Policing Asylum Seekers’ Flight Within Europe

A critical assessment of the Dublin regulation requires a look beyond its official function of allocating asylum seekers across EU Member States. This article argues it embodies the “hidden face” of Schengen insofar as it legally fixes them in the sole country responsible for their application. Because this responsibility lies primarily with the first country of arrival, it is consistent with the core-periphery axis of division of labor in the EU. The first part of this paper examines how the Schengen/Dublin dual regime of (im)mobility might respond to the constant need for bridled labor alongside free wage labor in the world-system. However, equally constant is labor power’s propensity to evade its subsumption under capital; this is exemplified by Dubliners’ appropriation of freedom of movement through irregularity. By deserting the “plantations” of the European peripheries, those “maroons” of our present time disrupt the European geography of power and contest their assigned position in it. But the widely acknowledged failure of this regime to deter “secondary movements” does not necessarily mean it is non-effective. Attention must then be given to mechanisms of “exclusion from within” experimented on Dubliners. The second part will offer an overview of the tactics of internal rebordering that have been recently deployed in core countries and question the extent to which those attempts to recapture their flight meet the conditions for the optimization of capital’s operations.

1 Data used in this paper are extracted from publicly available EU or national agencies statistics. It is generally admitted that their credibility is crippled by multiple shortcomings, bias, and persistent discrepancies between the methodologies adopted by each national authority. The pretense to produce objective knowledge on migratory flows from those data is both empirically and epistemologically questionable. But this consideration doesn't affect the pertinence of their usage for analyzing the way migrants are counted, identified, and controlled. Quite the contrary.

Analytic-Historical Background
The main achievement of world-systems analysis might have been the reformulation with and against Marx's thesis according to which capitalism emerges from the Industrial Revolution. In this literature, capitalism isn't singled out by a certain relation of production-namely, free wage labor-but in the principle of endless accumulation of capital and the international division of labor (Wallerstein 1976;. Drawing from this framework, Yann Moulier-Boutang (1998) contends wage labor does not precede capitalism but is the outcome of its reconfiguration imposed by struggles. Whereas James Scott (2013) investigated the historical challenge to state-building projects posed by the "exit option" (Hirschman 1970), others focused on how "flight as fight" (Lucassen and van Voss 2019) played a vital role in the transformation of global labor regimes. Among them, Boutang convincingly reasserted the almost (Wallerstein 1974) ignored role played by maroons in the abolition of slavery. In his view, the transformation of the slave market into a labor market constituted the less evil solution for recapturing fugitive labor. Nonetheless, he argued wage labor cannot become a universal condition since it relies on the persistence of other relations of production at its margins. Consequently, to this day, new forms of unfree labor have had to be constantly reinvented for securing the extraction of surplus value.
From that standpoint, the 1970s crisis of Fordism cannot be understood if we overlook the struggles over the liberation of non-wage labor. Decolonization placed in crisis the imperial extractive economy (often based on forced labor) on which postwar western prosperity rested on. In addition, while feminism contested the non-recognition of reproductive labor, the Civil Rights movement in America and the struggles of postcolonial subjects in Europe challenged the racialized wage differentiation. This all contributed to the disruption of the compromise between state, capital, and organized (white male) labor. In reaction, those demands have been partially reappropriated through the promotion of a meritocratic formal equality in the allocation of work and its reward. Liberation was strategically re-framed as flexibilization as the neoliberal turn unleashed an unprecedented fragmentation of labor contract categories. For our concern, we can acknowledge the articulation between new forms of racial-blind born again racism (Goldberg 2008) and this process of multiplication of labor. Indeed, stronger state regulation over access to residency and the deregulation of labor law witnessed since then have been far from antagonistic trends.
If there is one thing over which ex-colonial empires have managed to keep their hegemony, it might be in respect of the hierarchy of passports' value. Restrictive labor immigration policies adopted since the 1970s can be interpreted as the strategic substitution of the skin color by the passport one: "'immigration'…becomes the main name given to race within the crisis-torn nations" (Balibar and Wallerstein 1991: 52). Although unskilled migrant labor became undesirable with rising unemployment, it never ceased to be an essential component of the workforce (Fassin 2012: 104). The novelty resided in the paramount function endorsed by the juridico-administrative apparatus for constraining it. On the one hand, conditioning the residence permit renewal upon the possession of a labor contract intensified dependence towards the employer of those still holding one (Moulier-Boutang 1998). On the other hand, those having lost such benefit were jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 relegated into the informal labor market. Their condition of "deportability" (De Genova 2002;De Genova and Peutz 2010) could only but bolster their over-exploitation. In both cases, such labor relations "are unfree insofar as they feature coercive forces other than the economic need of selling one's labor power" (Scheel 2018: 272). With this late logic of modern capitalism also noticeable in Northern America or in the Pacific, the "world configuring function" (Balibar 2002) of borders came to lie in the drawing of "the very boundary between the wage contract and surplus labour" (Mitropoulos 2012). In late 1990s Europe, then, restrictive migration policies seemed to answer both to the rising xenophobic atmosphere and to the necessity to supply capital with precarious workforce: Marxism seemed pertinent to analyze the rising phenomenon of the sans-papiers [undocumented].
Since then, various changes may temper such analysis. Some of the sans-papiers eventually found the opportunity to be regularized, while others were deported. In parallel, new strategies of migration by sea loomed in response to the reduction of legal pathways. Accordingly, it is not merely residence but circulation which has become the object of criminalization. The incremental substitution of the figure of the sans-papiers entering with a visa by the one of the boat people pushed aside labor issues in favor of a humanitarian-security nexus. While pathways to regularization through labor were drastically reduced, asylum became the only one left. Theoretically, for this growing category of migrants, neither their right of residence nor the risk of being deported depended on the possession of a job contract. For all those reasons, the model explaining irregularization as a means for surplus value making can seem inappropriate in the current context. However the policing of labor migration and of asylum migration intertwine far more than people think. Assuming restrictions on the freedom of movement of labor inevitably contributes to its subordination, the Dublin regime can only be entangled in this blurring of the two policy fields. We shall then explore how this "apparatus of capture" (Scheel 2019) of flight may fuel the subjection of migrant labor. 2 A Core-Periphery Analysis of the Dublin Regime

The Hidden Face of Schengen
Paradoxically, the aspiration to make of Europe a borderless area has historically encouraged stronger securitization of mobility as "freedom" and "security;" mutual reinforcement became the new imperative (Bigo et al. 2010). Although the 1990 Implementing Convention of the Schengen Agreement consecrated the principle of lifting internal borders controls, its Title II discretely circumscribed freedom of movement to EU nationals (Huysmans 2000;Crowley 2001;Kasparek 2016). This implied the development of a whole range of so-called "compensatory measures" able to constrain the mobility of TCN without hindering that of EU nationals. Not only did the reinforcement of external borders became the prerequisite for the lifting of the internal ones, but their juridical (non-refoulement principle), political (forefront MS inability or unwillingness to protect them), and structural (migrants agency) flaws justified the development of complementary internal measures (Hailbronner and Thiery 1997;Geddes 2000). Internal borders were not abolished, but subsumed into new deterritorialized, digital, technological and transnational instruments of surveillance at a distance (Guiraudon and Lahav 2000;Bigo and Guild 2005;Broeders 2007).
The Dublin Convention comes therein into the picture. Put into force in 1997, transformed into an EU regulation in the aftermath of the Amsterdam Treaty (The Council 2003b), and subsequently reformed (European Parliament and The Council 2013b); it constitutes the core pillar of the Common European Asylum System (CEAS). It progressively encompassed all EU MS plus Norway, Switzerland, Liechtenstein, and Iceland. It was initially purported to solve the issue of "refugees in orbit," mutually rejected by each country. However, it progressively turned out to primarily target multiple asylum claims or "asylum shopping," in the neoliberal terminology (Moore 2013). Framed as a form of abuse, such practices could not be stemmed without an instrument to trace these so-called "irregular secondary movements": the EURODAC. 3 In this manner, incentives to stay in one place are not provided by expensive fences, social benefits or adequate work conditions, but by the simple threat of a biometric banishment from the whole of Europe in case of defection. According to Moulier Boutang (1998), the enlargement of any market enables to extend the area in which negative externalities provoked by the flight of labor can be neutralized. But he forgets an essential precondition for validating this hypothesis: the possibility to trace mobility within this market. This is the essential function endorsed by the EURODAC in the context of the European single market. By enabling the condition of "cyber" or "digital deportability" (Papadopoulos et al. 2008;Tsianos and Kuster 2012) within the market itself, it constituted the essential prerequisite for enforcing a dual regime of (im)mobility able to reproduce bridled labor alongside free wage labor.

Frontline vs. Core Member States.
While the outsourcing of coercion further into Africa and the Middle East through financial and political incentives garnered all the attention, another kind of externalization was inconspicuously taking place within Europe itself. We can find the origins of this process of "double externalization" (Staatsprojekt EUropa 2017) around the end of the 1980s with the emergence in the German context of the concept of "safe third country" (Hailbronner 1994), where asylum seeker having transited through should be returned. Since then, such rationale has been fully developed within and beyond the Dublin framework. The production of the distinction between EU citizens and TCN and the one between internal and external borders along the enlargement of jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 Schengen ran parallel to the (re)production of an asymmetry between core MS and weaker, peripheral ones. Although it may have enabled the latter's nationals access to free movement, it came with a price: to accept becoming "buffer zones for refugees" (Neuman 1993; see also Geddes 2000) coming from the peripheries.
From the onset of the Dublin regime, accession to Schengen has been conditioned upon the acceptance of the causation principle-according to which "the state that has 'caused' the entry of an asylum seeker is also responsible for processing the asylum claim" (Kasparek 2016: 62)-and the correlated alignment of each domestic migration policy with the ones of the most restrictive MS. This is first signaled by the fact that peripheral MS signed the first 1990 Dublin Convention even before being allowed to partake to Schengen negotiations (Hailbronner and Thiery 1997;Kasparek 2016). The territorial discontinuities still in place today between the Dublin, Schengen and EU areas further unravel these uneven power relations. Currently, aspirant Schengen members-Cyprus, Croatia, Bulgaria, and Romania-are compelled by the EURODAC regulation to register all TCN apprehended while irregularly crossing the external borders (European Parliament and The Council 2013c), notwithstanding that those borders are not properly Schengen ones. At the exact opposite end we find the UK (as well as Ireland, Norway, and Iceland), which until Brexit was part of the Dublin framework albeit outside the Schengen one; but for obviously different reasons. In short, peripheral MS have to protect Schengen before enjoying it whereas powerful ones can pick and choose the provisions of the European legal framework that best suits their interests.
Italy is the paradigmatic example to illustrate the demanding position left to MS at the geographical frontline of the Schengen space. Like Spanish or Greek ones, Italian officials were firstly excluded from intergovernmental discussions. It was only once the country had upgraded its security apparatus that they were invited to sign or leave (see Bigo 1992: 60). Since then, each time Italy was hit by new sea arrivals, the same infamous scenario has been repeating itself. While its government requests a quota system in place of the principle of first country responsibility, its partners threaten to reinstate their borders, by arguing that Italy is not fulfilling its legal obligations, leading to what has been referred to as Schengen intermittences (Garelli 2013). In the wake of the 2014 upsurge of sea arrivals, the first strategy of its authorities consisted of a "tacit acquiescence" (Heller and Pezzani 2016: 12) of migrants' refusal to give their fingerprints at disembarkation (Tazzioli 2015;Sciurba 2017). This "wave through approach" (Campesi 2018) enabled them to reduce the number of persons for whom Italy was held liable and to increase pressure on the EU to get more support. In response, the Commission imposed the "Hotspot Approach" (EC 2015). While the promised relocation scheme it foresaw broadly failed (European Council 2017), the deployment of European agencies on Italian and Greek islands succeeded in enforcing the systematization of the transmission of newcomers' fingerprints to the EURODAC (EC 2016a; Heller and Pezzani 2016). As the status quo remained unchanged until now, not less than 91% of the external borders crossings datasets transmitted to the Eurodac since 2003 were performed by Greece, Italy, and Spain alone. 4

Map 1: Net outgoing and incoming Dublin requests 2014-2019
Based on Eurostat incoming and outgoing requests. When outgoing data were not available for one country, it has been inferred from other countries incoming data, and vice-versa. This method is subsequently systematically adopted when dealing with Eurostat data.
As seen in the Map 1, the ratio between the number of requests of transfers sent and those received by each MS varies radically. The five original Schengen MS (Luxembourg, Belgium Netherlands, France, and Germany) have a clear positive ratio, whereas peripheral MS receive far more requests than they transmit. If one classifies countries according to the ratio incoming requests/total requests (Chart. 1), this spatial discrepancy is further confirmed. Except for Greece (see further), the analogue ratios for effective transfers are similar. Thus the Commission is globally incorrect when asserting that "net transfers in Dublin procedures are close to zero" (2016c: 12).
jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 That said, the EU is not merely a monolithic institution at the service of German, French, or British interests, but also a field of struggles between heterogeneous stakeholders. Noticeably, its Court of Justice has played a vital role in pushing towards a "collectivization of responsibility" (Guild 2006) and in limiting the excesses of the Dublin regime. Since a 2011 ruling (CJEU 2011) following a similar one of the ECtHR (2011), transfers to Greece have been suspended due to the systematic deficiencies of its reception and asylum system. Furthermore, besides the much more favorable stance of the Parliament, the Commission itself has questioned in various occasions the consistency of the Dublin system with the principle of solidarity and fair sharing of responsibility enshrined in the Article 80 of the of Treaty on the Functioning of the EU (see CEC 2007: 10). Without getting into the specifics of the 2015 relocation scheme, the Commission's long-standing commitment for such a quota system can be acknowledged (EC 2010).

Chart 1: Net EURODAC Hits, Dublin Requests and Dublin transfers 2014-2019
EURODAC hits data are extracted from EU-LISA annual Reports on the activities of the EURODAC. For request, data are the same as those used in map. 1. Data relating to transfers are drawn from Eurostat Incoming and Outgoing effective transfers. All charts include minors and are made with libreoffice calc.
Nevertheless, the deadlock of the relocation mechanism along with the envisaged reform of the CEAS since 2015 epitomizes how much these efforts were in vain. Alternative proposals would never reach consensus in the Council. Yet, it is the core countries bloc-rather than the Visegrad Group lead by Victor Orbán noisily opposing the quota system-that has the greatest interest and capability to impose the principle of causation, unaltered from the 1990s until the recent 2020 Migration and Asylum Package. Consequently, the current system plays in favor of core MS while eventually transforming southern peripheral ones into the "concentration camp"-to quote the former head of the Immigration office of the Italian Ministry of Interior (Comitato Schengen 2016: 7)-or rather the "plantations" of Europe.

Division/Multiplication of Labor
Incentives to secure southern MS consent are not reducible to the consecutive enlargement of EU budget for the reinforcement of external borders, commitment to preserve free movement, or promises of relocations. Because they are primarily addressed to those to be contained, those incentives culminate in substantial EU funds dedicated to reception, asylum, integration, and return programs. Parallelly, the three "harmonization directives" 5 were purported to improve asylum seekers' living conditions, especially in southern MS. If it is laudable to some respect, one shouldn't ignore that their supposed deterrent effect towards secondary movements has been essential in legitimizing this still unfinished harmonization objective. Meanwhile, the Commission's willingness to improve reception conditions in a country like Greece has eventually been driven by the ambition to resume Dublin transfers (EC 2016d). Crisis-hit countries have then turned into rentier states and asylum seekers into a commodity not only for states or NGOs, but also for private companies or criminal networks involved in the reception system (New Keywords Collective 2016: 27). Whether by dissuasion from leaving or incentives to remain, the CEAS contributes to this commodification in the sense that it partakes to the fixation of migrant labor in the semi-peripheries of the EU.
For thirty years now, where guest worker programs in core countries have become the preserve of southern and eastern Europeans, entire economic sectors of peripheral MS have come to rely on the over-exploitation of TCN precarious workforce. In areas like southern Spain or Italy, Schengen definitely fostered emigration in the north while, despite youth unemployment rates reaching 30 or 40%, the domestic workforce was unwilling to pick up tomatoes or oranges for three euros an hour in the "plantations." Consequently, endogenous labor was swiftly replaced by extra-EU exogenous ones in specific sectors. In this regard, the Dublin regime is coherent with the regime of [semi-]illegalized labor in southern countries, at least until 2008. Afterwards, while the financial crisis negatively impacted regional labor markets ability to absorb exogenous labor, southern MS started contesting the causation principle (Kasparek 2016).
Unsuccessful in this aspect, they have alternatively multiplied workfare programs embedded in the reception systems for reducing further the cost of migrant labor. For example, the Minniti-Orlando Law gave substantial capacity to local governments to enhance development projects through unpaid work of accommodation centers residents. Yet mechanisms for retaining migrants in the south are the paramount condition for implementing such schemes. When asked why he did not move north instead of accepting the underpaid agricultural job proposed by his reception center manager in Sardinia, a Ghanaian asylum seeker simply answered "because they would take me back." 6 While leaving his reception center would entail the closure of his application, he would not get a chance to register a new one anywhere in Europe. In other words, the availability of his workforce rested on a legal framework which inhibits the possibility to go search for better opportunities during the several years that the procedure lasts. Like deportability for irregular [ized] migrants or the conditionality of the residence permit renewal upon the possession of a work contract for labor-related immigration, the Dublin regime enables the fixation and subordination of asylum seekers labor in semi-formal local labor markets.
Here it is important to recall that the scope of the Dublin regime extends beyond the period of the procedure. If someone is granted a protection, the person may eventually be entitled to travel within Schengen for up to three months. But it does not entitle them to settle and work outside their country of residence before the hypothetical granting of a long-term residence status at least 5 years later. Conversely, if the application is rejected, the person finds themself under a double threat of deportation: from the MS which made the decision to her origin country and, in case of a "secondary movement", from the one of arrival to the former. In this regard, the Dublin regime is highly innovative in the sense it fixes one of the most mobile kind of workforce: the undocumented one.
If it would be excessive to conclude this regime aims at providing certain economic sectors of the semi-peripheries with bridled labor, it may contribute to the formation of semi-formal local labor markets at the margins of the EU-wide endogenous one. To grasp further the pretense of the single market, it is pertinent here to outline some mechanisms of differential inclusion/exclusion of migrant labor (Mezzadra and Neilson 2013;New Keywords Collective 2015) entrenched in the regulation of asylum seeker's access to the labor market.
As seen in Chart 2, whereas southern MS have adopted shorter delays and almost no additional barriers, northern ones kept longer time frames and have multiplied requirements to make this right effective. As the German authorities perfectly summarized: "Waiting periods for access to the labor market seek balance between the desire for integration and the desire to reduce social welfare expenses on the one hand and the problem of creating unwanted incentives for persons that apply for asylum in Germany merely out of non-asylum-related but purely economic reasons on the other hand" (EMN 2014: 4). In peripheral countries, the first argument prevails while in the core states, it is the later. There, whereas residence is granted upon the condition of possessing a work permit for work-related immigration, employment is prohibited for asylumrelated immigration exactly to deter inward (secondary) movements and eschew long-term settlement.

Chart 2: Minimum Delays Before Granting Access to the Labor Market to Asylum Seekers
Based on European Migration Network Ad Hoc Queries (2010, 2013, 2014, 2015 , 2016, 2019), Asylum Information Database Country Reports and a report on Denmark for the European Commission (Madsen 2016). Absence of columns means asylum seekers are de jure or de facto prohibited to work. Small ones that there are no foreseen minimum periods.
Against this background and in line with the ambition to harmonize asylum policies, the EU legislation rather tends to put limits to those barriers. Noticeably, the reception directive foresaw a maximum time limit of twelve months (The Council 2003a art. 11.2)-subsequently reduced to nine (European Parliament and The Council 2013a art. 15.1)-for granting the right to work. Therein we can observe a general trend towards the reduction of this time limit in the aftermath of 2015. But the ambition to standardize it at the EU scale has not been achieved at all. While the Commission has commonly praised for it in the hope of reducing "employment-related asylumshopping and incentives for secondary movements" (EC 2016b: 4), it may be for those same reasons that core MS are reluctant to lift obstacles (EMN 2014).
Even when northern MS significantly reduced the delay, other complex rules come to nuance this trend. In Germany for instance, it has been reduced from 12 to 9 (EMN 2014), and finally 3 months in 2016. While during the crisis, the lengthy procedures played a dissuasive function, in a second phase, Germany promoted swifter employment integration. As the asylum system got jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 overrun, it suddenly became appropriate to grant access to employment to those likely to get status. But in the meantime, the maximum period of residence in the newly created initial reception centers (Anker Centers) grew from 3 to 6 and finally 18 months in 2018 (AIDA 2020b). Insofar as their residents are mostly banned from work and the rules governing outward transfers are very individualized (ECRE 2019a), this inverse tendency reflects the ongoing flexibilization of the access to the labor market further enhanced through the possibility to refuse work permits on the discretion of the Federal Employment Agency (AIDA 2020b: 95).
Similarly, Netherlands, Austria, Sweden, or Norway have progressively fostered so much complex conditions for the access to the labor market that possible delay reductions lose their significance. The first kind relates to the type of requests; for instance, "inadmissible," "manifestly unfounded," "safe country of origin," fast track and, obviously, Dublin procedures (see further) are generally set apart. The second concerns administrative obstacles: labor market test, limitation to particular sectors, or hire promises for work permit. Lastly, those deemed responsible for extended procedure delays are likely to be denied the right to work even after 9 months.
In any case, the north/south gap undoubtedly manifests a genuine European division of labor with which the Dublin regime is strongly coherent. However, even in southern MS, restrictions on the labor market access can be acknowledged. Whether they are pushed in informal employment or granted the right to work, but under strict conditions, asylum seekers do not enjoy full access to the labor market. That is to say that it is not merely a question of division between MS but also of multiplication of labor within each of them. Whether it successfully contains migrants spatially or not, the Dublin regime plays an essential role in this formation of multiple local exogenous labor markets alongside the EU-wide endogenous one by formally denying migrant labor mobility within the single market.

The Government of "Secondary Movements"
Today, it is not uncommon to hear thinkers like Slavoj Žižek asserting that migrants "are offering themselves to become cheap precarious workforce, in many cases at the expense of local workers" (quoted in De Genova 2017: 16). Nor should we be surprised at his opinion that European hospitality should be conditioned upon their acceptance of "the area of living allocated to them by European authorities" (quoted in De Genova 2017: 15). What the self-proclaimed communist thinker misses is precisely the relation between his two statements. For it is because they do not benefit from freedom of movement within the single market that they are pushed to accept lower wages. The reserve army is not the natural outcome of the mythical invisible hand. Migrants are not imported by the cosmopolitan bourgeoisie nor are they "offering" themselves as cheap workforce. They have to be forcefully shaped so, through institutional heavy-handed intervention for limiting internal mobility-those same ones Žižek promotes. The Dublin regime is the name we can give to this intervention in the Schengen context.
At this stage, it is pertinent to recall the prerequisites identified by Moulier-Boutang (1998) for the validation of the reserve army model. First, an internal mobility between each subset (corporation, sector, area) of the market. Secondly, no defection to the informal market or under the form of emigration. Yet, those two conditions, whether the Dublin regime fails or not, are never fulfilled altogether. In the eventuality migrants comply with the rules-like the Ghanaian citizen quoted above-it implies their relegation into segmented subsets of the market distinct from the EU-wide one. Thus they don't represent a threat to domestic worker's wages. Conversely, a significant part of migrants persist in engaging in strategies of flight from those segmented markets. We thus propose to turn our gaze towards the continuous failure of the Dublin regime to stem the so-called "irregular secondary movements". We may then interrogate how this "incorrigibility of migration" (De Genova 2010) has been taken into account in the reconfiguration of the Dublin regime.
The Predictable Failure to Fix at the Peripheries Mobility might be the main asset of labor for escaping its commodification: "Migratory fluxes never let themselves be subjected to the law of offer and demand, neither are they totally manageable by the mechanisms of control" (Staatsprojekt EUropa 2017: 9). Migrants do not necessarily stay where there are more work opportunities or where they have been assigned to. For certain, the reintroduction of internal checks, the risk of Dublin deportability, and the hurdles to enter the labor market are dissuasive. But there is substantial evidence that some persist in leaving their country of first identification. "Border spectacles" (De Genova 2002) at various crossing points (Balkans, Calais, Bardonecchia, Ventimiglia, Como for Italy) vividly remind us of this. In fact, "secondary movements" are one of the most frequent arguments raised by MS in their notification of reintroduction of national borders checks (EC 2019). Albeit such supposed temporary reintroductions have become the new normal since 2015, they have hitherto failed to curb such movements, so as the Dublin framework, according to the Commission itself (2016c: 11).
The most notorious episode in the matter is the collective motion between Greece and Germany that unfolded in the summer 2015, which forced the German government to make concessions. By the end of August, Angela Merkel announced the suspension of Dublin for Syrians. Basically, the so-called "sovereignty clause" (European Parliament and The Council 2013b: art. 17.1) allows any MS to take charge of an application even if another one should otherwise be liable for it. This "generous" move should not mislead us to declare the death of the Dublin System. As a matter of fact, Germany never ceased to be the first MS in terms of outgoing requests and transfers (see Chart 3). Knowing that Syrians are the most likely to be granted a refugee status and, incidentally, also a relatively skilled category of asylum seekers, the lesson of this event is that the autonomy of migration can eventually be taken into account into governmental calculations.
Drawing from this example, it could be argued the Dublin Regulation-despite its apparent failure-and its sovereignty clause allows powerful countries to enforce selective mechanisms of differential inclusion of migrant labor. Skilled migrants would be incorporated in the core areas. Undesirable ones would be relegated to peripheral MS or to the informal labor market of the core. But to which extent does such a model fits with the reality? Do the policies faced by Dubliners jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 once they reach another country enable the extraction of surplus value from them? To explore those questions, we first need to clearly differentiate Dubliners who do not claim asylum but live and work semi-irregularly up north from those who engage in an asylum procedure once there.
For the first case, let's take the example of those circulating between southern Italy where they undertake seasonal agricultural work in the "plantations" and northern Europe in search for a better life. When their procedure in Italy is successful, they are often granted a precarious residence permit "not valid for expatriation" (non valido per l'espatrio): they do not benefit from the right to travel to another MS. Besides, they are compelled to come back from time to time to renew their documents, resulting in regimes of hypermobility (Fontanari 2016;Gambino 2017;Picozza 2017). At first glance, far from facilitating the harmonization of domestic policies, the Dublin regime reinforces the effects of the fragmentation of residence permits and legal statuses entitled to TCN within each national law, contributing so to their subordination within work relations.
Nevertheless, their probability to receive a legal status (independently from the Dublin issue) in another EU country is generally very low. Furthermore, whereas the Dublin regime extends the condition of deportability to categories of migrants otherwise spared from it, deportability within Europe spares those already facing this risk from being removed outside Europe. According to some Malians 7 , it is now common practice in their community to voluntarily register in Italy or Spain before moving to their destination so to be relieved from deportability to home country. Without contradicting the hypothesis of multiplication of labor, this manifests migrant's ability to take advantage of the Dublin regime contradictions and puts into question the later effect in terms of labor subjection. That said, we will now turn our focus to the second case. Namely, those claiming asylum subsequently to secondary movements.

Differential Exclusion of Dubliners from the Core
Despite or because of the improvement of the EURODAC, the administrative category of "Dublin procedure" has now come to represent more than 30% of the applications in countries like France (Bailo et al. 2019) or Germany (BAMF 2020: 32). From being an anomaly with regard to the refugee/economic migrant dichotomy, it has now become a mass phenomenon. Once registered, Dubliners face a procedure of determination of the responsible MS which can last several months. "Take charge" or "take back" requests can either be refused, accepted or ignored (resulting in an "implicit acceptance") by the requested MS. In the first case, the procedure is reconsidered as regular. Otherwise, a "transfer decision" is notified to the person. From this moment (or, in case of recourse, from the date of its rejection), the sending MS authorities have six months to implement the transfer. After this deadline, if the authorities are held responsible for this dead-end, the Dublin procedure is extinguished and the application reclassified as "regular." Conversely, if applicants are deemed to have obstructed their transfer, they are subsequently registered as "absconders." Then, there are three possible outputs of the Dublin procedure: 1) Removal; 2) Responsibility shift; 3) Absconding, which will be treated in the last section. jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 Effective transfers. As Chart 3 shows, some countries like Sweden are relatively efficient in their transfer policies. It might be related to their welfare system as much as with their police efficiency. Because the high level of dependency of the applicant upon the reception system makes it mostly impossible to breach its rules or to live at its margins, transfers are mostly enforced on a voluntary basis. However, the chart first and foremost illustrates the well-known global inefficiency of the Dublin regime in terms of transfer rates. For instance, in 2018, 27,769 transfers have been carried out across Europe 8 . This represents 28% of the accepted requests 9 and 18% of the total number of requests 10 . Besides, it is asserted that many returnees re-engage in a secondary movement just after their transfer (EC 2016e: 4). Yet, notwithstanding most Dubliners are finally not removed, they remain exposed to serious exclusionary mechanisms.

Chart 3: Average Transfer Rates Between 2014 and 2019 in Northern MS
Drawn from Eurostat, Outgoing, and Incoming transfers. The ratio is calculated in relation with the number of accepted outgoing requests and accepted incoming requests. The absolute number of effective transfers is indicated in the columns.

Temporal Borders of the Dublin Procedure.
In countries like Germany, France, or the UK, low transfer rates do not necessarily mean the Dublin regime is without consequences, if only because it significantly delays the examination in substance of the applications. For example, the British rate may not derive from a lenient application, but rather from the hardship to reach the island that leads Dubliners to prefer anything else than removal-including stifling procedures, lengthy detention, exclusion from the right to work or from the benefit of the material reception conditions, intrusive surveillance, mobility limitation, and temporary irregularization in case of "absconding." In practice, many exceptions infringe the principle enshrined in the reception directive of equal rights between Dublin and regular asylum applicants. Often, Dubliners' monthly allowances jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 are withdrawn or reduced. In France, even former Dubliners are often processed in fast track and consequently denied their material reception conditions (La Cimade 2019b: 55). Priority is frequently given to others in terms of housing. Or, when Dubliners are granted accommodation, it turns out to feature an "alternative to detention": compulsory accommodation, regional limitation of circulation, weekly or daily appointments at the police, or even house arrests. Any breach of those rules can be interpreted as "absconding" or "risk of absconding" justifying subsequent detention.
With regard to the right to work, even where it is granted quickly to regular asylum seekers, Dubliners are often banned from it (Chart 2). Whereas in Sweden (AIDA 2018) or Norway (EMN 2019), Dubliners are explicitly prohibited to work; in Austria (AIDA 2020a: 88), Denmark (Madsen 2016: 2), or France (ECRE 2019b: 94), the delay before which asylum seekers are granted the right to take up employment begins only once the country has assumed responsibility for the examination of the case. Meanwhile, in Germany, because their case is considered "inadmissible," they are compelled to remain in Anker centers (AIDA 2020b: 82). Finally, since they are generally assigned in remote accommodations, Dubliners are de facto isolated from their community, from urban areas, and thereby from the informal labor market as well.
Nevertheless, there is significant evidence that such sanctions fail to convince the majority of Dubliners to leave. There is no EU-wide data about the outcome of Dublin procedures. In France still, the ministry of interior has recently published statistics thereupon (see Chart 4). Since 2016, more than 50% of launched Dublin procedures have sooner or later been reconsidered as regular (Bailo et al. 2019;Del Biaggio 2020). Despite the incompleteness and contradictions of the available statistics, it is rather plausible that proportions are similar in other northern MS. Against this background, national and EU policies lean towards further sanctions against secondary movements in the hope to overcome Dubliners' stubbornness. At the EU level, whereas the Commission is now committed to further reduce to six months the maximum time-frame before granting access to the labor market (EC 2016b art. 15.1), it advocated for denying all Dubliners the material reception conditions, including the right to work (EC 2016b:20). 11 The way the majority of northern MS are handling Dublin does not result in territorial exclusion (division of labor) nor in differential inclusion (multiplication of labor). We should rather talk of life "suspension" (Brekke and Brochmann 2015;Oelgemöller 2011;Fontanari 2015;Picozza 2017) within the Dublin's "temporal borders" (Rigo 2011;Mezzadra and Neilson 2013) during which the person is kept forcefully unproductive. The purpose of this suspension that can last up to several years in case of "absconding" is firstly to proactively discourage secondary movements and to incite voluntary departure. This eventually enables a reduction of the financial and political cost of both reception and transfers policies. Finally, it also prevents any social integration that could subsequently open other regularization pathways. However, as we have seen, Dubliners' unrelenting determination frustrates those objectives, which in turn incite MS and EU jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 institutions to further increase those sanctions. The outcome of this vicious circle is greater social marginalization of Dubliners, to the point that their workforce is not exploited at all independently from their skillfulness. It is as if the deterrent rationality and the capitalist one had come into an irremediable contradiction and the former had superseded the later. This is the hypothesis we will explore further with the issue of "absconders." The Dubliner in a "Relation of Ban"? Dubliners have continuously improved their strategies to, in their words, "kill Dublin," just as the Dublin regime has improved its ability to recapture those strategies. The highest stage of this escalation is the re-framing of fugitives from transfers as "absconders." The Dublin Regulation provides that the six-month delay for carrying out a transfer "may be extended…to a maximum of eighteen months if the person concerned absconds" (2013b art. 29.2). The growing securitization in the recent years of those strategies for escaping Dublin transfers, far from stemming them, may have resulted in the significant increase of the administrative category of "absconder." There is no EU-wide data to assess the scope of the phenomenon. But according to twenty MS, "the absconding of the applicant [is] the primary explanation for delays" (EC 2016c: 10). In France, data transmitted by the ministry to the NGO la Cimade (2019a) corroborate this hypothesis. From 2015 to 2018, "absconders" have multiplied by ten (from 1566 to 15,000), whereas effective transfers by seven (from 525 to 3,533) and initiated Dublin procedures only four (from 11,657 to 45,538).

Chart. 4: The Outcome of Dublin Procedures in France
The number of regular applications initially registered as Dublin during the same year (yellow) and during the precedent years (orange) is extracted from the Ministry of Interior reports (2018,2019,2020). Effective (outward) transfer data are extracted from Eurostat and the number of "absconders" (not available for 2019) are data transmitted by French authorities to la Cimade.

Estinguished from the current year
Estinguishing from the past years

Extension of Dublin delays Effective transfers
Those statistics reflect the strategy adopted by both migrants and the French administration. From 2015, the latter was facing a sharp increase of Dublin procedures (chiefly towards Italy). The incapacity to transfer them resulted in numerous "shifts of responsibility," which in return attracted to France Dubliners willing to "kill Dublin" (Montel 2016). The administrative apparatus was consequently redrafted for trapping them into the dilemma between removal and absconding, inter alia by providing them (compulsory) housing for keeping track of them-and incidentally, for satisfying some NGOs demands.
Some Dubliners have reported having been given an order to present themselves at the airport early in the morning, when public transportation was unavailable. 12 Or they were simply asked to come sign up frequently with the police despite the absence of any organized removal. Such tricks suggest the apparent preference of the authorities for exclusion from within over exclusion from the territory. Because the former is less costly-or because it may be more dissuasive than a back and forth journey to the responsible MS-French authorities might have reversed the means and the ends of the Dublin Regulation by actively manufacturing absconders. Or rather, it becomes impossible to determine what is the means and what is the end. Meanwhile criteria for considering someone as an absconder multiplied, a legislative reform introduced a three years' prison sanction for returning to France after a transfer (French Parliament, 2018: art. 36). In other words, the probability and cost of both transfer and absconding from it have been steadily growing.
Absconders generally cannot have their temporary residence permit renewed and their material reception conditions-including the right to work-are withdrawn. This is why, in the camp-shifts of the northern areas of Paris, as even its municipal authorities regretted (Hullot-Guillot 2019), asylum-seekers waiting for an accommodation offer have given way to those already given an accommodation but subsequently expelled because they absconded from their transfer procedure. It can be added that many others living in these camps were former absconders in Germany, Austria, or Sweden struggling to have their case examined in France. 13 Lastly, in the eventuality they decide to return on their own to their responsible MS, absconders don't stand a chance to be reintegrated in its reception system.
In short, absconders who come floating in Parisian camps are subjected to a European-wide "Ban." 14 But this Ban does not arise from a "state of exception" like Agamben (1998) famously asserted. We contend here that Agamben's concept of Exception might describe the dream of some bearers of power, but not the complexity of social reality in which resistance plays a key role (Bigo 12 Personal interviews and field notes, 2017-2018 13 Field observation, Paris campshifts, 2015-2020. This trend became more evident in recent times. 14 As reminded by Jean-Luc Nancy, Ban is an old German term meaning both the exclusion from the community and the insignia of sovereignty. Reflecting on that double-meaning, Giorgio Agamben develops the concept of "relation of Ban" as a constitutive feature of modern sovereignty which stems from the "relation of Exception". He conceptualizes it as the paradoxical condition of inclusion/exclusion from the juridical order experienced by the subject abandoned and banished from the political community and at the same time directly exposed to punishment of the sovereign and included in the juridical order through its suspension (1998: 28-29). This subject-the homo sacer-is thus denied as a political subject while reduced to a biological (bare) life which "may be killed and yet not sacrificed " (1998: 9) jwsr.pitt.edu | DOI 10.5195/JWSR.2021.1024 2007: 12). This Ban is rather the effect of a petty exceptionalism compatible with liberal regimes (Bigo 2008), embedded in the routine administrative practices deployed for policing Dubliners "abuses" (Montel 2021). Circumscribed in time, this Ban unfolds beyond the space of the camp: absconders are not territorially but juridically banished. In a so-called "borderless Europe," exclusion takes place from within and in open-air. It could be referred to as "containment through mobility" (Tazzioli 2017). Here, "The Ban is the freedom to leave where one does not want to go. It is the time of endless circulation, of constant rotation. It is a way to bet on orbit and to manufacture a satellite population" (Bigo 2007: 31). Because of legal barriers, excessive precariousness and hypermobility, this satellite population is not even subjected to differential inclusion, but to pure exclusion from the sphere of production. Status-less, the reproduction of their biological life rests upon mere charity.
There is, though, a relevant nuance to it in Germany. Be it rejected applicants, visa overstayers or Dublin absconders (MacGregor 2020), some undocumented aliens can be granted a "tolerated stay" [Duldung]. This suspension of removal orders guarantees a paradoxical sort of limbo status of non-belonging which can last several years (Fontanari 2015). Duldung holders are permitted (and compelled) to stay in their reception center, while receiving minimal welfare assistance. Recently, legal pathways for the most deserving and desirable ones have been introduced. Some of them may apply for a work permit under very strict conditions six month after the delivering of the Duldung (AIDA 2020b: 94), paving the way for a work residence permit on the long run. However, since they are held liable for the non-implementation of their removal, absconders are likely to be refused this work permit (Handbook Germany 2020). Only in their case, the deterrence rationality prevails over the economic one. Rather than belying the hypothesis of the absconders as subjected to the Ban, the Duldung case thus confirms the peculiarity of the exclusionary policies they face compared to other categories of irregularized migrants.
One could expect that the irregularization of absconders contributes to their absorption in the informal labor market. In this hypothesis, the construction of this new category would be congruent with the need for bridled labor within the core area. But that would amount to neglect the question of the (re)production of labor force and its subjective dimension. As a matter of fact, the everyday precariousness faced by absconders encountered in France was so blatant that minimal living conditions necessary for working (even illegally) were not met. Meanwhile in Germany, it is plausible that only a small number the Duldung bearers integrate themselves into the informal labor market 15 . Furthermore, they may be doing it only for collecting the necessary amount to move somewhere else or, at worst, in waiting for the end of the 18 months' delay 16 . Broadly said, the subjective condition they live within these temporal borders makes them unlikely to work. They rather prefer spending the 18 months moving around Europe in search for a flaw within the Dublin regime. 15 According to field observation among asylum seekers in France previously living in Germany (2020). 16 Drawn from interviews with people on the move in Calais (French city where most of those wishing to join the UK are stranded in search for a flaw in the British border) who were previously in Germany (2020 Then, the banishment of the absconder comes with a price: abandoning any ambition to commodify migrant labor so as to maximize the relative surplus value appropriated from it. When "slaves" decide to escape the "plantations" of Europe, their refusal to become bridled workforce makes them unfit for surplus value making. As a last resort, then, "the capital prefers renouncing exploitation" (Balibar 2001: 198). Maroons must then be punished and reshaped into the Arendtian figure of the Stateless (Arendt 1982: 5) or the Agambenian homo sacer of which "sacredness of life [expresses] life's irreparable exposure in the relation of abandonment" (Agamben 1998: 83). We propose to use the term of "Schindler threshold" 17 to refer to this instant of dis-articulation between exclusionary policies and the rules of accumulation, where processes of inclusion/exclusion do not even take individual skills into account-not to compare the context discussed here with the Nazi Lager. Noticeably, the Duldung can be interpreted as the affirmation by a state haunted by its traumatic history of the unconditional sacredness of life even when forcefully kept worthless. In fact, if Agamben does not mention Auschwitz in his list of examples of the camp-form (Agamben 1998;Fassin 2012), maybe is it because the fracture he pinpoints between "qualified" and "bare life" firstly requires the reproduction of the later, and not its elimination.
Following some critics of Agamben (Mezzadra 2003;Tsianos and Karakayali 2010), we contend the camp itself-or Europe as an open-air "campscape"-assumes a function of spatiotemporal regulation of labor within global capitalism. It is only once this function is challenged by migration that exceptional means are deployed. The production of life "in excess" is not a "mean without ends" (Agamben [1995] 2006) inscribed in the immutable ontology of sovereignty. In the case discussed here, the end is dissuasion from secondary movements. And the making of the absconder is the inevitable conclusion of the ambition to constrain migrant labor: "Each time the labor market is bridled, i.e. the mobility of the dependent is constrained by a way or another, there is, in response to new channels of flight, an enlargement of the wage/labor transaction to the public and political sphere." Namely: "The limitations of the access to citizenry, to the city, to housing and exogamy" (Moulier-Boutang 1998: 19). If it was not for Dubliners perseverance to open those channels, the Schindler threshold would not be overcome. This is also why the unattainable horizon of "bare life" cannot be assimilated to a depoliticized life. As Agamben himself surprisingly concludes, the exile is always struggling to find "the best way to elude or deceive [sovereign power]." As a consequence, "no life…is more 'political' than his" (Agamben 1998: 184). This is precisely the case for absconders struggling to cross the extended temporal borders of the Dublin regime. Therein, Chart 4 confirms not only that more than 20% of applications registered as regular in France were formerly classified as Dublin, but also that the majority of them were firstly registered in the precedent years. Therefore, this chart suggests many ex-Dubliners were also ex-absconders and many are the absconders having finally managed to be registered as regular after the 18 months' period. And the political meaning of their resilience should not be underestimated. For it is the apparently most unqualified form of life which might be imperceptibly subverting the borders of Europe. Their persistence in encamping in the streets of Paris is not a no-choice, but an active strategy for overcoming Dublin's Temporal borders. This is exactly because of this infrapolitical nature of their action that they are denied even emergency housing, while being depoliticized through the humanitarian discourse.
Similarly depoliticizing is the re-framing of those strategies as "abuses" as the Commission does in its Dublin Regulation recast proposal: "Such shifts appear to have encouraged circumventing the rules and obstructing the procedure" (EC 2016: 16)-before calling for the deletion of all provisions enabling those responsibility shifts. Such a proposition would make the status of absconder everlasting. Admittedly, contrarily to physical ones, temporal borders feature an endless elasticity. But isn't their circumscription an essential safeguard for making petty exceptionalism compatible with liberal values? Furthermore, human subjectivity can't be reducible to mere biological life if not discursively: as political subjects, Dubliners will continue to find a way out. Therefore, the Commission proposal to make responsibility for an asylum application perpetual is likely to open unexpected channels of flight and result in hypermobility instead of fixation, unpredictability instead of security, suspension of life instead of its commodification. If adopted, this indefinite temporal extension of the Ban can only but deepen the overcoming of the Schindler threshold. In the meantime, the exacerbation of the tension between migrants' incorrigibility and tactics of rebordering thwarts the pretense to merge liberal principles and petty exceptionalism, fixation and deterrence, freedom of movement for some and its denial for others, liberty and security, and finally, the capitalist rationality and the "racist" one.

Conclusion
The Dublin regime is a paramount expression of the need for a differentiated freedom of movement for the processes of accumulation in the world-system. Designed and promoted by the wealthiest MS, its effective function outreaches mere asylum issues: once the European workforce could move freely, restrictions on the mobility of non-European one became an essential prerequisite for maintaining a spatial division of labor, but also for the multiplication of labor within each region. Since physical borders were inadequate for enforcing such a dual regime of (im)mobility, the EURODAC surveillance apparatus, understood as a "security dispositif" (Foucault 2004), was enhanced for complementing them with biometrical, administrative, and temporal borders. If this governmental program was fulfilled, undesired external borders crossings would be swiftly recaptured: the neoliberal and the reactionary hegemonic projects (Staatsprojekt EUropa 2017) would then be compatible.
Yet, the EU is not a monolithic power bloc able to overcome its inner contradictions, especially between its core and its peripheries. Newcomers do not hesitate to take advantage of those frictions, particularly made visible by the "wave through policy." Besides, the (relative) autonomy of migration structurally exceeds governmental calculations since the dispositif is not merely repressive but also normative and productive (Foucault 2001;Bigo 2008: 38). Accordingly, the more freedom of movement is valorized, the more it is desired and appropriated by those excluded from it. In fact, the Dublin regime failure to contain secondary movements is now unquestioned. The sovereignty clause might allow core MS to make of this failure an opportunity to select the most profitable ones. But this option is always more pushed aside for deterrent purposes. Then, most Dubliners are neither transferred nor (differentially) included, but exposed to an undifferentiated life suspension that denies them not only as political subjects, but also as productive ones.
According to Wallerstein, what he refers to "racism-sexism" (Balibar and Wallerstein 1991) helps overcoming the crisis of capitalism as it enables to increase surplus value and to partially redistribute it to the European working class; but this presupposes the differential inclusion of labor into the sphere of production. Yet, in the case of Dubliners, we are rather confronted to the situation where "we lose the labour-power of the person ejected and therefore that person's contribution to the creation of a surplus" (Balibar and Wallerstein 1991: 33). In our analysis, this happens because, when migrant labor contests its assigned position in the European geography of power by moving somewhere else, it cannot be subsumed under the form of bridled labor anymore. The balance between inclusion and exclusion is disrupted and the Schindler threshold eventually reached.
The making of the absconder is the final outcome of a Europe not able anymore to keep productive the tension between the exclusive determination of the national criteria and processes of accumulation (Mezzadra in: Staatsprojekt EUropa 2017) as well as between excessive deterrence policies and excessive migration. The absconder constitutes the borderline case that highlights this "irrational" dimension of racism-sexism "when they come too far" (Balibar and Wallerstein 1991: 35) But it is irrational only if we reduce the "jurisprudence"-rather than abstract Human Rights (Parnet and Deleuze 1996)-and moreover migration itself to a mere superstructure. Border struggles waged by migration do not result either in the status quo or an overhaul of the system, but rather to reactionary exceptional measures that migration still manages to partially overcome. Sovereign exceptional powers are not enacted indifferentially towards all, but only towards those who, like Dubliners, cannot be governed otherwise. The dis-articulation between capitalism and institutional racism observed here does not stem from ideology or the unhistorical abstract raison d'être of the state. It is rather the empirical and material conclusion of an endless conflict waged by living labor to escape its subsumption under capital. In a nutshell, it is not capitalism that produces superfluous population "in excess." It is migrant subjects which "exceed control" (Stierl 2017) through strategies of flight able to challenge the processes of exclusive inclusion of their workforce within the sphere of production. Vincennes Saint-Denis. This current project envisions to propose a broader analysis, instead based on archival work and institutional fieldground, of the genesis of the administrative category of "absconder" within the EU and national frameworks.
Disclosure Statement: Any conflicts of interest are reported in the acknowledgments section of the article's text. Otherwise, authors have indicated that they have no conflict of interests upon submission of the article to the journal.