DEPARTMENT OF POLITICAL SCIENCE DATA GOVERNANCE AND THE TWIN TRANSITION A qualitative analysis on the perception of the Data Governance Act by public actors in Italy Author: Marius Stoian Master’s Thesis: 30 credits Master’s Programme in International Administration and Global Programme: Governance (IAGG) Date: 21.05.2024 Supervisor: Olof Larsson Words: 16632 Abstract Harnessing the power of data is instrumental in developing innovative solutions to imminent environmental challenges. The European Union is making an unprecedented effort in advancing the digital and green - twin - transitions. New data governance regulatory frameworks, designed to maximise data’s full cross-sectoral potential, constitute the key puzzle piece of the EU’s digital agenda. This study has found there is a research gap when exploring the role of data within the green and digital nexus. While previous literature on European policy implementation has focused primarily on EU directives, this research contributes to the emerging focus on regulations. This thesis investigates the perceptions and responses of public actors towards the implementation of the Data Governance Act (DGA) Regulation within the wider context of the EU’s twin transition. Through a qualitative single-case study on Italy, the findings indicate that the DGA is positively received by the main public actors involved in data governance policy in Italy. Despite significant challenges, this thesis concludes that existing strategic national-level initiatives on data governance align well with the DGA, positioning the country for successful implementation. Keywords: Data Governance Act, EU regulation, DGA, policy implementation, digitalisation, twin transition, sustainability, Italy, European Union, green transition, digital transition, open data, public actors. 1 TABLE OF CONTENTS: Abstract ...................................................................................................................................... 1 Acknowledgements .................................................................................................................... 3 List of Tables ............................................................................................................................. 4 List of Abbreviations ................................................................................................................. 5 1. Introduction ........................................................................................................................ 6 2. Literature Review............................................................................................................. 10 2.1 Perspectives on the Twin Transition ................................................................... 10 2.2 Previous research on the Data Governance Act .................................................. 14 The European Data Strategy .................................................................................. 14 The Data Governance Act ..................................................................................... 16 3. Theoretical Framework .................................................................................................... 20 3.1 Differentiated Policy Implementation ................................................................. 21 3.2 Multi-level governance ....................................................................................... 25 4. Methodology .................................................................................................................... 27 4.1 Research problem, research question and its purpose ......................................... 27 Research problem .................................................................................................. 27 Research question .................................................................................................. 28 Purpose of study and research gaps ....................................................................... 29 4.2 Research design ................................................................................................... 30 Data collection ....................................................................................................... 32 Data analysis .......................................................................................................... 34 4.3 Ethical considerations and privacy ...................................................................... 35 5. Analytical Strategy........................................................................................................... 37 5.1 Analytical Framework ......................................................................................... 37 5.2 Sources of evidence ............................................................................................. 38 6. Results and Analysis ........................................................................................................ 42 6.1 Perceptions and Challenges ................................................................................. 42 Re-use of publicly held data .................................................................................. 43 Data intermediaries and data altruism organisations ............................................. 45 6.2 Institutional Fit & Administrative Behaviour ..................................................... 47 6.3. Alignment with National Strategies ................................................................... 50 7. Conclusion ....................................................................................................................... 52 References ................................................................................................................................ 55 Appendix .................................................................................................................................. 65 2 Acknowledgements I would like to thank my mother and father for their immense support throughout these years. I would not be where I am now without them. I remain grateful to the friends I have made in the last two years here in Gothenburg. The moments we’ve had together are special, and they’ll stay with me forever. Lastly, I would like to thank my supervisor for their guidance, and all the professors who have given me precious support during this academic journey at GU. 3 List of Tables Table 1: Analytical Framework Table 2: List of official documents 4 List of Abbreviations AgID - Agenzia per l’Italia Digitale (Agency for Digital Italy) AI - Artificial Intelligence DGA - Data Governance Act DI - Differentiated Integration DIS - Data Intermediation Service DMA - Digital Markets Act DPA – Data Protection Authority DPI - Differentiated Policy Implementation DSA - Digital Services Act DTD - Department for Digital Transformation EC - European Commission EGD - European Green Deal GDPR - General Data Protection Regulation GHG - Greenhouse Gas GPDP - Garante per la Protezione dei Dati Personali (Italian Data Protection Authority) IoT - Internet of Things MLG - Multi-Level Governance 5 1. Introduction “This generation wants to move fast – and they are right. Because in front of us are the major twin ecological and digital transitions. They will affect us all, wherever we live, whatever we do.” The words belong to the European Commission President, Ursula von der Leyen (2020), and were published on the 7th of March 2020 in a widely circulated op-ed aimed at marking and assessing the Commission’s first 100 days in office. Before delving into what the “twin ecological and digital transitions'' entails, the aim of the thesis, and the research gap I intend to address, it is appropriate to first provide some context around the EU’s policy efforts led by the Commission. Originally published in January 2020, the beginning of what was to become an extremely consequential year for the world, the European Commission’s annual Work Programme outlined the EC’s main initiatives for the next five years. At the top of the Commission’s agenda there are two ambitious priorities which have completely shaped recent policy making in the European Union: the "European Green Deal" and "A Europe fit for the digital age." (European Commission, 2020). The European Green Deal (EGD) aims to guide the EU's efforts in the policy area of environmental sustainability. It is the EU’s response to greenhouse gas (GHG) emissions, global warming and increase in natural disasters. However, the European Green Deal is not focused on the sole objective of pushing towards climate neutrality by 2050, and therefore enable the transition towards a sustainable society. It is also, as declared by the von der Leyen mandate, the EU’s “new growth strategy” (European Commission, 2019). This implies a dual purpose for the EGD: tackling the climate crisis, but also revamping European industry, pushing for innovation, and boosting EU competitiveness on the global level. The only way the European Green Deal can aspire to succeed at both climate neutrality and achieving the EU's new economic growth strategy is through its complementarity with a strong digital innovation strategy (European Commission, 2019). This is where the EU initiative “A Europe fit for the digital age” comes into play: with the goal of achieving synergies across both technology and environment domains, it aims to promote digitalisation and enable the digital transition to reach its full potential; its objectives encompass a wide range 6 of policy regulation areas, from Artificial Intelligence and digital markets to data governance and cybersecurity (European Commission, 2020). The two policy programs - the EGD and "A Europe fit for the digital age” - which were briefly contextualised above, constitute the unprecedented European Union’s policy efforts to advance a twin - green and digital – transition geared towards making our society more sustainable and just (Bianchini et al., 2023). Twin transition is an increasingly popular term in current political discourse that refers to possible synergies between the digital and green transitions; its focus lays on the role of new digital technologies in mitigating environmental damage and fighting climate change (Diodato et al, 2023). Although unveiled more than four years ago, it is only recently that those two strategic initiatives have started to significantly converge, as multiple efforts are being made to concretely move towards a “twin transition”. One of the more substantial actions made by the EU towards successfully achieving the twin transition is the adoption of the Data Governance Act (DGA) in 2022, which entered into force as the first legislative instrument of the European Data Strategy (von Ditfurth and Lienbemann, 2022). The European DGA, presented within a legal framework aimed at addressing environmental challenges by promoting data sharing, is arguably among the most exemplary EU actions within the broader domain of the twin transitions (Finck & Mueller, 2023). Moreover, as a key piece of the European strategy for data, the EU regulation is a policy highly representative of the twin transition: data itself is deemed an essential factor in bridging digitalisation and the quest for environmental sustainability (Jirka et al., 2022). However, the effectiveness of the DGA will depend on its implementation and on how public actors at the national level comply and adjust their behaviour to the new mechanisms introduced by the EU regulation. It’s important to note that, even though the DGA is an EU regulation and thus directly binding - unlike an EU directive - there will be substantial differences in how the Act is realised in practice at the member state level. The research project presented in this thesis employs a single case study design, utilising qualitative methods such as analysis of official documents (i.e., institutional reports, guidelines etc.) and semi-structured interviews with legal, technical, and policy experts to analyse the perception of the DGA implementation in Italy within the context of EU’s twin transition agenda. The interviews will be carried out with professionals from the public sector 7 involved with digitalisation and data governance policies in Italy, while the documents will be selected based on their relevancy in the context of the research design. The theoretical framework through which the collected data will be analysed draws from elements of policy implementation literature and multi-level governance (MLG) theories (Bondarouk & Mastenbroek, 2018; Eriksen, 2023; Hooghe et al., 2020; Thomann & Sager, 2017; Zhelyazkova et. al, 2024). Considering the potential implementation complexities of the DGA and its need for coordination across multiple governance levels, this theoretical framework is particularly relevant. More specifically, this thesis builds on previous research on Differentiated Policy Implementation (DPI) in the EU, as well as key elements from the field of MLG, which carry the potential to add theoretical depth to the analysis of the Italian case study. DPI provides insight into the diversity between EU member states in the implementation of EU policies (Zhelyazkova et al., 2024), while MLG helps contextualise how actors at different governance levels perceive the implementation process and interact within it (Hooghe & Marks, 2001). Considering that most academic research on policy implementation focuses on legal implementation and on whether member states comply, this study intends to contribute to more recent literature which is more concerned with practical implementation and how member states adapt or use discretion in fine-tuning EU policy (Zhelyazkova et. al, 2024). The aim of this thesis is to explore Data Governance Act policy implementation aspects within the context of the “twin” – green and digital – transition in the EU. In other words, the purpose of the research is to investigate and analyse the perception of the DGA of a variety of stakeholders, namely public actors operating in Italy. More specifically, it aims to carry out a single-case study, analysing the policy outputs - intended as response to legislation – of a select EU member state, that is Italy. An important conceptual distinction is useful here for understanding the structure of the thesis: the term policy outputs refers to actions adopted to respond to a specific law, and not to its outcomes or impact (Bondarouk & Mastenbroek, 2018). This methodological decision relates to the fact that it is rather premature to examine whether the DGA produces solutions to tackle the initial policy problems. The reason behind the selection of the case study country has been guided by previous research in the field and accessibility to data sources. In addition to feasibility factors, the unique pace of digitalisation and innovation policies in Italy has been a primary criterion in the case selection process. Moreover, Italy's complicated legal and institutional structure offers an 8 interesting case for analysing how exactly EU regulations are received in practical terms within the national landscape (Bibbee, 2007). The case selection is further justified by taking into account recent efforts carried by Italy in the field of data governance and digitalisation. Additionally, Italy's experience, considering its position within the EU, has the potential to provide insights which are relevant beyond the national context (EC, 2023a). By focusing on the perception and response of the DGA by Italian public bodies, this study seeks to identify and analyse the challenges presented by the EU Regulation in the wider context of the twin transition in the EU. Thus, the research question leading this thesis is formulated as follows: “How do public actors within the digitalisation field in Italy perceive and respond to the implementation of the Data Governance Act within the context of the 'twin' green and digital transition in the European Union?” The analysis conducted in this study finds that the DGA is being positively received by the main public actors involved in data governance policy in Italy. In addition, public bodies perceive the DGA as an appropriate instrument for enabling data re-use and improving data sharing. However, several challenges emerged in connection with the DGA implementation. They include the complex regulatory nature of Italian law; the overlap between data protection and data sharing legal frameworks; the uncertainty on how new mechanism such as data intermediation services will work in practice. Despite the outlined challenges, this thesis finds that strategic national-level initiatives, such as the National Interoperability Framework, align strongly with the DGA, making Italy ready to proceed with practical implementation. This thesis will continue as follows. Chapter 2 will provide a comprehensive review of previous literature on the twin transition and the Data Governance Act, contextualised within the broader European Data Strategy. Following that, the Theoretical Framework chapter will describe Differentiated Policy Implementation (DPI) and Multi-Level Governance (MLG) theories. Chapter 4 will discuss Methodology, detailing the qualitative research design employed by the thesis. The Analytical Strategy chapter will discuss operationalisation factors and explain how the data will be interpreted. Chapter 6 will present key findings, exploring the perceptions of public actors regarding the DGA. Finally, the conclusion will summarize the main findings, discuss their implications, and suggest areas for future research. 9 2. Literature Review This section will be structured as a two-part comprehensive review of previous literature. The first part will explore current perspectives on the relation between the green and digital transitions and the challenges posed by the phenomenon that is the twin transition. Furthermore, it will present relevant findings from previous research, highlighting the need for redirecting the research agenda towards concrete policy efforts while adopting a balanced approach to the issue. The second part of this review will focus on the Data Governance Act, which constitutes a crucial pillar of the European Data Strategy. It will explore the DGA’s role in incentivising data sharing and the reuse of data held by private and public actors, while at the same time will address some of its limitations in the context of the EU’s legal framework. Nonetheless, this review of previous research will outline the great potential held by the DGA, particularly when it comes to innovating through sustainable solutions in the context of the wider twin transition phenomenon. 2.1 Perspectives on the Twin Transition Before delving deeper into the complexities entailed by the twin transition, it is useful to define below its key concepts and underlying processes. The green transition represents a series of policy actions aimed at reversing environmental degradation, substantially reducing greenhouse gas emissions, promoting renewable resources of energy, as well as protecting biodiversity and preserving nature. The digital transition of society is about the entirety of computer power dependent processes which characterise society at every level; in other words, digital transition refers to how digital technologies are changing the ways we live, work, and interact; this includes transformation to our infrastructure, applications, services and interactive human behaviour. The term "twin 10 transition" refers to the question of how the digital transformation can contribute to the green transition, and how these two transitions can catalyse or reinforce one another (Dæhlen, 2023). Air quality management in London represents a practical example to better understand the twin transition: using data of air pollution measurements collected via dedicated sensors, an innovative project is employing machine learning algorithms and data science platforms with the goal to forecast and prevent air pollution across the city of London (Greater London Authority, 2023). Thus, the twin transition represents a broad and incredibly complex phenomenon. Its complexity and wide-ranging implications derive from the intricate relation between the green and digital transitions. Nonetheless, the twin transition can be defined as a policy approach aimed at coupling climate action to digitalisation efforts, with the goal to establish a synergic green-digital nexus (World Bank, 2023). The World Bank clarifies the twin transition’s ultimate purpose as “ensuring inclusive and sustainable digital foundations and applications are in place to accelerate mitigation and adaptation efforts [to climate change] while reaping wider benefits of digital transformation for growth, job creation, and better lives.” (p. 8). Concurrently, in their review, Rosário and Dias (2022) describe the digital tools’ application to environmental sustainability efforts as the strategic use of technologies such as artificial intelligence (AI), big data analytics and Internet of Things (IoT) to implement solutions focused on sustainable urban development, pollution control, and waste management. The uncertain nature of the synergies between the two phenomena has ignited an active academic debate across the innovation and sustainability areas of study (Bianchini et al., 2023; Gigler, 2020; Meijer, 2024; Muench et al. 2022). Moreover, the potential synergies within the twin transition depend on suitable data regulation frameworks. Data is at the core of the application of the digital technologies mentioned above, and that is why this thesis will focus on the EU Data Governance Act. As the European Union pushes through promoting the twin green and digital transition, with the stated objective of generating an increasingly prosperous and sustainable society, crucial issues are still being debated. There is still lack of certainty regarding whether the green and digital transitions are compatible with one another, or if one transition will hinder the other (Bianchini et al., 2023). Moreover, there are still doubts on how exactly the two transitions can manage to reinforce each other (Muench et al., 2022). The dominant line of thought, which can 11 be labelled as the optimistic perspective, maintains that global warming and environmental pollution are fit to be addressed through technological progress. Digital innovation, according to this hopeful view, will transform our society in a positive way, bringing us sustainable agriculture, green transportation, and clean energy, among other things (Meijer, 2024). In other words, this perspective argues that the fight against climate change can be sustained through the adaptation of innovative technologies, particularly digital ones, to environmental challenges in a multitude of sectors, from housing to manufacturing (Gigler, 2020). Examples of these include the development of digital environmental monitoring systems, or data-driven materials optimisation processes (Muench et. al, 2022). Recent literature has however challenged this mainstream hopeful line of thought and its excessive concentration on the successes of digital technologies (Bianchini et al., 2023; Fouquet & Hippe, 2022; Diodato et al., 2023; Meijer, 2024). This more balanced perspective highlights the environmental challenges posed by technological innovation and the digital transition, emphasising that new technologies can also carry negative effects and added costs for both our planet and its most vulnerable groups of inhabitants (Meijer, 2024). Diodato et al. (2023) find that there has been insufficient academic research exploring the interlinks between the green and digital transitions. It is only recently that scholars have started to analyse the potential outcomes of the twin transition, particularly when it comes to “scratching the surface of the dark side of innovation” (p. 756). Meijer (2024) characterises this “dark side” by presenting three particularly critical arguments for how the logic of technological innovation can be viewed as a root cause of environmental destruction. These arguments include perspectives derived from a variety of research fields, including environmental science, economics as well as philosophy. According to these perspectives, technology can be deemed as problematic due to i) its emphasis on continuous unsustainable economic growth ii) its short term focus and disregard for long term consequences, and iii) its perception as the primary solution for environmental challenges and as the primary agent within societal progress (Meijer, 2024). The pioneering research study presented in a paper by Bianchini et. al (2023) makes a valuable contribution to this perspective, countering the optimistic faith placed in digitalisation and tech innovation. By performing an econometric analysis on an original dataset combining information on environmental patents and pollution, the authors find that innovation and 12 technological development across Europe is occurring at an uneven pace. While green tech appears more evenly spread across member states, the development of digital technologies is unbalanced across many regions in Europe, particularly in the periphery and Eastern Europe. It is important to note that this study comes with some limitations as it only examines highly polluting plants, therefore it overlooks emissions produced by less intensive plants. Nevertheless, the research study reveals that while local green tech development reduces greenhouse gas emissions, the local development of digital technologies seems to negatively affect the environment. However, it is important to note that not all components of digital technology are equally detrimental to the environment; the most-energy intensive elements, for example data centres, are deemed to be the most harmful. Bianchini et. al (2023) conclude by establishing that the digital transition poses significant challenges for the environment, and by pointing out the need for future twin transition policies to be tailored on the technological capacities’ differences across Europe. Similar conclusions are presented in a study by Fouquet & Hippe (2022): by exploring the twin transition phenomenon from a broader historical perspective, they argue that the green and digital transitions are likely to progress at an imbalanced rate. The study analyses long run trends in communication intensity and energy intensity, starting from 1850, within a selection of key European economies. The authors conclude that, without policy intervention, the digitalisation process will continue ahead without decarbonising. This evidence reinforces the conclusions made by Bianchini et al.: twin transition policies must be designed with the goal of tackling regional disparities and actively boost development of green technologies (Fouquet & Hippe, 2022). Although the optimistic view makes clear the immense potential for innovative technologies to tackle environmental issues, the more prudent perspective underscores the risks of detrimental consequences of unchecked innovation and unsustainable growth practices. This implies that the success of the twin transition is dependent on a cautious balance between technological advancement and proactive initiatives to address potential environmental damages and social costs. Looking forward, researchers and policymakers should consider this complexity and push for solutions that keep digitalisation processes in check while anticipating potential downsides. 13 The ongoing debate in current and previous literature highlights the need for fresh balanced perspectives on the twin transition. This need becomes very clear especially when considering the degree of uncertainty surrounding the green and digital transitions. Nonetheless, uncertainty must not lead to policy paralysis nor overshadow the great potential carried within the twin transition. Previous research has shown that so far, and for the most part, the digital transition has progressed with little focus on sustainability. However, more recent literature highlights the key role of green technologies in successful digital transformation. While it appears that the topic of the twin transition is gaining traction in academia, research in this area remains somewhat fragmented (Burinskienė & Nalivaikė, 2024). There is a need for redirecting the research agenda on empirical studies exploring the reception and implementation of policy representative of the green and digital transitions. The European data governance regulatory framework is an ideal case of such policy making, and that is why it is placed at the forefront of this thesis. In other words, considering the leading role of policy making in this field, it is important to look at how specific regulations can effectively further the implementation of the green and digital transitions. Future research should investigate how policy can manage potential trade-offs between digital and green goals, while at the same time exploring ways in which policy can stimulate innovation and adaptation within businesses (Diodato et. al, 2023). To conclude, this thesis project seeks to contribute to a nuanced perspective on the twin transition by analysing how EU data policy, more specifically the Data Governance Act (DGA), could influence its overall trajectory and secure the realisation of its potential. This could be obtained by analysing potential benefits and challenges arising from the implementation of the DGA, as perceived by public actors at the national level in Italy. 2.2 Previous research on the Data Governance Act The European Data Strategy Accessible, interoperable, and high-quality data is instrumental in tackling environmental challenges through innovative solutions. That is the reason why data regulatory frameworks 14 represent a key prerequisite for the green and digital transition. For instance, new tech such as digital data-driven monitoring systems can generate valuable environmental data which can be used to prevent biodiversity loss and protect our climate. The adoption of the Data Governance Act, as anticipated in the introductory chapter, is one of the most significant steps made by the EU towards opening data and utilise it to achieve a successful twin transition. Before reviewing previous research on the DGA, and its key concepts and mechanisms, it is useful to briefly contextualise it within the broader European Data Strategy, which was launched in 2020 at the start of the von der Leyen mandate (European Commission, 2020). Data, and more specifically data protection, has been the EU’s centre of attention for a long time. This focus reached its pinnacle in 2016, with the enactment of the General Data Protection Regulation (GDPR). Nevertheless, more recent years have witnessed a change of direction from data protection to data sharing, and the promotion of data economy (Ruohonen & Micklesson, 2023). This is embodied in the European Data Strategy, which was launched in February of 2020 through an official Communication from the European Commission (EC). Its primary objective is to boost data-driven innovation, by building “on a strong legal framework – in terms of data protection, fundamental rights, safety and cybersecurity – and its internal market [characterised by a] varied industrial base.” (European Commission, 2020, p. 1). In other words, the strategy was designed to find new mechanisms to manage data, as well as to extract value from it. As part of this strategy, the EC has pushed for several legislative projects with the goal of improving data sharing within Member States and across various sectors of the economy (Portnoy & Nauwelaerts, 2024). This turning point marks the beginning of a modern and ambitious, post-GDPR law- making process designed to directly shape the data-driven digital economy (De Hert, 2023). The progressive evolution of EU data regulations, from the General Data Protection Regulation (GDPR) all the way to the Digital Markets Act (DMA), Data Governance Act (DGA), Digital Services Act (DSA), Data Act, and the AI Act, demonstrates a significant expansion of regulated data, accountable entities, and data management practices. This evolution represents the European Data Strategy's ambitions to secure a single data market and position the EU as a frontrunner on the global scene (Pathak, 2024; Del Castillo, 2021). 15 Through a comprehensive and thorough thematic analysis of the EU data strategy, Carvalho and Kazim (2021) explore various aspects for further academic consideration; these include disambiguating data, ethical implications regarding (cyber and physical) data infrastructures, data integration with emerging technologies such as AI, IoT and blockchain etc. Going in-depth into each one of these considerations risks falling outside the scope of this thesis; however, it is worth reporting the authors’ final point on why the EU should not be too confident in believing that its data policy and sharing strategies will become a global standard, as happened with the GDPR: the latter was much more focused on universal protection (e.g. introduction of Privacy by Design principles, such as pseudonymization; see Teixeira et. al, 2019) while, on the other hand, data-sharing envisions data as a valuable resource, making countries less likely to adapt to shared regulatory frameworks (Carvalho and Kazim, 2021). Amidst much of the debate surrounding EU data policies, there is however one component which, according to De Hert (2023), “went by a bit under the radar” and “might be the single most important piece of the new set of law when it comes to regulating the data driven society and in particular the 'Big Data-aspects': the Data Governance Act (DGA) proposal.” (p. 12). The next section of this chapter will delve more into the regulatory framework established by the DGA, focusing on its goal to consolidate common data spaces and encourage free flow of data. The Data Governance Act Having established that the Data Governance Act (DGA) is a key pillar of a broader regulatory framework put in place by the EC to advance its main strategic priorities, it is time to go deeper into previous literature on this regulation. The European Commission adopted the Data Governance Act (DGA), solidifying its strategic shift towards data use and sharing. Effective as of September 24, 2023, the DGA is the first new law in the context of the European Data Strategy. It is important to note here the difference between EU regulations and directives: the first is directly applicable and binding in EU member states, while directives are binding for what concerns the policy goals it sets, but not for the form and methods of application. Consequently, the DGA regulation - although 16 directly binding - still needs to be implemented by member states and entails a degree of flexibility at the national level (Blom-Hansen et al., 2023). Therefore, as the DGA establishes new concepts, legal definitions, and enforcement mechanisms for the reuse of data held by public and private actors, member states will still need to create, for example, the infrastructure for those mechanisms to effectively function. The Data Governance Act further outlines provisions encouraging voluntary data sharing by both individuals and organisations, instituting a regulatory framework for data- sharing intermediaries. Ultimately, the DGA aims to promote a new data governance model that allows all stakeholders to innovate through the re-use of data (Portnoy & Nauwelaerts, 2024). In other words, as outlined in its original text, this new legal instrument seeks to bolster data-sharing mechanisms within the EU by fostering data availability and increase public trust in data intermediaries (Regulation 2022/868). This suggests, as pointed out by De Hert (2023), that there is a certain degree of interaction between the DGA and other legislation concerning data use, namely the GDPR. This interaction has been a source of criticism in previous literature (De Hert, 2023; Ruohonen & Micklesson, 2023), primarily due to two challenges: i) discrepant terminology and definitions, particularly when it comes to concepts such as data, data user, and data holder; and ii) the DGA’s broader scope compared to the GDPR. These factors raise potential conflicts in compliance requirements and enforcement of both regulations (De Hert, 2023). The concept of data intermediation service (DIS), as outlined in the DGA text, deserves special consideration. It is defined as “a service which aims to establish commercial relationships for the purposes of data sharing between an undetermined number of data subjects and data holders on the one hand and data users on the other” (Regulation 2022/868). The main objective of data intermediaries is to facilitate data exchange through data infrastructures, such as databases and other platforms, by adopting standard formats and protocols which ensure data interoperability and security standards. Article 11 of the DGA clarifies that the institution of DIS providers needs to be formally registered with a competent public authority. Furthermore, data intermediation services must be supervised by relevant authorities designated by member states; similarly to the GDPR, the new national authorities have the faculty to fine or terminate data intermediation services in response to violations (Ruohonen & Micklesson, 2023). DIS providers, or data intermediaries, have emerged in recent 17 years as a new category of third-party services that facilitate data sharing between consumers and businesses (C2B) and between businesses themselves (B2B). These third-party services play various roles and hold a variety of different purposes; examples of these include data pools, orchestrators of data-sharing ecosystems, for example relating to European common data spaces, and data marketplace through which businesses can make data available to other subjects. (Picht & Richter, 2022; von Ditfurth & Lienemann, 2022). The other important element of the DGA is data altruism. This concept refers to the practice of sharing data, in all its forms, on a voluntary basis through the consent of the parties involved, namely data holders and data subjects; it works on a no compensation logic, and for the purpose of the collective good. Organisations that want to participate in data altruism can adhere voluntarily for many reasons, including to increase trust in their operations (Piachaud- Moustakis, 2022). One last point from the DGA deserving attention is the European Data Innovation Board, which will be tasked to facilitate coordination at the EU level. It is designed to carry its work through collaboration with the competent national authorities, data protection and cyber security institutions, as well as civil society groups among other actors. Moreover, the Board will deal with interoperability matters, devising industry standards and code of conducts on data sharing practices when necessary (Richter, 2023; Ruohonen & Mickelson, 2023; von Ditfurth & Lienemann, 2022). One flaw regarding the European Data Innovation Board relates to its ambiguity. Defined as nothing more than an expert advisor group informing the Commission on data sharing practices, a previous research article by De Hert (2023) argues that its label can be way too easily confused with the European Data Protection Board instituted by the GDPR. The issue of ambiguity extends to other elements within the DGA text. Ruohonen & Mickelson (2023) argue that the regulation is vague on the issue of data altruism, therefore it is uncertain how data altruism organisations will work out in practice. Moreover, the use of consent as the legal basis for data sharing, particularly personal data, remains problematic in the context of data altruism practices; it is often the case that users of data applications fail to understand exactly what they are consenting too. De Hert (2023) shares similar concerns, claiming that “the DGA fails to clarify how data altruism should work in practice in the light of these multiple consent formats in the GDPR” (p. 15). Another critical aspect of the DGA relates to the roles of data intermediaries. In an article published in 2022, Von Ditfurth & 18 Lienemann contribute to the debate by arguing on whether the regulation will restrict or promote data intermediaries. The authors conclude by expressing scepticism regarding the role of the DGA in promoting data intermediaries in the context of the emerging European single market for data. They justify their judgement by pointing out the notable risks (e.g. stifling innovation and competition) carried by policy interventions in rapidly shifting environments such as novel data markets. One last limitation within the DGA is particularly relevant in the context of this thesis: Finck and Mueller (2023) conclude that “the new data access and sharing regimes developed under the DGA [...] do not make special provisions for data that could be used for environmental purposes.” (p. 121). This critique suggests that the green-digital nexus stops when considering substantive norms; in the authors’ view, the overlap between data governance and environmental sustainability mainly occurs in terms of discourse. Despite Finck and Mueller's critique, it can be argued that the DGA's focus on data reuse, data-sharing, and data altruism will still significantly contribute to the promotion of environmental sustainability solutions through wider use of environmental data. This is evidenced by the numerous European initiatives in this policy area, such as the Green Deal Data Space; the INSPIRE Directive; the GreenData4All initiative; the Horizon Europe Strategic Plan 2025- 2027 (EC Website, 2024). 19 3. Theoretical Framework Data is at the core of a growing discussion on how to bridge digitalisation and environmental protection. Access to varied, interoperable and prime quality data is highlighted by the European Green Deal as having a major role in empowering climate action and addressing environmental sustainability challenges through policy informed by evidence-based decisions. To facilitate this, the establishment of a Common EU-level Green Deal Data Space, as part of the Digital Europe Programme (DIGITAL), aims to boost data-sharing and make high-quality data available in support of European Green Deal priority actions such as biodiversity zero pollution, and deforestation (Finck and Mueller, 2023). The Data Governance Act is instrumental in achieving these objectives, and that is why investigating its perception and implementation aspects among its member states is paramount. The actions taken in response to the adoption of the DGA must be explored as they can offer deeper insights into the larger twin transition implementation process. The theoretical framework through which the analysis of the collected data will be carried out in this study draws from academic literature on policy implementation and multi- level governance (MLG) theories. More specifically, this thesis builds on previous research and theories in Differentiated Policy Implementation (DPI), as well as relevant elements from the field of MLG which has potential to contextualise the single case study of Italy and add further theoretical depth to it. Large part of academic research on policy implementation focuses on legal implementation and on whether member states comply; this thesis intends to contribute to a more recent line in literature which is increasingly concerned with practical implementation and how member states adapt to new regulatory landscapes or use discretion in fine-tuning EU policy (Zhelyazkova et. al, 2024). This chapter will be structured in two parts: the first will discuss DPI, while the second section will provide relevant insights on MLG. 20 3.1 Differentiated Policy Implementation Differentiated Policy Implementation (DPI) can be defined as the existence of variations in how member states exercise discretion during practical and legal implementation in the EU, in terms of both processes and outcome. In other words, DPI seems to refer to a rather evident statement: EU law is not implemented in a uniform way across member states. Despite its apparent simplicity, DPI provides a valuable framework for studying implementation of EU policy through its central idea of differentiation interlinked with variation patterns (Zhelyazkova et. al, 2024). From a contextual perspective, the DPI research agenda has been revitalised in response to its connection with research on Differentiated Integration (DI) in the EU. DI literature conceptualises differentiation as an instrumental strategy to overcome the growing diversity of member states' capacities and integration preferences. Differentiated integration represents a pragmatic response to inherent challenges posed by the European Union’s multileveled structure. In other words, by allowing distinct policy areas and member states to be exempt from centralized rule, DI acts as a de facto stability mechanism (Eriksen, 2023; Zhelyazkova et. al, 2024). It is important to note the distinction between DPI and DI, as both instances of differentiation can develop in a variety of combinations: for example, EU-level legislation applicable to all member states (therefore, not providing for DI) could offer a lot of room (or little) for implementation differences; similarly, EU-level legal frameworks which offer opt- outs (therefore, providing for DI), could give little room (or a lot of it) for DPI within member states (Zhelyazkova et. al, 2024). Traditional EU implementation research puts a very strong focus on the issue of legal conformance with European Union rules. This approach has been however criticised for its shortcomings when it comes to fully grasping the implications of the EU’s multilevel governance system, and for not providing a full picture of implementation in the EU, particularly in relation to its range in diversity (Thomann & Sager, 2017; Zhelyazkova, Kaya & Schrama, 2016). Thomann and Sager (2017) distinguish between two types of implementations: 21 1. Conformance implementation: it refers to how the central EU-level policy is implemented at member states’ level. The primary aim of this top-down approach is to measure implementation success by how well policy outcomes match the original blueprint; any hint of deviation from the main plan is viewed as problematic. 2. Performance implementation: it refers to whether a certain policy produces solutions to resolve the original policy problem it was meant to address. This bottom-up perspective is primarily focused on the role of those who are closest to the policy issues, therefore on policy implementers as problem solvers; this approach is process-oriented, and it expects “that policy instruments and goals may undergo context-sensitive modifications during the process of policy implementation.” (p. 1254) Regarding performance implementation, there is one conceptual distinction that must be made in order to understand better the theoretical approach of this thesis. Previous literature outlines two different interpretations of performance: i) performance measured in terms of policy outcomes, by which is meant whether a specific policy produces solutions to tackle the initial policy problem; ii) performance in terms of policy outputs, referring to actions adopted to respond to a specific law (Bondarouk & Mastenbroek, 2018). The research question at the core of this study aims to look at Italy’s public actors’ perception and response to the implementation of the DGA, therefore it strives to understand policy outputs, as defined above. The reason behind this research decision relates to the fact that it is still too early to even attempt to measure the outcomes of the Data Governance Act; more on this will be discussed in the following chapter dedicated to methodology. Nevertheless, as multilevel implementation studies retain a solid focus on conformance implementation, analysing implementation beyond legal compliance carries substantial benefits and is increasingly relevant. This is particularly pertinent in a context of rapid technological innovation where policy problems are progressively becoming cross-sectoral and transnational (Thomann & Sager, 2017). 22 DPI represents such efforts in pushing the research agenda’s main focus from the issue of whether EU member states comply or not with EU rules, to the question of how exactly the adjustment, interpretation and fine-tuning occur when EU policies are being implemented. Moreover, most of the implementation literature has been focused on the process of EU directives’ transposition into national law. Part of the reason why this is the case has to do with the fact that legal implementation is methodologically easier to assess, particularly when it comes to measuring how quick the measures outlined are being transposed. Recent research on DPI widens its focus by addressing EU regulations, in addition to directives. This point is rather relevant, as regulations are often vague, and therefore require further specification at the national level (Zhelyazkova et. al, 2024). As previously outlined in the previous section, the Data Governance Act is a prime example of how EU regulations can be inherently ambiguous, and therefore needing further interpretation within the implementation process. The concept of DPI serves as an umbrella term containing various ways of describing how member states adapt EU law to their own contexts. These include terms like ‘customisation’, which highlights the liberties of member states when transposing EU rules, and 'domestication’ which emphasises adapting EU policies into existing national structures (Thomann, 2019 and Bugdahn, 2005, as cited in Zhelyazkova et. al, 2024). According to Zhelyazkova et. al (2024), five main drivers of DPI can be identified: - EU policy attributes: DPI is directly influenced by the degree of flexibility, coherence and clarity within EU policies. - Domestic institutional fit: member states align EU policies with their domestic preferences, legal systems, and styles of problem-solving. - Domestic politics: inner political demands such as populism, national interest groups, and bureaucratic styles are factors leading to differentiated implementation. - Functional needs: member states with shared policy interests can adopt similar implementation methods, sometimes exceeding EU's minimal requirements. 23 - Administrative capacity: National-level resources, and capabilities, contribute do practical implementation. Eriksen (2023), through his research on administrative behaviour, differentiated policy implementation and the issues of legal certainty, adds another layer to the framework provided by the DPI. The concept of administrative behaviour refers to the interaction between implementation practices and institutional context: while public administrative bodies are required to comply with the law, they are also provided with some degree of discretion, through interpretation and operationalisation of legal provisions. With the aim to analyse policy implementation in a multilevel context, Eriksen (2023) identifies three modes of administrative behaviour: - Instrumental mode: it refers to the technical application of EU law with an emphasis on procedural correctness. This mode operates in stable administrative environments. - Advocate mode: it emerges under politicised and conflictual conditions, where public bodies aim achieving desired policy outcomes and may interpret and apply rules selectively. - Conciliatory mode: it operates in a complex environment, where the policy goals and rules are numerous. This mode entails active and collaborative interpretation of the law, minding both the national and the EU level. These analytical categories have the potential to prove useful when investigating how the DGA implementation is being received by public actors in Italy. Overall the DPI framework could contribute to a better understanding of EU policy implementation, by focusing on practical implementation factors, and assessing how EU members use discretion in adapting to new regulatory landscapes. 24 3.2 Multi-level governance Although criticized in previous literature for its vague definitions and for being a catch-all concept, the dynamic nature of multi-level governance (MLG) theory and its analytical versatility can nonetheless add depth to the thesis’ main framework. (Alcantara et al., 2016; Piattoni, 2009). Moreover, MLG’s recent contribution to differentiation in the European Union makes it fit to supplement the differentiated policy implementation framework outlined previously (Hooghe & Marks, 2023). Multilevel governance refers to the dispersion of authority within national states jurisdictions, but also beyond, including federal systems and international organisations. There are two logics at the core of MLG, i) a demand for self-rule by diverse communities, states, regions etc. ii) a functionalist logic, which views governance as a means to provide public goods in the most efficient way. Moreover, while weighing the state-centric model of governance against the multilevel one in their contribution to European Union integration literature, Hooghe & Marks (2001) assert that “multi-level governance is prominent in the implementation stage” (p. 30). These considerations make MLG particularly relevant, as it provides useful theoretical perspectives on analysing the DGA within the context of the EU and the process of decentralisation to its members states (Hooghe et al., 2020). As anticipated previously, MLG is a wide and elusive body of theories. For this reason, MLG will serve this thesis as a complement to DPI and implementation literature, which remain as a primary framework; in other words, MLG will be used as a toolbox from which specific concepts will be extracted and used as analytical tools. The article on multilevel governance types published in 2003 by Hooghe and Marks presents a valuable distinction between Type I and Type II governance structures: - Type I: this governance structure relates to federalism, intended as the diffusion of authority on just a few levels, involving a limited number of governments. In other words, the focus is on the interactions between a central government and a tier of subnational governments. Type I governance is characterised, among other things, by 25 i) non-intersecting memberships, primarily territorial ii) durable, systemwide architecture of jurisdictions iii) jurisdictional levels are organised in few levels (eg. local, intermediate, and central) - Type II: the key feature of this governance structure is its radical departure from the traditional centralized state’s structure. Type II represents an alternative category of multilevel governance, where jurisdictions are task-specific, therefore flexible, and not all-purpose nor durable. Moreover, Type II governance structures are not distributed in a limited fashion but operate on numerous territorial scales diffusing authority in different ways. These two governance structure types portray radically different methods of political organisation. Type I governance represents membership based on (mostly) territorial communities. On the other hand, Type II governance entails overlapping roles determined by specific tasks or policy areas (Hooghe and Marks, 2003). This distinction is significant when looking at perceptions on the Data Governance Act’s implementation in Italy: the regulatory framework established by the DGA can be categorised as a form of Type II governance structure. Additionally, the context of EU policy implementation and its reception within national member states provides the ideal ground for appreciating different governance structures. 26 4. Methodology The following chapter will outline the methodological approach and research design employed in this study project. It will first briefly summarise the research problem, provide some considerations on the research question, and the aim of the thesis overall; following that, it will describe the qualitative approach to the study and its research design. The chapter concludes with relevant ethical consideration. 4.1 Research problem, research question and its purpose Research problem The proposed research problem revolves around the topic of the twin transition i.e. the interrelation between the green transition and the digital transition as EU-level policy domains. More specifically, this thesis is interested in whether the green and digital transitions are mutually compatible and reinforce one another, or if one transition is likely to obstruct the other. As discussed in the previous chapter, the debate on the potential outcomes of the twin transition is wide open, with one side highlighting the great potential for climate change solutions carried by innovative technologies, and the other adopting a more prudent line, underscoring the potential risks of technological innovation. This specific argument sparked a deep interest, and a personal drive, in exploring further the synergy between digitalisation and environmental sustainability, and how technological progress in this area could be catalysed towards innovative solutions for the fight against climate change and environmental degradation. Extensive preliminary research led to findings supporting the idea that sound data governance models hold a crucial role in determining whether the twin transition will be successful or not (EC Joint Research Centre, 2021; Jirka et al., 2022; Jurmu et al., 2023). Consequently, this research study is interested in how public actors within the digitalisation field in Italy are reacting to the adoption of the Data Governance Act (DGA) by the EU, and 27 what their perceptions entail in the context of the larger debate surrounding the twin transition; in other words, this issue is viewed as instrumental in gaining deeper insights into the problem of the twin transition, and that is why the DGA holds significant importance in this context and remains at the forefront of this thesis. Ultimately, this thesis is founded on a strong belief that the twin transition, and its far-reaching economic and social implications, will in many aspects define the decade we currently live in. Research question The primary research question is formulated as follows: “How do public actors within the digitalisation field in Italy perceive and respond to the implementation of the Data Governance Act within the context of the 'twin' green and digital transition in the European Union?” Acknowledging the broadness of the research question and the vast nature of the research problem, the formulation of the main RQ went through a process of refocusing, from broad (twin transition policy) to narrow (Data Governance Act). This process culminated with identifying mid-range theories within current literature with potential to bridge the micro and macro levels. Moreover, factors such as data collection, case selection, and data analysis influenced the flexible path towards consolidating the research question at the core of this thesis. As mentioned in the literature review chapter, the DGA is central to the EU’s digital agenda and a key piece of the European strategy for data: its goals include increasing “trust in data sharing, strengthen mechanisms to increase data availability and overcome technical obstacles to the reuse of data” (European Commission, 2023c). This makes the regulation an EU policy highly representative of the twin transition, as data itself can be viewed as an essential factor in bridging digitalisation and the quest for environmental sustainability. All things considered, the research question is crafted by keeping DGA implementation aspects at the forefront, while the twin transition element informs the wider background; evaluating the twin transition as a comprehensive phenomenon and measuring to which degree it will be successful or not is outside the scope of this master’s thesis. The research question is structured to investigate whether analysing the reception of the Data Governance Act in EU member states 28 could deepen our understanding of the twin transition. The focus on perception and response to the DGA relates to the fact that the DGA has come into effect only very recently (September 2023). This early phase makes it premature to identify concrete steps towards implementation and compliance factors. As for the choice to analyse public actors in Italy, this is justified as an effort to narrow down the research problem’s scope and defining a feasible single case study within the constraints of this master's thesis. Purpose of study and research gaps The main purpose of the research question is to lead the investigation process into uncovering how Italian public authorities involved in the country’s digitalisation efforts will receive and adapt to the Data Governance Act. This thesis hopes to shed light on issues such as: who are the main public actors overseeing the implementation of the DGA, what will be the implementation challenges posed by the DGA, how concerned are Italian public actors with the environmental aspect of open data and data sharing practices. Overall, this research project is set on finding out what actions public authorities in Italy are taking regarding the Data Governance Act, how are they perceiving the introduction of this new European regulatory framework, and what implications are there concerning the green and digital transitions, if any. Moreover, this thesis aims to build on previous implementation literature in the EU and contribute to the ongoing academic debate on data governance by providing valuable insights from one of its member states, Italy. As mentioned previously, large part of academic research on policy implementation focuses on legal implementation and on whether member states comply. This thesis’ purpose is to contribute to the line of study which is more concerned with practical implementation and how member states use discretion in adapting to new regulatory landscapes (Thomann & Sager, 2017; Zhelyazkova et. al, 2024). In addition, while past literature has primarily focused on EU directives’ implementation, this research on the DGA Regulation aims to contribute to a shift in direction in academia, which includes EU Regulations (Blom-Hansen et al., 2023). 29 Following preliminary research, which uncovered published literature on the twin transition and on data governance regulations in the European Union, it became evident that there is a scarcity of published academic papers investigating the role of data within the green and digital nexus; the one exception is an article published by Finck and Mueller (2023) analysing access to environmental data in relation to the recent changes in EU Data Law. The authors are hopeful their paper will be a contribution to a conversation, which has yet to flourish, on how to bridge environmental sustainability and protection to digitalisation in an effective way. Nonetheless, data availability on the topic is constantly growing due to its emergent nature and relevancy in the EU and beyond; public and academic interest towards the twin transition is also increasing. Furthermore, literature on the Data Governance Act and on the implications of the European strategy for data in general (e.g., Data Act, DSA, DMA etc.) is becoming more and more available. 4.2 Research design The research design involves employing a qualitative research method and developing a single case study. As Yin (2014) suggests, a research design entails way more than a plan for structuring the research work: its main purpose is to act as a blueprint informing on which question to tackle, what data to collect, which evidence is relevant, and consequently how to analyse the results. In other words, it is important to avoid a scenario where the results and the initial research question are disjointed. For that reason, as mentioned above, the research design employed in this case was adaptive. After dedicating substantial time to formulating and refining the research question, while simultaneously remaining open to adjustments along the way, the case study research method is viewed as the most fitting. The latest reformulation of the research question is indeed a “how” question, and it is about contemporary and timely events over which I do not have control as a researcher; these characteristics, as outlined by Yin (2014), make this research question suitable for a case study method. Furthermore, as mentioned previously, the goal is to develop a single case study, which in this instance will be Italy, a member state of the European 30 Union currently in the process of implementing the Data Governance Act. Moreover, a single case study design focused on Italy enables this research to investigate deeper into administrative contexts and institutional structures related to the EU level. The selection of the case study country is explained primarily by feasibility factors: accessibility to data sources played a major role in this phase. Being aware of the significant differences between EU member countries, in terms of both the structure of their economies as well as the diversity of their political systems, sampling has proved to be a lengthy and complex task. Initially, the research design was structured around the selection of two EU member states, and therefore the plan involved carrying out an extensive comparative case study project. This early methodological choice was because, traditionally, evidence gathered from multiple cases is usually regarded as more compelling, thus the research is perceived as more robust overall (Herriott & Firestone, 1983, as cited in Yin, 2014). However, the preliminary comparison between the two countries revealed itself to be unproductive in terms of generating compelling results, therefore the research design was reconsidered, and then restructured into a single case study. Italy’s selection, as already mentioned, relates to its quality of having the most available and accessible data sources relative to the data collection process carried in this research. Moreover, Italy's historically below EU average digitalisation performance makes it an interesting case study, particularly when noting the significant progress in recent years in terms of digital infrastructure and increased efforts on interoperability of data across public institutions (European Commission, 2023a). Considering that the DGA has been enacted only very recently, and consequently EU member states are still in the process of adjusting under the new regulatory framework, this research design is well suited for gaining interesting national level insights. Considering that this research project is not purely exploratory, this study project aims to insert itself within a theoretical debate, and ideally contribute further to it. The theoretical framework described in the previous chapter will help set the direction of the thesis, guide its focus when collecting relevant evidence, and inform the entire research project (Yin, 2014). More specifically, the theoretical framework provided by Differentiated Policy Implementation (DPI), supplemented by conceptual insights drawn from Multi-level Governance (MLG) literature will be operationalised in order to fit the single case study on the response to the DGA by Italian public actors. Concepts from DPI theory, including institutional 31 fit and policy attributes such as flexibility and coherence, will allow for an analytical framework which is integral to the research design. In terms of data collection, as part of the qualitative research design, this thesis will present qualitative data extracted from interviews carried with experts employed in the Italian public digital sector; they include legal experts, public officials, and policymakers. Aside from interviews, official documents (e.g., policy briefs, press statements by relevant public actors, governmental reports etc.) will be subject to data collection and analysis. Brussels-based stakeholders, EU public institutions, public advocacy groups, industry associations could constitute interesting open sources of EU-level data relevant to the research. Data collection This section will discuss some data collection aspects in relation to this thesis’ single case study research project. In the initial phase, the process of data collection will be preceded by a thorough preparation process; poor preparation has the potential to put at risk the entire research design and the validity of the qualitative data (Yin, 2014). Good preparation entails a variety of factors, but for what concerns this study, in the context of a master’s thesis, the primary focus will lay in developing a well-structured interview guide. Moreover, refining personal research skills and prior academic knowledge is necessary before starting to concretely collect data. In other words, this thesis will actively engage in self-reflection, considering both the researcher's academic strengths and personal values as inherent factors shaping the research process. Furthermore, previous research and a thorough literature review on the topics of twin – green and digital - transition and EU data governance issues are deemed instrumental in crafting an effective list of questions for prospective interview subjects; this will also prove useful further down in the process, when analysing the collected data (Yin, 2014). One of the primary reasons behind choosing to use a qualitative approach and case study methods is its variety when it comes to data collection. The opportunity to employ multiple and different sources of evidence is without a doubt a vital asset of case study methods; it allows for better triangulation and crystallization practices, giving further depth 32 and complexity to the research (Tracy, 2010). For this project, as mentioned in the research design section, the plan is to collect data from i) official documents and ii) interviews. Concerning the latter method, the thesis intends to carry out semi-structured, in-depth interviews following specific lines of inquiry according to a written guide. Understanding the complexity of gathering useful data through the method of expert interviews, the evidence gathering objectives will be prioritised through the adoption of a friendly, conversational approach (Yin, 2014). The interviews will be conducted with legal specialists, international relations professionals and senior policy experts in the field of digitalisation, as well as with officials dealing with policy implementation and digital governance issues in Italy. Moreover, the interviews will be carried with the aid of the Interview Guide (see Appendix) which contains a list of open semi-structured questions, leaving room for custom follow-up queries. Potential interview subjects will be contacted via e-mail, following meticulous research on their work experience history; after the initial agreement, the interviews will be carried out virtually through a preferred video conference platform. The number of interviews is dependent on multiple factors, such as feasibility, access to interview subjects and their availability, and time constraints. Moreover, the interviewees will be sampled based on their role and functions within Italy’s public authorities; furthermore, priority will be given to interview candidates with a background in public law and international relations. The reason for this is the complex nature of the Data Governance Act. Even though interviewing has the potential to offer compelling evidence and vast amounts of information, which often cannot be retrieved from alternative sources, this thesis is well aware of its significant limitations; as Lilleker (2003) suggests, “one has to be aware that you cannot base a conclusion upon the recollections of any one individual any more than you would base history on a single autobiography” (p. 212). Moreover, some insights gathered through the interview tool can be inconsistent or distorted by the person being interviewed (often, distortion can also be caused by the interviewer). To be clear, the role of interviews is to add depth to the analysis, while remaining aware of potential biases. For this reason, interviews will not be the one and only method of data collection on which the research will be based on. Therefore, the collected data will be deliberately backed up by other sources of empirical material (Lilleker, 2003). 33 That is the main reason for using documents (e.g., governmental reports, policy briefs, or legal documents by Italian institutions, public authorities, and policy actors) as empirical material for this research. The Italian language documents will be translated into English during the analysis phase; for transparency motives, the relevant textual excerpts and evidence will be included in both languages. At this stage of the research work, the specific documents have not been fully selected; they will be however presented in the following chapter. Nonetheless, they will be chosen and analysed based on criteria such as feasibility of analysis (or length of document), recency, accessibility, and ethical reasons. The last point implies that documents which have not been publicly presented or made available to a wide audience yet will not be considered due to privacy and confidentiality. Data analysis Prior to developing an analytical strategy, the collected data will be processed and organised in a way that will serve the purpose of the single case study. For what concerns documents such as policy briefs, official reports, and official guidelines, processing them should not be too complex; dedicated software will be employed for cataloguing the files. On the other hand, interviewing will entail additional work: the conversations will be electronically recorded after gaining informed consent; after that the audio files will be transcribed in full, possibly using computer software as a tool; subsequently, the raw interview transcripts will be edited for clarity and brevity, while maintaining their original meaning, intent, and form. For that reason, the empirical material subject to analysis will be entirely in text form, categorised in specific themes, and then coded for the purpose of facilitating research and analysis tasks. It is important to acknowledge that linguistic analysis of a text inherently implies an examination of the corresponding discursive practice; that is why the research intends to adopt a critical lens in the quest to investigate the perception of the DGA by Italian public bodies (Jorgensen & Phillips, 2002). The analytical strategy employed in this project relies on theoretical insights which are drawn from previous literature on the research topic, as already mentioned. In other words, the plan consists in looking for patterns or indicators in the collected data that match the theoretical concepts at the core of the framework guiding the analysis. Moreover, the theory described in 34 the initial research design, and supported by the collected evidence, will ideally form the basis for an analytic generalisation (Yin, 2014). Regarding analytical techniques, the pattern- matching tool stands out as the most appropriate for this research project; according to Yin (2014), one of the most advisable techniques in case study analysis is in fact pattern-matching. In the political science field, this technique has been labelled congruence method (George & Bennett, 2004, as cited in Yin, 2014): the results achieved through such techniques can aid the single-case study and “strengthen its internal validity” (Yin, 2014, p. 222). Nevertheless, the thesis will strive for high analytic quality, ensuring the collected data will be thoroughly examined without omitting relevant details. Furthermore, the analysis will include as much evidence from the collected data as possible, and the subsequent interpretation will take it into account without leaving room for ambiguity. 4.3 Ethical considerations and privacy Good qualitative research demands authenticity and honesty towards the self, the research, and its dedicated audience. One must be aware of their own biases, academic inclinations and what guides them. Stated differently, self-reflexivity and transparency are integral part of sincere research work, which adheres to the highest ethical standards (Tracy, 2010). Robert Yin (2014, p. 442) emphasises that the “highest ethical standards [...] include having a responsibility to scholarship, such as neither plagiarizing nor falsifying information, as well as being honest, avoiding deception, and accepting responsibility for one’s own work.” Focusing in on the issue of ethics, there is one aspect in particular which deserves special attention, particularly in relation to the research context of this thesis: that is procedural ethics. These refer to “ethical actions dictated as universally necessary by larger organizations, institutions or governing bodies” (Tracy, 2010, p. 847). Procedural ethics comprise the importance of data accuracy and avoiding falsifying information or omitting relevant evidence; agreeing on informed consent and ensuring privacy and confidentiality of participants are also encompassed by what we defined as procedural ethics. 35 The last point, ensuring participant privacy and confidentiality, will be a high priority throughout this research due to the work on interviews with individuals who will most likely be bound by confidentiality agreements. 36 5. Analytical Strategy The following chapter will outline the thesis’ analytical strategy by presenting the framework employed to analyse the empirical material. In addition, it will detail the sources of evidence from which the data was collected, which include interview transcripts and official documents. 5.1 Analytical Framework As previously mentioned in discussing methodology aspects, this strategy utilises concepts and insights drawn from Differentiated Policy Implementation (DPI) and Multi-level Governance theories. These concepts have been selected, evaluated, and operationalised into an analytical framework, as illustrated in the table below: Analytical Framework Concept/ Focus Operationalisation Indicators Codes Area Statements indicating Identify public officials' understanding of DGA’s Perceptions of perceptions of the DGA: its legal framework and Positive Perception, DGA and objectives and challenges to implications, in terms of Concerns, Expectations Challenges practical implementation both challenges and opportunities References to specific laws, regulations, or Examine DGA – public Alignment, Conflict, Institutional Fit institutional practices; institutions compatibility Adaptation references to degree of institutional adaptation Statements on cooperative Investigate relations between efforts or tensions among Administrative Italian public bodies in the public institutions on Collaboration/Competition Behaviour digitalisation field DGA/data governance issues; Alignment with References to national Assess DGA in the context National policies or strategies Alignment/ of Italy's national digital Strategies and related to digitalisation or Divergence/Synergies strategy Goals environmental goals Table 1: Analytical Framework 37 The analysis, guided by the framework above, will look for and identify specific language relating to four themes: i) Perceptions of the DGA and its challenges: this thematic area will analyse excerpts from the collected data relevant to understanding how public bodies involved in digitalisation policy perceive the legal framework introduced by the Data Governance Act. ii) Institutional Fit: this section will look at elements which are indicative of compatibility between the DGA and Italy’s legal structure, including institutional disposition. iii) Administrative Behaviour: here the focus will lie on horizontal relations between public actors involved in Italy’s digitalisation policy landscape. iv) Alignment with national strategies and policy goals: this section will analyse references relevant to national digital policy and data practices, and how those align with the European level data strategies, including the green-digital nexus. 5.2 Sources of evidence Before delving into each area of focus, it is important to provide further details on the sources of evidence which are subject to analysis. As already mentioned, the data collection methods for this thesis consist in i) official documents and ii) semi-structured expert interviews. Regarding the first method, the documents were selected based on their relevancy to the research topic, their recency in terms of date of publishing and their accessibility. The publishers and authors of the analysed documents include the main public actors responsible for Italy’s data governance, digital transformation, and overall policy strategy. Furthermore, the selection of documents was informed by theoretical concepts contained in the analytical framework (e.g. references to multi-level adaptability, policy fit within national legislation etc.). Considering the scope of this master’s thesis, and a range of feasibility factors (e.g., accessibility, time constraints etc.), the document list has been refined, and then categorised as such: 38 Year of Document Type Original Title Publisher/Author Language release Piano triennale per l’informatica nella Pubblica Agenzia per l’Italia Multi-year Plan 2023 Italian Amministrazione Digitale (AgID) (2024-2026) Linee Guida recanti regole Guidelines/ tecniche per l’apertura dei Agenzia per l’Italia Technical dati e il riutilizzo 2023 Italian Digitale (AgID) Rulebook dell’informazione del settore pubblico Guida operativa sulle serie Agenzia per l’Italia Operational Guide 2023 Italian di dati di elevato valore Digitale (AgID) Digital Public European Factsheet Administration factsheet 2023 English Commission 2023 - Italy Garante per la Relazione Annuale/ Italian/ Annual Report Protezione dei Dati 2022 Annual Report 2021 English Personali (GPDP) Table 2: List of official documents Each of the documents will be analysed through the thesis’ analytical framework; relevant excerpts will be reported in corresponding focus area sections. Consequently, the evidence will be interpreted in context of the research. A few clarifications on the Italian language documents are needed, specifically regarding their content, and their authors or publishers: - Agenzia per l’Italia Digitale (or Agency for Digital Italy), from now on referred to as AgID, represents the main public body responsible for ensuring the implementation of objectives set by the Italian digital strategy at the national level. The agency is tasked with coordinating public administrations in the process of promoting Italy’s digital transition. Additionally, it represents the main body in charge of interoperability in the 39 country, ensuring the integration of national information systems at the EU level. (AgID, 2024; EC, 2023b). The following documents published by AgID will be included in the analysis: 1. Piano triennale per l’informatica nella Pubblica Amministrazione (2024-2026) is a multi-year plan for IT in Public Administration. This key strategic document is part of the broader framework defined by the EU’s strategic programme 'Digital Decade 2030'. 2. Linee Guida recanti regole tecniche per l’apertura dei dati e il riutilizzo dell’informazione del settore pubblico is a rulebook containing guidelines aimed at supporting public institutions (and other stakeholders) in the ongoing process of opening data and incentivise its reuse. 3. Guida operativa sulle serie di dati di elevato valore is an operational guide on the re-use of high value data (HVD), as outlined by the European legal framework on data. High value datasets include environmental data; therefore, the Guide is useful in providing insights on how Italian public actors view and perceive the DGA in the sustainability context. - Garante per la Protezione dei Dati Personali (or the Italian Data Protection Authority), from now on referred to as GPDP, is the independent authority in charge of protecting fundamental rights in connection with the processing of personal data. The GPDP is tasked to ensure that data is processed as required by the law and in accordance with the European GDPR, both in the public and private sectors (GPDP, 2024). The analysis will include the latest national reports published by the GPDP. 1. The Relazione Annuale 2021 (or 2021 Annual Report) represents a precious source of evidence for generating insights on the perception of data governance legal frameworks in Italy, including the DGA, by public actors. The 2022 edition of the Report was initially considered for the purpose of analysis, however preliminary research revealed a lack of relevancy to this thesis project, thus it was decided to 40 exclude it. The 2023 edition of the GPDP Annual Report will be published presumably in June of 2024; unfortunately this thesis did not succeed in getting access to its draft version. - Dipartimento per la Trasformazione Digitale (or The Department for Digital Transformation), from now on referred to as DTD, although not listed in Table 2, contributed to most of the documents published by AgID, as well as to the Digital Public Administration factsheet 2023 – Italy document published by the European Commission (EC, 2023b). The DTD is in charge of assisting the coordination of government actions aimed at promoting the modernisation of the country through digital technologies (DTD, 2024). The three public institutions outlined above – the AgID, the GPDP, and the DTD – are the top authorities involved in digital policy and data governance in Italy. The thesis evaluated the possibility of including other public actors in the selection, such as Undersecretary for Technological Innovation and National Agency for Cybersecurity. However, considering their wide array of functions (i.e., from digital divide monitoring to cyber defence), their selection would have risked leading the analysis process outside the scope of the thesis. Concerning the second source of evidence, three interviews were conducted with experts working with the DTD and the GPDP. Their professional expertise includes Data Law, Data Governance and Policy, and International Relations. Two of the interview participants have backgrounds in politics and academia, respectively. The interviews took place under conditions of confidentiality; this entails that the recorded content and its full transcription will only be utilised for the purpose of this research project. Moreover, the participants were promised a certain degree of privacy, and above all, anonymity. Therefore, with the exception of their professional background, extensive details about the interviewees will not be disclosed. As mentioned above, the analysis will be executed by extracting key statements or references from the full transcripts, which will be then categorised and interpreted in the context of the analytical framework. 41 6. Results and Analysis The following chapter will analyse the collected empirical material and present the results relevant to the thesis’ research question: “How do public actors within the digitalisation field in Italy perceive and respond to the implementation of the Data Governance Act within the context of the 'twin' green and digital transition in the European Union?”. To facilitate the comprehension of the analysis, it is useful to summarise some of the key elements of the Data Governance Act. There are three main topics on which the DGA is focused. Firstly, the re-use of available data held by the public sector, including local, regional and state institutions; this includes both personal and non-personal data. Secondly, the governance of data intermediation services which hold the purpose of facilitating data sharing; these services establish commercial relationships between relevant parties (i.e. data holders and data users). Lastly, the introduction of a data altruism framework with the objective of promoting data sharing on a voluntary basis, for the collective good. The overarching theme of the DGA is promoting open, interoperable data sharing. Informed by the previously outlined analytical framework, the analysis presented in the following sections will be structured in focal areas and will touch on the aforementioned topics relating to the DGA. 6.1 Perceptions and Challenges The first thematic area of focus revolves around the legal framework outlined by the DGA regulation: this section analyses how relevant public actors in Italy perceive the concepts and legal definitions outlined by the Data Governance Act. It will carry out the task through careful examination of official documents and interview transcripts, thus looking for statements, references and implications indicating the understanding, in the form of either expectations or concerns, of the legal provisions introduced by the DGA. 42 Re-use of publicly held data Public institutions hold vast amounts of protected data (e.g., personal data, intellectual property and other sensitive data) that could be reused for a multitude of purposes across a variety of sectors, such as environment, health or energy to name a few. The DGA aims to increase data availability and data sharing to stimulate its re-use for the purpose of stimulating cross-sector innovation (Regulation 2022/868). At first glance, the Data Governance Act appears to be well received at the national level, with public actors involved in digitalisation and data governance expressing positive expectations about the DGA. Findings reflect that national public bodies generally view the Regulation as an appropriate legal instrument for increasing data availability and its re-use. Interviewee #1, a legal expert in the field of data governance, underlines the great potential carried by the DGA in ensuring the efficient re-use of data through high levels of interoperability, while also addressing cybersecurity concerns. In their view the EU Regulation contributes significantly to the establishment of a European single market for data, a key objective of the European Data Strategy: “[...] by creating such a level playing field in the data economy through openness, companies will be incentivized to compete not on the mere quantity of data available but on the quality of services offered. [...] The re-use of available public sector information is an asset, as the public sector handles vast amounts of sensitive data, such as personal information and commercially confidential data, which cannot be re-used as open data. However, the possibility exists to re- use them in accordance with specific EU or national laws. The richness of knowledge inherent in such data can be extracted without compromising its protected nature, and the DGA establishes its own rules and safeguards to facilitate such reuse whenever possible [...].“ (Interviewee #1, 2024) Interviewee #2, a senior official and international relations expert, also expresses positive views from within Italy’s Department for Digital Transformation (DTD). They highlight some areas, namely corruption and tax evasion, which could benefit from the new mechanisms introduced by the DGA: 43 [...]if you want to have a productive government, you need data sharing. And also, we know this, for instance, there is a lot of data that the Court of Auditors has about the government, as well as its citizens, and they cannot reuse it. The data could be important to reduce corruption or to identify this tax evasion that in Italy is very severe.” (Interviewee #2, 2024) The observations made by Interviewee #2 underscore possible practical applications derived from the opening of data under the DGA. Similar institutional positions on the re-use of data can be interpreted from the analysis of the three documents published by AgID. In the context of the Multi-year plan document for the 2024-2026 period, the governmental agency views the DGA as a key instrument in promoting multi-sector innovation in Italy and achieving shared strategic objectives such as enabling the full potential of data. As reported in an excerpt from the official document: Original: “Con l’adozione dell’atto sulla governance dei dati (Data Governance Act), inoltre, sono stati definiti e rafforzati i meccanismi per aumentare la disponibilità dei dati e superare gli ostacoli tecnici al riutilizzo di alcune particolari tipologie di dati altrimenti non disponibili.” (AgID, 2023c, p. 77) Translated: With the adoption of the Data Governance Act (DGA), mechanisms have been defined and strengthened to increase data availability and overcome technical obstacles to the reuse of certain types of data that would otherwise not be available. Conversely, one of the public actors in Italy adopting a more prudent and critical line is the national Data Protection Authority (GPDP). In their 2021 Annual Report, the GPDP conveys specific concerns regarding the DGA, calling for the new regulations on data governance to align with the existing EU and national legal frameworks on data protection: Original: [...] il futuro regolamento sulla governance dei dati sia pienamente in linea con il quadro giuridico dell’UE in materia di protezione dei dati personali, in particolare per quanto riguarda la competenza delle autorità di controllo, i ruoli dei diversi attori coinvolti, la base giuridica per il trattamento dei dati personali, le garanzie necessarie e l’esercizio dei diritti degli interessati. [E’ necessario] che il nuovo regolamento affermi in modo chiaro e inequivocabile che esso non incide sul livello di protezione dei dati personali, né modifica i diritti e gli obblighi stabiliti nella legislazione in materia di protezione dei dati. (GPDP, 2022, 44 p. 222) Translated: [...] the future regulation on data governance must be fully in line with the EU legal framework on personal data protection, particularly concerning the competence of supervisory authorities, the roles of the various actors involved, the legal basis for the processing of personal data, the necessary safeguards, and the exercise of data subjects' rights. [It is necessary] that the new regulation states clearly and unequivocally that it does not impact the level of personal data protection, nor does it modify the rights and obligations established in the legislation on data protection. In their answer to a question regarding concerns posed by the DGA, Interviewee #1 noted that: “[…] national data protection authorities will need to undergo an important educational effort in order to be conscious of the interplay and potential overlap of the DGA and the data protection legal framework. Especially, national DPAs will have to develop a “data governance culture” and learn how to valorise the principles of free flow of personal data [including the re-use of data]” (Interviewee #1, 2024). The statement by the GPDP and the remarks by Interviewee #1 represent a crucial trend in European policies on data governance which have recently seen a shift in direction, from data protection towards data sharing as the enabler of the new data economy (Ruohonen & Micklesson, 2023). Balancing the two spheres of interests is a key challenge posed by the DGA during its implementation. Data intermediaries and data altruism organisations Some challenges emerge when investigating perspectives on the introduction of institutions such as data intermediation services (DIS) and data altruism organisations, which are all elements meant to facilitate data sharing and the opening of data, as well as coordinate new data governance mechanisms. When questioned about the role of data intermediaries introduced by the DGA, interviewee #1 was hesitant to give an opinion as there is still need for clarification on some points, namely: 45 “I believe that it will be necessary to clarify whether or not the notion of “data brokers” falls within the “data intermediary” definition. This raises important questions on the conformity of monetisation practices also under GDPR and is notoriously a debated topic these days.” (Interviewee #1, 2024). Similarly, concerns were raised about data altruism: “I think that it will be necessary to await the implementation of the DGA at the national level to ascertain the kind of legal persons authorized to conduct data altruism practices and, therefore, exempted from the requirements applicable to the service intermediaries, as well as the type of organisational standards required to qualify as data altruism organization. (Interviewee #1, 2024). Moreover, the GPDP recommends further clarification on the notion of data altruism, underlining the importance of consent mechanisms in data sharing: Per quanto riguarda il cd. altruismo dei dati il parere congiunto raccomanda di definire meglio le finalità di interesse generale comprese in tale nozione e di permettere alle persone interessate di prestare e revocare facilmente il consenso all’utilizzo dei loro dati per queste finalità. (GPDP, 2022, p. 223) Translated: With regard to the so-called altruism of data, the joint opinion recommends better definitions of the general interest purposes included in this notion and to allow data subjects to easily give and withdraw consent to the use of their data for data altruism purposes. The findings presented above are in line with previous research on the DGA, which argues that despite their value in promoting open data, it is still unclear how the new concepts introduced by the EU regulation will work in practice (De Hert, 2023; Ruohonen & Micklesson, 2023). Moreover, the observations made by interviewee #1 are particularly relevant to this thesis as they relate to key aspects of practical implementation, as defined by Differentiated Policy Implementation theories. Specifically, in the context of EU’s multilevel governance system, the points on data altruism practices highlight potential variations or differences in how the DGA regulation will be implemented at the national level (Zhelyazkova et. al, 2024). 46 Finally, Interviewee #3 adds a compelling perspective on the concept of data altruism. In their view, as a senior technical expert on data governance, data altruism does not actually mean open data by default; in other words, data altruism practices refer to closed data which is subsequently shared or rendered as open data through informed consent. Interviewee #3 makes a further point: “I am happy to know that there's this data altruism concept [in the DGA text], because I think that the sharing of data is always seen from public administrations towards the market, which is a very nice thing as it can enable value. But I think the private sector, the associations, the non-governmental organisations, could [also] share data. And they should lead by example, in some way.” (Interviewee #3, 2024) The statement provides a significant insight: public actors view data altruism primarily as a unidirectional flow from the public to the private sector. However, Interviewee #3 points out a need for a mutual framework where data sharing flow in both directions. This would be necessary for the practical implementation of the data altruism framework, as outlined by the DGA. 6.2 Institutional Fit & Administrative Behaviour In this section, the thesis will analyse compatibility factors between the DGA and public institutions in Italy; additionally, due to their nexus, this section will also include analytic insights into administrative behaviour. More specifically, this part aims to present and interpret evidence which is indicative of how well the regulation aligns with existing Italian legal structures and practices, in the context of Italy’s inter-institutional relations within data governance and digitalisation. This proved to be a rather difficult task, as the Italian legal system is incredibly complex and difficult to navigate. As one of the interviewees reports: “[...] Italian law is not really linear. It's really chaotic. Yes, we have actually a few problems with legislation, especially lawmaking. We make too many laws. And often they're also badly 47 written. And we have a proliferation of laws at the national level, but also at the regional level.” (Interviewee #1, 2024) The non-linear aspect of Italian law and the practice of overregulating at different levels of government is indicative of a complex policy environment, characterised by a constant need for an active and collaborative interpretation of the law, not too dissimilar from the conciliatory mode of administrative behaviour identified by Eriksen (2023). However, despite this complexity, this research has found that there is likely some interplay between the DGA, and the legal framework enacted by Italy to transpose the 2019 EU Open Data Directive (Interviewee #1). This EU Directive laid down the initial framework for regulating the re-use of data held by public bodies; the legal framework contains various categories of data, including high-value datasets such as geospatial and mobility data (Directive 2019/1024). The existing national legal framework on data governance has been determined by the Agency for Digital Italy (AgID), which remains one of the core public actors operating within the digitalisation field in Italy; it is important to note that AgID is not an independent agency, but responds to a political authority (Interviewee #2, 2024). The structure of the legal framework is built primarily on two documents, which are also included in the document analysis of this research study: the Guidelines on opening data and its reuse & the Operational guide on high value data (for more details on the documents, consult previous chapter). As reported by AgID, the key objective of the framework is: Original: “[...] supportare le pubbliche amministrazioni e gli altri soggetti interessati nel processo di apertura dei dati e di riutilizzo dell’informazione del settore pubblico [...]” (AgID, 2023b, p. 10) Translated: supporting public administrations and other stakeholders in the process of opening data and reusing public sector information. However, achieving that objective will require significant coordination efforts, particularly at the national level, under the continuous support of the AgID. When questioned on this issue, Interviewee #1 responded that: 48 “[...] it will be necessary, though, to have a consistent effort to ensure coordination among the different pieces of legislation and governance practices. I think that the role of AGiD (Agenzia per l'Italia Digitale) will be also to issue guidelines, probably, guiding public administrations on how to keep all data governance practices aligned with all the different pieces of legislation that we currently have now [with reference to the EU level].” (Interviewee #1, 2024) Concerning relations between public actors involved in data and digitalisation policy in Italy, the institutional landscape remains complex. Evidence emerged from interview transcripts reveals critical points of tensions in this area. A recent example, although outside the scope of this thesis but relevant for understanding inter-institutional relations and the practical implementation of EU law in Italy, is the current debate on the implementation of the AI Act. The decision to put AgID and the Agency for Cybersecurity in charge of the EU regulation on Artificial Intelligence is receiving wide criticism from independent authorities, such as the GPDP. One interviewee noted that: [...] probably AGiD jointly with the National Agency for Cybersecurity will even be competent authorities for the implementation of the AI Act. [...] And this is not really welcomed by many, especially NGOs and civic organisations. This is a hot topic now in the data landscape in Italy, because the problem is that AgID, like, also the Agency for cybersecurity, are subject to the spoils system mechanisms, because they are governmental agencies. So basically, whenever the government changes, it happens that also the directors of the different offices are removed or substituted.” (Interviewee #1, 2024) Another interviewee provided additional insights on institutional relations among Italy’s public actors in the context of the Data Governance Act. When questioned about cooperative efforts or tensions between the GPDP (the Italian Data Protection Authority) and the DTD (Department for Digital Transformation), they acknowledged the existence of frictions, characterising the relationship between the two as a “thin line” between collaboration and competition. The GPDP remains firm about protection of personal data, while the DTD pushes ahead in improving and establishing platforms for data sharing. Therefore, in practice, when the DTD drafts a new piece of legislation that might have effects on personal data, the GPDP issues an opinion and evaluates its compliance with existing legal frameworks, for example the GDPR. (Interviewee #1, 2024; Interviewee #2, 2024). 49 6.3. Alignment with National Strategies This last section will analyse the alignment of the Data Governance Act with the Italian national strategy on digitalisation, within the broader context of EU-level twin transition efforts. More specifically, it will examine policy strategies aimed at integrating digital and environmental objectives adopted by Italy´s public authorities. As reported in the Digital Public Administration factsheet on Italy (2023), there are several strategic initiatives aimed at addressing data and digital services in the public sector: - Cloud Italia (or the Italian Cloud Strategy): established in 2021 by the DTD, outlines a plan for public administrations to migrate their infrastructures to the cloud by 2026; this strategy is a key component of the wider Italia Digitale 2026 plan, which shares objectives consistent with the EU 2030 Digital Compass targets. - Modello di Interoperabilità (or the National Interoperability Framework): launched in 2021, it addresses organisational and technological barriers to data sharing between public bodies and agencies. The Framework constitutes a critical part of the Multi-year Plan document published by AgID. - National Recovery and Resilience Plan (or PNRR): the Recovery and Resilience Plan includes several strategic projects aimed at strengthening data governance. These initiatives align with the DGA and focus on establishing an efficient regulatory framework that facilitates secure and timely data reuse practices (European Commission, 2024). Interviewee #3, a senior technical expert working on the national interoperability framework, highlights Italy’s efforts in improving data governance: 50 “I would say that we are focusing on data interoperability, and we are focusing in making public data interoperable between public administrations. This is the first focus, as we had to make sure that public administrations were at ease in sharing data between them knowing that it was secure, that they could choose who had access and who didn't.” The interviewee continues by providing additional insights on the role of digital data within the green transition: “We are dealing with green interests. For example, we are trying to help the national authority ISPRA, which is the National Environmental Protection Agency to gather data in a timely manner and interoperability with all the regional environmental protection agencies. This could lead to a better understanding of the environment, because as you may know, Italy is in a very bad position regarding hydrogeological risks, floods etc.” (Interviewee #3, 2024) These findings are substantiated by the Operational guide on the re-use of high value data (HVD) in Chapter 5, section 5.2 of the document (AgID, 2023a). This extensive section outlines in great technical detail the state of the art on environmental and climate data in Italy, within the context of the EU-level framework. In conclusion, the national digitalisation strategies adopted in Italy prove a strong alignment with the DGA, particularly when considering the progress on data interoperability and reuse made by public actors. Evidence emerged from this analysis demonstrate that there is a strong data governance foundation on which can be built on. As one of the interviewees noted: [...] we’ve been working on interoperability, and now we have a very strong data sharing platform, which we built, and we are currently using. [...] this is a real revolution, because we are enabling data sharing between different administrations, and it is not an easy thing to do […]. (Interviewee #3, 2024) 51 7. Conclusion This thesis set out to explore the perceptions and responses of public actors in Italy towards the implementation of the Data Governance Act (DGA) within the wider context of EU’s efforts to advance the twin - green and digital - transition. Following a comprehensive qualitative analysis through in-depth interviews with digital and data governance policy experts and the examination of relevant policy documents, this thesis identifies several insights. Key findings from the thesis’ analysis are summarised as follows: Public bodies in Italy acknowledge the significant potential carried by the DGA for the purpose of promoting and increasing data re-use across different sectors, such as the energy and the environmental ones. Therefore, this thesis finds the DGA Regulation to be positively received by the main public actors involved in digitalisation and data governance policy at the national level in Italy. However, several challenges are foreseen ahead of the implementation of the EU Regulation. These include the complex regulatory nature of Italian law and the practice of overregulating at multiple levels of government; the need to re-educate the national DPAs to navigate the overlap between data protection and data sharing legal frameworks; the uncertainty on how data altruism and data intermediaries will function in practice. Despite these obstacles, strategic initiatives like Cloud Italia and the National Interoperability Framework align strongly with the DGA making Italy ready and fit to proceed with practical implementation. This research contributes to previous existing literature by providing valuable insights on the state of the art on data governance and digitalisation policies within the EU, in the broader twin transition context, from one of its member states, Italy. Moreover, this thesis contributes to the line of study on EU policy implementation literature, which is more concerned with practical implementation, and how member states use discretion in adapting to new regulatory landscapes (Thomann & Sager, 2017; Zhelyazkova et. al, 2024). Additionally, since previous literature has almost exclusively focused on EU directives’ implementation, this research on the DGA, which is an EU regulation, aims to contribute to a new EU policy implementation and compliance field of studies which includes regulations (Blom-Hansen et 52 al., 2023). Finally, as the nexus between data governance and the green transition is generally not very apparent, this master’s project hopes to contribute to raising awareness about the role of data in generating innovative solutions for tackling environmental challenges. The findings of this thesis indicate significant policy implications that warrant further discussion. First, as observed in the single-case study of Italy, the relationship between public actors involved in digitalisation is characterised by a delicate balance between competition and collaboration; the situation is revealed to be even more complicated when considering Italy’s complex legal system. Therefore, it is important to pay attention to country-specific interinstitutional and legal variables when analysing the implementation of EU regulations in individual member states, as they can impact policy coordination efforts. Second, there is a need for further clarification on novel data governance mechanisms such as the ones on data intermediaries and data altruism. Lastly, national-level Data Protection Authorities will need to develop a new legal approach to data regulation, learning to adapt to the DGA by accepting the principles of open free-flowing data, and reconcile it with the existing data protection regulations. Although the analysis generated additional insights that might suggest other promising policy recommendations, the three policy implications discussed above are deemed to be the most pressing at the moment. Although the case study of Italy provides some compelling insights, there are several limitations to it which must be pointed out. The single-case study approach, while aiming to offer in-depth qualitative analysis, may limit the generalisability of the findings. Moreover, this thesis would have greatly benefited from additional empirical data, particularly when it comes to the number of interviews with experts. Collected data from a wider pool of sources would have allowed for better triangulation practices, thereby increasing the validity and credibility of the findings emerged in this study (Tracy, 2010). However, given the scope of the research problem and the complexity of the DGA within the twin transition phenomenon, this thesis hopes to have generated interesting insights for additional and more comprehensive studies in this area. Future research should consider focusing on more ambitious multiple-case studies, potentially adopting a quantitative or mixed-methods approach. This would provide more insights into the successes and failures of the green and digital transitions and the practical role of data regulations in the process. 53 In conclusion, this thesis project has highlighted the key role of data governance in bridging environmental sustainability and digitalisation, while promoting the EU’s twin transition. By addressing the challenges and leveraging the opportunities identified in this study, policymakers can enhance the effectiveness of the DGA and contribute to a more sustainable and digitally advanced future for the EU. 54 References Agenzia per l’Italia Digitale (AgID). (2023a). Guida operativa sulle serie di dati di elevato valore. https://www.agid.gov.it/sites/default/files/repository_files/guida_operativa_hvd_- _ver._1.0.pdf Agenzia per l’Italia Digitale (AgID). (2023b). Linee Guida recanti regole tecniche per l’apertura dei dati e il riutilizzo dell’informazione del settore pubblico. https://www.agid.gov.it/sites/default/files/repository_files/lg-open-data_v.1.0_1.pdf Agenzia per l’Italia Digitale (AgID). (2023c). Piano triennale per l’informatica nella Pubblica Amministrazione. 2024-2026. https://www.agid.gov.it/sites/default/files/repository_files/piano_triennale_per_linformatica_ nella_pa_2024-2026.pdf Agenzia per l’Italia Digitale (AgID). (2024). About Us Webpage. https://www.agid.gov.it/it/agenzia/chi-siamo Alcantara, C., Broschek, J., & Nelles, J. (2016). Rethinking Multilevel Governance as an Instance of Multilevel Politics: A Conceptual Strategy. Territory, Politics, Governance, 4(1), 33–51. Almeida Teixeira, G., Mira Da Silva, M., & Pereira, R. (2019). The critical success factors of GDPR implementation: A systematic literature review. Digital Policy, Regulation and Governance, 21(4), 402–418. https://doi.org/10.1108/DPRG-01-2019-0007 55 Bianchini, S., Damioli, G., & Ghisetti, C. (2023). The environmental effects of the “twin” green and digital transition in European regions. Environmental and Resource Economics, 84(4), 877–918. https://doi.org/10.1007/s10640-022-00741-7 Bibbee, A. (2007). Making Federalism Work in Italy", OECD Economics Department Working Papers, No. 590, 1-41. https://doi.org/10.1787/017785106355 Blom-Hansen, J., Christensen, J. G., Grøn, C. H., Jensen, M. H., & Mortensen, P. B. (2023). ‘Transposition’ of EU regulations: The politics of supplementing EU regulations with national rules. Journal of European Public Policy, 30(12), 2786–2806. https://doi.org/10.1080/13501763.2022.2118355 Bondarouk, E., & Mastenbroek, E. (2018). Reconsidering EU Compliance: Implementation performance in the field of environmental policy. Environmental Policy and Governance, 28(1), 15–27. https://doi.org/10.1002/eet.1761 Burinskienė, A., & Nalivaikė, J. (2024). Digital and Sustainable (Twin) Transformations: A Case of SMEs in the European Union. Sustainability, 16(4), 1533. https://doi.org/10.3390/su16041533 Carvalho, G., & Kazim, E. (2022). Themes in data strategy: Thematic analysis of ‘A European Strategy for Data’ (EC). AI and Ethics, 2(1), 53–63. https://doi.org/10.1007/s43681-021- 00102-y 56 Dæhlen, M. (2023). The Twin Transition Century: The role of digital research for a successful green transition of society? (The Guild Insight Paper No.5). The Guild of European Research-Intensive Universities and Bern Open Publishing. De Hert, P. (2023). Post-GDPR Lawmaking in the Digital Data Society: Mimesis Without Integration. Topological Understandings of Twisted Boundary Setting in EU Data Protection Law. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4601974 Del Castillo, A. P. (2021). Europe’s digital agenda: People-centric, data-centric or both? Social Policy in the European Union: State of Play 2021, 81. Diodato, D., Huergo, E., Moncada-Paternò-Castello, P., Rentocchini, F., & Timmermans, B. (2023). Introduction to the special issue on “the twin (digital and green) transition: Handling the economic and social challenges”. Industry and Innovation, 30(7), 755–765. https://doi.org/10.1080/13662716.2023.2254272 Dipartimento per la Trasformazione Digitale (DTD). (2024). La Struttura: Nascita, obiettivi e organizzazione del Dipartimento per la trasformazione digitale. https://innovazione.gov.it/dipartimento/la-struttura/ Directive 2019/1024. Directive (EU) 2019/1024 of the European Parliament and of the Council of 20 June 2019 on open data and the re-use of public sector information (recast). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32019L1024 57 Eriksen, E. O. (2023). Three modes of administrative behaviour: Differentiated policy implementation and the problem of legal certainty. Journal of European Public Policy, 30(12), 2623–2642. https://doi.org/10.1080/13501763.2022.2125047 European Commission. (2024, January 11). European Data Governance Act | Shaping Europe’s digital future. https://digital-strategy.ec.europa.eu/en/policies/data-governance-act European Commission. Directorate General for Communications Networks, Content and Technology. (2023a). 2030 Digital Decade: report on the state of the Digital Decade 2023. Publications Office. https://data.europa.eu/doi/10.2759/999709 European Commission. DG DIGIT & DG CONNECT. (2023b). Digital Public Administration factsheet 2023 – Italy. https://joinup.ec.europa.eu/sites/default/files/inline- files/DPA_Factsheets_2023_Italy_vFinal_rev.pdf European Commmission. (2023c, September 26). European Data Governance Act | Shaping Europe’s digital future. European Union official website. https://digital- strategy.ec.europa.eu/en/policies/data-governance-act European Commission. (2022). Towards a green & digital future: key requirements for successful twin transitions in the European Union. Joint Research Centre. Publications Office. Regulation (EU) 2022/868. Regulation (EU) 2022/868 of the European Parliament and of the Council of 30 May 2022 on European Data Governance and Amending Regulation (EU) 58 2018/1724 (Data Governance Act). https://eur-lex.europa.eu/legal- content/EN/TXT/?uri=CELEX%3A32022R0868%20&qid=1673517089264 European Commission. Joint Research Centre. (2021). INSPIRE, a public sector contribution to the European green deal data space: a vision for the technological evolution of Europe’s spatial data infrastructures for 2030. Publications Office. https://data.europa.eu/doi/10.2760/062896 European Commission. (2019, July 16). The European Commission’s priorities 2019-2024. European Union official website. https://commission.europa.eu/strategy-and-policy/priorities- 2019-2024_en Finck, M., & Mueller, M.-S. (2023). Access to Data for Environmental Purposes: Setting the Scene and Evaluating Recent Changes in EU Data Law. Journal of Environmental Law, 35(1), 109–131. https://doi.org/10.1093/jel/eqad006 Fouquet, R., & Hippe, R. (2022). Twin transitions of decarbonisation and digitalisation: A historical perspective on energy and information in European economies. Energy Research & Social Science, 91, 102736. https://doi.org/10.1016/j.erss.2022.102736 Garante per la Protezione dei Dati Personali (GPDP). (2022). Relazione Annuale 2021. https://www.garanteprivacy.it/documents/10160/0/Relazione+annuale+2021.pdf/7a98bf21- 9150-26ae-c5ce-eeeb4edb02b2?version=2.0 59 Garante per la Protezione dei Dati Personali (GPDP). (2023). Relazione Annuale 2022. https://www.garanteprivacy.it/documents/10160/0/Relazione+2022+del+Garante+per+la+pro tezione+dei+dati+personali.pdf/125d413d-9ec3-ea46-8e01-643d585a8425?version=1.0 Garante per la Protezione dei Dati Personali (GPDP). (2024). The Italian Data Protection Authority: Who We Are. https://www.gpdp.it/web/garante-privacy-en/the-italian-data- protection-authority-who-we-are Gigler, B. (2020). 7 steps enabling the twin transition to a sustainable and digital economy. https://doi.org/10.13140/RG.2.2.11472.94729 Greater London Authority (2023). Improving air quality in London through data science and machine learning. https://dt4regions.eu/dt-book/stories/enhancing-air-quality-london- through-data-science-and-machine-learning Hazas, M., & Nathan, L. P. (Eds.). (2018). Digital technology and sustainability: Engaging the paradox. Routledge, Taylor & Francis Group. Herriott, R. E., & Firestone, W. A. (1983). Multisite Qualitative Policy Research: Optimizing Description and Generalizability. Educational Researcher, 12(2), 14–19. Hooghe, L., & Marks, G. (2001). Multi-level Governance and European Integration. Rowman & Littlefield Publishers. https://books.google.se/books?id=xZmMb5EJni8C 60 Hooghe, L., & Marks, G. (2003). Unraveling the Central State, but How? Types of Multi- level Governance. American Political Science Review, 97(02). https://doi.org/10.1017/S0003055403000649 Hooghe, L., & Marks, G. (2023). Differentiation in the European Union and beyond. European Union Politics, 24(1), 225–235. https://doi.org/10.1177/14651165221127885 Hooghe, L., Marks, G. & Schakel, A. H. (2020) Multilevel governance, in D. Caramani (ed.), Comparative Politics, Oxford: Oxford University Press, 2020, 193-210. Jirka, S., Granell, C., Mooney, P., Rieke, M., van den Broecke, J., Sarretta, A., Verhulst, S., Dencik, L., Oost, H., Micheli, M., Minghini, M., Kotsev, A., & Schade, S. (2022). Emerging approaches for data-driven innovation in Europe: Sandbox experiments on the governance of data and technology. Publications Office of the European Union. Jorgensen, M. W., & Phillips, L. (2002). Discourse analysis as theory and method. SAGE Publications, Limited. Jurmu, M., Niskanen, I., Kinnula, A., Kääriäinen, J., Ylikerälä, M., Räsänen, P., & Tuikka, T. (2023). Exploring the Role of Federated Data Spaces in Implementing Twin Transition within Manufacturing Ecosystems. Sensors, 23(9), 4315. https://doi.org/10.3390/s23094315 Lilleker, D. G. (2003). Interviewing the Political Elite: Navigating a Potential Minefield. Politics, 23(3), 207–214. 61 Meijer, A. (2024). Perspectives on the twin transition: Instrumental and institutional linkages between the digital and sustainability transitions. Information Polity, 29(1), 35–51. https://doi.org/10.3233/IP-230015 Muench, S., Stoermer, E., Jensen, K., Asikainen, T., Salvi, M., & Scapolo, F. (2022). Towards a green & digital future: Key requirements for successful twin transitions in the European Union. Publications Office of the European Union. Pathak, M. (2024). Data Governance Redefined: The Evolution of EU Data Regulations from the GDPR to the DMA, DSA, DGA, Data Act and AI Act. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4718891 Piachaud-Moustakis, B. (2022). Updated Guidance to the European Data Governance Act. Pharmaceutical Technology Europe, 34(11), 8–9. Business Source Premier. Piattoni, S. (2009). Multi‐level Governance: A Historical and Conceptual Analysis. Journal of European Integration, 31(2), 163–180. https://doi.org/10.1080/07036330802642755 Picht, P. G., & Richter, H. (2022). EU Digital Regulation 2022: Data Desiderata. GRUR International, 71(5), 395–402. https://doi.org/10.1093/grurint/ikac021 Portnoy, A., & Nauwelaerts, W. (2024). What You Should Know About the EU Data Governance Act. The Global Regulatory Developments Journal, 1(1), 43–50. 62 Richter, H. (2023). Looking at the Data Governance Act and Beyond: How to Better Integrate Data Intermediaries in the Market Order for Data Sharing. GRUR International, 72(5), 458– 470. https://doi.org/10.1093/grurint/ikad014 Rosário, A., & Dias, J. (2022). Sustainability and the Digital Transition: A Literature Review. Sustainability,14(7), 4072. https://doi.org/10.3390/su14074072 Ruohonen, J., & Mickelsson, S. (2023). Reflections on the Data Governance Act. Digital Society, 2(1), 10. https://doi.org/10.1007/s44206-023-00041-7 Thomann, E., & Sager, F. (2017). Moving beyond legal compliance: Innovative approaches to EU multilevel implementation. Journal of European Public Policy, 24(9), 1253–1268. https://doi.org/10.1080/13501763.2017.1314541 Tracy, S. J. (2010). Qualitative Quality: Eight “Big-Tent” Criteria for Excellent Qualitative Research. Qualitative Inquiry, 16(10), 837–851. Von Ditfurth, L., & Lienemann, G. (2022). The Data Governance Act: – Promoting or Restricting Data Intermediaries? Competition and Regulation in Network Industries, 23(4), 270–295. https://doi.org/10.1177/17835917221141324 63 World Bank. (2023). Green Digital Transformation: How to Sustainably Close the Digital Divide and Harness Digital Tools for Climate Action. Washington, DC: World Bank. https://doi.org/10.1596/40653 Yin, R. K. (2014). Case study research: Design and methods (5th Edition). SAGE Publications. Zhelyazkova, A., Kaya, C., & Schrama, R. (2016). Decoupling practical and legal compliance: Analysis of member states’ implementation of EU policy. European Journal of Political Research, 55(4), 827–846. https://doi.org/10.1111/1475-6765.12154 Zhelyazkova, A., Thomann, E., Ruffing, E., & Princen, S. (2024). Differentiated policy implementation in the European Union. West European Politics, 47(3), 439–465. https://doi.org/10.1080/01402382.2023.2257963 64 Appendix Interview Guide Potential interview subjects were contacted first via e-mail. Following an initial agreement, the interviews were carried out through the Zoom video conference platform. The number of conducted interviews were three. The interviews were conducted with experts working within the DTD and the GPDP. Their professional expertise includes Data Law, Data Governance and Policy, and International Relations. Two of the interview participants have backgrounds in local and national politics, and academia, respectively. The interviews took place under conditions of confidentiality. Therefore, the recorded content and its full transcription will only be utilised for the purpose of this research project. Moreover, the participants were promised a certain degree of privacy, and above all, anonymity. Except for their professional background, extensive details about the interviewees will not be disclosed. The guide’s questions served as an initial structure for in-depth semi-structured interviews; the open-ended nature of their formulations allowed for custom follow-up queries. 1. Could you describe your role and responsibilities in relation to data governance issues? 2. Can you provide an overview of your understanding of the Data Governance Act (DGA) and its objectives? 3. How would you describe the role of the DGA in relation to the larger regulatory framework pursued by the EU in digitalisation? 4. How frequently do you encounter discussions about environmental data specifically in relation to the DGA or data-sharing practices in general? 5. How do you see the DGA interacting with Italy's national digital strategy and its goals for the green transition? 6. How do you think the DGA aligns (or conflicts) with Italian public authorities’ existing data governance policies and practices? Could you name specific policies? 7. How do you view the concept of data altruism, as defined in the DGA, working in practice? 8. What are your thoughts on data intermediation services? 9. In your view, is the European Data Innovation Board suitable for coordinating EU- level data governance efforts? 10. How do you perceive the interaction between the DGA and the General Data Protection Regulation (GDPR)? 11. Are you concerned about the fact that the duties of national data protection authorities appear to substantially increase with the implementation of the DGA? 65