I.  Introduction

This article presents the results of a comparative research study, which was carried out in the context of the AFRADE project co-funded by the EU Anti-Fraud Programme (EUAFP) and the University of Pisa.¹ It involved Italy, Poland, Bulgaria, Slovakia, and Romania. The project aimed at providing insights to help improve analytical methods for detecting fraud and irregularities concerning funds in Europe’s common agricultural policy (CAP).

AFRADE’s outline followed three stages, each corresponding to a key area of investigation, namely: CAP payment mechanisms, criminal patterns, and inter-authority information exchange strategies. These topics were respectively addressed in three dedicated focus groups. The project involved five legal experts from each country: Dr. Claudia Cantisani (IT), Prof. Celina Nowak (PL); Prof. Minko Georgiev (BU); Prof. Libor Klimek (SK); and Dr. Monica Mihaela Tudor (RO). They prepared a report on their national legal system concerning the payment mechanisms for CAP funds, the most common fraud schemes/offences, and the detection and reporting of agricultural frauds. The findings in each national report were then discussed at the project's final conference. In addition, the research team used these findings to develop common guidelines for the detection and reporting of agricultural fraud.

The study was initiated because the European legal framework sets rules for reporting irregularities and suspected frauds but lacks uniformity in risk indicator assessment.² This in turn leads to varied evaluation methods across Member States that impede cooperation between national and supranational investigative bodies.³

The fragmentation among national legal systems in detecting and reporting fraud is due, in particular, to inconsistent definitions of offenses as well as administrative irregularities across EU Member States.⁴ It is also due to the complexity of fraudulent schemes in the CAP sector. Furthermore, data on the rates of fraud are ambiguous, as they rely on investigative authorities’ capabilities in detecting not just fraud per se but also other linked activities, such as corruption and conflicts of interest; the detection of fraud and irregularities also often depends on transparency and efficient information exchange between authorities.⁵ In addition, public authorities involved in the payments’ mechanisms (paying agencies) often operate in opaque environments – a circumstance that reinforces the need to simplify administrative procedures.

The research project addressed these issues and suggested two key strategies:

• Enhancing inductive methodologies for fraud risk analysis, focusing on case-based strategies rather than on the legal discipline of each offence;

• Increasing the integration of IT systems, and improving data accessibility and interoperability.

Before these key strategies are explained in more detail (IV. below), the article will first describe the CAP paying system (including the role of paying agencies) and the concrete functioning of control activities (II.) and then present the broader comparative results concerning CAP fraud schemes (III.).

II.  The CAP Funding System

CAP funding is basically made up of two main segments: the European Agricultural Guarantee Fund (EAGF) and the European Fund for Rural Development (EAFRD).⁶ Payments deriving from both funds are managed at the national level by each EU country. To protect the Union’s financial interests in this sector, Member States are required to set up a management and control system for payments that complies with EU rules. Moreover, they must ensure that this system functions effectively and is able to prevent, detect, and correct irregularities. Last but not least, Member States are required to use IT systems to collect and report performance data on expenditure under the CAP strategic plans.⁷

1.  National paying agencies

Accredited national paying agencies and coordinating bodies⁸ are the public entities entrusted with ensuring the eligibility of all fund applications and the correct execution of payments to beneficiaries of CAP funds. Since they are required to provide sufficient guarantees that a claim is authorised for payment, they must undertake sufficient checks to ensure compliance with EU rules. Moreover, they must correctly and fully record the payments and submit the requested documentation to the European Anti-Fraud Office (OLAF).⁹

Paying agencies allocate each specific kind of fund on the basis of different requirements, depending on the type of intervention. It is important to clarify that all direct payments are granted only to active farmers. The relevant definitions, e.g., ”active farmer”, ”agricultural activity”, “agricultural area”, “eligible hectare”, “young farmer”, and “new farmer”, are provided by each national CAP Strategic Plan¹⁰ in accordance with the framework given by EU legislation.¹¹ In practice, this means that these definitions may differ among Member States.¹²

The application for funding is highly digitalised, which simplifies both the application process and the verification of the farmer’s declarations. For area-based measures and measures implemented within the framework of CAP strategic plans, in particular, the application must be submitted using the geospatial application form provided by the competent authority.¹³ Furthermore, the declarations must be verified through the area monitoring system (AMS).¹⁴ This system is used to observe, track, and assess agricultural activities and practices on agricultural land, making use of information provided by the Sentinel satellites of the European Copernicus programme, supplemented by European Geostationary Navigation Overlay Service (EGNOS) and Galileo automatically processed data.¹⁵

2.  Means of detection and reporting

According to Art. 72 of Regulation 2021/2116, paying agencies shall annually conduct administrative checks on aid applications and payment claims to ensure their legality and regularity.¹⁶ These checks are carried out on all applications for direct payments through the Integrated Administration and Control System (IACS),¹⁷ which exchanges and cross-references certified information with other databases.

On-the-spot checks are conducted on only a sample of applications. According to Regulation 2021/2116, they may be executed remotely with the use of technology.¹⁸ On-the-spot checks usually involve a physical visit to the farm in order to verify the accuracy of the declarations before the full aid amount is paid.

Ex-post checks, however, apply only to those measures that require the commitments be maintained after the full amount has been paid. These checks are conducted on a sample of applications and may also include a visit to the farm.¹⁹

A key role in the context of controls of applications is played by the Land Parcel Identification System (LPIS). A geographic information system established and periodically updated by Member States on the basis of aerial or spatial orthophotos,²⁰ LPIS makes it possible to geolocalise, visualise, and spatially integrate the constituent data of the Integrated Administration and Control System (IACS) at the agricultural parcel level. In this way, it enables paying agencies to determine the land’s use and maximum eligible areas under the various Union aid schemes.

III.  CAP Fraud Schemes

As far as CAP fraud schemes are concerned, the study highlighted that each EU Member State (which was part of the study) more or less incorporated three main types of fraud (outlined in the PIF Directive) in its legal system:²¹ falsity, non-disclosure, and misapplication of funds.²² These types of fraudulent conduct can be divided into the following two groups:

• Undue receipt of funds, mainly based on false declarations or falsification of documents as well as on the non-disclosure of obligatory information;

• Distorted use of funds, mainly based on the misapplication of purposes the funds were granted for.

While the first type of offence requires treacherous conduct to obtain EU funds, the second type concerns legally obtained funds that are successively used for purposes other than those they were originally planned for.

The analysis revealed the following problematic issues: First, a large number of legal provisions in some countries may be ineffective, because judges struggle with establishing the elements of crimes. This is often the case in the context of misapplication of funds. Second, national legal orders include several similar provisions, as a tendency towards overlapping exists; this causes delays in the definitive application of sanctions and leads to risks of infringing the ne bis in idem principle.²³ Third, criminal sanctions can only be imposed if there is strong evidence that the crime occurred; however, fraud often follows very complex patterns, especially in the CAP sector, that depend on several factors related to the type of funds, territory, national payment mechanisms, and eligibility conditions, all of which make proving the crime difficult.²⁴

The study found that administrative measures, such as recovering misallocated funds or excluding beneficiaries from further payments, may offer a more efficient solution than criminal sanctions. Indeed, measures like pecuniary administrative sanctions or disciplinary actions can enable authorities to intervene in the payment process earlier and are more effective in curbing fraud. They could also better target corporate compliance strategies, as companies often play a central role in fraudulent activities.

Indeed, the study’s ability to identify and develop more effective solutions for preventing CAP fraud largely depends on its focus on corporate activities and the most common fraudulent strategies employed by cross-border criminal organisations. For this reason, it also included a brief analysis of the most common criminal patterns, which can be summarized as follows.

Common criminal patterns

To understand the structure of fraudulent offences in the CAP sector, it is important to consider that aid requirements have a significant influence on fraud patterns. The study found, in fact, that they represent a key element in understanding the mechanisms of CAP fraud. Given the historical development of the EU’s common agricultural policy, the content of the aid requirements has changed over time,²⁵ with significant impact on fraudulent strategies. For example, the latest CAP reforms relate the disbursement of funds to the accomplishment of sustainability requirements (according to the conditionality principle) instead of production rates (i.e., quantitative thresholds of agricultural production).²⁶ Making the disbursement of funds dependent on the achievement of productive results makes it more difficult to resort to fraudulent strategies, because production results, in terms of agricultural yields, are quantifiable data and more easily verifiable. But, declarations of compliance with sustainability requirements call for assessments, the control of which is increasingly problematic.

It has been shown that the most common criminal patterns related to CAP shared-management funds (i.e., funds that are implemented and managed by the European Commission and the EU Member States together) are falsification or alteration of the conditions requested for disbursement of agricultural funds (e.g., false declarations regarding the farmers’ land or the farmers’ personal circumstances).²⁷ For example, applicants requesting direct payments may request aid for plots of land they are not entitled to, due to false agreements, or they may artificially create conditions for receiving aid and financial support. Indirect payments, such as rural development funds, may encourage applicants to submit false invoices or falsely declare equipment as new, even though it is not. This can involve manipulated information and misrepresentations regarding compliance with the financing conditions.²⁸ Violations and falsifications may involve eligibility criteria for receiving advance payments, submitting aid requests, or accessing support schemes. Furthermore, beneficiaries may breach procurement rules, seek reimbursement for inflated costs or non-existent transactions, or even request reimbursement for costs already covered elsewhere. Notably, this last type of fraud is common in cross-border corporate crime, often carried out by organised criminal groups that establish shell companies at the same address, each with its own bank account tied to the same financial institution.

IV.  Proposals for Improving CAP Fraud Prevention and Detection

A first key point raised by our study is the need to shift the focus from legal harmonisation to a more practical, case-by-case strategy in order to develop more effective ways of combating agri-frauds. Legal discipline will always differ from one country to another, as each Member State is free to choose how to deal with the criminalisation obligations imposed on it to protect the EU's financial interests. Fraud patterns, however, tend to display recurrent elements.²⁹ In practice, this means that they tend to be predictable to a certain extent, which makes it possible to formulate common risk assessment criteria. Consequently, a key point in the development of effective protection of the EU's financial interests is to improve the inductive methodology used to analyse the risks of CAP fraud.³⁰ Practically, this means, for example, focusing on recurrent elements of frauds as a starting point for the development of common guidelines for fraud detection.

A second key aspect emerged is the improvement of the information exchange activity and, more generally, the use of IT tools. As our study demonstrates, the early detection of fraud depends to a large extent on the quality of the information exchange systems adopted at the national and supranational levels as well as on the timeliness with which information-exchange is implemented. Depending on the Member State, digital strategies have already proven effective domestically, especially in the case of direct payments. As illustrated above (section II.1), paying agencies use IT tools to quickly check applications for CAP funding. In addition, the use of such tools enables agencies to exchange data easily with other administrations and public entities, allowing for smooth cross-checking. At the cross-border level, however, much remains to be done. A starting point might be to increase the use of ARACHNE, a risk scoring and data mining/enrichment tool developed by the European Commission,³¹ and to simultaneously make it more efficient and effective. Its universal use could prove decisive for the EU-wide effective prevention end: early detection of fraud. Indeed, when several countries are involved, it is crucial to rely on a data mining tool to identify red flags when processing data from more than one EU Member State.³²

At the time being, many Member States already use ARACHNE.³³ However, it is still perceived as the least effective detection tool, especially when compared to other approaches, such as on-the-spot checks and audits, internal fraud reporting mechanisms, and fraud risk assessments of applicants and/or beneficiaries. This perception is largely corroborated by the fact that managing authorities face difficulties in collecting data (excessive administrative burden, also related to the multiplication of IT systems), accuracy issues (high number of false positives), and legal barriers (for instance, national data protection laws).³⁴ In addition, data interoperability among ARACHNE, the Irregularity Management System (IMS), and EDES (Early Detection and Exclusion System)³⁵ as well as OLAF’s and other national databases should be further developed.

To properly address these points, the introduction of a distinct EU regulation in this field seems necessary. Only a broader application of ARACHNE and a consistent increase in the available data can ensure the system’s proper functioning, in turn reducing the shortcomings in the accuracy of the results. This would require a specific legal duty to make the use of ARACHNE compulsoryand clear, binding rules on data interoperability among EU and national databases.³⁶ Moreover, such a regulation should also provide for the extension of the use of EDES to the area of shared management funds, as this would greatly contribute to the early exclusion of unreliable entities from accessing EU funds.³⁷ Finally, well-defined rules would also be essential to ensuring full compliance with criminal procedural guarantees and with principles governing the use of artificial intelligence.³⁸