The quiet politics of migration supranationalization – Commission entrepreneurship and the intra-Corporate Transferee Directive

ABSTRACT Migration is widely considered a sensitive and contested area of EU policymaking, with Member States reluctant to cede control. Yet, supranational rules on ‘legal migration’ increasingly reduce Member State discretion to control migration. While Commission entrepreneurship has been proposed as an explanation, its importance for supranational outcomes remains unclear. To understand the role of Commission entrepreneurship in enabling supranational policy outcomes, this study traces the negotiations of on one central labour migration policy, the Intra-Corporate Transferee Directive. I demonstrate how the Commission was able to defuse opposition by proposing a narrow policy instrument and framing it as a matter of business needs, international obligations and competitiveness rather than labour migration. This enabled an environment of ‘quiet politics’ where concerns over sovereignty and control were tuned down to a preference for administrative flexibility. The findings have implications for our understanding of the dynamics of supranationalization in politically sensitive policy areas.


Introduction
Migration policy is an unlikely field for supranational policy outcomes.While the European Commission has been striving for common policies since the 1990s, Member States are said to be reluctant to relinquish national sovereignty and discretion over the admission of thirdcountry nationals (Caviedes 2016;Menz 2008).However, since the mid-2000s as many as ten directives governing 'legal migration' have been adopted (European Commission 2019, 7).Gradually, they have expanded the degree of supranationalization in migration policies, that is the pooling of sovereignty at the EU level in a manner that constrains Member States' discretion by eliminating pre-existing national policies or prescribing common requirements (cf.Quaglia and Spendzharova 2017).EU directives determine the admission conditions of students and researchers, intra-corporate transferees, seasonal workers, and family members of third-country nationals, not allowing Member States to maintain separate national schemes for these groups of migrants (European Commission 2019, 64).As a result, ca.38% of all labour migration permits issued since 2018 were governed by EU admission rules. 1 Existing research on Member State preferences focusses on the reasons for Council opposition to supranational migration policies rather than on enabling conditions (Cerna 2014;Roos 2019), and thus cannot explain this supranational expansion.Several studies stress the importance of Commission entrepreneurship and strategic framing in the area of migration (Baracani and Sarotto 2022;Menz 2015;Roos 2013).However, two important factors are hitherto missing in this literature: first, studies on supranational entrepreneurship have mainly focussed on the Commission's strategy rather than the process by which strategic action leads to a successful adoption of a policy in line with (or close to) the Commission's preferences.This is however crucial to understand the impact of Commission entrepreneurship, especially given Member State's reluctance to supranationalize migration policies.Second, the most pertinent studies on Commission entrepreneurship in the area of labour migration focus mainly on the Blue Card Directive where no supranational policy could be achieved in the sense that Member States are still free to maintain national labour migration schemes for highly skilled migration (Menz 2015;Roos 2013).
This paper adds to existing accounts by theorizing and testing a causal mechanism to connect Commission strategy to a supranational outcome.I argue that Commission entrepreneurship can enable an environment of 'quiet politics' (Culpepper 2010), where salient conflict is defused by discussing migration as a technical and economic matter in governance fora dominated by bureaucrats and business actors.I test this theory by process tracing the negotiations of the 2014 Intra-Corporate Transferees Directive (ICTD; Directive 2014/66/EU).The ICTD is one of three directives directly regulating labour migrant admission, next to the Blue Card Directive for highly skilled employees and the Seasonal Workers Directive.It introduced a supranational migration scheme for employees of multinational companies, who represent a small but substantial part of labour migration from third countries in Europe (OECD 2019, 127;Salt and Brewster 2022).Combining frame analysis and process tracing, I first identify Commission entrepreneurship, consisting of strategic policy choice, framing of the ICTD as a matter of business needs and international obligations rather than migration, and the inclusion of business representatives and trade policy-makers who supported the Commission's framing.I then trace the impact of Commission entrepreneurship on Member State preferences and the negotiation process, finding that the Commission's strategies enabled a low-salience negotiation environment in which sovereignty and control concerns were downgraded to technical questions of administrative flexibility.This enabled the adoption of an innovative supranational policy despite the Council's initial opposition, diverging Member State preferences and extensive implications for labour markets and migration control.The findings have implications for policy-making in other contested areas as they suggest that next to strategic framing, low salience is a key enabling condition of supranationalization.

EU migration policy between reluctant Member States and Supranational Entrepreneurs
The strong intergovernmental legacy of migration policy-making in the EU underpins the still dominant view that '[l]abour immigration remains a predominantly national prerogative' (Farcy 2020, 198).Initially, migration was an issue of trans-governmental coordination between interior ministers (Lavenex 2001).Gradual institutionalization only came with Treaty reforms starting in the 1990s, which provided first for intergovernmental policy-making in the Treaty of Maastricht's third pillar, and later for gradual communitarization since the Treaty of Amsterdam.Since the Treaty of Lisbon, migration policy is an established area of EU policy-making with full co-decision between the Council and the European Parliament (EP), yet current scholarship is still shaped by the policy area's strong intergovernmental legacy and emphasises both Member States' dominance vis-à-vis supranational institutions and their reluctance to give up national discretion (Caviedes 2016;Trauner and Ripoll Servent 2016).
The existing literature offers four main reasons for Member State reluctance to integrate migration policy: first and foremost, concerns of sovereignty and administrative control (Roos 2019, 570;Zaun 2022, 199).Member States want to control who is admitted to their territory and under what conditions, and thus oppose -sometimes out of principle -ceding these powers to the EU (Brinkmann 2012, 354;Maricuț-Akbik 2018, 170;Menz 2008, 98).This applies especially to labour migration, where Member States are not bound by strong international obligations, in contrast to asylum or family reunification (Roos and Zaun 2014, 61).Second, migration policy is prone to salient domestic politicization (Dennison and Geddes 2019;Grande, Schwarzbözl, and Fatke 2019).While highly skilled migration is generally less controversial than that of refugees or workers in low-skilled jobs (Cerna 2014, 83), public concerns over migration and labour market competition remain significant, as the discussions around the Blue Card Directive illustrated (BBC News 2007;Gümüs 2010, 448).Third, the goal of labour migration policiesattracting labour that is deemed scarce and 'wanted' -fosters competition rather than cooperation (Cerna 2014, 74;Menz 2008, 12;Messina 2007, 151).Finally, Member States are unwilling to incur the high administrative adaption costs implied by new EU rules, especially when pre-existing national policies differ from one another (Cerna 2014, 76).Consequently, if Member States support common labour migration policies at all, they strive for maximum flexibility and continuity of national rules (Brinkmann 2012, 355).Thus, received wisdom is that EU policies will only entail minimal changes to national policy, and reflect either pre-existing domestic policies of powerful Member States (Roos 2013) or a lowest common denominator (Cerna 2014).
However, this scholarly presumption of powerful and reluctant Member States cannot explain the growing number and reach of EU migration directives.Of the five directives that currently govern the admission of 'legal migrants' (the Family Reunification Directive, the Students & Researchers Directive, the Blue Card Directive, the Seasonal Workers Directive and the ICTD), only one -the Blue Card Directive -allows Member States to retain or design parallel national policies (European Commission 2019a, 64).In addition, the administrative procedures to obtain a residence permit are considerably influenced by EU rules through the Single Permit Directive.Hence, even if Member States often enjoy leeway as to how they implement EU legislation on labour migration, their discretion as to whether their migration policies are governed by EU rules is limited.
To address this puzzle, scholars have turned to research on supranational entrepreneurship which illustrates how the Commission strategically mobilises ideas, discourses and actors to shape policy outcomes (Rhinard 2018;Saurugger 2013, 898), relying on three main strategies: shaping the decision-making process in favour of the proposal, mobilising coalitions around the proposed solution, and strategic framing of the problem and the proposed solution (Capano and Tullia Galanti, 2021;Vesan and Corti 2021).Shaping the decision-making process includes 'strategies intended to reduce the risk aversion among decision-makers' (Vesan and Corti 2021, 7).In migration policy, a central strategy of the Commission has been policy choice.After the Council's swift rejection of a directive covering all types of non-EU labour migration in 2001 (Carrera and Formisano 2005, 2), the Commission started partitioning migration policy through a set of 'sectoral' proposals covering specific aspects of migration that were more tolerable to Member States, thus hoping to alleviate Member State aversion against EU-level legislation (Roos 2013, 72).Mobilising coalitions in favour of the proposed solution can entail involving different Commission Directorates-General (DGs;Blom-Hansen and Senninger 2021, 12) and stakeholders supporting the issues and ideas highlighted (Rhinard 2018).Existing research in the area of migration highlights the involvement of actors who depart from a narrow migration perspective.In the early 2000s, an expansion of involved Commission DGs, Council configurations, and NGOs contributed to migrant rights being framed as a matter of social inclusion and anti-discrimination in the single market, rather than Justice and Home Affairs in the realm of interior ministries (Guiraudon 2003;Lavenex 2006).With full co-decision in migration since the Treaty of Lisbon, the EP has become another important ally to the Commission's efforts to supranationalize migration policy (Bonjour et al. 2018;Ripoll Servent & Trauner, 2014).
Lastly, strategic framing consists of shaping the definition of a problem to privilege some solutions over others and underscoring their necessity in light of the identified problem (Hadj Abdou 2020) or shifting the focus from a political problem perception to a technocratic one (Zaun and Nantermoz 2023).In the area of migration, the Commission shifted to more strategic framing after the failure of the 2001 proposal on economic migration and has increasingly referred to the economic needs of the single market since the adoption of the sectoral approach (Lavenex 2019, 580;Menz 2008, 30;Roos 2013, 72).Menz (2015) finds that regarding the Blue Card and Seasonal Workers Directives, the Commission aimed to circumvent concerns over migration control or sovereignty by framing migration as a matter of competitiveness and economic growth.Such frames take 'a functionalist, apolitical stance' (Lavenex 2019, 579) and emphasize a"competition state logic" where the goal is to maximise competitiveness vis à vis other states 'by assuring a business-friendly climate' (Menz 2008, 9).
While we thus have considerable insights into the strategies employed by the Commission to limit Member State opposition, so far we do not know how these result in supranational policy outcomes despite Member States' persistent insistence on national control.The two most pertinent studies on Commission entrepreneurship in the area of labour migration (Menz 2015;Roos 2013) focus on the Commission's strategies rather than on the decision-making process.They argue that in the case of the Blue Card Directive, strategic framing equipped Member States with arguments to justify adoption domestically (Menz 2015, 564) and 'successfully linked immigration to the competitiveness of the single market in order to reinvigorate Member States' interest in common labour migration policy' (Roos 2013, 73).However, more detailed studies of the negotiation process are necessary to establish if the Commission's policy entrepreneurship was indeed the main cause for the successful adoption of common migration policies, or if other factors, such as interest alignment among Member States independent of the Commission's strategy, played a more important role.This is crucial since the Blue Card Directive did not produce a supranational outcome and most Member States continue to apply separate migration schemes for highly skilled immigrants (Kolb 2017).Hence, the Blue Card is arguably a most likely case for a successful adoption even among reluctant Member States.In the next section, I address this gap by hypothesizing that strategic action creates a depoliticized environment of 'quiet politics' (cf.Culpepper 2010) in which Member state opposition is minimized and compromise is facilitated.

The quiet politics of migration: how low salience facilitates supranationalization
The concept of quiet politics (Culpepper 2010) is helpful to explain how entrepreneurship can overcome opposition by pointing towards the changed rationales of decision-making under low salience.Quiet politics describes a mode of governance on topics of high complexity and low public salience, originally developed for the study of business influence in politics (Culpepper 2010).When the public does not care about an issue and issue complexity is high, politicians have few incentives to get involved, partisan contestation is less likely, and as a result both politicians and media defer to business actors with expertise, increasing their influence on policy outcomes (Culpepper 2010, 177-78).Culpepper differentiates between formal and informal modes of governance under quiet politics.Quiet politics under formal rules in legislative or regulatory arenas takes the shape of bureaucratic network governance, where interest groups 'work to establish ongoing ties with the bureaucracy in charge of regulating them' and are able to influence outcomes mainly through their expertise (Culpepper 2010, 182).Business influence is greatest in informal governance arenas that are neither legislative nor regulatory, where business groups can set or shape rules (Culpepper 2010, 180-81).
EU politics have long been marked by the sort of decision-making which Culpepper describes as bureaucratic network governance, with the Commission as a strong supranational bureaucracy and an originally mainly regulatory policy agenda.Bureaucratic network governance was also characteristic of the origins of migration policy in the EU, albeit with a strong dominance of national bureaucracies vis-à-vis supranational and non-state actors (Guiraudon 2000;Lavenex 2001).The growing communitarization of migration since 1999 enabled the Commission to form coalitions with policy-makers from other policy areas and non-state actors to further the integration of migration policies (Guiraudon 2003;Lavenex 2006, 206).As a result, up until the refugee reception crisis of 2015, EU migration policy-making was largely shielded from public contestation and national majoritarian institutions (Zaun 2022).However, low salience decision-making is not a stable state and needs to be carefully maintained (Morgan and Ibsen 2021), especially in contested areas like migration (Mach et al. 2021).Even under predominantly non-majoritarian decision-making, migration policy is prone to politicization, both by domestic opposition or media and by governments themselves who aim to benefit from publicly voicing opposition to EU rules (BBC News 2007;Roos 2019, 278).
Recent studies on active depoliticization on the side of the Commission have advanced our understanding of how a de-politicized environment can be created or maintained in contested policy areas (Bressanelli, Koop, and Reh 2020;Vlasiuk Nibe, Meunier, and Roederer-Rynning 2023;Zaun and Nantermoz 2023).In studying Commission entrepreneurship on the issue of investment screening, Vlasiuk Nibe et al. find that the entrepreneurial strategies of policy choice, purposeful actor involvement and an active steering of negotiations were instrumental to defuse potential political 'mines' at all stages of the negotiations (2023).Zaun and Nantermoz (2023) find that shifting the problem frame of refugee policy by promoting development aid as a 'technocratic' solution enabled the Commission to minimize conflict between Member States.Drawing on these insights, I hypothesize that strategic action in the policy initiation phase facilitates supranationalization by creating an environment of quiet politics at the negotiation stage.The hypothesized causal mechanism illustrated in Figure 1 contains the three elements of entrepreneurship identified in the literature: strategic policy choice, the selective inclusion of supportive actors, and the framing of the proposed policy as a market-enhancing, technical issue.Together, this facilitates low salience negotiations in which seemingly technocratic arguments voiced by businesses and the Commission figure strongly while potentially controversial aspects are muted, opening a window of opportunity for Member States to prioritise economic benefits over traditional concerns for sovereignty.Compromise is facilitated as the need to communicate a tough stance to domestic audiences is reduced (Caviedes 2010, 15).The ensuing reduction of Member State opposition enables compromise with the EP when the latter prioritizes supranationalization.

The case of intra-corporate migration
To test the hypothesized causal mechanism, I conduct an in-depth case study of the negotiations on the 2014 Directive on Intra-Corporate Transferees.The ICTD is a typical case for studying the mechanism linking Commission entrepreneurship to supranational policy outcomes since 'the hypothesized cause, outcome, and contextual conditions are all present' (Beach and Pedersen 2019, 90).Typical cases are particularly suitable for theory-building process tracing, which aims to establish the causal mechanism connecting a cause to an outcome (Beach and Brun Pedersen 2012, 19).The directive, negotiated between 2010 and 2014, created supranational migration rules for third-country managers, specialists and trainees employed by multinational companies who move to another branch of the same company inside the EU while keeping their initial employment contract.Previous studies support the assumption that at least two elements of Commission entrepreneurship were present: strategic policy choice, as the ICTD was part of the Commission's partitioning strategy for labour migration, and strategic framing, as the proposal was found to put forward a predominantly economic rationale (Maricuț 2016, 224;Roos 2013, 72).
As one of three legal migration directives regulating the admission of labour migrants, the ICTD is also a typical case in substantive terms: Although the directive only covers a comparatively small group of labour migrants, it is a central piece of EU migration legislation, has considerable implications for Member States' labour markets and implies high adaptation costs.Throughout the EU, more ICT Cards than Blue Cards were issued between 2017 and 2019 and in 2022 if the outlier Germany is not taken into account. 2While intra-company transfers are intended to be temporary, the ICTD allows for stays of up to three years, and the vast majority of Member States allow for the transition into another residence permit based on local employment (Deloitte 2018).What is more, the intra-EU mobility scheme introduced in the ICTD allows ICTS to move to -and work ina second Member State for at least 90 days without needing a second permit.This was a key innovation for EU migration policy (Klaus 2015, 1), and has since served as a blueprint for the enhanced intra-EU mobility scheme in the revised Blue Card and Students and Researchers Directives (Peers 2021).
Additionally, ICT migration has implications for local labour markets.ICTs can be either managers, specialists or trainees, but the great majority (ca.86% in 2021-2022) are specialists and thus in a comparable position to highly skilled local employees. 3Since ICTs retain their employment contract in the country of origin, transfers can be a way for employers to reduce labour costs for several reasons.First, ICTs remain subject to the social security system of their country of origin if there is a bilateral social security agreement with the host state (Verschueren 2021, 295).Second, it is possible that ICTs continue to work under the employment regulations of their home country (Verschueren 2021, 291), and equal treatment is only guaranteed at the level of the Posted Workers Directive, which at the time of the ICTD's adoption only covered the 'hard core' of minimum workers' protection (Verschueren 2021, 285), and is a lower standard than in other EU directives (the Single Permit Directive, the Blue Card Directive and the Seasonal Workers Directive, Friðriksdóttir 2017, 315).Finally, while the ICTD requires equal renumeration between ICTs and employees in 'comparable positions', it does not define what such comparable positions are nor what is included in renumeration (e. g. if sickness pay counts as salary, Verschueren 2021, 285-86).As a result, in some Member States ICTs only need to earn minimum wage or the average gross national salary (Deloitte 2018).Possibilities for using ICTs as a cheaper workforce arise especially when country of origin wages and social security provisions are lower, which is relevant given that Indian and Chinese nationals made up ca.57% of all ICT permits granted in 2021 and 2022. 4As such, ICTs are comparable to intra-EU posted workers, who are subject to long-standing and highly controversial debates (Lens, Mussche, and Marx 2022).Finally, the directive's transposition entailed high adaptation costs: only 14 out of 25 Member States 5 that later adopted the Directive had pre-existing ICT schemes, with different rules for admission conditions, length of stay, rights granted to ICTs and possibilities for family reunification (European Commission 2010a, 71).Summing up, the case of the ICTD is well suited to test whether the hypothesized causal mechanism of quiet politics enabled through Commission policy entrepreneurship is present in EU migration policy-making.
When tracing typical cases, it is important to consider the explanatory power of potential confounding and contextual factors (Beach and Pedersen 2012, 19).An important factor is the enhanced decision-making power of the EP since the Treaty of Lisbon, if we assume that the EP will use its veto power to push for supranational policies.Previous research has found that the EP has favoured strengthening migrant rights over supranational admission rules (Roos 2015) and has behaved more consensually towards the Council since gaining more power (Servent 2011).Thus, pre-existing knowledge on the causal effect of the EP's increased legislative power for the supranationalization of immigrant admission is inconclusive.
Another contextual factor which is specific to the ICTD is that intra-corporate migration is subject to international trade commitments, including in the 1994 General Agreement on Trade in Services (GATS) and in the EU's preferential trade agreements (PTAs; Lavenex and Jurje 2015).Under these agreements, sector-specific commitments allow the entry of ICTs for a fixed duration, and provide exemptions from labour market tests ensuring priority for EU nationals (WTO 2019, 17).However, these provisions explicitly rule out any commitments regarding long-term residence or access to the labour market (WTO 2009, 6), and apply to the national level (Jacobsson, 2013), and as such do not mandate EU legislation.In fact, other types of trade-related migration commitments regarding contractual service suppliers and independent professionals have not translated into EUlevel policies.Furthermore, how exactly Member States implement the GATS commitments is often unclear (de Lange, Tans, and Azhar 2021), and their factual enforceability is low as international legal disputes are almost non-existent (Tans 2015, 326).Thus, trade commitments by themselves do not provide an explanation for supranational outcomes.However, they can be strategically invoked by actors who want to frame the ICTD as a matter of international obligations.

Data and method
The analysis relies on a total of nine expert interviews and 80 manually coded documents from EU institutions (N = 74) as well as from media and interest groups (N = 6) spanning from 1994, when ICTs are first mentioned in EU policy documents, until the ICTD's adoption in 2014.This includes key strategy documents in the area of legal migration issued by the Commission and the Council before the ICTD proposal was issued (N = 11) and documents directly related to the negotiations such as Council minutes and trilogue tables, press statements and articles (N = 69).The expert interviews were conducted with policymakers from the European Commission, national ministries in three Member States (Germany, the Netherlands and Sweden), representatives of Business Europe and the European Services Forum and with one legal expert working for a Brussels-based immigration law firm.The main purpose of the expert interviews was to provide additional insights into Member State positions and the Commission's strategy, including the involvement of other DGs and external stakeholders.Secondary literature, including accounts of practitioners involved in the negotiations, further informs the analysis (Lazarowicz 2013;Loschert 2020;Maricuț 2016;Töttős 2018).
To analyse the gathered evidence, I combine process tracing methods (Beach and Pedersen 2019) with document and frame analysis.The analysis reconstructs the ICTD negotiation process to juxtapose the observable implications of the hypothesized causal mechanism outlined in Figure 1 with in-depth within-case evidence, based on a mix of deductive and inductive qualitative coding.A complete list of the coded documents and interviews as well as a detailed description of the coding process and coding reports can be found in the Appendix.

Tracing entrepreneurship: Policy Choice, Coalition Building, and Problem Framing
The analysis proceeds in two steps: first, I establish Commission entrepreneurship at the policy initiation stage, consisting of strategic policy choice, the involvement of favourable actors and strategic problem framing highlighting the market character of the ICTD while downplaying migration and labour market implications.Second, I analyse the impact of these strategies on the negotiation process, focussing on the salience of the negotiations as well as Member State and EP preferences and their justifications on key issues.Table 1 in the Appendix summarizes the main empirical observations.
The first element of Commission entrepreneurship -purposeful policy choice in order to minimise opposition -was clearly present in the case of the ICTD.Policy documents and interview evidence show that ICTs were considered a comparatively small group where 'very few people would have (. ..) security concerns' (Interview 4) or worry about labour market implications (Interview 3).This assumption was supported by previous developments, as ICTs -like seasonal workers -had been singled out as a category of labour migrants to be granted more favourable treatment in the Commission's 2001 proposal and in the 2005 policy plan introducing the sectoral approach, as well as in a non-binding Council resolution of 1994 concerning national-level policies (Council of the EU 1994;European Commission 2001, 2005).Member States had not contested this in previous negotiations (Interview 3), supporting the Commission's assumption that they were unlikely to oppose the creation of an EU directive (European Commission 2010a, 6)even though the negotiations turned out to be much more difficult than expected.
The purposeful involvement of favourable actors both within and outside the Commission also confirms the Commission's entrepreneurial approach.According to a DG HOME official, there was strong support and a 'push' (Interview 3) for a proposal on ICTs from DG Trade, whose officials were involved in drafting the proposal (Interview 4).Before issuing the proposal, the Commission conducted targeted consultations of 'relevant stakeholders' and commissioned a study by Ernst & Young (European Commission 2010b: 4).Business and employer representatives constituted four out of seven consulted stakeholders, including Business Europe, the Work Permit Foundation (an NGO backed by multinational corporations lobbying for work permits for ICTs' family members), the FEACO-Pendo group (a Federation of Management Consultancies Associations) and the American Chamber of Commerce.The other three were the European Trade Union Confederation (ETUC), the International Organization for Migration (IOM) and the Mission of Japan to the EU (European Commission 2010a, 4).Interviews with a Business Europe representative and a Brussels-based immigration law firm confirmed that large multinational companies were both the main beneficiaries of the directive, and its strongest supporters (Interview 1, Interview 6).The overrepresentation of such highly concentrated interests who would directly benefit from the ICTD arguably enhanced their ability to shape the proposal and its framing to a considerable extent (cf.Olson 1971).
In terms of policy content, the Commission proposal put forward ambitious rules for admission and intra-EU mobility, reflecting the interests of businesses.The proposal foresaw a supranational ICT admission scheme, leaving no room for parallel national schemes (European Commission 2010a, 6).On intra-EU mobility, the Commission proposed that ICTs could live and work in a second Member State for up to 12 months without an extra permit (European Commission 2010b, 28).Both suggestions would have substantially limited Member State discretion to regulate migration, and thus touched on a central aspect of Member State opposition.Conversely, multinational companies considered them the key benefits of the proposal (Euractiv 2011).
Regarding migrant rights, the Commission adopted a more minimalist stance after an internal discussion on which standards should apply to ICTs.The first option was granting ICTs full equal treatment with local employees concerning salary, working conditions and access to social security.This raised fears of re-opening the long-standing debate on reforming the 1996 Posted Workers Directive which did not grant full equal treatment (Interview 4).The second option was to apply the standards for posted workers in the EU to ICTs, implying only a minimum salary requirement and reduced workers' rights guarantees.Ultimately, DG HOME opted for the Posted Workers standard despite the potential for unequal treatment (European Commission 2010b, 26) in order to minimise Member State opposition (Interview 4).
In the Commission proposal and the accompanying impact assessment, strategic framing is evident in the way the policy problem is presented, in the treatment of evidence to underline the problem and assess the ICTD's potential impacts, as well as in the absence of migration frames.Hereby, the Commission discursively shifted the issue from migration to trade and economic policy.The proposal clearly emphasized the market rationale for the ICTD when describing its main objective: In face of the obstacles encountered by businesses in relation to the complexity and diversity of rules, the aim of this Directive is, in particular, to facilitate intra-corporate transfers of skills both to the EU and within the EU in order to boost the competitiveness of the EU economy, and to complement the set of other measures the EU is putting in place to achieve the goals of the EU 2020 Strategy.(European Commission 2010b, 2) The directive was thus presented as business-oriented instrument embedded in a wider strategy for economic competitiveness, rather than an migration policy instrument.Figure 2 gives an overview of the arguments in the proposal and the accompanying impact assessment to justify the need for the directive.
The main problems the ICTD sought to address according to the proposal were a shortage of highly skilled employees, the bureaucratic barriers which corporations encounter when they want to send their employees to another country, and the need to enhance the EU's competitiveness and attractiveness for businesses.Next to business needs, trade (GATS or PTA) obligations regarding ICT admissions were mentioned most often as a reason for the ICTD's necessity.In the Impact Assessment accompanying the proposal, the Commission frames intra-corporate migration as a matter of trade policy by stating that 'the conditions under which ICTs can perform their economic activity fall under the Common Commercial Policy (which is an area of exclusive EU responsibility)' and argues that differences across national ICT schemes may 'hamper the uniform application of the international commitments' of the EU (European Commission 2010a, 19).The proposed solution to this problem however goes much further than the specific GATS and PTA commitments, implying that the aim of uniform application is invoked strategically, by making regulation at the EU level seem inevitable because of international commitments.
The Impact Assessment strongly emphasises the differences between national schemes as a problem the ICTD should address.In presenting the ICTD's economic and social impacts and how it contributes to the goal of supporting EU businesses' development, evidence is treated selectively, clearly illustrating the deference to business 'expertise': the complexity and rigidity of national legislation is illustrated with a single example of a Japanese manager having to wait more than 12 months to receive all documents for his transfer to Italy (European Commission 2010a, 16).Strikingly, where the Impact Assessment refers to trade union concerns about the potential for unfair competition between EU workers and third-country ICTs, these are dismissed as 'based on anecdotal evidence from Trade Unions and mainly fuelled by the experience drawn from the EU posting of workers legislation.However, as ICTs are typically qualified and well-paid workers, exploitation is usually not a concern for social partners' (European Commission 2010a, 17).In the proposal, labour market competition is only alluded to in a cursory fashion (European Commission 2010b, 3).This inconsistent treatment of evidence is another example of strategic framing, especially given that the Commission extensively discussed this matter internally (Interview 4) and was well aware of potential concerns.
Even more striking is the complete absence of migration-specific frames.A text search of the proposal shows that the words '(im)migration' or '(im)migrant' are only used in the introduction and the recitals, mostly when referring to other policy instruments -in the legislative text, the words do not appear once.The predominance of economic frames and business interests is also reflected in the distinction made between ICTs and 'typical migrants' in the impact assessment, which stresses that ICTs 'are typically specialists and managers, possessing uncommon knowledge specific to the company', and that 'the decision to post an ICT is not the result of a personal decision from the migrant but is purely demand-driven and depends on business needs' (European Commission 2010a, 6).Oddly, though described as highly skilled and wanted, ICTs are denied any agency; the fact that they often stay for several years and might -like other migrants -want to stay longer is entirely muted.Security and migration control or enforcement, usually central notions of EU migration policies, are all but absent, except for references made to the temporary nature of intra-corporate migration and the 'efficient management of migration flows' (European Commission 2010b, 13).This testifies to the strategy to present ICT migration as a non-salient, specific and technical matter, rather than a matter of sovereignty, control or labour markets.

The impact of entrepreneurship: negotiating in a quiet politics environment
During the negotiation process, the market framing of the ICTD put forward by the Commission and the understanding that the Directive concerned a technical, rather than a political matter, largely prevailed in the Council.This facilitated the maintenance of low salience and a high degree of concern for business interests characteristic of the bureaucratic network governance mode of quiet politics.Quiet politics prevailed even as the negotiations turned out long and difficult due to strong disagreements between several Member States, as well as between the Council and the EP.Negotiations mostly took place at the technical rather than the political level: most Council negotiation documents emanate from the Working Party on Integration, Migration and Expulsion, composed of experts from national ministries.There was little involvement of labour ministries or trade unions. 6The market framing and disproportionate representation of business needs are also reflected in the discussions in media and wider policy circles: while the European Trade Union Federation (ETUC) released a press statement in opposition of the Directive (ETUC 2010), business advocacy groups and employer associations weighed in more actively through statements, press interviews, and an open letter to MEPs (AmCham et al. 2011;Euractiv 2010aEuractiv , 2010bEuractiv , 2011;;POLITICO 2012).Press coverage in Brussels based outlets emphasized the business rationale and echoed lobbying efforts for the inclusion of family members' right to work in the ICTD (Euractiv 2010a, 2010b; POLITICO 2012) -though one article also cited trade unions' wage competition concerns and fears of migration backlash (Euractiv 2011).The limited coverage of the directive supported the businessoriented framing of the policy and increased negotiators' ability to compromise without facing significant public scrutiny.This environment did not erase Member State opposition to a supranational ICT scheme.However, Member States by and large recognized the economic added-value of the Directive, and contestation focused on lower salience aspects such as flexibility and administrative control, rather than higher salience issues like public opposition and sovereignty.An early Council press release emphasises a strong preference for flexibility in implementation as well as concerns for labour market implications (Council of the EU 2010b, 8), whereas concerns for public security or public opposition to migration or its supranationalization are completely absent from the analysed Council documents.Sweden, Spain and the Netherlands strongly supported the directive as a way to attract highly skilled migrants and elevate their national policies to the EU level (Loschert 2020, 149;190, Interviews 7, 8).Still, agreement in the Council was far from self-evident, as two Member States -Austria and Hungary -opposed the directive and questioned EU competence to even regulate ICT admission rules (Council of the EU 2011a, 2014a, 2-3).However, the opposed camps did not bring the controversies to the level of the European Council or the public (Maricuț 2016, 222-23).
In contrast to sovereignty concerns, the notion of competition for companies and skilled labour impacted Member States' positions significantly but was both a hindrance and a facilitator for supranationalization.The Commission's argument that a supranational ICT scheme would enhance competitiveness of the EU as a whole did not convince Member States, who preferred the possibility to maintain national schemes (Council of the EU 2013, 41) in order to keep more permissive rules as a competitive advantage over other Member States (Council of the EU 2011b: 7; Interviews 7, 8; Loschert 2020, 149).Thus, similar to the Blue Card negotiations, Member States remained eager to preserve their own competitiveness, rather than elevating this concern to the EU level (Roos 2013, 74).The competition argument did catch on for the issue of intra-EU mobility, however.The Commission proposal implied extensive mobility rights and a considerable loss of control over migration amid significant differences in wage and working conditions between Member States.This would have required a high level of mutual trust in the issuance of residence permits and the examination of wage and working conditions, of which Member States were sceptical (Interviews 4, 7).However, in contrast to the question of parallel schemes, the Commission was able to convince most Member States that intra-EU mobility was the core added-value of the directive (Interviews 3, 4, 7, 8), and a key innovation where negotiators 'for the first time since the creation of Schengen [had] the feeling they are doing something new' (Interview 4).The possibility to work in several Member States was the main selling point for an EU-level directive rather than separate national rules, as it made 'the European Union as a whole more attractive for intracorporate transferees' (Interview 7).The advantages of freer mobility ultimately outweighed concerns over control and mutual trust and convinced Member States to agree on a compromise between the Commission's liberal proposal and the Council's initial preference to allow second Member States to reject even short-term mobility of up to 90 days (Council of the EU 2013, 89;Lazarowicz 2013, 3).The final compromise foresees stays of up to 90 days without a new permit, and leaves Member States the option to require a new permit for longer term mobility (Verschueren 2021, 288).
Similarly, while Member States initially wanted to restrict ICTs' access to family benefits (Council of the EU 2011d, 9), the Netherlands -after lobbying efforts by the Permit Foundation, with which officials from the Dutch Ministry of Justice had been in regular contact -brought forward a proposal to allow family members to work as a way of making the scheme more attractive (Council of the EU 2010a, 33, Interview 7).This was accepted without much discussion, testifying to the prevalence of the market over the migration frame, where economic rights are a tool to increase the ICTD's attractiveness for companies.
Though strategic market framing alleviated sovereignty concerns, Member States maintained their strong preference for reducing adaptation costs not only on the central question of parallel schemes, but also for the more detailed rules of the ICTD.The strong emphasis on flexibility in the application of new rules is illustrated in numerous 'may clauses' introduced by the Council, especially concerning admission conditions (Lazarowicz 2013, 2).These included the possibility for more restrictive rules, such as a 'cooling off' period of up to 3 years before a new application could be launched (Council of the EU 2013 79), but also for more permissive rules such as reducing the required time of previous employment within the company from twelve to three months (Friðriksdóttir 2017, 277).The preference for flexibility also shaped the Council position on ICTs' rights as employees and migrants.Even though ICTs' access to work and social benefits and the risk of abuse of the ICT scheme for wage dumping took up a substantial part of the discussions, Member States accepted the application of the Posted Workers standards which allowed for more national discretion given that the Posted Workers Directive at the time was confined to minimum guarantees.
The EP was in favour of a supranational ICT scheme and extensive intra-EU mobility rights, but was less convinced by the strong business-friendly framing put forward by the Commission.The draft report centred the question of migrant rights, demanding full equal treatment of ICTs and local employees (Friðriksdóttir 2017, 349).However, the EP sided with the Commission in its stance not to allow for parallel national schemes for ICTs so as not to undermine the goal of a supranational policy (European Parliament 2011, 20).The question of equal treatment remained an outstanding issue until the final negotiation phase (Council of the EU 2014b, 13).For the EP, this was also driven by its ambition of pushing for a reform of the Posted Workers Directive, which meant it was not willing to take it as a benchmark for the ICTD (Lazarowicz 2013, 2, Interview 4).Positions were split between the two involved committees: while the Employment Committee (EMPL) insisted on equal treatment (European Parliament 2011b), the Committee for Civil Liberties and Justice and Home Affairs -(LIBE) whose rapporteur was in charge of the trilogue negotiations -prioritized the adoption of a supranational migration scheme with facilitated intra-EU mobility, which the Council was willing to compromise on in the end (Maricuț 2016, 228-30).The EP won some concessions on rights -the final compromise foresees equal treatment in terms of salary and access to employment-related social benefits, as well as family benefits after nine months of stay.However, in other respects the Posted Workers standards apply, and the ICTD allows for ICTs to remain subject to the employment and social security law of the country of origin in many cases (Verschueren 2021).In sum, the LIBE committee in charge ultimately prioritised the adoption of a supranational ICT scheme and compromised on migrant rights.The directive was however only narrowly adopted in the EP plenary: while most liberal and centre right MEPS voted in favour, centre-left, left and green MEPs mostly rejected the Directive (Parltrack 2014).
Overall, the ICTD negotiations confirm the presence of the hypothesized causal mechanism, where Commission entrepreneurship provided the conditions for the Council and the EP to negotiate in a mode of quiet politics, confined mostly to the expert level with little exposure to public discourse and considerable weight given to business interests.As such, the effect of strategic action was not so much to change Member State preferences or provide justifications for domestic audiences as suggested by previous research, but to create an environment where public opposition was less relevant, making it difficult for opposing voices to politicize the labour market or sovereignty implications of the ICTD.

Conclusion
This article has analysed the negotiations leading up to the adoption of the ICTD to understand how Commission policy entrepreneurship can enable supranationalization in the contested policy area of migration.I have argued that to answer this question we must not only identify strategies, but also study their effects on negotiations.The analysis thus builds on existing studies on Commission entrepreneurship in the area of labour migration by studying the mechanism through which policy entrepreneurship facilitates supranational outcomes.I find that while Commission strategic action only changed Member State preferences in a limited way, it provided for a low salience environment in which disagreements between Member States were not politicized, facilitating compromise with the EP.The findings have two main implications for the study of Commission entrepreneurship in the area of migration: first, they show that paying more attention to the decision-making process can show how the Commission's strategies reverberate in the policy formulation phase -whether this is through persuasion, by providing justifications for domestic audiences or by reducing the need for such justifications and the possibilities for opposed actors to politicize the issue, as was the case for the ICTD.Second, the findings imply that the mode of decision-making, including the level of salience and the involvement of business interests, are crucial factors for the supranationalization of labour migration policies in the EU.
The findings also point to avenues for further research.First, the divisions within the EP show that it cannot be taken for granted as an ally for supranationalization, and that attention needs to be paid to party and inter-committee dynamics.Second, further research is needed to establish whether the same causal mechanism identified here also applies to other cases.While the present single case study cannot make claims to generalizability, a cursory look suggests that low salience and the absence of competing policy frames could well be important factors in other cases too: the Seasonal Workers Directive also excludes parallel national schemes (Costello and Freedland 2016, 56), and here too, the Commission actively framed the policy as a 'rational and easily manageable tool' in service of the EU's competitiveness, while emphasizing its temporary character and catering to both the left and the right in an effort to de-politicize the issue (Menz 2015, 565).In contrast, the 2021 Blue Card Directive reform was negotiated in a context of heightened politicization of migration since the refugee reception crisis of 2015, and still falls short of a supranational migration scheme (Peers 2021).Importantly, its scope was expanded to recognized refugees, and hence beyond the 'technical' realm of highly skilled migration in service of the economy.Perhaps not surprisingly in this climate, the latest legislative proposals on legal migration focus on reforming already existing directives on the administrative procedures for work and residence permits and on rights of long-term residents, while the admission of new migrants is dealt with in policy initiatives short of legislative proposals (European Commission 2022).
The present account of labour migration policy-making centres on the conditions for supranationalization, whereas a discussion of whether and when such supranationalization is normatively desirable is beyond the scope of the analysis.However, the study reveals two conditions that should be further scrutinized: first, the success of economic frames for propelling migration policy seems to imply the necessity of discursively constructing migrants as a special group of 'expats' and company assets rather than human beings with agency.Second, the fact that successful supranationalization required limiting migrant rights underlines the risk that the pursuit of supranationalization might come at the cost of equal rights for non-EU nationals, even the most economically 'desirable' ones.

Figure 2 .
Figure 2. Commission proposal and impact assessment: justifications of the need for the ICTD.Note: the figure displays the share of mentions of each reason within the total number of coded arguments in the respective document.The arguments are shown in descending order by the sum of shares in both documents.Gr = Groundedness; number of times the code was attributed to a quote throughout all coded documents.

Notes 1 .
Own calculation based on Eurostat [migr_resocc], [migr_resict1_1] and[migr_resfirst]  for the years 2018-2021 and 27 Member States (excluding the United Kingdom).The calculation method follows an earlier calculation by the European Commission(European Commission  2019b, 209).2018 is chosen as a starting point since the ICTD and the Seasonal Workers Directive had to be transposed in 2017.2.Own calculation based on Eurostat ([migr_resbc] and [migr_resict1_1]).Germany has beenissuing around 80% of all Blue Cards in recent years, and is the only big Member State using the Blue Card as the main entry channel for highly skilled migration(Kolb 2017, 16).More Blue Cards than ICT cards were issued in the years 2020 and 2021; however this is likely a consequence of the COVID-19 pandemic.3.Own calculation based on Eurostat[migr_resict1_3]. 4. Own calculation based on Eurostat [migr_resict1_1]. 5.The United Kingdom, Ireland and Denmark did not adopt the ICTD because of their opt-outs from EU immigration policies (see recitals 47 and 48 of the ICTD).6.The Social Questions Working Party in the Council was invited to draft an opinion, which was then integrated by the Presidency into a compromise suggestion in April 2011 (Council of the EU 2011c).