Introduction

Few would question the sense in encouraging quick social integration of migrants in general, and refugees in particular. It is not only better for the refugees themselves, and has many advantages for the receiving societies, but it also reduces transition costs in all senses of the term. Even though it is a priority of integration policies, we widely observe that the asylum adjudication procedure and the subsequent integration of people who have been granted international protection are often tardy processes characterized by conditionalities, temporalities, and geographical segregation.Footnote 1 Consequently, refugees are easily politically framed as a problem or a challenge to receiving societies, or as a category that poses a problem instead of having a problem they can be helped with. Furthermore, they are not viewed as people who, in the short or long run, are a boon to society. In the eyes of most European governments, and the European public in general, the costs of refugee reception ought to be lowered.

It was interesting to see how the reception of Ukrainian refugees developed in an extremely different and much more accelerated manner, skipping the laborious asylum decision procedures, giving Ukrainians the choice to move to any country in the EU that they wished, and the right to work immediately. What was also striking was that this mass reception of refugees was not framed by politicians and media as something problematic or impossible to cope with, while the opposite happened when Syrians, Afghans, and other asylum seekers entered the EU in 2015–2016 (Dražanová & Geddes, 2023).

About a year after the first arrival of Ukrainian refugees, we compare the labor market integration of Ukrainians to those of Syrians and Afghans. This paper aims to provide another perspective on how refugees’ integration process could be speeded up, to the advantage of the receiving societies. Doing so, we look at the present EU practices relating to and stemming from the Common European Asylum System (CEAS). Thus far, these do not come close to the ideal of optimal allocation and responsibility sharing. We also look at the practices that result from the invocation of the Temporary Protection Directive to address the mass displacement of Ukrainian citizens (and legal residents) to the EU’s member states and again compare these to the ideals formulated in the initial section. In the conclusion, we can ask what lessons we can draw from this comparison.

The central question of this paper is: which refugee regime (TPD or CEAS) leads to better integration outcomes for the labor market, and how does the procedure help (or not)?

The scientific contribution of this paper is that we add a critical comparison of two refugee reception systems to the literature on integration policy outcomes. As for the contribution to policymaking, we evaluate what the different outcomes in terms of labor market integration mean for a more efficient policy.

Theoretical Starting Point: Integration

As the central question indicates, in this paper we are looking at the integration outcomes of refugees under the two regimes. Garcés-Mascareñas and Penninx define integration simply as: “the process of becoming an accepted part of society” (Garces-Mascarenas et al., 2016, p. 14). They state that integration has legal, socio-economic, and cultural-religious aspects. Integration in the labor market is part of the socio-economic aspect of integration.

The Migrant Integration Policy Index Project (MIPEX) compares integration policies in 56 countries and rates them on a scale of six levels, from one (“critically unfavorable”) to six (“favorable”) (MIPEX, 2020). For example, Saudi Arabia is level one and Portugal, Sweden, and Canada are level six (MIPEX, 2020). The Netherlands is ranked at level four (“halfway favorable”). When we compare the Dutch integration policy to other European countries, it is relatively well developed on paper, but often without the required positive outcomes in practice (Pasetti & Lebon-McGregor, 2023). For instance, in theory, there are housing facilities for newly arriving asylum seekers, but in practice these facilities are full.

Even though Garcés-Mascareñas et al.’s definition points to the two-sidedness of integration, in Dutch policy documents about refugees, a more limited approach is chosen, namely the perspective that integration outcomes depend on an immigrant’s effort and adjustment, ignoring that adjustment should be two-sided. The integration concept has become contested as it is typically a policymaker’s perspective that does not take the aspirations of migrants themselves into account (Van Heelsum, 2017). Additionally, as Tamar de Waal (2019) points out, migrants deal with a long list of integration requirements. After feeling unwelcome because of the many difficulties in crossing European borders, and then bumping into the invisible borders of racism in the receiving societies, refugees feel reluctant to get such a list of integration requirements (Politi et al., 2023).

Dutch integration policy places emphasis on integration into the labor market and knowledge of Dutch language and culture. For refugees, this feels contradictory as they are stopped from working until they are granted refugee status. The elements are interconnected, since someone who is integrated into the labor market (working) may automatically get to know Dutch people and Dutch culture while learning the language is easier in a Dutch-speaking environment. Even in the cosmopolitan, English-speaking environments of international offices in the Dutch capital Amsterdam, it is easier to get acquainted with Dutch culture than being in an isolated asylum shelter or unemployed at home (De Lange et al., 2019).

We argue that fast integration into the labor market is easier when recent work experience and existing networks can be employed, and that working enhances contacts in the receiving society and language acquisition, which are important aspects of integration. In this paper, we focus on the speed of integration into the labor market. As we will show, Ukrainians are able to use their diplomas, work experience and networks immediately after arriving in the Netherlands, while Syrian and Afghan refugees under the traditional asylum system are stopped from integrating rapidly because it takes a long time before they are even allowed to look for jobs. Data on the labor market position of Syrian and Afghan refugees show how their integration potential is severely hampered. Not just for the refugees themselves, but for Dutch society as a whole, it would be beneficial if refugees’ educational and work experience, and networks, could be put into practice immediately (Vervliet & Heyma, 2024).

Outline of the Paper

We start this paper with a comparison of the labor market participation of Ukrainians and other (Common European Asylum System) refugees. This shows that there is an extreme gap between the two. Then, we proceed with information on the policies, and analyze which policy lines could explain the difference. First, we elaborate on the reception of Ukrainians in the Netherlands and what effect this had on their labor market participation. Then, we explain how the lower labor market participation of other refugees relates to the asylum procedure. We distinguish two main problems of the traditional system, namely a temporal problem and a spatial problem. We conclude by returning to labor integration and discussing the road ahead for a solid legal system that avoids the negative side effects for integration.

The Problem: The Differential Labor Integration of Ukrainians and CEAS Refugees

Work on this paper started after the observation of a stark difference between the percentage of working Ukrainians and other refugees. In this section, after considering the available data, we analyze the two policies and their implementation in the Netherlands, and explain this difference.

Labor Market Participation of Ukrainians in the Netherlands

The Russian invasion of Eastern Ukraine started on February 24, 2022. It is striking that already in June 2022, 35% of Ukrainians in the Netherlands, 18,000 people, had found paid employment (Swagerman, 2022). In other words, over a third of all Ukrainian adults who fled to the Netherlands at the start of the invasion were already in employment within 4 months. According to a UNHCR survey towards the end of 2022, the Dutch employment figure is considerably higher than in other countries, where on average, 28% are (self-)employed (UNHCR, 2022, p. 4). An OECD report on January 6, 2023 concluded that “labor market inclusion of Ukrainian refugees has been faster compared to other refugee groups. In a few European OECD countries, the share of working-age Ukrainian refugees in employment is already over 40%, including the Netherlands, Lithuania, Estonia, and the Czech Republic, if short-term jobs and informal employment are also considered” (OECD, 2023, p. 1).

Data from Statistics Netherlands from November 1, 2022, shows that 45% of Ukrainian refugees of working age who came to the Netherlands after the Russian invasion were working within half a year after their arrival (see Fig. 1). For men aged 25–45, the percentage is even higher, namely 55%. As is the case in other European countries, many have part-time jobs and/or jobs that do not suit their educational level, while a relatively high number are working in two part-time jobs. So, their labor market position resembles that of workers from Poland, Bulgaria, and Romania and is likewise at risk of exploitation (Doomernik et al., 2020; Rusinovic et al., 2023).

Fig. 1
figure 1

Employment rate Ukrainian refugees (15–64 yrs) in the Netherlands, 1 November 2022. Source: CBS website, retrieved on 26 May 2023: https://www.cbs.nl/en-gb/news/2023/05/higher-share-of-ukrainian-refugees-in-employment

Full size image

Labor Market Participation of CEAS Refugees

A comparison with refugees who arrived earlier and under the rules set by the CEAS now becomes extremely interesting. Dutch statistics show for a cohort of refugees who arrived in 2014 (more than half of which hail from Syria – see Fig. 2) that 42% were employed in 2021, seven and a half years after having received refugee status (see Fig. 3).Footnote 2

Fig. 2
figure 2

Nationalities of asylum seekers taken into COA reception 2014–2021. Source: CBS

Full size image
Fig. 3
figure 3

Percentage of employed refugees, 1, 3, and 6 years after their residence permit (2014 cohort). light blue = working; dark blue = not working; green = left/passed away. Source: CBS Cohort study Asylum 2014; https://dashboards.cbs.nl/v4/asielenintegratie/

Full size image

Fifty-seven percent of those who were employed worked part-time and are likely in need of additional governmental support. There are differences between figures for nationality/country of origin (see Fig. 3) but only after 6 years is some refugees’ employment rate on a par with the Ukrainians. The fact that, during their first years in the Netherlands, few refugees are in employment is most likely attributable to the mandatory integration regime they are subjected to. After 3 years (when refugees ought to have passed the mandatory Dutch citizenship exam), the figures markedly increase. The question that arises out of this comparison is whether the policies around the asylum procedures and the mandatory integration process delay integration in the labor market. In the next section, we investigate what these processes look like. The method used is background and policy analysis.

Background and Policy Analysis: Ukrainian Refugees and the Temporary Protection Directive – An Interesting New Policy for Refugees

The Introduction of the Temporary Protection Directive

Shortly after the Russian invasion of Ukraine, the European Council activated its Temporary Protection Directive (TPD). Ukrainians already had the right to travel to the EU with a 90-day visa before the start of the war, and the number of labor migrants in Poland, Greece, and Italy was considerable (Düvell & Lapshyna, 2022). The reason for the EU’s fast reaction was that the number of refugees who entered Ukraine’s neighbors reached one million within a month, and this was considered a clear case of the “mass influx” that the provision requires (Duvell, 2023; Carrera & Ineli-Ciger, 2023). The direct political interest of the EU was of course related to the reason that Putin presented for the invasion, namely Ukraine’s request to join NATO. The war was considered a direct threat to the EU.

Although the Directive left some room for national differences, all member states unconditionally opened their borders and labor markets to all Ukrainian citizens (İneli Ciger, 2022; Carrera & Ineli-Ciger, 2023). Given the sheer numbers, it was seen as unrealistic that the bureaucracy would be able to cope with individual asylum requests, so asking for asylum was not necessary and de facto discouraged (Parusel & Varfolomieieva, 2022). This means that the cornerstone of the European Union’s (EU) Common European Asylum System (CEAS) – i.e. the Dublin Regulation – was not applied. This Regulation dictates that, in principle, asylum seekers are and remain the responsibility of the first EU member state they arrive in. To enforce this rule, arriving asylum seekers are fingerprinted. In contrast, Ukrainians can move anywhere in the Union where they happen to know people, expect to find work or feel welcome and cared for (Ovacik, 2023). Interestingly, unlike in 2015/2016, the situation with Ukrainians was not referred to as a refugee “crisis”, and was presented as something the EU could manage.

Number of Ukrainians in EU Countries

According to a EUROSTAT update of April 21, 2023, the largest number of Ukrainians are in Germany (1,034,630), followed by Poland (993,755), the Czech Republic (447,830), Spain (168,970), Bulgaria (151,590), Italy (148,060), and Romania (117,395). The Netherlands is eighth on the list, with 112,700 Ukrainians (EUROSTAT, 2023). However, these numbers are very unstable, and there seems to be a lot of movement within the EU. For instance, in September 2022, the Dutch newspaper NRC reconstructed the trip of seventeen Ukrainian refugees in the EU. On September 23, it printed an interview with a 36-year-old IT specialist who fled with his wife and three children to a suburb of Copenhagen, Denmark. He was receiving governmental support (welfare and a room in shared accommodation) and was hunting for a job, ideally in his profession. However, he discovered that employers expect Danish language skills and many years of experience. The family was not happy with their housing situation and wanted to have more privacy but could not move as they lacked the money to do so. Consequently, “[t]he family wants to relocate to a cheaper country, Romania or Poland perhaps” (Meischke, 2022).

Movement within the EU, but also back and forth to Ukraine, causes instability in the figures about the numbers of Ukrainians in different European countries (Düvell & Lapshyna, 2022). Many make short trips to visit those still in Ukraine, for instance the elderly, and husbands fighting in the army. We also noticed that the figures in Europe change rapidly: for instance, Poland and Germany exchanged first and second place between March and April 2023. An analysis by the Clingendael Institute (Sie-Dhian-Ho et al., 2023) demonstrated through well-argued scenarios that refugees who initially went to an easy-to-reach country or to stay with someone in their network would likely end up in the country that was their favorite before the war (Lapshyna & Franck, 2015; Duvell and Lapshyna, 2018).  Both Sie-Dhian-Ho et al., (2023) and Vitalii et al., (2021) forecasted correctly that Germany would become more popular than Poland.

The Clingendael analysis also showed that the number of refugees in Europe might increase further (Sie-Dhian-Ho et al., 2023). If Kyiv cannot reclaim the Donbas region, some of the current 7 million internally displaced Ukrainians may give up on returning there. It is also probable that the husbands who are now separated from their wives and children would join their families when the state of emergency that forbids them from doing so is lifted.

Rush (2022, p. 8) provides data from the UNHCR survey “Lives on hold” from July 2022 among Ukrainians about migratory intentions. 26% said that the main reason for moving onwards to another country were family ties, 27% asserted safety, and 23% said that the main reason was to get access to employment (UNHCR, 2022).

Ukrainians in the Netherlands

The number of Ukrainians in the Netherlands is relatively high, even though it is not a priori an obvious destination as it does not border Ukraine or lie nearby like Poland, Hungary, and Romania, it is not known as immigrant-friendly like Germany, and it does not offer easy access to housing. Rusinovic et al.’s (2023, pp.14–15) survey of the Ukrainian population of the city of The Hague found that most of them moved there because they knew people: family and friends but also vague acquaintances. A few respondents ended up in the city by accident. According to Statistics Netherlands, on January 1, 2022, before the start of the war, the Netherlands was home to 21,000 Ukrainians, with two-thirds first-generation migrants, and the rest Dutch-born children. By the end of June 2022, 71,000 Ukrainian TPD beneficiaries had registered their stay in a municipality, and 1 year on (March 2023) their number had increased to 91,000 (of 112,700 Ukrainians in the population register). Just like in Spain and Italy, the reason for these high numbers is probably that many of those arriving had networks of family or acquaintances already, and maybe specific knowledge of the opportunities in the Netherlands.

Two-thirds of the arriving refugees in the Netherlands were women. In other European countries, the proportion tended to be just as high if not higher: from 66% in Latvia to 85% in Poland and Italy (OECD, 2023, p.3).Footnote 3 In the Netherlands, it became the task of municipalities to arrange accommodation, and additionally, NGOs and private parties supported the reception. In March 2023, 71,000 Ukrainians were housed in municipal reception facilities. If required, municipalities paid for their living costs. Municipalities in turn are compensated for their support by the national government.

Based on a multitude of sources, the OECD investigated the educational qualifications and skill profiles of Ukrainians. The percentage of highly educated people in Ukraine exceeds the figures in most European countries. One might therefore assume there is a high potential for their integration into the labor market. However, a high education level does not mean that they can work in higher qualified jobs, as many do not speak English (or Dutch). Brücker (2022) finds that only 38% of Ukrainians (surveyed in Germany) speak English. Ukrainians in Germany are also relatively highly educated, women even more so than men (Brücker, 2022). Koudijs & Eliza’s research for the Dutch refugee health organization PHAROS shows that language is a big problem for Ukrainians in the Netherlands, and large groups depend on one English-educated acquaintance (Koudijs & Elize, 2022). In cities like Amsterdam, it is possible to cope in English, as English has become rather common, but it is, of course, unlikely that newcomers (already) read and write Dutch, so the types of jobs available to them are limited.

Problems Encountered by Ukrainians in the Netherlands

Although Ukrainians’ integration into the labor market has seemed to take place at high speed, stories from representatives of an NGO that supports Ukrainians, OPORA, provide a less positive picture. In a meeting of the authors with representatives of OPORA, one representative stated that many Ukrainians were not working, and they were initially just waiting to go back, remaining isolated from Dutch society, without learning the language, and only speaking to each other. Shaidrova et al. (2022) wrote a report about Ukrainian children in Dutch schools and saw a worrying trend of parents who did not know anything about the Dutch educational system and children being emotionally unstable and unable to concentrate due to trauma. Needless to say, NGOs first support the problematic cases, while the successful cases might be less salient for them.

Tessorio (2023) shows after interviews with Ukrainians in a remote village in the Netherlands that the TPD arrangement has considerable disadvantages. The biggest problem is that the status is temporary and people remain insecure about their future, while a quick fix to the Ukraine war cannot be expected (Houtkamp & Drost, 2022; Rush, 2022). Another problem is that the TPD is implemented in different ways throughout Europe, and arrangements even differ between municipalities in the Netherlands. Shaidrova et al. (2023a, 2023b) show that the housing situation of Ukrainians in the Netherlands is difficult. Women with children, who are not working, are often staying temporarily in hotels, shelters or private homes, while those who are working are living in very expensive apartments. The second mentioned problem is the lack of structural language courses (Tessorio, 2023). Additionally, the rights to health care are limited to a basic level (Shaidrova et al., 2023b). And last but not least, we should also be aware of the precarity in the employment situation, as many Ukrainians have temporary and poorly paid jobs, and/or may be working below the level they are qualified for (Tessorio, 2023).

Analysis of the Asylum Procedure as Part of the Traditional (CEAS) Refugee Policy

First Arrival of Asylum Seekers in the Netherlands

In this section, we provide information on the Dutch asylum procedure and what seems so problematic about it (for more detailed information, see ECRE 2023). In almost all instances, refugees need to go through an asylum adjudication process that can be as brief as 8 days but in practice can easily take much longer (see below), including waiting periods ahead of the actual procedure. By far, the largest number of refugees apply in the Netherlands after arriving via Germany and Belgium by train or car, or by plane from another country with a tourist visa. They have not registered or applied for asylum in another EU country, even though they could have done so. In the cases where they arrived in Greece and Italy, this happened because they managed to avoid having their fingerprints taken. If they have already registered in another EU country, the Dublin rule applies, and they should be sent back to this country. Asking for asylum outside the EU is rarely an option, being reserved for a limited number of extremely vulnerable people in camps, and in exceptional cases for specific reasons, for instance, for Afghans who worked for the Dutch Army or Embassy at the moment that the Taliban took over. This can result in resettlement.

The adjudication of asylum requests can have several legal outcomes. First, the authorities can decide that the applicant fulfils the requirements as laid out in the Refugee Convention (fearing persecution on the grounds of race, religion, nationality, political opinion, or membership in a particular social group) and the EU’s Qualification Directive (which is more inclusive as it explicitly names gender-based persecution). From the applicant’s viewpoint, this is the best possible outcome as it offers access to the labor market and a measure of equal treatment in other respects such as the right to family reunification (even though discriminatory practices do occur in practice). Normally speaking, refugee status is indefinite, or at some point in the future – 5 years appears common, according to the AIDA database (ECRE, 2016) – can result in a permanent permit or naturalization.

In the Netherlands, legal conditionalities are imposed in the form of integration requirements that must be met before this can occur. Accepted refugees must pass a mandatory exam with cultural and language components. Although the measure was ruled illegal in court, the Dutch government tried to impose conditionalities regarding the right to family reunification in October 2022 by only allowing family reunification once suitable independent housing became available. The political salience here is that the Dutch state is responsible for the accommodation of asylum seekers as well as refugees, with the latter being entitled to social housing (which is in very short supply, leading to long waiting times).

The asylum request may result in refugee status, but also in the less favorable subsidiary protection status. This usually pertains to people who do not individually run the risk of persecution but cannot be returned to a situation in which their lives and livelihoods are not secure, e.g., in case of a (civil) war. Here, governments may impose restrictions on the application of certain rights, like the German government has done since 2018 by imposing a quota on the number of applications (1000) that are processed per month (ECRE, 2022b). Presently, Dutch policies do not differentiate between these protection statuses, but the outgoing government contemplated introducing them to curtail family reunification (Volkskrant, May 16, 2023).

We can distinguish between two major problems that characterize the asylum procedure in the Netherlands, and that could be the reason that Ukrainian refugees find employment so much faster. As we will show below, the first problem is time: going through the asylum procedures and meeting the integration requirements take at least 4 years. These years are unproductive, both from the refugees’ perspective and from the perspective of the state. The second problem relates to space: while Ukrainians have the right to move to where they have networks and expect job opportunities, CEAS refugees are stuck in the EU country where they first apply for asylum for a long time.

The Temporal Aspect of Mobility of Refugees

Invariably, the asylum adjudication process is time-consuming, both for migrants and for the authorities. For the authorities, this has obvious reasons: legal proceedings require careful consideration and may be prolonged by objections and appeals. Longer procedures are more expensive, but even when the proceedings are organized as efficiently as possible, capacity issues can easily slow matters down. In the Netherlands, for instance, the first registration applications take 8 days to process (a period that can be extended should the case require extensive fact-finding). After the large influx of 2015 and 2016 was dealt with, the Dutch authorities decided to down-scale the provisions, since it was assumed inefficient to keep facilities open with a lower influx. Yet, understaffing is now a problem: in 2023, there was a waiting time of up to 2 years before the application could be submitted to and processed by the IND, which is the Immigration and Naturalization Service (VluchtelingenWerk Nederland, 2024).

The temporal dimension is not just a problem for the authorities, which must pay for migrants’ shelter and sustenance, but moreover from the perspective of the migrants themselves, who feel they are wasting time and get frustrated (Van Heelsum, 2017). Refugees cannot start a ‘normal’ life and cannot start to work without suitable papers and are waiting for years.

For many asylum seekers who arrive in the Netherlands, it is not the first time that they spend time waiting that could have been used more productively; many have already been stuck on the way, for instance in Iran, Turkey, Libya, or in the hotspots on the Greek islands or in Italy (Linekar & Achilli, 2022). The problem of lost time multiplies again if an asylum seeker has moved after applying for protection in one member state to another member state (secondary movement). The so-called Dublin claim by the second state on the first one takes long to process, thus adding waiting time to the overall procedure. When asylum seekers fear such a claim, some disappear into illegality so that they are not returned to a country where they do not want to stay (Kuschminder & Dubow, 2022). If they remain invisible to the authorities for 18 months, their Dublin claim expires, which is the only way they can stay where they want to be. This is time not spent in the adjudication process and living in illegal shelters, and so is far from living a ‘normal’ life.

Asylum seekers often arrive alone, leaving dependent family members in their country of origin or transit. So, these family members suffer from the delays too, as their right to reunification depends on their refugee status. The rights of children to family life are not respected. ECRE (AIDA) reports indicate that those staying behind and hoping for family reunification face long waiting times, e.g., at Dutch and German consular offices. Although the right to family reunification is part of a refugee’s entitlement, several countries limit the option to request it to the first months, while beneficiaries of subsidiary protection can be made to wait for years. Austria does not even allow reunification before 3 years (ECRE, 2016, p. 18).

In some instances, the long wait causes additional problems when deadlines have been introduced as hurdles in the process. In the Netherlands, for instance, an integration examination must be passed within 3 years of being granted refugee status. As the website of the organization which administers all educational matters, Dienst Uitvoering Onderwijs, states: “If you do not integrate in time, you will be fined” (Dienst Uitvoering Onderwijs, 2023). This may mean you cannot get unconditional and permanent residence status.

The negative consequences of the waiting time have strong effects on the ability and willingness of refugees to ‘integrate’. While initially highly motivated to work and earn, the idle waiting not only leads to lethargy and institutionalization effects but also stress-related psychiatric disorders (Hvidtfeldt et al., 2019; Hvidtfeldt et al., 2020) and a dislike of the country and its rules. Insecurity about the well-being of family members is likely to (strongly) distract people. This also undermines their ability to integrate, as may the feelings that arise from conditionalities imposed by formal integration requirements. Intentionally or not, waiting times become a perverse tool of refugee migration governance. Though not always stated officially, the policy aim is often to deter any new migrants in the countries of origin or transit countries. Another side effect is that refugees are depicted as a burden to society who have little to offer (Bueno Lacy & Van Houten, 2023). For the Dutch authorities, there is a financial incentive to send people back to Italy and Greece, since the CEAS assumes that the country responsible for the asylum process is also the country where accommodation and integration support is arranged and paid for.

After asylum seekers finally have their refugee claims accepted, they usually start to strive towards acquiring citizenship, because this makes their position finally secure. In the Netherlands, the number of years before the entitlement to naturalization arises is 5 years. Other EU member states have different rules, and differences may also arise according to the type of protection status, i.e., those who hold a subsidiary protection status can be made to wait longer (in the Austrian case 9 years more) (ECRE, 2016, p. 20; Sredanovic, 2022). In the unlikely case that accepted refugees do not want to naturalize, the alternative is to apply for long-term resident status (under the EU’s long-term resident directive) which does include rights that come close to those of EU citizens, including the freedom to take up employment in any EU member state. However, there are conditions to be met such as having resided in one member state for a minimum of 5 years, being financially independent, and having fulfilled relevant integration requirements. In any event, it will take a minimum of 5 years before accepted refugees have the choice to go where they can make optimal use of their work experience and networks.

Spatial Mobility of Refugees: Opportunities and Barriers

The second problem of the asylum system in Europe is the spatial limitation: it limits free movement and therefore limits flexible access to support networks or labor opportunities (Ovacik, 2023). Since the passing of the Dublin Convention (the predecessor of the Regulation) in 1997, the intention has been that asylum seekers remain in the country where they arrive on the border of the EU, so often in Italy, Greece, and Spain, and since 2019 increasingly also in Poland (Jeandesboz et al., 2020). The extremely problematic conditions on the Greek islands were a consequence of the “hotspot” approach that followed the 2015 refugee “crisis”, in which reception centers close to Greek and Italian borders were introduced, that immediately became overcrowded:

It is due to this practice of indiscriminate and en masse imposition of the geographical limitation measures to newly arrived persons on the islands that a significant deterioration of the living conditions on the islands has occurred. Newly arrived persons are obliged to reside for prolonged periods in substandard facilities, where food and water supply is reported insufficient, sanitation is poor and security highly problematic, while their mental health is aggravated (ECRE, 2022a, p. 154).

Economic and social opportunities in the EU are highly diverse and spatially sorted. Employment opportunities and salaries are better in north-western Europe than in Greece, Italy, and Poland. It is therefore not surprising that a lot of potential asylum seekers try to avoid submitting their asylum request in Greece, Italy, or Poland, and attempt to move on irregularly to the country where they have connections and/or where they see better chances of finding a job.

Even after having gone through the asylum procedure in Greece and Italy, it is easy to understand that the now-accepted refugees would try to move to a country where jobs are available. Since 2021, several EU member states (for instance Germany and the Netherlands) have been confronted with the arrival of recognized refugees who left Greece legally as tourists hoping for opportunities to find employment or support elsewhere. Greece’s labor market offers very few prospects, and the Greek state does not offer access to welfare. The states where these refugees hope to resettle are not keen to assist them in their ambitions and in June 2021, their governments wrote a letter to the European Commission to bring this unforeseen secondary movement to the Commission’s attention. This letter was written on behalf of the German, Dutch, French, Belgian, Luxembourgian, and Swiss governments. ECRE (2022b: 15) mentions that 39,000 accepted refugees from Greece applied for asylum in Germany up to the end of 2021. German courts have ruled that these refugees cannot be expected to return since their fundamental rights would not be secured. Currently, discussions appear to be underway with the Greek government to create conditions under which these rights would be met. It is difficult to envisage how this could mean more than basic shelter and sustenance.

Not only the demand for labor and accessibility of (affordable) housing but also the existing refugee communities and the language, vary across the EU. While French, Spanish, and English may be manageable, languages like Hungarian, Slovak, and Bulgarian are difficult. One might be tempted to add Dutch or the Scandinavian languages to the latter group, but the native speakers of these countries pride themselves on being fluent in English. In any event: language matters and makes integration into society relatively easy or difficult.

Choosing where to live is not possible for refugees in the Netherlands. The high rents make it nearly impossible for refugees to find affordable houses themselves. Therefore, recognized refugees are distributed to all municipalities throughout the country, based on a percentage of the number of inhabitants, and these municipalities are obliged to find them affordable housing. Local labor offices are asked to absorb refugees into their labor market, and the refugees’ children are admitted to local schools. The liberalization of the Dutch housing market in the past two decades has caused a decline in social housing and refugees may sometimes pass locals on the waiting list for social housing, which may cause unhappy reactions among the latter.

Although houses might be easier to find in rural areas with shrinking populations, towns and cities offer better opportunities for employment in public administration, services, building, and technology whereas employment in rural areas is only available in the (low-paid) agricultural sector. This is a dilemma for policymakers: it would be better if the demand for labor is taken into consideration when choosing the place to live since nobody wants the refugees to become unemployed. Additionally, it would be better if the work experience and networks of the refugees were matched to the prospect of integration into the labor market, but besides a few experiments, there is little evidence of this happening. This becomes especially problematic when refugees are prevented from being mobile beyond their place of residence by government policy or practical impediments. Practical impediments to mobility are clear when refugees end up in rural areas with little if any public transport. This limits their employment opportunities even further compared to those who end up living in towns and cities. Access to education and chances for social integration likewise can significantly differ between settings (Doomernik & Bruquetas-Callejo, 2016). In other words: opportunities for successful integration are strongly correlated to where a refugee ends up living.

Given the differences between national settings, it should come as no surprise that refugees seek to move from one member state to another. However, this is only possible after a minimum of 7 years after arrival, assuming 2 years in the procedure, and 5 years to take on the citizenship of a member state. It is interesting to find out to what extent naturalized former asylum seekers use their freedom to move or why they would do so, but there is relatively little systematic research available. The reason is that population statistics only register that people leave, but it is often not known where they move to. Nevertheless, a few case studies are available. For instance, Van Liempt (2011) showed how between 10,000 and 20,000 Somalis moved from the Netherlands to the UK between 2000 and 2010. They moved after between 8 and 15 years (so an average of 11 years) in the Netherlands. The interviews showed that they were attracted by the larger Somali communities in certain neighborhoods of the big cities in the UK, which include Somali shops and businesses, and because of that, they felt that it was easier to live according to Somali culture. Perceived undue government intervention in cultural and religious affairs in the Netherlands was mentioned, while in the UK they would feel freer. Van Liempt concluded that initial plans to move to the UK were interrupted by the EU’s rules, and the relocation was the finalization of that initial plan (Van Liempt, 2011, p. 254).

Five years later, Ahrens et al., (2016) reported on Dutch Somalis, Swedish Iranians, and German Nigerians who had chosen to resettle in the UK. The authors distinguished several types of movers, namely career movers, student movers, family movers, political movers, diaspora movers, and cosmopolitan movers. Career movers are for instance highly educated Swedish Iranians. Even though they often have jobs in Sweden and have naturalized, they feel they are not taken seriously and are discriminated against. It is striking that the experience of discrimination is an important reason for all movers. The high intra-European mobility patterns that one can observe for young people in general, as visible among student movers and cosmopolitan movers, are also visible among the groups in this study. Brexit has changed this situation, and a certain number of CEAS refugees who moved to the UK have decided to move on or back to the EU country where they were initially registered (Guma & Daffyd Jones, 2019).

More recently, Leerkes & De Hoon (2019) looked more systematically at the Dutch population register to establish how many migrants who had been granted refugee protection during the second half of the 1990s were still residing in the country in 2015. It transpired that 4.3% had gone back to their country of origin, and 12.4% were known to have migrated onward, mostly to neighboring countries. 27.3% had been removed from the population register because their presence was unknown. It is likely that they too moved to another country. It would be extremely interesting to find out whether the relocated migrants were indeed more successful in the labor market. But for now, we conclude that it is likely that about a third had identified better opportunities outside of the Netherlands.

After putting forward the limitations of the refugee reception system in the Netherlands (and elsewhere in Europe), we can conclude that the spatial and temporal limitations keep refugees from optimally using their valuable experience and networks. This has a particular impact on (unnecessarily) slow integration in the labor market and not building up relationships to integrate further into Dutch society. It also causes frustrations about not feeling accepted.

Conclusion

The central question of this paper was: which refugee regime (TPD or CEAS) leads to better integration outcomes for the labor market and how does the procedure help (or not)?

As we have seen, multiple serious problems slow down the (labor market) integration of asylum seekers under the CEAS, while Ukrainians under the Temporary Protection Directive can (1) move flexibility through the space of EU countries, and (2) can apply for jobs directly after arrival. Figure 4 provides a summary of what we have described above.

Fig. 4
figure 4

Flow diagram of the two reception procedures in the Netherlands

Full size image

Related to the spatial dimension, the Ukrainian influx shows that people who flee tend to move to a country if they have informed reasons to do so, where they have networks and (through these networks) access to support and economic opportunities. As Parusel & Varfolomieievasay (2022, p.12) write, “[s]urprisingly perhaps, even if the reception of people fleeing from Ukraine has so far been highly unequal across the EU, there has as yet been relatively little discussion about unbalanced burden-sharing or a lack of solidarity. Countries that received many Ukrainians generally took pride in it rather than complaining. To some extent, this could be explained by the fact that they could count on financial solidarity as the EU made considerable resources available for reception, registration, and integration support.” Parusel & Varfolomieievasay (2022) conclude that the reception of Ukrainians is an example of good cooperation and good resource sharing, as well as an example of how the CEAS could be made more effective.

We conclude that the EU mechanism applied to CEAS refugees, geared towards permanent integration into the first member state of entry, has counterproductive outcomes. Secondly, the temporal dimension is problematic. In the Netherlands, refugees are forced to depend on government accommodation and income for years and involuntarily lose valuable time before they can enter the labor market. In terms of Bourdieu (1986), one could say that human capital – in terms of their initial educational achievements, work experience, networks and energy – that could be used to the advantage of Dutch society is wasted, while it takes years before CEAS refugees can build up work experience and social networks again.

Discussion

In this final section, we return to the relevance of this analysis. What do our results mean for Ukrainian and CEAS refugees themselves, and for integration policy?

As we have shown, Ukrainians actively choose the Netherlands, which gives them a positive starting point. Fifty per cent integrate rapidly into the Dutch labor market, and this is a very good result. However, the temporariness of the TPD status makes it difficult to proceed to a more stable condition. Secondly, they are often working on a level that is lower than they are qualified for and under precarious conditions. Shortages in the lower segment of the Dutch labor market are filled, but there is also potential in the higher segment. Thirdly, Dutch language courses are only provided in some municipalities, so in this regard, the Ukrainians depend on luck. To summarize, although their integration potential is very good, more stability and language courses would provide better long-term integration. This is certainly true for the labor market, but the effect would broaden to other fields, like being able to pay for better housing.

For CEAS refugees, one can observe that frustrations build up during the asylum procedure (Motlaq, 2024; Van Heelsum, 2017; Van den Braak et al, 2023). As Motlaq (2024) shows, waiting idly, not being allowed to work, not being able to send money to families back home in Syria, Iraq, and Lebanon, and being forced to take an integration exam, leaves these asylum seekers with negative feelings of not belonging. Supporting the not-working asylum seekers is also expensive for the Netherlands, and therefore the approach can also enhance negative feelings and racism among the locals, particularly because they see the refugees jumping on the waiting list for social housing. Integration through social contact between refugees and Dutch people only occurs towards the end of the process, in the municipality where they are placed, and when they start to work, or maybe at school for those with young children.

For policymaking on integration, we conclude that: (1) though there is good and fast integration into the labor market, the rest of the integration policies for Ukrainians are not sufficiently developed (language learning needs attention) and lack long-term perspective; (2) for CEAS refugees, faster participation in the labor market would prevent a lot of problems with integration later on. Learning the Dutch language and culture and creating social networks would then occur naturally. The financial costs for the Dutch state could also considerably diminish, as there would be fewer people unemployed and waiting for status who need support.

It is important to mention that this is not just a Dutch problem but also visible in the rest of Northern Europe, where integration regimes are based on welfare traditions, as well as in Southern Europe, where provisions are insufficient or missing altogether. Moreover, expenses are either overly high (under a welfare regime) or means are insufficiently available for refugees in need of income, housing, and other support. Refugees only regain their agency to move after naturalization which in many cases means years of wasted time. This offers a strong argument for earlier mobility opportunities.

This is an issue around European arrangements. Since solidarity between the EU’s member states cannot be taken for granted, some mechanism to achieve a sharing of responsibility is a political necessity. The Dublin regulation is not fit for this purpose, yet it is the modus vivendi Europe is presently stuck with. Instead of a common distribution/sharing system, one could envisage a regime in which responsible member states swiftly adjudicate asylum requests, ideally in close cooperation with the European Union Agency for Asylum (EUAA, a recent addition to the CEAS). This ought to ensure harmonized outcomes. The swift processing should then be followed by a labor market-driven distribution mechanism by granting refugees the very same freedom of movement presently reserved for EU nationals.

As a final remark, we note that refugee regimes not only in Europe, but worldwide, are costly and not very sustainable as they often lack durable solutions. We observe a diminishing tendency to welcome refugees and an increasing tendency to block refugees through closed borders. Camps in Africa and Asia are full of people who are stuck without access to legal status, with the UNHCR actively encouraging countries to allow refugees to integrate fast (Filippo Grandi at the presentation of the Global Compact on Refugees).Footnote 4 When studying the integration of refugees, it is important to use a comparative timeline and a longitudinal approach. And as Hein de Haas (2023) argues, connecting labor market policy and refugee policy would offer a way forward. Solutions like the Canadian points system for immigrants have never been considered in the Netherlands but could diminish some of the pressure on the reception system, while at the same time providing faster access to the labor market.