EFS stakeholders preferred this option the most when asked for the evaluation because they saw it as having the highest chance of fulfilling their needs and wishes. The respondents to the public consultation of the impact assessment considered the impacts of this option on data providers and producers as neutral to positive and on data users as neutral to very positive.

In all, this option would results in improvements in the relevance and effectiveness of EFS through the new streamlined legal framework. It would ensure the availability of more high-quality EFS for wide user groups and improve the efficiency and coherence of fishery statistics and other fisheries data. Such a major reform, while entailing a certain workload and time commitment, would represent a chance to solve many long-standing problems at once instead of trying to resolve them over time with less effective approaches that could in the end amount to an even higher expenditure of work and time.

6.4.Option 4: New legal basis for aquaculture and compilation of other EFS from EU level administrative sources

This option would mean that the current EFS aquaculture statistics regulation would be replaced by a new streamlined regulation, in line with the REFIT programme, while the legal basis for EFS catch and landings statistics would be discontinued. European fishery statistics would from then on only be derived from EU level administrative data to the extent possible, similar to the current approach for fishing fleet statistics, which are derived by Eurostat from the EU fleet register without an Eurostat regulation that would, for example, require additional data.

Technically, this option is relatively straightforward in that the Commission would repeal the EFS regulations on catches and landings while developing a new regulation for aquaculture statistics, something the responsible Eurostat unit has recent experience in. However, streamlining definitions and classifications could be technically challenging, and politically, both data producers and users as well as the European Parliament and the Council could oppose repealing EFS for catches and landings, but with little power to stop the repeals.

The likely impacts for this option would be a mixture of those of option 3 for aquaculture statistics and those of option 2 for catch and landings statistics:

·Relevance, meeting EFS user needs better and more flexibly and resolving EFS confidentiality issues increased for aquaculture statistics, similar place of EFS in the EU and global fisheries data landscape: the new aquaculture regulation would entail a lower burden and costs for comparable and high-quality aquaculture statistics and better meet current and future user needs. For example by including information on aquaponics or using insects as feed, creating less confidential data through more suitable data aggregations, and better aligning classifications with the latest CWP classifications. EFS data producers would thus continue providing aquaculture data to Eurostat, which would continue compiling, validating and disseminating aquaculture statistics. 

However, there would likely only be modest efficiency gains because of this new regulation, as the streamlining efforts and the new data needs could balance each other out. As stated previously, the amounts involved are small in any case as EFS production is comparatively very cheap across the EU. These statistics would also continue adding EU value through a common methodological framework creating comparable, high-quality aquaculture statistics. Moreover, EFS aquaculture statistics are an important basis for DG MARE to analyse socio-economic and environmental impacts of the sector.

·Relevance, meeting EFS user needs better and more flexibly and resolving EFS confidentiality issues reduced for catch and landings statistics, similar place of EFS in the EU and global fisheries data landscape: data producers would not provide catch and landings statistics directly to Eurostat anymore and Eurostat would stop compiling, validating and disseminating them beyond extracting them from available administrative sources to the extent possible and publishing those data as EFS. However, as explained in the impact analysis of option 2, “EFS-like” data would most likely continue to be produced by Member States anyway for national and international purposes. Therefore, there would likely be only modest gains in efficiency, if any, both for data providers, data producers and Eurostat. Aggregation and comparison costs would increase because Member States would choose different approaches in the absence of an EU-wide harmonised legal basis. The use of administrative data without significant additional validation steps also entails risks in ensuring coverage, quality, data availability, timeliness and long time series. The EU added value of these catch and landings data produced directly from administrative data would be lower than that of EFS because of the lack of additional validations and a common, harmonised legal basis that also serves as an inspiration for other producers of fishery data in the world. Handling catch, landings and aquaculture statistics differently would also forgo opportunities to harmonise all EU fisheries statistics and to enable a more holistic view of the fisheries sector that could support better policy-making in this area; for example, a lot of aquaculture feed consists of catches of wild fish. Lastly, the legal obligations to provide fisheries data to various international organisations would have to be met by EU Member States and potentially DG MARE only instead of also by Eurostat.

Repealing the catch and landings EFS regulations would also reduce double reporting by Member States to the EU in line with the principle “collect once, use multiple times”, creating a small efficiency gain. In short, there would be a loss of efficiency because Eurostat legislation and the ESS offer harmonisation, comparability and standardised procedures.

·The objectives of solving the quality problems of EFS source data and reducing gaps, overlaps and discrepancies in the EU and global system of fisheries data would suffer under this option because there would be no regulations for catch and landings statistics anymore imposing common concepts, definitions etc. both on EFS themselves and on EU Member States producing them, no integration of catch and landings statistics with aquaculture statistics, no in-depth cooperation with the Control Regulation system, and no more coordination with international organisations.

·Only few EFS stakeholders preferred this option as their first choice when asked for the evaluation because they thought that their data needs and requirements e.g. for more flexibility of EFS would not be met by a partial repeal. The respondents to the public consultation of the impact assessment considered the impacts of this option on all stakeholder groups to be negative to neutral.

In summary, although option 4 improves EU aquaculture statistics, it entails a partial loss of reliable, accessible and comparable statistics on catches and landings.

7.How do the options compare?

Given the objectives of streamlining and simplifying EFS as described in section 4, the options as described in section 5 and the impacts as analysed in section 6, the following option comparison table 1 can be constructed, with thehe following rows as measured against the baseline:

·Time horizon: short term (ST) and long term (LT) to show the different impacts of adhering to each option over time. In general, as the fisheries sector will continue to change as outlined in section 2.2 of this report and EU fisheries policy legislation will also change in response, the available data and users’ data needs will over time drift further away from EFS if the legislation they are based on remains inflexible, and EFS will thus get progressively worse.

·Effectiveness with regard to the objectives

·Cost-efficiency

·Coherence with EU policy objectives

·Impacts on stakeholders, i.e. EFS data users, data producers and data providers.

Table 1: Option comparison against the baseline. Source: Eurostat.

It can be seen that option 2 has very negative impacts in the short term and even worse ones in the long term. Option 3 is promising in the short term and can be expected to create even more positive impacts in the longer term. And option 4 has some negative impacts in the short term which can be expected to get worse over time.    

8.Preferred option

In light of the impact analysis in section 6 and the option comparison in section 7, the results of the consultation activities outlined in Annex 2 and in the EFS evaluation, and discussions with the stakeholders, the preferred option is clearly option 3 – a new streamlined legal framework for EFS. It is supported by the main data users (DG MARE, OECD, FAO, ICES and several RFMOs). A large majority of the representatives of the national statistical authorities responsible for fisheries statistics expressed a clear preference for option 3 in the meetings of DGAS in June 2019 and again in June 2020. Occasional users of EFS such as the Commission Directorate-General for Trade, which mostly uses trade statistics and only rarely fisheries production data via EUMOFA, also prefer option 3 to have reliable, accessible and comparable fishery statistics that better serve users’ needs.

Option 3 was also the preferred option in the public consultation. It was supported by seven of 15 respondents. The supporters of option 3 were regional, national and international authorities and NGOs. The respondents also gave their opinions on the expected impacts of the options on five stakeholder categories: data producers (statistical authorities), data providers (fishers, fleet and aquaculture plant managers), policy makers, professional users (e.g. NGOs and business associations), and citizens and others. Option 3 scored the most positive impacts on all stakeholder groups. In addition, only very few respondents identified negative impacts for option 3. Table 2 below offers an overview of the preferred option by type of stakeholder group by summing up the answers from the various consultations.

Table 2: Ranking of preferred options by type of stakeholder group. Source: Eurostat.

Option 3 responds best to the objectives of the REFIT programme by simplifying and streamlining the five legal texts currently governing EFS into one coherent legal framework. Option 3 is also most in line with the mission and strategy of Eurostat. Independent EFS are inherently important for the design, implementation, monitoring and evaluation of the CFP and related EU policies. Option 3 allows aligning statistical requirements flexibly with evolving user needs and to develop a coherent Commission-wide strategy and renewed architecture for fisheries data instead of the current partly uncoordinated and duplicated approaches. This is important for efficient policy delivery and resource use.

The expected cost savings of pursuing option 3 are detailed in table 3 below. The option will lead to the reduction of costs and burdens in several ways. Firstly, it will cut double data collection for catch statistics, as the data will be taken directly from an EU-level administrative source. This reduces the administrative and data collection costs and also the time expenditure for data respondents, i.e. fishers and aquaculture plant managers. Secondly, other international organisations can take the data, which better meet their needs, directly from Eurostat without also asking EU Member States to deliver it to them. Thirdly, more data will become available to data users when the data structure for aquaculture will be simplified, and consequently less data will be confidential. And fourthly, the simplification of the data will ease the burden on data providers and data producers.

Lastly, option 3 is a proportional response to achieve the objectives outlined in section 4 and solve the problems described in section 2. This is because a less comprehensive legislative and organisational reform of EFS would not be able e.g. to meet user needs better and in a structurally more flexible fashion, nor to redefine the place of EFS in the fisheries data landscape, as explained in section 6. And as described in section 3, EU Member States could not achieve the necessary harmonisation and coordination to produce coherent EFS acting on their own. At the same time, a new streamlined legal framework for EFS would not go beyond what is necessary to modernise the data collections for catches, landings, aquaculture and fleet while respecting the principle of subsidiarity, e.g. by leaving the choice of data sources to Member State authorities.

The preferred option is also proportional to the general and specific objectives and does not go beyond what is needed to address the identified problems because in principle, legislative flexibility requires implementing and delegated acts because those can amend important and non-essential elements, respectively, of legislation faster and with less overhead than an ordinary legislative procedure. But there should still be a harmonised, permanent act for elements that need to be standardised across secondary legislation and are unlikely to change much over time, such as scope, coverage or quality requirements. From these constraints, a framework structure with a basic regulation and several implementing and delegated acts follows naturally, so it can be seen as proportional to solve the problems. At the same time, there will be safeguards and limits so that subsidiarity and proportionality will be respected and burdens and costs kept in check.

8.1.REFIT (simplification and improved efficiency)

Table 3: REFIT cost savings of preferred option 3. Source: Eurostat.

9.How will actual impacts be monitored and evaluated?

The existing monitoring and evaluation tools which are in place and being used for the statistical production and dissemination of European statistics will also be used to monitor and evaluate the impacts of EFS. These tools will enable analysis of the effectiveness and efficiency of the new statistical initiative and of the quality of the data produced. In detail, they are:

·The ESPs foresee systematic mid-term and final evaluations. Fisheries statistics are part of this reporting mechanism 44

·The Eurostat Management Plan foresees the follow-up of key performance indicators, which also applies to fisheries statistics 45

·User satisfaction surveys are carried out on an annual basis 46

·EFS 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 national statistical authorities will further improve the standard metadata and quality reporting system for fishery statistics. This will allow a more sophisticated monitoring and evaluation of the statistical processes used in the Member States and of the output disseminated. For example, more detailed information will be available on the use of administrative data sources by Member States, which can lead to a reduced burden. The results and the effects of the EFS legislative initiative will regularly be monitored via annual compliance and quality reports. In order to measure the progress towards achieving the objectives of the initiative, the list of monitoring indicators in table 4 below has been defined. These indicators are juxtaposed against a list of operational objectives which have been derived from the specific objectives. The indicators will be measured against the benchmark targets indicated in the last column of the table.

Eurostat needs to rely on user surveys and direct contacts with data producers and data users in order to monitor and evaluate the impacts of its statistics since, as explained, the statistics do not have a direct effect in the world by themselves. But Eurostat’s cooperation with data producers and data users is long established, trustful and takes place in regular fora and frequent informal exchanges; in fact, most problems of EFS and the main requests to improve the statistics were brought up within this cooperation, so it can be expected that the actual effects of a modernisation of EFS, whether positive or negative, will also be comprehensively raised during future exchanges with the data producers and users.

Compliance is usually not a large problem for EFS because the statistics are either based on administrative data which are obligatory under the Control Regulation or on surveys where the response rate is generally high. Sometimes the data are sent to Eurostat with a short delay, for example due to a change of information technology systems. The new legislation in the preferred option 3 is not expected to affect compliance in a negative way; on the contrary, compliance should improve as less data are expected from Member States and more statistics can be drawn directly from Control Regulation data by Eurostat.    

Table 4: Monitoring and evaluation indicators. Source: Eurostat.

Annex 1: Procedural information

Lead DG, Decide Planning/Commission Work Programme references

Lead DG: DG EUROSTAT

Decide planning reference: PLAN/2020/6889

Organisation and timing

Eurostat set up an Inter-Service Steering Group (ISG) for the impact assessment of EFS by inviting the interested DGs to join on 18/02/2020. The following DGs nominated representatives: DG Environment, DG Maritime Affairs and Fisheries, DG Trade, and the Secretariat-General.

The ISG met five times:

·27/02/2020 to discuss the inception impact assessment,

·11/05/2020 to discuss the consultation strategy,

·29/06/2020 to discuss sections 1-4 of the draft impact assessment report,

·10/12/2020 to discuss the results of the consultations and sections 5-9 of the draft impact assessment report, and on

·23/04/2021 to discuss the draft final impact assessment report.    

The inception impact assessment was published for feedback on the European Commission’s “Have your say” platform from 21/04/2020 to 19/05/2020 47 . There was no feedback.

Consultation of the Regulatory Scrutiny Board

The draft final impact assessment was submitted to the Regulatory Scrutiny Board on 03/05/2021. The Board met with Eurostat representatives on 02/06/2021. It issued a positive opinion on 04/06/2021, noting that the report should further be improved with respect to the following aspects: 1) the impact analysis and the comparison of options should be presented more clearly, and 2) stakeholder views, including dissenting ones, should be more fully integrated into the report. In addition, the board gave a number of more technical recommendations. Eurostat has implemented these recommendations. The ISG approved the changes in a written procedure concluding on 22/10/2021.

Evidence, sources and quality

The specific evidence sources used are referenced throughout the document whenever they are used. In particular, the main problems of EFS are evidenced by substantial, regular, in-depth stakeholder consultations and events for several years, by reports of the European Court of Auditors, and by the evaluation of EFS. The main quantitative analysis of administrative burden and cost impacts is based on stakeholder information (targeted questionnaires, financial statements, legal reporting obligations etc.) This information is considered to be robust and complete enough to estimate the magnitude of costs and benefits.

Stakeholders' issues with EFS are considered to be represented and evidenced faithfully and completely in the EFS evaluation and the stakeholder consultations and events. In fact, these seminars and meetings mainly exist to create a regular exchange between Eurostat and the main data users and producers.

External expertise

The ISG created in the context of this impact assessment, in which the above-mentioned DGs were represented, constituted a very relevant source of external expertise for the preparation of the impact assessment because these DGs are the main users of EFS, in addition to the ISG’s already established function of steering the impact assessment process and contributing to the collective preparation of the impact assessment report.

The expertise of the NSIs has also been profoundly relied upon. Namely, there have been extensive discussions and exchanges of views as part of regular domain-specific working groups, task forces and the directors’ group on agricultural statistics.

Otherwise, this impact assessment report was written internally in Eurostat.

Annex 2: Stakeholder consultation

Stakeholder analysis

The main categories of stakeholders of EFS are:

·Data providers, i.a. fishers, aquaculture producers and others providing source fisheries data either as administrative data e.g. from logbooks and landing declarations or in surveys and censuses. They have a moderate influence over EFS and are moderately interested in them.

·Data producers, i.e. national statistical institutes and other national authorities collecting and treating fisheries statistics and transmitting them to Eurostat and other organisations. They have a high influence on EFS via the relevant Eurostat expert groups and a high interest in them.

·Data users (see Figure 4 below):

oInstitutional users are ones that are directly involved in EU policymaking. At EU level, they are the actual policy makers, and at international and national levels, they support policymaking and contribute to it. Institutional users include EU bodies, international organisations, national ministries, other fisheries authorities and national research institutes linked to the DCF. These users have a high interest in and influence on EFS.

oRedistributors are users that depend on EFS. They create and share publicly products based on EFS. The European Marine Observation and Data Network (EMODnet) and EUMOFA are identified as redistributors. At international level, FAO, OECD and the World Trade Organization (WTO) are at the same time institutional users and redistributors. They have a high interest in, but only a moderate influence on EFS.

oOther professional users contribute occasionally and indirectly to the policymaking process at EU level or structurally provide fisheries information. These are EU professional fisheries organisations, advisory councils, NGOs with marine programmes, sea conventions, specialized fisheries media, universities, research institutes, national fisheries organisations and individual private companies. They have a high interest in, but a low influence on EFS.

oThe general public and media only has a low interest in fisheries statistics and very limited influence.

Figure 4: Data user classification. Source: Eurostat.

According to the online survey and the public consultation organised in the context of the evaluation, there are a wide range of groups using EFS. The largest groups identified were national public authorities mainly responsible for fisheries policies (21%), NSIs (16%), and researchers working for the DCF, for fisheries research in other contexts and for other purposes (14%). The other half is distributed between specialised users (international organisations, fisheries associations, NGOs, EU services, and businesses) and the residual class of 'other users', including private persons, regional bodies etc. (11%). This analysis may not be exhaustive as the on-line survey was sent to a contact list of known potential users identified by Eurostat. However, the links to the survey and the open consultation were also available on the Eurostat website and social media, so they reached a wider user group.

Most users said that they use EFS for more than one purpose. The most common purpose was linked to the CFP at either national or EU level (36%). Research (16%) and commercial purposes (14%) were the second and third most common uses before environmental policy, blue economy policy, other policies and media uses.

Consultation strategy

The evaluation of EFS, which preceded this impact assessment, was supported by a large range of consultation activities. The evidence collection for the evaluation included:

·a workshop with Member States on the strengths, weaknesses, opportunities and threats of EFS from national statistical institutes’ point of view. The first major consultation was a one-day workshop organized back to back with a meeting of the Eurostat expert group "Fisheries Statistics Working Group" in October 2018. It focused on the strengths and weaknesses of the current EFS and the future opportunities and threats from the national data providers' point of view (NSIs and other national authorities). In addition to national members of the Fisheries Statistics Working Group, who are responsible for providing EFS, representatives from FAO, OECD and ICES took part in the workshop as well.

·16 in-depth interviews with key stakeholders, for example policymakers and contributors to the CFP, on their EFS use, needs and wishes. The in-depth interviews were principally undertaken to consult specific stakeholders: redistributors (organisations that redistribute EFS through their own databases, adding information from other countries or areas, here e.g.: FAO, OECD and EMODnet) and regular professional users (organisations that need EFS for the execution of their main professional activities, here e.g. DG MARE, EUMOFA, the JRC and ICES). The aim of the interviews was to gather detailed information on how EFS are used by these organisations, how they rate the quality of EFS, and how EFS could be improved in the future.

·Six national case studies and one cross-country aquaculture case study to provide overviews and detailed analyses of different data collection and collaboration approaches. The purpose of the case studies was to provide an overview on national set-ups for collecting EFS and to analyse them in a more detailed manner. They also served as a basis for understanding how various fisheries-related data collaborations are organised in Member States and how organisations cooperate. In addition, the case studies aimed at analysing how national data users use EFS and to assess whether EFS are meeting their needs from a national perspective. The national case studies took place in Denmark, France, Greece, Ireland, Italy and Poland. In addition, the supporting horizontal case study on aquaculture with particular emphasis on the data confidentiality issue covered the above-mentioned countries and Germany.

·an online survey as a targeted expert-oriented consultation with general and specific questions about EFS; an online questionnaire targeting EFS data users and producers collected feedback about their views of EFS covering a range of topics including the utility, ease of use, cost of collection, statistical quality, efficiency, effectiveness and coherence relevant to EFS. It ran from 13/12/2018 to 28/01/2019 and was circulated to 353 organisations or individuals identified by Eurostat as potential users of EFS. The questionnaire included general questions addressed to all respondents and three routings categorizing respondents as users, producers, or combined users and producers of EFS statistics. 135 respondents answered the questionnaire, representing 38.2% of the identified target audience. Responses were received from 33 out of 36 EU Member states, EFTA countries, candidate and potential candidate countries.

·a public consultation to gather information from professional users, citizens and other stakeholders on their experiences with EFS; the consultation 48 was conducted through the Commission’s dedicated website for consultations (EU Survey) and conformed to the Commission’s general principles and standards for consultation. It was launched on 18/01/2019 and closed on 12/04/2019. In total, 24 respondents answered the questionnaire.

The consultation strategy for the impact assessment was twofold:

1)As the impact assessment followed immediately after the evaluation, the evidence collected for it was also used where applicable for the impact assessment.

2)Impact assessment specific consultations of the general public (public consultation covering all target groups: data providers, producers and users), experts (expert consultation targeting mostly data users) and data producers (consultation of the DGAS, which also covers fishery statistics) were organised.

It was decided not to conduct further case studies or interviews for this impact assessment. The reason is that the public and expert consultation as well as the recent consultations for the evaluation cover a wide range of stakeholders and perspectives (see also table 5 below), enabling a detailed fleshing out of the context of EFS, its problems and the objectives. The development of the impact assessment report also did not reveal any substantial gaps in required elements, evidence or logic to be filled with more consultations. For these reasons, additional consultation activities were not seen as cost-effective.

In sum, the consultations for the EFS evaluation and this impact assessment provide a good overview of the views of data providers, data producers, institutional users and redistributors. However, the response rate of NGOs, the business sector, academia, media and the general public was low, so that no representative conclusions can be drawn from them by stakeholder group. This was expected, as fisheries statistics in general and in particular their legal reform do not attract much interest outside the management of the fishery statistics sector and their immediate main user circle.

Table 5: Impact assessment consultation methods to reach the stakeholder groups of EFS. Source: Eurostat.

Impact assessment public consultation

The public consultation focused on whether EFS meet respondents needs, let them comment on and rank the objectives and possible options, and let them give feedback on potential impacts of the options. It was online from 20/07/2020 to 23/11/2020 and translated into 23 EU languages. The consultation was promoted widely on the EU’s “Have your say” platform 49 , the Eurostat website 50 and social media with monthly reminders, and a mail was sent to Eurostat fishery stakeholder contacts. All results are published on the “Have your say” platform 51 .

The consultation garnered 15 responses. Almost half of the respondents were public authorities, two represented NGOs, two business associations, two citizens, one businesses and one an international organisation (other) (see Figure 5). 11 EU Member States and the UK were represented.

Figure 5: Respondents to the public consultation. Source: Eurostat.

The respondents said that they use EFS moderately because there are other sources and because EFS lack certain breakdowns and the non-confidential aquaculture data they require. The objectives of the EFS modernisation were seen as important or very important and as “about right” by 11 respondents, 3 wanted more ambition and mentioned known problems, and 1 wanted less. 11 respondents also liked the range of options offered.

The preferred option of 7 respondents was option 3: a new streamlined legal framework for EFS, followed by 4 preferring option 1: continue as now (baseline), and 3 option 4: a new legal basis for aquaculture, abolish catch and landing statistics. Nobody preferred option 2: discontinue EFS, and 1 respondent did not support any of the proposed options (see Figure 6).

Figure 6: Preferred option. Source: Eurostat.

The supporters of option 3 were regional, national and international authorities and NGOs. Option 1 was preferred by business organisations, one representative of a national authority and a non-EU citizen. Option 4 was preferred by a public authority, a business association and an EU citizen.

The respondents were also giving their opinions on the impacts of the options on five stakeholder categories: data producers (statistical authorities), data providers (fishers, fleet and aquaculture plant managers), policy makers, professional users (e.g. NGOs and business associations), and citizens and others (see Figure 7).

Option 3: “A new streamlined legal framework for EFS” scored the most positive impacts on all stakeholder groups. In addition, only very few respondents identified negative impacts for option 3. Option 2, “Discontinuation of EFS”, was considered as the most negative in particular for the data user category. Option 1 “Continuation of the current policy” was considered rather neutral in its impacts. Lastly, option 4: “ New legal basis for aquaculture and compilation of other fisheries statistics from EU level administrative sources” had similar positive impacts as option 3, but to a lesser extent, and at the same time many respondents also identified negative impacts. This is explained by the different treatment of aquaculture and other fisheries statistics in this option.

Figure 7: Public consultation opinions on the impacts of the options on stakeholders. Source: Eurostat.

In conclusion, there was a range of responses to the public consultation despite the comparatively low number of respondents. They mostly confirmed issues already known from other stakeholder consultations, they approved of the EFS modernisation objectives, and they preferred option 3. Given the limited participation to the consultation, these results need to be treated with caution, but they point in the general direction the other consultations do.

Expert consultation

The expert consultation targeted data needs. It ran from 16/07/2020 to 31/08/2020 on EU Survey. The invitation to participate was sent to more than 300 contacts and promoted on Eurostat’s website. The consultation received 35 responses from 14 countries, of which 22 were from professional organisations (4 EU, 14 national, 4 regional), 9 from data producers, 3 from redistributors, and 1 from a NGO. The respondents gave detailed responses on their data needs for catches, landings and aquaculture, their frequency and timeliness preferences, and the purposes they use EFS for (e.g. market and traceability analyses).

The summarised results are:

Catches

·Expert users rely on catches to be derived from weighted landed products rather than catches estimated on board.

·The part of the catches that is not landed or is not the target species is interesting for more than half of the respondents.

·More than 70% of the respondents use landings for fish market analysis, while more than 52% use it for traceability.

·The most demanded breakdowns for catches (in addition to species) are the flag of the vessel, the FAO fishing region and the country of landing.

·77% of the respondents are interested in bycatch; out of those, almost 90% are interested in bycatches at species level and 70% on both fish and non-fish species (marine mammals, birds, reptiles).

·Most of the respondents would appreciate that catch statistics for the reference year n become available in the second quarter of year n+1 at the latest.

Landings

·Almost 70% of the respondents interested in landings use it for analysis of the fish market, while traceability is interesting for more than half of the respondents.

·Weight of fishery products and their value stand out as the most interesting variables for expert users.

·As regards coverage, more than half of the users need data on fishery products fished by EU and EFTA vessels and landed anywhere in the world.

·For breakdowns to present variables, the higher interest is on fishing area, country where products are landed, flag of the vessel, intended use of the product, and presentation and preservation status.

·Expert users are interested in the breakdown between landed fishery products for economic purposes, and those landed for conservation purposes (landing obligation).

Aquaculture

·A majority of users is interested in covering all aquaculture (for example extensive aquaculture) rather than only commercial aquaculture.

·Remaining stocks are interesting for the respondents (69%).

·For a majority of users, official statistics are the only source for aquaculture.

·Some professional organisations (including at EU level) are running their own questionnaires.

·On the structural dimensions and variables proposed, production capacity scores the highest, together with number of companies, geographical location, number of production sites, and environments. Environmental measures and data on the structure of the aquaculture sector score the lowest.

·Aiming at the second or third quarter of year n+1 for data delivery would satisfy most of the users.

The results of the consultation thus confirmed that there is a need to update the legislation in order to accommodate new data needs.

Consultation of DGAS

DGAS organised a workshop on “Future options for European fisheries statistics” on 5-6/06/2019 in Luxembourg. It was attended by all Member States and EFTA countries as well as by some candidate and potential candidate countries. In that way, all data producers were reached (NSIs and other national authorities).

The workshop focused on the preparation of the EFS impact assessment. The objective was to brainstorm potential impacts of the draft options, propose other options, rank the preliminary options and contribute to the impact assessment. The four preliminary options discussed in splinter sessions were the following:

Option 1: Baseline scenario: Continuation of the current policy

Eurostat continues producing the EFS under the current legal framework, but streamlines and simplifies the legal framework through delegated acts.

Option 2: Discontinuation of EFS

The EFS regulations are repealed, and Eurostat would stop producing EFS.

Option 3: New streamlined legal framework for EFS

A new streamlined framework regulation for EFS covering catches, landings, fishing fleet and aquaculture statistics is adopted.

Option 4: Integration of streamlined aquaculture statistics into SAIO and compilation of other fisheries statistics from EU level administrative sources

Integration of streamlined aquaculture statistics into the SAIO regulation; the legal basis for catch and landing statistics is discontinued, and EFS are derived from EU level administrative data to the extent possible.

The outcomes of the DGAS workshop were as follows:

Option 1: Baseline scenario: Continuation of the current policy

Member States and data providers have a good understanding of the current policy and legal framework for EFS. This is a strength of the current system, and the risk of upheaval and the transformation costs in data collection and preparation of statistics under any alternative approach should be considered more fully in the impact assessment. The main benefits that would be retained under the baseline option are the continuation of the long time series of statistics which have been compiled, and the industry, user and producer familiarity with the system.

However, several significant disadvantages to continuing with the current approach were identified, and importantly, the working group considered that the most significant of these could not be solved through delegated acts to the current legal acts. Retaining the current approach would not address:

·the need for consistency in definitions and classifications used under the EFS;

·double reporting of data to different users; and

·the need for flexibility in the statistics to allow a response to new and emerging user needs, and technological developments in the sector, especially in aquaculture.

Retaining the current approach would enable small improvements to streamlining quality reporting and methodological updates. The group concluded that the time and effort required to make minor amendments to the current legal basis for this purpose would not be justified as it would not respond to the primary weaknesses of the current system. The burden of double reporting would remain due to the lack of harmonisation.

This was the least preferred of the “do something” options based on the poll of DGAS attendees, but still preferred to discontinuing EFS altogether.

Option 2: Discontinuation of EFS

Within the ESS, Member States produce high quality and comparable data. The prospect of being unable to benefit from these harmonised standards across the EU would jeopardize data comparability and availability.

Discontinuing EFS also poses a risk to data users: unavailability of long time series or, in the best-case scenario, breaks in time series could be foreseen. The lack of independent statistical data could lead to data from administrative sources only being collected to respond to short-term policy needs, which could result in the meaning of variables changing over time. With regard to data availability for users, the risk is higher for aquaculture, as catches and landings data will continue to be collected due to the CR.

In the case of discontinuation at EU level, national regulations on aquaculture statistics risk being repealed, and instead rough estimates could potentially be used by statistical offices, resulting in a decrease in quality.

There are marginal benefits to not reporting catches and landings data twice; the data has to be collected and produced anyway.

Fisheries statistics are needed to build national accounts by NSIs. Some countries conduct sample surveys to complement administrative data, for example to cover small-scale fishing fleets in the case of catch statistics; discontinuing EFS would result in incomplete data.

This was the least preferred option in the poll. Only one participant identified discontinuation of EFS as the preferred option, while 19 of the 26 respondents identified option 2 as the least preferred option of the four presented.

Option 3: New Streamlined Legal Framework for EFS

There was a strong view that any simplification of EFS should be aligned with administrative data. In particular, it was suggested that EFS adopt the same level of aggregation as administrative data, i.e. not request additional information which leads to additional work and inconsistencies. It was noted that it is difficult to collect detailed species data for catch and landing statistics and that there is a desire to move to more aggregated data (i.e. genus, family level) for some species that are of minor economic importance.

Simplification of EFS should ensure the harmonisation of data collection and avoid the double reporting that occurs under the current system. It is considered essential that other organisations such as FAO and OECD are involved in the process to ensure that their needs are addressed and to secure alignment of definitions and approaches so that the needs of these users and the Commission are met without the need for double reporting. Furthermore, wider user needs should be reviewed to ensure that the statistics under the “new” EFS are useful now and for the future.

Regarding aquaculture, simplification is needed particularly with respect to data collected on methods of production, which are considered difficult to collect. More generally, there is a need for less disaggregation of aquaculture data to avoid the current problem of data being flagged as confidential. There is also a need to ensure greater harmonisation of EFS aquaculture data with DCF data.

In addition to simplification, any new legal framework should ensure the streamlining of definitions, classifications, concepts and timings of data transmission compared to the current situation. This would allow for one single point of data reporting and eliminate the need for double reporting.

Other out-of-date and limiting aspects of the current legal framework could also be addressed under this option to better meet user needs. For example, the geographical coverage of catches and landings (global catches in marine and inland waters and global landings by EU and EFTA vessels) could be extended.

Option 3 received the highest number of first choice votes in the informal poll taken during the event. Of the 26 responses received, 21 (80%) chose option 3 as their preferred option of the four presented.

Option 4: Integration of streamlined aquaculture statistics into SAIO and compilation of other fisheries statistics from EU level administrative sources

Integration of aquaculture into SAIO

The general understanding is that there is a need to develop a new streamlined and simplified regulation for aquaculture, which could be integrated into SAIO or as an alternative option, as a stand-alone regulation.

The integration into SAIO seems difficult for marine aquaculture, mainly because aquaculture belongs to the same policy area as fisheries in many organisations, and would need more harmonisation, for example on species, than agriculture statistics. Wild catches are also a particular case which is different to agricultural production. From the production point of view, especially for inland aquaculture, there are more similarities. The group identified a potential issue linked to the availability of funding instruments for the production of aquaculture statistics if they were to be covered under SAIO.

Integration into SAIO could potentially require an increased effort to achieve harmonisation between domains, while keeping the link with catches and landings. However, integration into SAIO could be an opportunity to collect information on innovative ways of producing food, such as aquaponics or using insects or algae for feed.

The participants discussed how to improve the problems of aquaculture data availability due to confidentiality. Streamlining and simplifying data collection through targeting user needs could help, for example by reducing the breakdowns of the data to be disseminated and using scientific use files for research-related users. Another alternative could be using perturbation techniques instead of suppression.

Compilation of other fisheries statistics from EU level administrative sources

The participants also indicated that the use of administrative data involves some risks in ensuring coverage, quality, data availability and timeliness. In particular, when other statistical domains such as accounts need fisheries data, their needs could potentially not be met.

The participants of the DGAS seminar stressed that administrative data must be used to produce statistics where possible. To do this, the European Commission was asked to ensure that administrative data sources used meet the quality criteria required to produce reliable official statistics.

This was the preferred option for two respondents to the poll, and the second choice option for 14 of the 26 respondents (all of whom ranked option 4 second to option 3).

Additional options

One additional option for consideration in the impact assessment was proposed, a hybrid of options 3 and 4 whereby there would be a single legal framework developed for EFS fisheries statistics (catch, landings and fleet), while aquaculture would be integrated into the SAIO regulation.

Annex 3: Who is affected and how?

Practical implications of the initiative

The practical implications of implementing option 3, a new streamlined legal framework for EFS, would mainly affect three stakeholder categories:

·Data providers: this category includes not only the respondents to surveys (i.e. fishers, vessel owners and aquaculture businesses) but also national institutions that are keepers of administrative data files, such as national fisheries authorities keeping fleet, catch and aquaculture business registers.

·Data producers: this category comprises the authorities responsible for the collection and compilation of fisheries statistics. It mainly includes NSIs, other national authorities such as fisheries agencies, and Eurostat.

·Data users: this category includes institutional users such as Commission services, other EU institutions, international organisations such as FAO, ICES, OECD, and RFMOs, national administrations, businesses, NGOs, research institutions and academia, media and the general public.

The practical implications for these groups would be as follows:

·The implications for the data providers would mainly touch aquaculture businesses and fisheries authorities keeping the administrative registers. The less complex data collection for aquaculture would slightly decrease the burden on aquaculture businesses. Most of these enterprises are small and medium sized. There are no impacts on their competitiveness. The other fisheries businesses are not directly influenced by the new streamlined legal framework for EFS, as the statistics are to be drawn from the administrative registers kept primarily for other than statistics purposes.

The implication for the national fisheries authorities would materialize in closer co-operation with statistical authorities as more administrative data would be used directly for providing statistics. This would require fine-tuning the validation of administrative data.

·Data producers would have to adapt their data collection, production and dissemination systems to new or different variables in the data collections. National data producers would benefit from reduced costs of data collection, as Eurostat would draw the statistics for catches directly from administrative data used for managing the CFP and national fisheries administrations. In the short term, national statistical authorities would need to invest in closer co-operation with fisheries authorities for improving the validation of the administrative data. The impact on Eurostat is estimated to be cost-neutral.

·The most significant implications would touch data users as the new streamlined legal framework for EFS would increase the availability of data and improve the coherence of the data. This would enable them to use better statistical data on which to base their work and decisions.  

International organisations, in particular FAO and OECD, could directly use EFS without conducting separate data collections from the countries. This would save them time and effort and the countries as well.

The summary of benefits and costs is presented in table 6 below.

Table 6: Benefits and costs of the preferred option. Source: Eurostat.

Annex 4: Analytical methods

Analysis of the potential impacts of the different policy options for the future EFS has been based on the methodology proposed in the Better Regulation Guidelines of the European Commission. As a first step, potential impacts were identified in relation to the different options. This was followed by a primarily qualitative analysis in order to identify the more important impacts of the various options. This analysis was complemented with information available from a number of monitoring and evaluation documents linked to the implementation of the current EFS, as well as from a series of consultations with stakeholders of EU fisheries statistics, the results of the public consultation, and several other documents outlined above. When analysing the impacts of the different options, factors such as relevance, effectiveness, efficiency, coherence, EU added value, statistical quality, EFS stakeholders’ opinions, and the technical feasibility of implementing the respective option were taken into account.

In line with the Better Regulation guidelines, impact assessments should also provide details for all options on the information obligations for businesses, citizens and national/regional/local administrations which are likely to be added or eliminated if the option were implemented. In those cases in which the change in administrative burden is likely to be significant, the effects should be quantified using the EU Standard Cost Model. However, this approach is difficult to apply while analysing the impact in terms of costs and burden of EFS. To make this analysis for the different policy options, Eurostat therefore relied, inter alia, on the evidence gathered from previous and ongoing quantitative and qualitative assessments of costs and burdens related to statistical production in the field, in order to assess the expected costs and burdens of data collection for respondents and producers of fishery statistics.

Cost estimates

The costs of the data collection are regularly collected from the Member States as part of the activities of Eurostat’s Resource Directors group. Costs estimation methodology is based on the costs declarations done by countries to the ESS. It is often difficult for the countries to assess the costs for catches and landings separately, as the two data collections are to some extent intertwined – catches are mostly calculated from landings using conversion factors as “nominal catches”. In addition, the indirect costs are difficult to assign per data collection. A further challenge for fisheries statistics is posed by data collections carried out outside of NSIs in other national authorities; in some cases, NSIs were not able to report on costs incurred in other institutions. For these reasons, the cost savings are estimations. The saving of 1.2 million € is estimated on the basis of the plan to collect catch statistics directly by Eurostat from the administrative data collected by Member States and provided to DG MARE. Currently Member States spend 1.5 million € on catch statistics annually. This cost is estimated to drop to 0.3 million € as only the costs resulting from the cooperation on the data quality with the fisheries management administration would stay.

Assessing the relevance of statistics

The relevance of statistics is normally assessed when the legal basis is adopted, and hence it is not part of the regular assessment of statistical quality. EFS are produced under the expectation that the main users are policy users, mainly those working on the CFP, and also those working with environmental and trade policies. Inside the European Commission this includes DG MARE, the JRC (in support tasks to DG MARE), DG Environment and DG Trade. EFS are also assumed to be used by other EU Institutions (e.g. the European Parliament, the European Court of Auditors and the Council of the European Union as linked for example to fishery quotas), international organisations (FAO, OECD and ICES), regional fisheries bodies (e.g. RFMOs), national administrations (ministries), national statistical institutes (NSIs) and research institutes, the business sector, media and the public in general.

Eurostat regularly carries out user satisfaction surveys; EFS are clustered together with agricultural statistics. Due to the limited number of replies on these areas, it is not possible to draw substantive conclusions for EFS except rough user sentiments. Partly because of this, Eurostat uses the number of downloads of EFS from its online database as an indication of their relevance. In a five-year period (2013 to 2017), the number of downloads almost tripled, reaching 24.000 downloads in 2017. The downloads are relatively evenly spread between dataset groups, with catch statistics being slightly more used than the other dataset groups.

These downloads do not however reveal what the statistics are used for. Hence, more contextual and qualitative information needs to be used to assess statistics’ relevance and impact.

(1)

Statistics in the ESS context are defined according to Article 3(1) of Regulation (EC) No 223/2009 as “quantitative and qualitative, aggregated and representative information characterising a collective phenomenon in a considered population”.

(2)

   European statistics Code of Practice, revised version endorsed by the European Statistical System Committee on 16 November 2017

(3)

  Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics and repealing Regulation (EC, Euratom) No 1101/2008 of the European Parliament and of the Council on the transmission of data subject to statistical confidentiality to the Statistical Office of the European Communities, Council Regulation (EC) No 322/97 on Community Statistics, and Council Decision 89/382/EEC, Euratom establishing a Committee on the Statistical Programmes of the European Communities, OJ L 87, 31.3.2009, p. 164–173

(4)

  Better regulation: why and how

(5)

  The European Commission’s priorities

(6)

  Regulation (EU) No 1380/2013 of the European Parliament and of the Council of 11 December 2013 on the Common Fisheries Policy, amending Council Regulations (EC) No 1954/2003 and (EC) No 1224/2009 and repealing Council Regulations (EC) No 2371/2002 and (EC) No 639/2004 and Council Decision 2004/585/EC, OJ L 354, 28.12.2013, p. 22–61

(7)

  Council Regulation (EC) No 1224/2009 of 20 November 2009 establishing a Community control system for ensuring compliance with the rules of the common fisheries policy, amending Regulations (EC) No 847/96, (EC) No 2371/2002, (EC) No 811/2004, (EC) No 768/2005, (EC) No 2115/2005, (EC) No 2166/2005, (EC) No 388/2006, (EC) No 509/2007, (EC) No 676/2007, (EC) No 1098/2007, (EC) No 1300/2008, (EC) No 1342/2008 and repealing Regulations (EEC) No 2847/93, (EC) No 1627/94 and (EC) No 1966/2006, OJ L 343, 22.12.2009, p. 1–50

Commission Implementing Regulation (EU) No 404/2011 of 8 April 2011 laying down detailed rules for the implementation of Council Regulation (EC) No 1224/2009 establishing a Community control system for ensuring compliance with the rules of the Common Fisheries Policy, OJ L 112, 30.4.2011, p. 1–153

(8)

  Regulation (EU) 2017/1004 of the European Parliament and of the Council of 17 May 2017 on the establishment of a Union framework for the collection, management and use of data in the fisheries sector and support for scientific advice regarding the common fisheries policy and repealing Council Regulation (EC) No 199/2008, OJ L 157, 20.6.2017, p. 1–21

(9)

  Saving our heritage, our future: The worrying state of Mediterranean fish stocks

(10)

  Agriculture, forestry and fishery statistics, 2020 edition, Eurostat

(11)

  The state of world fisheries and aquaculture, 2020, FAO

(12)

  Eurostat Statistics Explained - Fishery statistics

(13)

  Scientific, Technical and Economic Committee for Fisheries: The EU aquaculture sector – Economic report 2020. Publications Office of the European Union, Luxembourg. 2021.

(14)

Scientific, Technical and Economic Committee for Fisheries: The 2020 Annual Economic Report on the EU Fishing Fleet. Publications Office of the European Union, Luxembourg. 2021.

(15)

  Facts and Figures on the Common Fisheries Policy 2020

(16)

  Regulation (EC) No 1921/2006 of the European Parliament and of the Council of 18 December 2006 on the submission of statistical data on landings of fishery products in Member States and repealing Council Regulation (EEC) No 1382/91, OJ L 403, 30.12.2006, p. 1–8

Regulation (EC) No 762/2008 of the European Parliament and of the Council of 9 July 2008 on the submission by Member States of statistics on aquaculture and repealing Council Regulation (EC) No 788/96, OJ L 218, 13.8.2008, p. 1–13

Regulation (EC) No 216/2009 of the European Parliament and of the Council of 11 March 2009 on the submission of nominal catch statistics by Member States fishing in certain areas other than those of the North Atlantic (recast), OJ L 87, 31.3.2009, p. 1–41

Regulation (EC) No 217/2009 of the European Parliament and of the Council of 11 March 2009 on the submission of catch and activity statistics by Member States fishing in the north-west Atlantic (recast), OJ L 87, 31.3.2009, p. 42–69

Regulation (EC) No 218/2009 of the European Parliament and of the Council of 11 March 2009 on the submission of nominal catch statistics by Member States fishing in the north-east Atlantic (recast), OJ L 87, 31.3.2009, p. 70–108

(17)

 Fish catches cover fish, molluscs, crustaceans and other aquatic animals, residues and aquatic plants that are taken for all purposes, by all types and class of vessel, gear and fishermen, operated in all marine areas: high-sea fishing areas, offshore, inshore or brackish water areas. The production from aquaculture and catches in fresh water are excluded. EFS data on fish catches are recorded in tonnes live weight. Landlocked EU countries without a marine fishing fleet are not included (Czech Republic, Luxembourg, Hungary, Austria and Slovakia).

(18)

 EFS landings statistics relate to fisheries products (product weight and value) landed in a country by vessels of the country or vessels from other EU/EEA countries (excluding landings by non-EU/EEA vessels), and to fisheries products landed by the country’s vessels in non-EU ports and then imported into the EU (excluding non-imported fisheries products). Landlocked EU countries without a marine fishing fleet are not included (Czech Republic, Luxembourg, Hungary, Austria and Slovakia).    

(19)

Aquaculture is the farming of aquatic organisms under controlled conditions. It includes the cultivation of fish and fish eggs, the production of juveniles, capture-based inputs, and production for human consumption e.g. of fish, mussels and crustaceans.    

(20)

 The EFS catch regulations currently cover seven FAO fishing areas near the EU: 21 – Northwest Atlantic, 27 – Northeast Atlantic, 34 – East-Central Atlantic, 37 – Mediterranean and Black Sea, 41 – Southwest Atlantic, 47 – Southeast Atlantic, and 51 – West Indian Ocean.    

(21)

  Commission Implementing Regulation (EU) 2017/218 of 6 February 2017 on the Union fishing fleet register, OJ L 34, 9.2.2017, p. 9–17

(22)

Administrative data are data collected for other than primarily statistical purposes. In the case of fisheries, they are used e.g. for the management of the CFP and in some cases reused and treated to compile EFS.

(23)

  Commission Implementing Decision (EU) 2019/909 of 18 February 2019 establishing the list of mandatory research surveys and thresholds for the purposes of the multiannual Union programme for the collection and management of data in the fisheries and aquaculture sectors, OJ L 145, 4.6.2019, p. 21–26

(24)

  Farm to Fork strategy

(25)

  Commission Staff Working Document Evaluation of the European Fishery Statistics, SWD(2019) 425

(26)

Source: Eurostat own calculations. Data available on request.

(27)

These cost estimates are provided by the countries in the statistical cost data collections carried out by the ESS Resource Directors’ Group for the reference years 2015/2016 and 2020. The Member States’ cost for aquaculture statistics is 2.1 million €, for landing statistics 1.8 million €, and 1.5 million € for catch statistics. The estimated cost for Eurostat is 0.3 million €. The evaluation and subsequently the inception impact assessment erroneously quoted as the total cost of EFS the national costs of only catch statistics and the Eurostat costs.

(28)

  Report from the Commission to the European Parliament and the Council on the implementation of Regulation (EC) No 762/2008 of the European Parliament and of the Council of 9 July 2008 on the submission by Member States of statistics on aquaculture and repealing Council Regulation (EC) 788/96, COM(2017) 747

(29)

  https://www.eumofa.eu/  

(30)

  Special report No 7/2007 on the control, inspection and sanction systems relating to the rules on conservation of Community fisheries resources together with the Commission's replies, OJ C 317, 28.12.2007, p. 1–33

Special report No 10/2014 – The effectiveness of European Fisheries Fund support for aquaculture

Special report No 8/2017 – EU fisheries controls: more efforts needed

(31)

  Legislative train schedule – revision of the Fisheries control system  

(32)

  Commission Delegated Decision (EU) 2021/1167 of 27 April 2021 establishing the multiannual Union programme for the collection and management of biological, environmental, technical and socioeconomic data in the fisheries and aquaculture sectors from 2022, OJ L 253, 16.7.2021, p. 51–90

(33)

 The Brussels effect can be defined as a process of unilateral regulatory globalisation caused by the European Union de facto, but not necessarily de jure, externalising its laws outside its borders through market mechanisms. This is because the combination of market size, market importance and relatively stringent standards and regulatory capacity of the EU can have the effect that firms trading internationally find that it is not economically, legally or technically practical to maintain lower standards in non-EU markets, and instead adopt standards set by the EU uniformly throughout their business. See e.g. Bradford, Anu (2020): The Brussels Effect. How the European Union Rules the World. Oxford University Press.

(34)

  Regulation (EU) No 1350/2013 of the European Parliament and of the Council of 11 December 2013 amending certain legislative acts in the field of agricultural and fishery statistics, OJ L 351, 21.12.2013, p. 1–14

(35)

  Report from the Commission to the European Parliament and to the Council – Annual report to the Discharge Authority on internal audits carried out in 2017, COM/2018/661

(36)

The source for the numbers in this paragraph are Eurostat and DG MARE own calculations. Data available on request.

(37)

 Blue growth is the European Commission’s long term strategy to support sustainable growth in the marine and maritime sectors as a whole, as seas and oceans are seen as drivers for the European economy and to have great potential for innovation and growth. See https://ec.europa.eu/maritimeaffairs/policy/blue_growth_en  

(38)

 The six tuna RFMOs are ICCAT, IOTC, WCPFC, IATTC, AIDCP and CCSBT, and the nine non-tuna RFMOs are CCAMLR, CCBSP, NEAFC, NAFO, NASCO, SEAFO, SIOFA, SPRFMO and GFCM. See https://ec.europa.eu/fisheries/cfp/international/rfmo_en

(39)

  Consolidated version of the Treaty on the Functioning of the European Union, OJ C 326, 26.10.2012, p. 47–390

(40)

  Regulation (EU) 2019/2152 of the European Parliament and of the Council of 27 November 2019 on European business statistics, repealing 10 legal acts in the field of business statistics, OJ L 327, 17.12.2019, p. 1–35 

(41)

  Proposal for a Regulation of the European Parliament and of the Council on statistics on agricultural input and output and repealing Regulations (EC) No 1165/2008, (EC) No 543/2009, (EC) No 1185/2009 and Council Directive 96/16/EC

(42)

  2012/504/EU: Commission Decision of 17 September 2012 on Eurostat, OJ L 251, 18.9.2012, p. 49–52

(43)

  Regulation (EU) 2018/1091 of the European Parliament and of the Council of 18 July 2018 on integrated farm statistics and repealing Regulations (EC) No 1166/2008 and (EU) No 1337/2011, OJ L 200, 7.8.2018, p. 1–29

(44)

 See https://ec.europa.eu/eurostat/web/quality/general-evaluation-results

(45)

The five key performance indicators are the following ones: 1) number of data extractions made by external users from the Eurostat reference database via the Eurostat website; 2) percentage of users that rate as "very good" or "good" the overall quality of European statistics; 3) percentage of users that rate as "very good" or "good" the timeliness of European statistics for their purposes; 4) percentage of users that rate as "very good" or "good" the comparability of European statistics among regions and countries; and 5) residual error rate.

(46)

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

(47)

  https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12344-European-fishery-statistics  

(48)

  https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/1857-Evaluation-of-the-European-Fishery-Statistics/public-consultation  

(49)

  https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12344-European-fishery-statistics/public-consultation  

(50)

  https://ec.europa.eu/eurostat/about/opportunities/consultations/european-fishery-statistics-2020

(51)

  https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12344-European-fishery-statistics/public-consultation