Policy diffusion across political ideologies: explaining Denmark’s desire to externalise asylum

Abstract

In 2021, the Danish Social Democratic government tabled a bill allowing asylum seekers to be transferred to another country to process claims and provide protection. Witnessing a Social Democratic government embracing this highly controversial idea, even though less than a handful of right-wing governments outside Europe had previously outsourced asylum, is puzzling. Policy diffusion theory is used to explain the underlying mechanism and analyse the justification strategies. In the face of public attention focused on the refugee protection crisis between 2014 and 2016, the Social Democrats embraced the idea of externalising asylum in order to be recognised as responsive problem solvers. Essentially, they employed three strategies to justify the policy and to mask its origin and its inconsistency with the party’s ideology: (1) referring to policy advisors as the original source of the idea; (2) reframing externalisation as a humanitarian project; and (3) shifting the narrative about Social Democratic identity.

Keywords:

• Asylum
• Denmark
• externalisation
• policy diffusion
• social democracy

The idea of externalising refugee protection has regained traction among Western European governments over the past few years. Externalisation usually describes the aim to transfer responsibility for asylum processing and, in several cases, the subsequent provision of protection from a host state to another country to which the person often has no personal connection.¹ In April 2022, the United Kingdom, for example, closed a bilateral agreement with Rwanda to transfer asylum seekers arriving in Britain to the East African country (The Economist 2023). Similarly, in November 2023, Italy announced an agreement with Albania to set up detention and processing centres in the latter country, primarily for people trying to reach Italy by boat (Carrera et al. 2023). Nevertheless, while analogous externalisation models have been tested overseas for several decades (Ghezelbash 2018; Koh 1994), Denmark was the first European state to adopt a bill allowing the government to direct asylum seekers to another country to process their claims and subsequently grant them protection (Rausis 2022a, 2022b). Bill L226, which introduced the possibility of externalising refugee protection, passed the Danish Folketinget (parliament) with a clear majority of votes in June 2021.² In clear opposition to the bill, the United Nations High Commissioner for Refugees (UNHCR) (2021), the European Commissioner for Home Affairs Ylva Johansson (2021), and the African Union (2021) declared that such externalisation policies undermine the principle of responsibility-sharing among states and run against the spirit of the 1951 Refugee Convention. The most recent diffusion of externalisation policies in Europe has raised widespread concerns that more states could follow Denmark’s example and end the traditional asylum system. This apprehension is based on the observations that restrictive asylum policies, which gained popularity in Europe in earlier decades, eventually spread across states of the Global North and, ensuingly, also became popular in states of the Global South (Rausis 2023). Generally, the politicisation of refugee protection and the rise of illiberal asylum policies has been viewed as threatening the liberal international order (Börzel and Zürn 2021).

From a political science perspective, witnessing a Social Democratic government embrace a highly controversial policy that had previously only been introduced by a handful of right-wing governments outside Europe,³ and stands in conflict with the party’s universalist ideological foundations, is puzzling. For the Danish Social Democrats, Bill L226 signalled a major policy U-turn, as leading party members have repeatedly taken a clear stance against the idea of outsourcing asylum in the past. In 2014, Mette Frederiksen, then minister of justice, qualified the idea put forward by the Danish People’s Party to transfer responsibility for asylum processing to an East African country to be ‘at best a bit comical and at worst very, very problematic’ (Frederiksen 2014). Yet, after the Social Democrats lost power in 2015 and Frederiksen became opposition leader, she brought the idea of externalising refugee protection to the top of the party’s immigration programme dubbed ‘Fair and Realistic’ (Socialdemokratiet 2018).

The idea of externalising refugee protection dates back to the 1980s but has been revived regularly since the early 2000s (Lemberg-Pedersen 2019; Noll 2003). However, in Europe, normative constraints and the unwillingness of prospective partner countries to play their supposed role have hindered attempts to turn the desire to externalise asylum into actual legislation or agreements (Léonard and Kaunert 2016). The environment shifted when Denmark’s government tabled Bill L226 in April 2021, and the United Kingdom clinched a bilateral deal with Rwanda one year later. Nevertheless, Denmark and the United Kingdom have not sent any asylum seekers for asylum processing and refugee protection outside of Europe to the date of writing. By contrast, Australia, the United States, and Israel have managed to give effect to their policies of externalisation. Most of these arrangements, however, have been terminated in the face of practical problems and legal challenges (Shachar 2022).

The proliferation of externalisation policies suggests a dynamic of policy diffusion across jurisdictions. Policy diffusion occurs when policy-making in one country is influenced or inspired by policy-making in others (Braun and Gilardi 2006; Graham et al. 2013). In the Danish case, externalisation policies abroad have influenced Denmark’s attempt to externalise refugee protection during all phases of the policy process. In the policy formulation phase, the Australian externalisation model served as the primary source of inspiration (Immigration and Integration Committee 2016; Interview AL). In the policy adoption phase, both the Australian and US externalisation models were cited to justify Denmark’s plans (Tesfaye 2021). In the policy implementation phase, finally, the ‘EU-Turkey deal’ and the bilateral agreement between the United Kingdom and Rwanda have become particularly relevant (Interviews AF; TF1; TF2).

While policy diffusion seems to have been at play in the Danish policy U-turn, it remains puzzling that a Social Democratic party took inspiration from right-wing parties in adopting a policy that runs against its ideological foundations. Researchers of party policy diffusion emphasise that decision-makers are generally unwilling to adopt policies by their ideological counterparts (e.g., Böhmelt et al. 2016; Grossback et al. 2004; Pereira 2021). This reluctance is arguably driven by the insight that deviating from a previous ideological position potentially confuses voters and causes voter loss (Budge 1994; Robertson 1976). Policy diffusion across political ideologies is generally considered explicable only when policies are particularly successful in solving pressing policy problems (Butler et al. 2017). Specifically, Social Democratic parties adopting policies of populist radical-right parties face the risk of losing credibility (Bale et al. 2010).

Viewed through a policy diffusion lens, these developments prompt the questions of how and why the Social Democrats turned to the idea of outsourcing asylum even though only its ideological counterparts abroad had previously introduced the policy. In this article, I tackle this research puzzle by addressing the following two questions:

1. How can we explain the Social Democrats’ decision to table the proposition to externalise refugee protection in the Danish parliament in 2021?⁴

2. What justification strategies have the Social Democrats used to adapt the idea of externalising refugee protection to their political ideology?

Both questions carry important theoretical and empirical implications for the literature on policy diffusion and migration studies. Theoretically, studying Denmark’s attempt to externalise refugee protection offers new insights because it ostensibly contradicts the accepted wisdom in the party policy diffusion literature. On the empirical level, the Danish case is highly relevant because externalising refugee protection runs counter to the basic idea of responsibility-sharing and the spirit of solidarity in refugee protection. These ideas are central to the Global Compact on Refugees, which was approved by the United Nations General Assembly in 2018 and aims to govern refugee protection at the global level.

As a first step to solving the research puzzle, I detail the process and the underlying motivations that led to the adoption of the externalisation policy by the Social Democrats. I employed the process-tracing method (Beach and Pedersen 2019; Hall 2006; Starke 2013) to identify the mechanism that led to the adoption of the externalisation policy in Denmark. To this end, I conceptualised and operationalised every possible explanation before the exhaustive collection of data that was followed by the systematic comparison of the theoretically deduced expectations for each potential explanation with the empirical fingerprints of the case at hand. The analysis of all four potential explanations of the dominant set of diffusion mechanisms (Simmons et al. 2006), which consists of three mechanisms conceptualised referring to rationalist assumptions, reveals the limitations of rationalist accounts to explain why the Social Democrats embraced the externalisation policy. Therefore, I turn to a novel policy diffusion theory (Blatter et al. 2022) that provides a nuanced understanding of both rationalist and social constructivist mechanisms and pathways to solve the research puzzle.

As a second step, I analyse the justification strategies applied by the Social Democrats to mitigate the risks associated with the policy change. To this end, I compare the public justification for the outsourcing of asylum with the actual driver. For both research steps, I draw from a rich body of empirical sources, including parliamentary debates, committee reports, newspaper articles, and expert interviews with observers and key actors.

The remainder of the article is structured as follows. First, I summarise the substance of Bill L226. Next, I present the theoretical frameworks of policy diffusion that I use to examine the process and the motives that made the Social Democrats turn to the idea of externalisation. Then, I outline the strategies used by the Social Democrats to justify the highly controversial policy. Finally, in the conclusions section, I discuss how this case study feeds into policy diffusion theory and current explanations on the rise of restrictive asylum policies in liberal democracies. The Online Appendix provides detailed information on how the methodological challenges of policy diffusion studies was addressed and how this study can be located within qualitative research traditions (part A); moreover, a systematic overview of the conceptualisation, operationalisation, and the most relevant empirical evidence for every diffusion mechanism and pathways that could be at play (part B) are provided along the list of all expert interviews conducted for this case study (part C) and an overview of the empirical sources that have been analysed for the case in point (part D).

The substance of Denmark’s Bill L226

Denmark’s Bill L226 foresaw the possibility of transferring asylum seekers to another country for asylum processing and refugee protection based on a future bilateral agreement with that country. Exemptions from this rule were only anticipated when such a shift of responsibility for an asylum seeker would be ‘in conflict with Denmark’s international obligations’ (Folketinget 2021). According to the legal note, two models of cooperation with a partner country are possible. In model 1, the asylum processing and the reception of people seeking protection in the partner country are provided by Danish authorities and under Danish law. In model 2, the processing and reception are provided by the authorities and within the legal framework of the partner country.⁵

Remarkably, the bill passed parliament even though Denmark had yet to conclude agreements with any country to take over responsibility for people seeking asylum in Denmark. That being said, two days before the bill was presented in parliament on 29 April 2021, Mattias Tesfaye, then minister of immigration and integration, as well as the minister for development cooperation, Flemming Møller Mortensen, signed a Memorandum of Understanding with Rwanda promising to strengthen cooperation on asylum and migration between the two countries. However, it took until September 2022 for the Danish and Rwandan ministeries of foreign affairs to announce in a joint statement that they would work to sign a bilateral agreement allowing asylum seekers arriving in Denmark to be transferred to Rwanda.

Several actors drove the process that led to the parliamentary approval of Bill L226. In the policy formulation phase, Anders and Morten Lisborg, two policy advisers operating the consultancy firm Migration Management Advice, were hired by the Social Democrats and tasked with analysing the Danish asylum system and proposing alternative policies (see Bruun 2017). In the policy adoption phase, the political process was driven primarily by members of the Social Democratic Party. In the policy implementation phase, the process was led by an inter-ministerial task force consisting of members of the ministry for immigration and integration, the foreign affairs ministry, and the justice ministry (Interview TF).

Policy diffusion mechanisms

Externalisation models tested in other countries influenced Denmark’s externalisation plans during all phases of the policy process (Immigration and Integration Committee 2016; Interviews AF, AL, TF1, TF2; Tesfaye 2021). Therefore, I turn to policy diffusion theory⁶ as an analytical framework to identify the underlying processes and drivers by which foreign policies have affected the Danish externalisation plans. In doing so, I focus on the processes that predate the tabling of Bill L226 but do not examine the role and relevance that foreign policies and international court rulings had on the implementation of the law.

Policy diffusion studies are well-suited for this analysis because they are based on the presumption that policy choices in one state are not only determined by domestic actors and processes within that state but substantially influenced by transnational actors and processes that take place beyond national borders (Braun and Gilardi 2006; Graham et al. 2013). Because diffusion studies focus on the process of policy-making, the mechanisms are at the centre of this research field. The dominant set of diffusion mechanisms consists of four now widely accepted mechanisms: coercion, competition, learning, and emulation (Simmons et al. 2006). Whereas three of these were traditionally conceptualised concerning rationalist assumptions, only one – emulation – refers to perceptions of appropriateness and is viewed to have roots in social constructivism (Gilardi and Wasserfallen 2019).

The empirical analysis reveals, however, that rationalist mechanisms cannot fully explain why Denmark’s government has turned to the idea of externalising refugee protection. Therefore, in the empirical sections, I complement the possible explanations provided by the dominant set of mechanisms with the social constructivist mechanisms of the diffusion theory by Blatter et al. (2022). The authors take rationalist and social constructivist approaches equally into account, and diffusion processes that take place in symmetric constellations are distinguished from those taking place in asymmetric constellations. By adding a third analytical dimension, each of these diffusion mechanisms is further divided into two pathways, distinguishing processes in which diffusion effects are limited to one policy field from those that have effects across policy fields. Thus, applying both sets of diffusion mechanisms allows a balanced and nuanced set of explanations for rationalist and social constructivist approaches. Each mechanism and pathway used for the analysis is introduced in the empirical section before it is applied to the case at hand.

In the following sections, I first test the explanatory power of the rationalist mechanisms and then turn to the social constructivist mechanisms.

Missing preconditions and effects: the limits of rationalist explanations

Do rationalist mechanisms of the orthodox diffusion theory explain why the Social Democrats turned to the idea of externalisation? While the empirical evidence provides, at first glance, support for some rationalist explanations, others can be ruled out because the necessary preconditions are missing. In the Danish case, we can safely exclude that either coercion or competition explains the diffusion process across political ideologies. Technocratic or political learning, by contrast, are more promising explanations for the Social Democrats’ decision to adopt an exclusionary asylum policy that stands in conflict with the universalist foundations of the party.

Political conditionality or pressure from a supranational entity such as the European Union (EU) that underpins the mechanism of coercion (Simmons et al. 2006) could explain why a candidate- or member-state adopts a policy that is not in line with the ideology of the ruling party. However, in the case of Denmark, we can exclude coercion as a possible explanation because the key precondition – the demand for externalisation from the EU or some other international actor (Gilardi and Wasserfallen 2019) – is absent. On the contrary, all relevant international or supranational actors have generally taken a position against Denmark’s externalisation plans.⁷ In any event, Denmark’s opt-out clause from the EU asylum acquis means it is hard to see how the EU could pressure Denmark to adopt a particular asylum policy.

Likewise, we can exclude the possibility that the policy change of the Social Democrats results from competition. In the context of asylum, competition reflects a process during which the adoption of a restrictive policy by one state is motivated by externalities resulting from the prior adoption of that policy by another state (Ghezelbash 2014). In order to stand in competition with each other in this area, the relevant states would need to lie on the same migratory route. Yet, because Denmark was the first mover in seeking to externalise refugee protection in Europe, this precondition is not met. Moreover, before Denmark, only non-European states had outsourced asylum, namely Australia, the United States, and Israel. Importantly, there is neither an indication that the adoption of outsourcing asylum by these non-European states had a significant impact on the number of asylum seekers in Denmark nor that decision-makers in Denmark held such beliefs.

We can also dismiss the possibility that Denmark’s government tabled the externalisation policy because a neighbouring state adopted a similar restrictive asylum policy that had proven bad for Denmark before the Social Democrats took interest in the idea of outsourcing asylum. In fact, the states with arguably the greatest likelihood of generating policy externalities affecting Denmark – the neighbouring countries Sweden and Germany – demonstrated more openness between 2014 and 2016 (Hagelund 2020; Ostrand 2015).

Similarly, none of the empirical evidence lends credence to the notion that the inner circle of the Social Democrats anticipated a widespread adoption of externalisation policies by European states in the future. Only Denmark’s opt-out from the EU asylum acquis offers the country the avenue to potentially transfer asylum seekers to another country with whom the applicants lack any substantial connection (see Tan and Vedsted-Hansen 2021). All other EU member states are bound by the directives and regulations that form the Common European Asylum System.

Trumping ideology with efficiency?

Technocratic learning could explain the spread of policies across political ideologies. Research has shown that policies that are particularly successful in solving pressing policy problems can travel across party ideologies (Butler et al. 2017). Simmons et al. (2006) define the diffusion mechanism of learning as the process by which decision-makers update their beliefs about the effectiveness and efficiency of policies. In its minimal version, the causal mechanism of learning consists of three elements. First, an external shock lays bare the insufficient effectiveness and efficiency of a state’s current policy and the responsible authorities’ and decision-makers’ observation of this objective problem pressure. Second, an exchange with other state authorities that are expected to have policies better suited to solve an existing problem pressure. Third, the adoption of this state’s policy because of its superior efficiency and effectiveness is expected to have desired policy consequences by reducing the problem pressure for the adopting states.

At least at first glance, all the necessary conditions that underpin technocratic learning can be found in the Danish case. First, the sharp increase in asylum applications in Denmark between 2014 and 2016 (Eurostat 2022) came as an unexpected external shock and put tremendous pressure on the Danish asylum system. The ineffectiveness of the policies adopted by the incumbent Danish government was readily observable to public officials and political decision-makers but also to the general public. In search of policy alternatives, members of the Danish parliament’s immigration and integration committee, including Dan Jørgensen, a Social Democrat, visited Australia in August and September 2016 to learn more about its practice of externalising refugee protection (Immigration and Integration Committee 2016).⁸ Adding the last missing element to this causal chain, one could argue that the Social Democrats embraced the idea of outsourcing asylum because they had learned through this transnational knowledge transfer that the Australian way of governing asylum is more efficient and effective.

From the administrative perspective, an efficient and effective asylum policy should fulfil several criteria. First, it should be effective in reducing the problem pressure and, thus, the number of asylum seekers arriving in the country. Moreover, it should be cost-efficient so that the costs do not exceed the foreseen budget item for asylum. Finally, it should be in line with a state’s legal obligations in refugee protection. Yet, when it comes to the externalisation of refugee protection, an evaluation of the Australian model and other attempts to outsource asylum shows that these criteria are not always fulfilled.

At first sight, the Australian externalisation model in place when members of the Danish delegation visited in 2016 (the ‘Pacific Solution II’) appeared to have worked. The policy (established in 2013) involved outsourcing of asylum processing and subsequent protection from Australia to Papua New Guinea and Nauru (see Ghezelbash 2018). The number of boat arrivals had steadily increased in the years before its implementation, surpassing 20,000 people in 2013 alone. Yet after 2013, the number of those arriving by boat in Australia seeking asylum plummeted to just a few hundred per year (Phillips 2017).

On closer examination, however, the track record of Australia’s externalisation model is questionable. Researchers have shown that intercepting boats, not outsourcing asylum, has been the decisive factor in reducing the number of boat arrivals in Australia (Gleeson and Yacoub 2021). In addition, studies have put the financial costs of externalising refugee protection to Papua New Guinea and Nauru at more than A$1 billion per year on average (Gleeson and Yacoub 2021). In addition, Australia suffered significant reputational damage. For example, the ‘Nauru files’ (The Guardian 2016), consisting of over 2,000 leaked reports from the Nauru detention facilities, document severe human rights violations. Published shortly before the Danish delegation arrived in Australia, the leaked documents were a powerful reminder that such a policy can harm the international reputation of the externalising state and, as a consequence, potentially also its material interests.

Likewise, politicians interested in drawing lessons from Australia’s externalisation model must consider potential legal challenges. In 2011, for instance, the High Court of Australia qualified a ministerial declaration that aimed to authorise the extraterritorial processing of asylum claims in Malaysia and subsequent protection in Australia to be invalid (Ghezelbash 2018: 117–120). However, questions regarding the legality of externalisation are not limited to Australia but have also been critical elsewhere, such as in Israel, whose model of externalisation has been challenged (see Bar-Tuvia 2018). Crucially, a combination of legal challenges, practical problems, and normative constraints have seen bilateral agreements based on the externalisation of refugee protection struck down in all states that have tested them (Léonard and Kaunert 2016).

While there is no ‘proof of concept’ for externalisation policies as efficient and effective instruments of asylum governance that generate the desired policy outcomes, one could argue that the Social Democrats introduced it because they held such beliefs. However, this reasoning does run against two basic presumptions of technocratic learning, namely a careful and systematic manner of evaluating policy consequences by decision-makers and the factual success of a policy under consideration (see Gilardi and Wasserfallen 2019). Nonetheless, the empirical evidence also does not suggest that the Social Democrats were fully convinced of the efficiency and effectiveness of the externalisation policy. Furthermore, there is no indication of a careful and systematic evaluation of alternative asylum policies by the Social Democrats before they embraced the idea of externalising refugee protection. One could also ponder why Lars Løkke Rasmussen’s government (Jun 2015–Jun 2019), representing the conservative ‘blue bloc’ and known for its strict immigration policy, did not follow the path of externalisation as it was in power during the time of the fact-finding mission. Instead, representatives of the ‘blue bloc’ questioned the practicability of externalisation (Knuth 2016) – much as Mette Frederiksen (2014) qualified such plans as unrealistic when she was Minister of Justice. Finally, even the travel reports of the Immigration and Integration Committee (2016) to Australia do not reveal clear indications that the Australian model has fully convinced the Danish delegation.

In sum, the empirical evidence does not support the view that technocratic learning has been decisive for Denmark’s Social Democrats to embrace the idea of externalising asylum.

Expecting political gains?

Gilardi and Wasserfallen (2019) stress that decision-makers not only consider if policies tested in other polities successfully solve a pressing problem. Instead, they argue that decision-makers evaluate policies through ideological lenses, assessing whether policies secure votes or offices. Indeed, political learning could explain why the Social Democrats introduced an externalisation policy despite its questionable track record. For the case at hand, three conditions are necessary to make a convincing case for political learning. First, the Social Democrats’ asylum policy must have failed to secure the expected votes or offices in elections. Second, the Social Democrats must have screened asylum policies of ideological allies in other countries that have introduced an externalisation policy. Third, the Social Democrats introduced the externalisation policy because they learned that ideological allies had benefitted politically from its introduction.

This explanation resonates with the timing of when the Social Democrats decided to change their asylum policy. In fact, the Social Democrats turned only to the idea of externalising refugee protection after they lost power in the general elections of June 2015. As the elections took place during an emerging refugee protection crisis, immigration has been one of the main issues in the election campaign (Hagelund 2020). While the parties of the ‘red bloc’ (on which the Social Democrats depended) saw their voting share decline, the centre-right parties of the ‘blue bloc’ prevailed (Kosiara-Pedersen 2016a). After the elections, the new Social Democratic core team around Mette Frederiksen took over and quickly toughened up on immigration (Kosiara-Pedersen 2016b).

Nevertheless, the Social Democrats have not taken lessons from ideological allies to restrict access to asylum in Denmark. The second Rudd Labour government in Australia could be considered an ideological ally and a possible role model for the Danish Social Democrats. Even though it has not introduced the idea of outsourcing asylum in Australia, it revived its externalisation model (known as ‘Pacific Solution II’) in July 2013. Only a few months later, however, they lost power in the Australian elections in September 2013. Therefore, an indispensable condition for making a convincing case for political learning – the existence of an ideological ally that has gained votes or offices with an externalisation policy – is missing.

Anyhow, even a right-wing government must not necessarily benefit politically from externalising refugee protection. Until the date of writing, the Australian conservative coalition government under John Howard and the Israeli government under Benjamin Netanyahu are the only governments that stayed in power after outsourcing asylum. Yet, the Howard government lost power in the national elections in 2007 – and the Netanyahu government had to terminate bilateral agreements with Uganda and Rwanda in the face of legal challenges and political pressure in 2018 (Bar-Tuvia 2018).

Besides the uncertainty of the electoral gains, left-wing parties that embrace the idea of externalising asylum must expect additional costs. Specifically, they risk confusing or losing voters (Budge 1994; Robertson 1976). Furthermore, Social Democratic parties that turn to ideas that are promoted by populist radical-right parties could damage the party’s credibility (Bale et al. 2010).

In sum, the most promising rationalist explanation – that decision-makers embraced the idea of externalisation due to technocratic or political learning processes – falls short of explaining why even right-wing governments turned to this policy. Therefore, I examine next if social constructivist mechanisms or pathways provide a convincing explanation for why Denmark’s Social Democrats embraced the idea of externalisation.

Social constructivist explanations: responsibility or responsiveness?

Social constructivist explanations that focus on the perception of policies rather than on the cost-benefit calculations of decision-makers could offer leverage to explain the Danish case. Nevertheless, because the only mechanism of the orthodox diffusion theory with roots in social constructivism – emulation – presumes that decision-makers select policies they view as most appropriate, the mechanism itself can hardly explain why the Social Democrats in Denmark opted for externalisation. However, a social constructivist mechanism of the diffusion theory by Blatter et al. (2022) – recognition – posits two diffusion pathways that provide a potential explanation for why decision-makers adopt a policy that does not necessarily align with their normative belief system: policy expertise and popular attention.

In the dominant set of diffusion mechanisms, emulation serves as a residual category that is supposed to capture all diffusion processes that are not predominantly driven by perceptions regarding the appropriateness of policies (Gilardi and Wasserfallen 2019). In diffusion processes underpinned by this mechanism, decision-makers are generally viewed to be driven by the will to adopt policies that represent their normative convictions (Simmons et al. 2006). Yet, a highly exclusive asylum policy promoted primarily by right-wing parties can hardly be viewed as an ideal reflection of the Social Democrats’ ideology.⁹

Consequently, the mechanisms posited by orthodox diffusion theory do not explain why the Social Democrats in Denmark turned to the idea of externalisation. For this reason, I use Blatter et al.’s (2022) theory of policy diffusion, which foregrounds two social constructivist pathways that potentially explain why decision-makers adopt policies that are not necessarily consistent with their ideology. Accordingly, such incongruent policy adoption can be explained either by decision-makers’ aspirations to act as responsible leaders or because of their desire to be recognised as responsive doers.

Demonstrating responsible leadership by following an epistemic community?

The policy expertise pathway (Blatter et al. 2022) explains the willingness of decision-makers to strive for responsible leadership and the ability of epistemic communities to persuade them based on three main suppositions. The first is the existence of an epistemic community that is setting the agenda in the policy field and singling out ‘best practices’ in systematic evaluations of international policies. The second is that policy experts have access to decision-makers and are able to persuade them of particular policies. The third is the willingness of decision-makers to act as responsible leaders informed by advice from an epistemic community.

At least at first glance, the policy expertise pathway provides a convincing explanation of the puzzle of the Danish Social Democrats’ policy U-turn. In December 2017, the party journal presented Anders and Morten Lisborg, two brothers with experience in the field, as ‘migration experts who will share their knowledge with the Social Democrats’ (Bruun 2017: 26). This cooperation between the Social Democrats and the Lisborgs was formalised in a working agreement that lasted from 2017 to 2019, which commissioned an analysis of current problems in Denmark’s asylum system and the development of policy alternatives (Interview AL). This shows that the Lisborgs had access to decision-makers and suggests that policy experts had the key role in producing policy change.

Well before they were contracted by the Social Democrats, Anders and Morten Lisborg had started discussing their ideas for broad-based asylum reform. From 2015 onwards, they contributed analyses criticising the Danish and wider European asylum systems in newspapers and other media (e.g., M. Lisborg et al. 2018). They described the European asylum system as dysfunctional, expensive, and unfair (A. Lisborg 2015; Socialdemokratiet 2017). By setting up extraterritorial asylum centres in the region of origin while accepting quotas of resettlement refugees, they concluded, European countries could end the system of ‘spontaneous asylum’ (A. Lisborg 2015). Such an asylum system, they claimed, would help more people more effectively, reduce the number of deaths in the Mediterranean, and lower the costs of asylum for European states (A. Lisborg 2015). These assumptions were informed by Australia’s model of externalisation and the ‘EU-Turkey deal’ (Interview AL; Socialdemokratiet 2017).

This sequence of events supports the view that the Lisborgs were indeed able to persuade the Social Democrats to opt for the externalisation of refugee protection. As touched on above, both the vision and rhetoric of their argumentation were published well before they had been contacted by the Social Democrats, most notably in an article entitled ‘The European asylum system is dysfunctional’, published by Anders Lisborg in the Danish newspaper Politiken in November 2015 (see A. Lisborg 2015). In the following years, their ideas, as well as the line of argumentation, were reproduced – almost perfectly – by leading members of the Social Democrats (e.g., Frederiksen 2018; Stoklund 2021; Tesfaye 2021). Like the Lisborgs, the Social Democrats claimed that fundamentally transforming the current asylum system would also allow more people in need to be helped.¹⁰

On a closer examination, however, the empirical evidence does not support the assumption that the Social Democrats embraced the idea of outsourcing asylum because they wanted to demonstrate responsible leadership informed by expert knowledge. As noted, the professional background and the mode of knowledge generation of the Lisborgs do not correspond to the understanding of a transnational epistemic community. Their expertise is primarily based on practical experience with the Danish immigration service and international organisations in the field of migration and human trafficking, and their knowledge of alternative asylum systems builds on background reading (Interview AL). Hence, their personal experience in this field convinced them that the asylum system needed to be radically transformed: while they had privileged insights into the asylum field and gained expertise in this issue, they have not systematically evaluated alternative policies that characterise transnational epistemic communities.

Moreover, even though the Lisborgs have ostensibly been successful in selling a particular asylum policy to the Social Democrats, this came after the party began to look for alternative asylum policies to toughen up on immigration. Martin Rossen, then personal adviser of Mette Frederiksen, contacted the Lisborg brothers at the beginning of 2016, much to the brothers’ surprise (Interview AL). As it turned out, the 2015 Politiken article had drawn Rossen’s attention and propelled his outreach to the Lisborgs. Nevertheless, there is no indication that the Lisborgs played any direct role in initiating the policy change as established and dominant actors in the field of asylum. In sum, the empirical evidence does not support the view that the policy expertise pathway of diffusion drove this particular policy change.

A window of opportunity, a tempting alternative, and the desire for responsiveness: explaining the Danish case of externalising refugee protection

The pursuit of decision-makers for responsiveness is essential in diffusion processes that follow what Blatter et al. (2022) call the popular attention pathway. In the face of highly medialised issues, decision-makers can be tempted to weigh responsiveness to their constituency higher than loyalty to their party ideology. A diffusion process must fulfil at least three necessary conditions to achieve such a particular configuration. First, the process is initiated by significant media attention on an issue that quickly gains political salience, irrespective of whether it is based on actual or perceived problem pressure. Second, policy advocates must make a foreign policy solution available and transferable to the local context. Third, the choice to adopt the policy by decision-makers is motivated by their will to be perceived as responsive doers. Because all elements are given in the Danish case, the popular attention pathway is an accurate explanation of why the Social Democrats turned to the idea of externalisation.

Between 2014 and 2016, the steep increase of people seeking protection in Europe (Eurostat 2022) led to extensive media coverage in Denmark. Having become a central topic for national and local media, immigration quickly became a vital issue in the political debate in Denmark (Hagelund 2020). Specifically, the salience of the immigration issue and the public perception of national administrations being unable to cope with the number and speed of people coming to Europe have questioned the appropriateness and sustainability of existing policies. Moreover, the extensive media coverage was particularly relevant for politicians at that time since the general elections took place in 2015, at a time when the refugee protection crisis dominated front-page news.

The high salience of immigration in the political and public debate and the apparent ineffectiveness of existing asylum policies in Denmark have opened a window of opportunity for alternative approaches. It was this context of high mediatisation of the asylum issue and the general perception of overburdened state capacities in Denmark that laid the ground for the Lisborgs’s aforementioned public campaign of criticism against the established system after 2015. In doing so, the Lisborgs made the idea of outsourcing asylum available to a broader public in Denmark, including the Social Democrats. Notably, they provided not only a policy solution (externalisation) but also a problem definition (the asylum system is dysfunctional) and a policy framing (externalisation strengthens the system’s humanitarian foundations).

Anders and Morten Lisborg began to publish their critique of the established asylum system and promote externalisation at a time when the Social Democrats were on the hunt for a way to burnish their credentials as responsive to their constituency. Even though the party slightly increased its own share of the vote in the June 2015 elections, it lost its bid for another term in office as the parties of the ‘red bloc’ ceded the majority to the ‘blue bloc’ (Kosiara-Pedersen 2016a). With an anti-immigration programme, the Danish People’s Party¹¹ became, for the first time in history, the strongest party of the blue bloc and the second biggest party in Denmark (Kosiara-Pedersen 2016a). After the elections, the Social Democrats replaced Helle Thorning-Schmidt as party leader, and Mette Frederiksen and her team took over. In the face of continuously growing numbers of asylum applications, the new Social Democratic core team wanted to demonstrate responsiveness to their constituency and find a new position on immigration (Interview AL). Consequently, they quickly communicated to their political allies that they would toughen up on immigration (Interview NR).

Martin Rossen, a personal adviser to new party leader (and future prime minister) Frederiksen, recognised the perfect match between the need of the Social Democrats for more restrictiveness and the value of Anders and Mortens Lisborg’s policy alternative and framing. At the beginning of 2016, he contacted the Lisborgs, and a few months later, they were allowed to present their analysis and vision in front of the inner circle of the Social Democrats, including Frederiksen and Mattias Tesfaye, who later became immigration minister (Interview AL). Being persuaded by the Lisborgs’ take on externalisation, the Social Democrats’ new leadership eventually decided to make the idea a central pillar of their immigration policy and their electoral campaign for 2019 (Socialdemokratiet 2018, 2019). When the party returned to office in 2019, they tried to turn the vision to externalise asylum into reality, and in April 2021, Tesfaye tabled Bill L226 in the Danish parliament.

Given that the government of Denmark’s externalisation plans have yet to be implemented at the time of writing, one could argue that Bill L226 represents a symbolic policy. Symbolic policies are means that allow decision-makers signalling desirable outcomes, not measures meant to be fully implemented (Gustaffson 1983). Yet, the fact that the Social Democratic government set up a task force consisting of about a dozen people working full-time on the plan to outsource asylum suggests that the Social Democrats have indeed tried to put the policy into practice. Therefore, Bill L226 is characterised best as a pseudo policy – a policy that decision-makers want to put into practice but for which they are reluctant to consider all the necessary conditions for its successful implementation (Gustaffson 1983).

However, Bill L226 is not only a pseudo policy and, therefore, more likely to fail in the implementation process but also inherently paradoxical. First, it is not a solution that seeks to solve a problem but a solution in search of a problem. When the Social Democrats began to promote the policy in December 2017, the number of asylum applications had already shrunk below the level of 2014 (Eurostat 2022). Subsequently, the Social Democrats not only had to advocate a solution but also construct and cultivate the problem it promised to solve. Moreover, the policy is a paradox because its ultimate aim is non-use. Because externalisation is a tool that aims to deter people from seeking asylum in a host country, the policy’s aim is not the transfer of asylum seekers from Denmark to another country en masse but rather to make sure they do not travel to Denmark in the first place. As much was made clear by Prime Minister Frederiksen herself when she declared the objective of Denmark’s asylum policy as having ‘zero’ asylum seekers (McPartland 2021).

Justification strategies: adapting the policy to the party ideology

In the previous section, I demonstrated that the Social Democrats turned to the idea of externalising asylum because they were motivated to present themselves as responsive problem solvers. Such a motivation characterises decisions described by the social constructivist popular attention pathway of Blatter et al.’s (2022) diffusion theory. Identifying the relevant mechanism and unveiling the motivation behind a policy adoption is an established task for diffusion scholars. In this section, diffusion studies are brought one step further by investigating the justification strategies for policy adoption by decision-makers. To this end, I apply the diffusion theory by Blatter et al. (2022), which is suitable for this endeavour as it provides a balanced and nuanced set of alternatives for analysing the communicative strategies employed by Social Democrats for embracing the idea of externalising refugee protection. Therefore, in this section, I focus on the justification strategies used by the Social Democrats to legitimise the adoption of a policy that, up until then, had only been introduced by a handful of right-wing governments outside Europe. I find that the Social Democrats have avoided referencing the actual motivation to legitimise externalisation. Instead, they justified the policy with motives that relate to all three social constructivist pathways posited by the theory of Blatter et al. (2022) that were not decisive for adopting the policy.

While scholars generally limit the use of policy diffusion theory to identify the drivers of policy change, I demonstrate that such a conceptual framework is likewise well-suited to analyse the justifications for the import of ideas. Specifically, I test whether the public reasons given by the Social Democrats to legitimise externalisation relate to specific motivations that describe decision-making processes of the diffusion pathways posited by the theory of Blatter et al. (2022). By shedding light on the causation and justification – as well as the discrepancy between them – I provide a more complete understanding of how the Social Democrats ended up tabling Bill L226. However, every policy change requires not only an actual driver but also a compelling legitimisation, at least in liberal democracies. Therefore, examining the justification strategies that decision-makers use will provide diffusion scholars with a deeper understanding of the processes they study.

Because the Social Democrats embraced a highly controversial asylum policy that had previously only been introduced by their ideological counterparts in other countries, they risked damaging the credibility of the party, confusing voters, and undermining intra-party coherence (Bale et al. 2010; Budge 1994; Robertson 1976). These risks result from potential perceptions of ideological inconsistencies between the universalist foundation of Social Democratic parties and the highly restrictive character of externalisation policies. To reduce these risks, the Social Democrats have mainly used three strategies: referring to policy advisers, reframing the restrictive policy as a humanitarian project, and changing the narrative about the Social Democrats’ identity.

Referring to policy advisers

Often, the engagement of political advisers takes place behind closed doors. In the Danish case, however, the Social Democrats openly communicated their partnership with Anders and Morten Lisborg to the public. By introducing the policy advisers as ‘migration experts’ who inform the party (Bruun 2017: 26), the Social Democrats suggested following the lead of experts. By contrast, the empirical evidence collected indicates that the policy advisors have not initiated this policy change and unveils that embracing the idea of externalisation has not been the result of a systematic evaluation of policy alternatives. While consistently referring to their policy advisers, the Social Democrats have sidelined open questions regarding the practicability and ignored critical views expressed by the UNHCR (2021) regarding the appropriateness and the expertise of legal scholars regarding the legality (e.g., Bar-Tuvia 2018). In doing so, they disguised their responsiveness to the perceived demand for restrictive policies as openness to expert knowledge and evidence-based policy-making. Hence, they claim to have changed their asylum policy because of motivations that generally describe decision-making processes along the policy expertise pathway of recognition-driven diffusion (Blatter et al. 2022).

The Social Democrats stressed the role and relevance of the Lisborgs by giving them ample opportunities to present their vision of a new asylum system in front of different audiences. In December 2017, the party journal ran a profile of the Lisborgs for party members (Bruun 2017), and they took centre stage at the party’s national congress (Socialdemokratiet 2017). Through several articles in national newspapers (e.g., Birk 2018), the collaboration between the party and the brothers has also been communicated to a larger audience in Denmark. Moreover, the Lisborgs joined the Social Democrats in their meetings with other Danish parties (Interview NR) and could be found promoting the Social Democrats’ plan to outsource asylum in the pre-election campaign trail of 2019 (Interview AL).

By pointing to the collaboration with the Lisborgs and their consultancy firm Migration Management Advice, the party was able to obscure that Bill L226 is inspired by non-European policy models criticised widely for their poor humanitarian outcomes. Moreover, pretending that the policy recommendations they were following were expert advice allowed the Social Democratic core team to present themselves as responsible leaders. This imitation of responsible leadership can be viewed as an attempt to mitigate the risk of losing credibility, which is acute for Social Democratic parties that take ideas of their political adversaries in other states.

Reframing externalisation as a humanitarian project

Representatives of right-wing parties have usually embedded the idea of outsourcing asylum in nationalist discourses, signalling the need to protect nations from the threat of immigration. The Social Democrats, by contrast, have applied – somewhat paradoxically – a humanitarian framing, foregrounding the claim that externalisation protects people from the risks of dangerous migration journeys (Socialdemokratiet 2018). By reframing externalisation as a humanitarian project, they suggest that the highly restrictive policy represents the universalist foundation of the party ideology better than the current asylum system. In doing so, the Social Democrats indicate to have embraced the idea because of motivations that underpin decision-making processes along the policy beliefs pathway of ideology-driven diffusion (Blatter et al. 2022).

Before the Social Democrats decided to move the externalisation policy to the top of their new immigration programme ‘Fair and Realistic’ (Socialdemokratiet 2018), they had harshly criticised the idea of sending asylum seekers arriving in Denmark to countries outside of Europe. For example, in 2014, Mette Reissman, the Social Democrat’s spokesperson, branded the idea ‘inhumane’ (Jessen and Arnfred 2014) and Mette Frederiksen called it ‘very, very problematic’ (Frederiksen 2014). Yet, following the framing used by the Lisborgs, the Social Democrats suddenly came to the inverse conclusions and claimed that externalisation establishes a ‘more humane asylum system’ (Socialdemokratiet 2019: 15).

Professing to have opted for externalisation because it better reflects their general approach to asylum held two major advantages for the party. First, the party leadership could reassure – at least rhetorically – the part of the Social Democratic voter base that identifies itself strongly with the universalist foundations of the party. Simulating consistency with previous positions could therefore be viewed as an attempt to avoid confusing (and potentially losing) voters. Second, the humanitarian framing enabled the Social Democrats to face domestic and international criticism, such as the criticism that outsourcing asylum is ‘irresponsible and lacking in solidarity’ (Danish Refugee Council 2021). Suggesting that the policy could produce better humanitarian outcomes allowed the party to feign that their programme is in line with international calls for more solidarity in refugee protection, a view that dominated the political discussion around the United Nations Global Compact on Refugees of 2018.

Changing the narrative about Social Democratic identity

Generally, Social Democratic parties are committed to universalist principles and favour more liberal positions on immigration compared to their conservative and right-wing counterparts (Abou-Chadi and Wagner 2020; Kitschelt 1994). However, after Mette Frederiksen took over the Social Democratic leadership in June 2015, she positioned the party as being almost as restrictive on immigration and refugees as the Danish People’s Party (Kosiara-Pedersen 2016b). Subsequently, the party embraced the highly restrictive plan to outsource asylum, even though the preceding conservative-liberal government dismissed the idea of externalisation. The new inner circle of the Social Democrats pretended that the turn towards externalisation was motivated by the will to find a policy that accurately reflects Social Democratic ideology. In doing so, they suggested that the tabling of Bill L226 was based on motivations common for decision-making along the principled beliefs pathway of ideology-driven diffusion, in which parties adopt policies because they are viewed as most appropriate within their party family (Blatter et al. 2022).

Mattias Tesfaye, who tabled Bill L226, played a notable role in this process. In 2017, Tesfaye authored the book titled Welcome Mustafa: 50 Years of Social Democratic Immigration Policy, wherein he describes the debates within the Social Democratic Party after the Danish government opted for a more open immigration policy in 1967. In his book, Tesfaye outlines that leading Social Democrats swiftly recognised the drawbacks of immigration and advocated for restrictive measures in order to safeguard the welfare state and protect the interests of the Danish blue-collar workers. Subsequently, he argues, the party deviated from this trajectory, only to return to it when Mette Frederiksen rediscovered and revitalised what she perceived as a ‘more traditional’ (mere klassisk) Social Democratic identity (Tesfaye 2017: 362). By changing the dominant narrative about Social Democratic identity, Tesfaye and Frederiksen adapt, at least to some extent, the party to the externalisation policy.

Promoting the externalisation of asylum meant some party members could argue that the new leadership around Mette Frederiksen was disloyal to the party ideology. Against this backdrop, claiming the prerogative regarding the dominant narrative on Social Democratic identity reflects also an attempt to forestall internal criticism. Furthermore, by changing the dominant narrative about the party ideology, party representatives signal to their ideological allies in other countries that Social Democrats must not necessarily be bound by normative constraints when it comes to immigration. In doing so, they present themselves as pragmatic but authentic Social Democratic pioneers, inviting other Social Democratic leaders in Europe to follow their example.

Conclusions

In this article, I argue that the particular configuration of (1) high media attention on immigration and apparent policy failure in the face of the refugee protection crisis between 2014 and 2016, (2) the transfer of foreign externalisation models to the Danish context and its translation for Social Democratic ideology by policy advisers, and (3) the desire of the new leadership around Mette Frederiksen to present themselves as responsive doers after the party lost power in the 2015 general elections explains why the Social Democrats turned to the idea of externalising asylum. This configuration is expressed by the social constructivist popular attention pathway of the policy diffusion theory by Blatter et al. (2022). Nevertheless, the Social Democrats used three communicative strategies that suggest that the decision-making process was motivated by reasons that characterise all three social constructivist pathways that have not been decisive for the adoption of the policy to justify the policy change.

The findings from the Danish case study carry implications for policy diffusion theory and method as well as for migration studies. Focusing on a rare case of policy diffusion across political ideologies, I show, against the conventional view, how policies can also travel in this way, so long as they are not expected to be particularly efficient or effective in solving policy problems. Moreover, I demonstrate that diffusion theory serves not only as a theoretical framework to identify the drivers of policy change but also to analyse the strategies used to justify policy choices. By illustrating that a conceptually coherent framework of policy diffusion allows the study of both the causation and the justification of policy change, I offer a way for diffusion scholars to gain a fuller understanding of the salient political processes. Specifically, this case study points to the relevance of systematically distinguishing the justifications from the motivations and offers a methodological framework to investigate the potential discrepancies between them.

In the realm of migration studies, the article illuminates that within liberal democracies, the adoption of restrictive immigration policies does not invariably align with the pursuit of national interests or the agenda of nationalist-oriented political parties. Furthermore, it provides insights into the rationale behind Denmark’s adoption of one of the most controversial policies, even in the face of vehement opposition from civil society organisations and the prospect of damaging the country’s international reputation as a pioneer in refugee protection. Because party policy diffusion appears to be especially pronounced within the Social Democratic party family (Senninger et al. 2022) further studies could examine whether the Danish Social Democrats have inspired Social Democratic parties in other countries to pursue a similar course on immigration. Future research could also unveil the dynamics between the national plans to externalise refugee protection in Europe and the reforms of the EU asylum and migration system or investigate the interplay between the recent diffusion of illiberal asylum policies and the increasing contestation of the liberal international order.

Acknowledgements

I thank Joachim Blatter, Sandra Lavenex, Jens Vedsted-Hansen, Zachary Whyte, and the anonymous reviewers for their valuable comments and suggestions on earlier drafts; participants of the IMISCOE Annual Conference 2023 for their helpful feedback; Konstantin Kreibich for the excellent research assistance; Cecilie Odgaard Jakobsen for the translation of Danish quotes into English; and the Centre for Advanced Migration Studies (AMIS) at the University of Copenhagen for hosting me as a visiting scholar and providing me with an ideal environment for this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung [grant number: 51NF40-182897].

Notes on contributors

Frowin Rausis

Frowin Rausis is a postdoctoral researcher in the Department of Political Science and International Relations at the University of Geneva and an associated research fellow at the Swiss National Centre of Competence in Research for migration and mobility studies nccr – on the move. His research focuses on the diffusion of policies and the politics of refugee protection. [[email protected]]

Notes

1 More generally, the term ‘externalisation’ describes the ‘process of shifting functions that are normally undertaken by a State within its own territory so that they take place, in part or in whole, outside its territory’ (Refugee Law Initiative 2022: 114). While various asylum policies and forms of international cooperation have been associated with the term, externalisation generally includes a territorial dimension reflected by the attempt to move persons outside of the territory of a host state as well as a legal dimension as it also implies the will to bring people seeking protection outside of the jurisdictional responsibility of a state (Lavenex 2022: 28–29). In this article, I use the term specifically in reference to the shifting of responsibility for the processing of asylum claims and subsequent refugee protection from a host state to another one to which the person has not necessarily any connection, and that has also been dubbed ‘Safe Fourth Country Policy’ (Rausis 2022b: 7).

2 Bill L226, a legislative amendment to Denmark’s Aliens Act which transposed into the Law No. 1191, passed parliament on 6 June 2021 with 70 votes in favour, 24 votes against, and zero abstentions out of 179 possible votes.

3 In 1991, the Republican administration of George H. W. Bush was the first to introduce extraterritorial processing by allowing asylum claims to take place at the United States naval base in Guantanamo Bay, Cuba. In Australia, a conservative coalition government under John Howard allowed the extraterritorial processing of asylum claims in Papua New Guinea and Nauru in 2001. In Israel, a multi-party government led by Benjamin Netanyahu of the right-wing Likud party concluded secret agreements with Uganda to send asylum seekers there in 2014. Finally, in 2019, President Donald J. Trump, a Republican, completed bilateral agreements with Guatemala, Nicaragua, and El Salvador, allowing the United States to send asylum seekers to Central America.

4 Seeking to answer why the Social Democrats have turned to a specific asylum policy (externalisation) that has previously been adopted by other states must not be confounded with asking why the party has taken a general position (more restrictive) on immigration in Denmark in the past few years. Theoretical perspectives focusing on shifts in public opinion and issue salience or on party competition and intra-party dynamics are useful to examine the latter question (see Green-Pedersen and Otjes 2019; Kosiara-Pedersen 2020; Rathgeb and Wolkenstein 2022). Policy diffusion theory, however, is best suited to answer the former question.

5 For a legal analysis of Denmark’s Bill L226 in light of international and EU law, see Tan and Vedsted-Hansen (2021). For an account on its evolution and on international reactions to the bill, see Lemberg-Pedersen et al. (2021) and Tan (2022).

6 In this article, I use ‘policy diffusion’ as an umbrella term for referring to studies that examine how and why policies travel across polities. Specifically, I encompass research conducted under the label ‘policy transfer’ (e.g., Dolowitz and Marsh 1996). In recent decades, these different strands of research studying the mobility of policies have, to a considerable degree, converged (de Oliveira 2022).

7 In fact, it was Denmark (together with Austria) that requested the EU to make the externalisation of refugee protection a key element of a European asylum policy (Federal Ministry of Interior Austria & Ministry for Immigration and Integration Denmark 2018). Yet, the European Commissioner for Home Affairs Ylva Johansson declared that attempts to outsource asylum sends ‘a strong and wrong signal to the outer world’ (2021).

8 However, the committee members did not travel to Nauru as planned because the government of Nauru decided not to issue travel visas for some of the members. Therefore, the committee members, receiving this information only after they landed in Australia, limited their fact-finding mission to Australia (Farell 2016; Immigration and Integration Committee 2016).

9 Likewise, the mechanism of ideology posited by the diffusion theory of Blatter et al. (2022) on social constructivist grounds is unable to explain why the Social Democrats embraced the idea of externalising refugee protection because it presupposes the party shares a normative foundation with the other transnational actors that have outsourced asylum.

10 The Lisborgs envisaged externalisation as a European project (Interview AL). Nevertheless, even after determining that other European countries would not follow Denmark’s lead, the Social Democrats decided to move forward with their externalisation plan after they returned to office in 2019.

11 Studies suggest that the rise of the Danish People’s Party, among other factors, could have incentivised the Social Democrats to take a more restrictive position on immigration (see Kosiara-Pedersen 2020; Rathgeb and Wolkenstein 2022). Nevertheless, the empirical evidence collected for this case study does not attribute the Danish People’s Party to a key role in the diffusion of the externalisation policy to Denmark. Even though representatives of the party had called for the introduction of the Australian asylum model to Denmark (e.g., Henriksen 2014), these calls may have only played an indirect role, if at all. The temporal disconnect between the calls and the interest of the Social Democrats in externalisation, for example, does not qualify them as a relevant element in a causal chain or as the initial stimulus. Furthermore, representatives of the party have never sketched out externalisation plans in detail and, therefore, have not become actual agents of diffusion.