When the State Stepped into the Arena: The Swedish Welfare State,Refugees and Immigrants 1930s–50s

This article examines the encounter between welfare state and immigration policy that took place during the Second World War. It focuses on what happened to those who were not employed or who were incapable of working, and the extent to which the emerging Swedish welfare state took responsibility for them. The paper demonstrates that although the Swedish authorities were not overly generous, they ensured that all refugees had access to basic necessities. Refugees were granted the same economic and social assistance that Swedish citizens also had access to – in certain cases even to a greater extent than they did. When temporary social institutions for refugees were dissolved around 1950 and their work transferred to the state's ordinary agencies, refugees and foreign nationals largely came to enjoy the same rights and responsibilities as assistance-seeking Swedish citizens. In a similar fashion, they also – at least formally – came to be included in the same social citizenship that has been associated with the emerging welfare state.

The Second World War is a natural point of departure for historical research on refugees and immigration in Sweden. It was during this period that Sweden encountered the world and became a country of immigration rather than one of emigration, as it had been since the middle of the nineteenth century. This process, which can be understood in terms of Sweden moving from a restrictive to a more generous refugee policy, suggests that there was a specific ‘turning point’ in refugee policy during the war. In quantitative terms, the change is indisputable: in September 1939, there were approximately 5000 refugees in Sweden; when the war ended in Europe, in May 1945, the number had risen to approximately 200,000. ¹ But does this quantitative fact mean that there had also been a qualitative change in Sweden's attitude and policy towards refugees? And if so, was this change an effect of external world events or was it a result of a domestic political reorientation? During the 1930s, Sweden's refugee policy corresponded to the broader European trend: borders were difficult to cross; refugees were subjected to harsh scrutiny; and the state took minimal responsibility for their social and economic condition. But as one of just a handful of states to escape being drawn into the Second World War, I will argue that Sweden's refugee policy was distinct from that of other states. In Switzerland, another country which remained neutral, labour camps were used, refugee families were split up according to gender-based labour requirements, and the state had very limited responsibilities towards refugees. ² In contrast, the Swedish authorities chose to open the labour market to the refugees, making it possible for refugee families to remain intact, and also provided economic assistance to those who could not support themselves. This more progressive policy culminated in the 1954 Aliens Act which henceforth determined the Swedish state's responsibility for refugees and other migrants.

In order to understand how, compared to other countries, Sweden accepted state responsibility for refugees and other foreign citizens relatively early, it is essential to examine the parallel development of the welfare state. Around the turn of the twentieth century, Sweden was a poor, peripheral European country. Millions of Swedes had emigrated to the Americas in search of a better future. Declining population, poverty and the relatively rapid process of industrialization, brought with it social and economic upheaval and cast a long shadow over late-nineteenth and early-twentieth century Sweden. Ambitious political projects were launched in an attempt to tackle these problems, and with the social policy reforms of the 1930s the foundations were laid for the Swedish welfare state. During the Second World War, forward-looking reformist policies were set aside in the face of a more crisis-oriented political agenda, but the project of building the so-called ‘People's home’ (Folkhemmet) was not entirely abandoned. It remained present in the political rhetoric of the war years, and to a more limited extent even in political practice. Even more importantly, as historical research has shown, the war years laid the groundwork for the comprehensive and efficient bureaucracy on which the emerging postwar welfare state would be built. By the 1950s and 1960s that state would ensure all citizens had social security and could claim their basic social rights.

Research on the history of the welfare state provides the basic theoretical perspective of the present article. ³ This perspective prompts several important questions. What role did the welfare state project play in what has been termed ‘the turning point’ in Swedish refugee policy, and how did it affect the refugees’ position once they had arrived? The article focuses on the economic, social and legal aspects of the integration-segregation dynamic in so far as the treatment of people who were considered ethnic or national ‘others’ in the emerging welfare state were concerned. What opportunities did they have in terms of employment and of becoming self-supporting? Although the discussion that follows will necessarily be conducted on a general level, questions of ethnicity, nationality and gender will also be addressed.

This article outlines how Swedish refugee and immigration policy underwent significant change between the 1930s and 1950s, as it went from a restrictive to a liberal, welfare state-oriented one which would predominate until the end of the 1960s. I will highlight some essential factors that explain this transformation. The first of these is what I have elsewhere called the ‘Nordic prerogative’, ⁴ by which Sweden primarily saw itself as responsible for ethnic Nordbor (‘Nordics’, that is, people from the Scandinavian countries, Finland and Iceland). The notion of an ethnic obligation towards refugees was seen as more important than factors such as the ideological views of the refugees, their need for protection and other humanitarian concerns. As well as influencing policy towards Nordic refugees, which was highly accommodating, the ‘Nordic prerogative’ also inadvertently shaped policy towards non-Nordic refugees. Secondly, I will argue that the reception of tens of thousands of Baltic refugees in 1944–5 is a key factor to consider when trying to explain the extent of state responsibility for refugees in postwar Sweden. Since in contrast to the Nordic refugees the Baltic refugees were expected to stay, their integration and reception truly was a challenge – especially as they had no support from refugee committees or homeland organizations. The experience which the Swedish state apparatus gained as a consequence of this challenge during the war, then became the model of how state responsibility for refugees could and should be designed. Thirdly, as has already been indicated, during the war the state's capacities expanded, not least due to an increasingly professionalized corps of civil servants and a new effective and flexible bureaucracy. I will argue that the same patterns are visible in the emerging state apparatus surrounding the handling of refugees. This suggests that the state's bureaucratic and administrative development is of importance when accounting for the relative integration of refugees and other foreigners into the Swedish welfare state.

In what follows, I will first briefly describe the exterior framework of the Swedish reorientation on refugee issues, by focusing on the development of legal regulations and by describing in general outline refugee and immigration policy. I will then discuss the state's responsibility for refugees and immigrants from a general welfare state perspective, as well as identify certain ethnic hierarchies that this perspective highlights. I will focus the discussion on refugee policy during the Second World War and the political discourses of migration that evolved from the end of the war until the early 1950s, but I will also look both back and forward in time to give the issue some necessary contextualization. ⁵

When the Second World War began in September 1939, Sweden was one of several countries which declared themselves neutral. A coalition government was formed, comprising all parties in parliament except the communists. Sweden was one of few countries that was able to remain a non-belligerent to the end of the war. Consequently, the debate and research about the Second World War have had a different slant in Sweden than in many other countries. For quite some time the Swedish trajectory during the war was rarely questioned, and there was little research in this area. It was not until the end of the 1960s, when the research project Sverige under andra världskriget (SUAV, ‘Sweden during the Second World War’) was launched that the theme came to be more closely scrutinized by historians. SUAV focused mainly on military preparedness, security and foreign policy, domestic conditions and the formation of public opinion. ⁶ Only one study dealt with refugee issues. ⁷ To a great extent, the SUAV project confirmed what had to that point been the received view: the goal of the Swedish government had been to keep the country out of the war, and in that effort it had been successful. Critical views of Swedish policy or alternative perspectives were few and far between. By the end of the 1980s and early 1990s, a succession of studies had appeared which criticized considerable parts of the historiography on Sweden and the Second World War, while at the same time there was a growing interest in new research perspectives among scholars. ⁸ Steven Koblik's The Stones Cry Out: Sweden's Response to the Persecution of Jews 1933–1945 was the first work to examine the Holocaust in a Swedish context. ⁹ In the wake of this study, important new questions began to be asked about what Swedes knew about the Holocaust and if more could have been done to rescue Jews. ¹⁰ These questions also placed refugee policy and the treatment of refugees during the war onto the research agenda, though still in a rather fragmentary and peripheral manner.

More recently, another large scale research project about the Second World War has been completed. ‘Sweden's Relations to Nazi Germany, Nazism and the Holocaust’ addressed questions of a moral nature: how did Swedish society relate to Nazism; and how was the violent development of antisemitism perceived? The project resulted in several monographs, anthologies and a comprehensive synthesis by its project coordinator, Klas Åmark. ¹¹ However, the refugee angle was notably absent from this project as well, although discussed at length in Åmark's book. That large scale projects such as these have overlooked refugee research is all the more remarkable considering the fact that Sweden never came into closer contact with the war than through its encounter with war-time refugees. ¹²

The amount of research on Swedish migration, especially on Second World War refugees, has nevertheless increased considerably in recent years. ¹³ Current research can be divided into two areas of interest, each identified by its different empirical focus. The first area concentrates on how laws and their implementation affected the chances of refugees, and Jewish ones in particular, being granted asylum in Sweden. ¹⁴ In an earlier study, I referred to this as ‘the reception aspect’ (that is, who was admitted, to what extent, and who was excluded). The second area of research interest deals with what happened to refugees in Sweden, what I have called ‘the treatment aspect’ (what rights and obligations the refugees had, and what possibilities and limitations determined their situation). ¹⁵ Research in this latter area usually focuses on how individual groups of refugees were perceived and their general treatment. ¹⁶ Overall, the perspective in refugee research has commonly been that of ‘Sweden’, but some work has been conducted from the viewpoint of individual groups of refugees. ¹⁷ Closely related to refugee research, have been the studies into Swedish antisemitism during the period in question. During the 1930s, Sweden denied entry to many Jewish refugees at a time when Swedish refugee policy could have been saving lives. The scope for providing refuge diminished by the early 1940s, when the Jews already had fled to other countries or had been killed. ¹⁸

Considering its relation to the labour market and to the social development of modern society it is not surprising that research on postwar labour immigration has been more comprehensive than that on refugees. ¹⁹ Yet the focus of this reserach on labor immigration has been relatively narrow, dealing almost exclusively with the numbers and national origins of the immigrants coming to Sweden. It is only more recently that researchers have started exploring this area in more depth and examining the conditions in which the immigrants lived and the way they were received. ²⁰ This new research has also begun to discuss the question of whether the admission of refugees during the Second World War was related to labour immigration. Whereas historians used to draw a clear line between these two phenomena, the tendency today is to speak of ‘two parallel processes’ or an ‘overlapping process’. ²¹

It can be argued that this dual process began to emerge when it became apparent that foreign labourers could be controlled more easily than their Swedish counterparts, and as a consequence came to be seen as a valuable resource. When the war ended, for example, the possibility of exchanging labour was formalized among the Nordic countries. This agreement above all benefited Sweden, which was enjoying economic growth, while the other countries were still rebuilding their war-ravaged economies. The agreement rested on the principle that labourers were to be recruited via the labour market authorities in the respective countries. When the recruitment of non-Nordic labourers was introduced there was therefore already a functioning system that could be utilized: evidence of the ‘overlapping process’ in action. However, in the Swedish context of the 1940s it is problematic to argue that war refugees and labour migration should be seen as part of the same phenomenon. Labour migrants were ‘desirable’ while war refugees, as well as postwar refugees and individual immigrants, were not necessarily so – even if there was a labour shortage. This is a research topic that remains to be studied in more detail, and there is wide scope for doing so owing to the convergence between the extensive field of research on the welfare state, labour unions and labour on the one hand, and the decidedly more limited research on refugees and migration on the other.

An important remark to be made about the existing historiography is the ‘gender blindness’ that inflicted both the historical actors themselves as well as later historians. Although the majority of the war refugees were men, there were a significant number of women among them, and this has not been adequately represented in histories of the period. Research about postwar labour migrants has largely overlooked issues of gender (and class). The focus on men who came to Sweden in order to work within the metallurgical and manufacturing industry presents an over-simplified picture of the situation. Until the middle of the 1950s, approximately 55 per cent of immigrants were women, most of them single. Some came ‘accompanying’ their partners, but others should be seen as labour migrants in their own right, working in a strictly gender-segregated labour market. Women – mainly from Norway, Finland and Germany – were predominantly employed in low-wage gender-specific jobs, such as house maids or within the textile and clothing industry. By the mid-1950s, the proportion of female immigrants fell to approximately 40 per cent. With the exception of the mid-1960s, when the female share decreased even further, women until the 1990s made up between 40 and 50 per cent of immigrants to Sweden. ²² Nevertheless, they have been almost invisible as a group in the wider debate about immigration and within the research agenda, with the result that an incomplete picture of war refugees and postwar immigrants has emerged. ²³

Research on the welfare state which examines the concept of citizenship offers a useful point of departure, especially Thomas Marshall's thesis about the introduction of ‘social citizenship’ guaranteed by the state, and its importance for the historical development of social rights. ²⁴ In the following discussion, I shall address the important question of how Swedish authorities during the war were confronted with the problem of deciding whether individuals who had no legal or political rights – refugees – were to enjoy such social rights. ²⁵

From 1860 onwards, Sweden followed the liberal state trend of the ‘free interchange of people’, which came to be the guiding principle in the so-called ‘alien issue’. In practice, this meant that foreign citizens were free to enter and work in Sweden for a period of time, even if a very limited and highly fragmentary legislation allowed the local police authorities to keep ‘vagrants of foreign ancestry’ out. But these undesirable individuals were very vaguely defined. Strictly speaking, ‘gypsies’ constituted the only racially-identified category to be expelled for ‘loitering’ and ‘vagrancy’. ²⁶

The period 1914 to 1926 saw several attempts to regulate immigration through the formulation of laws that specifically referred to foreigners. The motives behind the introduction of these laws varied. Even though Sweden managed to remain neutral during the First World War, the situation was seen to demand tighter regulation of immigration. During the early 1920s, Sweden, as with many other countries, was beset by economic instability and high unemployment which gave rise to the notion that Swedish jobs had to be protected from foreign labour. The 1920s were also a time of political instability, with 10 governments in as many years, and it was difficult to mediate between the different interests involved – in particular the divided interests of the employers and right-wing parties on the one hand, and the growing influence and power of the labour movement and the Social Democratic Party on the other. On the whole, immigration policy therefore continued to be based upon provisional proclamations and a weak and fragmentary legislation. But the principle of free migration was gradually buried under restrictive proclamations. From 1917 onwards, the great majority of migrants were subjected to passport controls as well as visa restrictions; and in December 1918, a new bill strengthened the government's power to expel foreigners on even the flimsiest of pretexts when the interests of the state were perceived to be under threat. In practice, the principle of free interchange of people was abandoned, and in addition foreigners staying permanently in Sweden were subjected to greater controls. ²⁷

The main consequence of the 1927 Aliens Act – the first of its kind in Sweden – was that the administrative practice that had already developed became formalized in law. The law emphasized labour market issues – Swedish workers were to be protected from the competition of foreign workers. For other groups the impact of the new legislation was limited. ²⁸ In earlier legal texts and proclamations, notions about ‘race’ had not been referred to in explicity, exept in the case of ‘gypsies’. The same can be said about public debates concerning foreigners. By the time the bill was introduced in 1927 the racial issue had become more prominent. For instance, the bill mentioned how ‘the importance of our country's population being of a uniquely homogeneous, unmixed race can hardly be overstated’. ²⁹ With the 1927 Aliens Act, Sweden followed an international trend, as other European countries and the United States also introduced their first laws governing immigration during the first decades of the 1900s. Although the Act has been characterized as restrictive, passport controls and visa regulations were gradually eased up for Scandinavian citizens and west and central Europeans. The Act was supposed to be temporary, but was extended until it was revised in 1937.

The concept of ‘political refugee’ was introduced during this same period within the framework of the League of Nations, but Swedish legislation had very little to say on such matters. In fact, although Sweden loyally stood behind the League and acted in accordance with the overall objectives of the organization, it was less interested in getting involved in the League's specific commitments on refugee issues. ³⁰ However, the amended Aliens Act of 1937 aimed to enhance the protection of political refugees. ³¹ Opinion as to the wisdom of doing this, however, was polarized: most Social Democrats, Liberals and some Communists wanted more generous legislation; while the Right and the Farmer's Party, but also some Social Democrats, demanded stricter regulations. As a consequence, the new law was in many ways a compromise which tried to accommodate these two different viewpoints. ³² In retrospect, the 1937 Act has been the subject of scrutiny as it has been seen as a means of denying Jewish refugees the right to be recognized as political refugees.

Although the definition of a political refugee was vague, people who claimed to be political refugees could no longer be summarily turned away at the border but were entitled to have their case heard. Another central part of the 1937 Act was the preservation of the paragraph which gave the government practically unlimited power over immigrant and refugee issues, because, the Act explained, in case of war, threat of war, or other ‘special circumstances’ it was the prerogative of the government to issue whatever regulations were found necessary. ³³ The paragraph allowed for a considerable amount of arbitrary ruling and had a direct bearing on the foreigners who were placed in so-called detention camps during the war. ³⁴

The 1937 Act was supposed to last for five years but because of the war it was extended and kept with minor changes until 1945, when a new Aliens Act was passed, to which only minor revisions were made in subsequent years. Major changes to the Aliens Act were not made during this period owing to the uncertainty surrounding how to resolve future immigration issues as Europe faced a postwar displaced persons crisis and then the Cold War added a new dimension to European politics. The next Swedish Aliens Act – of 1954 – showed how international refugee policy had a bearing on the formulation of national legislation. ³⁵ This Act was of an entirely different character to the ones preceding it and was clearly influenced by and adapted to the 1951 UN Refugee Convention. It states that as far as possible the same rights and obligations should apply to foreigners as to Swedish citizens. The Act also spoke about the ‘rights’ of the refugees, stating that refugees could not be denied asylum unless there were special circumstances. The 1954 Act could be considered the first ‘modern’ Swedish Immigration Act in the sense that it closely adhered to international law, including Sweden's broader humanitarian responsibilities and the promotion of human rights. ³⁶ The 1954 Act introduced several changes that remained in force until it was rewritten in the 1980s.

At the international conferences that attempted to resolve the ‘European refugee problem’ in the 1930s, Sweden remained passive in the face of every attempt at reaching an international agreement. When Denmark pushed for the creation of a joint Nordic approach on refugee issues, as these countries had similar legislations and policies regarding immigration, Sweden was again unreceptive and a Nordic policy on refugees never materialized.

The Swedish experience of refugees – or other migrants – was limited before the Second World War and nationally there was a marked fear of an ‘invasion of foreigners,’ and of Jewish refugees in particular. ³⁷ The fear of Jewish refugees was displayed in several concrete ways: in a national census of foreigners in early 1939 in which respondents were asked, among other things, if they or either one of their parents were Jews; in student demonstrations against the decision to allow a dozen or so Jewish doctors to come to Sweden; and last but not least in a hostile debate in parliament in which the Farmers’ League MP Otto Wallén declared that he was an antisemite. ³⁸ However, it should be emphasized that there was also a considerable pro-refugee contingent which has seldom been given much attention in historical accounts. Wallén, for example, was immediately challenged in the parliamentary debate: it was simply not acceptable to proclaim oneself an antisemite. ³⁹ Voluntary refugee committees, organizations like the Red Cross, and private citizens all made considerable practical contributions, and there were plenty of commentators and politicians who professed their ardent support for refugees.

At the outbreak of the Second World War there were only some 5000 refugees in Sweden, a consequence of a restrictive immigration policy. By and large, this restrictive policy remained in effect until the German hold on Norway hardened in autumn 1942, after which almost 5000 Norwegian refugees escaped to Sweden in the space of six months. Henceforth there was a steady flow of Norwegians to Sweden. By the end of the war they were almost 30,000 of them. During autumn 1943, some 10,000 Danish citizens fled to Sweden, the majority of which were Danish Jews. In autumn 1944, some 25,000 individuals fled from the Baltic States, most of them from Estonia. These, too, were admitted into Sweden. On 6 May 1945, there were almost 200,000 foreign citizens in Sweden, most of them refugees from Norway, Denmark and the Baltic States, but also some 80,000 evacuated children and adults from Finland. ⁴⁰ In addition, the Swedish Red Cross conveyed some 20,000 ex-concentration camp prisoners from the Continent to Sweden (Folke Bernadotte's so-called ‘White buses’). The majority were Nordic citizens. After an international inquiry from United Nations Relief and Rehabilitation Administration (UNRRA) Sweden also admitted another 10,000 ex-concentration camp prisoners, mainly Polish Jews, most of whom were women. ⁴¹ If the number of refugees before and after the war is compared, the ‘turn’ in Swedish refugee policy during the period is unmistakable.

During the war, Swedish authorities had to decide whether refugees should be entitled to social and economic rights. The central problem was what means the refugees had for providing for themselves. During the 1920s, the attitude of the Swedish authorities had been clear: refugees or other foreign citizens were not allowed access to the Swedish labour market, which was to be reserved for Swedish workers; a view which was reinforced in the 1937 Aliens Act. ⁴² At the same time, access to state allowances and other forms of economic support were practically non-existent. Refugees were to be provided for by entities other than the Swedish state, for instance family, friends, or – and in particular – the hard-pressed voluntary refugee committees. During the 1930s, these committees had a major responsibility for the reception of refugees because the funds that they raised to support refugees financially while they remained in Sweden were a pre-condition for the refugees being admitted into the country. The committees mostly specialized in providing support for a specific category of refugees: Labour Movement Refugee Aid supported Social Democrats and trade-unionists; Red Help aided Communist refugees; and refugee aid of the Mosaic Congregation supported Jewish refugees. ⁴³ ‘The foreigner’, however, was not totally excluded from the Swedish labour market. The Socialstyrelsen (Board of Social Affairs), which was responsible for work permits, did not hesitate to issue these to foreign workers who were in considerable demand even in the years of unemployment in the 1920s. ⁴⁴ It should also be noted that in spite of the firm rhetoric of the 1930s, most of the Central European refugees (although few in number) were well integrated into the Swedish labour market during the boom of 1937 and 1938. ⁴⁵

The formal attitude toward the gainful employment of refugees also underwent changes during the war. At the time of an upward economic trend around 1938, the International Federation of Trade Unions and the international Matteotti Committee – which included Labour Movement Refugee Aid – recommended that refugees being assisted by the committees should as far as possible support themselves. Labour Movement Refugee Aid acted accordingly, and due to the close relationship between the trade unions and the Social Democratic Party which was in power at the time, this policy became influential. ⁴⁶ Against the background of a shortage of manpower during the war, it became common practice for refugees to be self-supporting. ⁴⁷ A special division of Statens arbetsmarknadskommission (SAK, The National Labour Market Commission) – which was responsible for the national labour market – was created to secure employment for refugees. Although minor attempts at such coordination had taken place earlier, the project was launched in earnest in summer 1941, when increasing numbers of Norwegian refugees found their way to Sweden. This enterprise subsequently grew considerably. SAK did not only organize and monitor actual employment offices, it also financed vocational training and courses for refugees, and even took on some social responsibilities, such as arranging travel to the refugees’ new employers, for instance, or providing emergency financial aid. ⁴⁸ This new governmental commitment and the new directives, which entailed that refugees had to support themselves by working, was rational in many respects: the number of refugees was constantly growing as was the demand for workforce, and not least importantly, self-sufficient refugees cost the Swedish state nothing.

The norm in Sweden at the time was that of a male provider, and the job market was gender segregated. This applied to refugees as well. Male refugees found employment mainly in forestry and farming, while women worked as maids or – in the latter part of the war – in the textile industry. The large group of Baltic refugees that arrived in autumn 1944, however, presented a challenge to these norms. The number of women among them was relatively high, which caused the Swedish state to deviate from the hitherto strictly held norm of the male provider. Women were sometimes given traditionally ‘male’ jobs and it was made clear that husband and wife were jointly responsible for the family's income. The reason for this seems to be that the support of the Baltic refugees was, as shall be seen, the responsibility of the Swedish state, making the authorities particularly keen on ensuring that the refugees could support themselves. ⁴⁹

Working conditions for foreign labourers were formally the same as for Swedes but with clear elements of compulsion. At certain points, it was more or less a requirement for male refugees to have worked a stipulated time in forestry in order to be eligible for other jobs. ⁵⁰ There is no doubt that the foreign workforce was a major resource in war-time Sweden. For example, 25 per cent of forestry workers were Norwegians in the cold winter of 1943. The importance of the foreign workforce, however, did not prevent the discussion concerning its role in the labour market from sometimes becoming heated.

For refugees who for different reasons were not able to provide for themselves, the refugee committees continued to play an important role as an economic and social guarantor. ⁵¹ In addition to the committees, a solution for dealing with refugees emerged during the war which was specific to the Swedish context. Many of the countries occupied by Nazi Germany had legations in neutral Sweden. These legations represented the interests of their respective governments-in-exile and its nationals, and usually contained a refugee office. The purpose of these offices was to offer support and assistance to co-nationals who had come to Sweden as refugees. From an early date, the Swedish authorities established a far-reaching collaboration with the refugee office of the Norwegian legation, which was consulted over labour market issues, and initiated educational programs and courses for refugees that the Swedish state financed to varying degrees. The refugee office also became the economic and social authority for the Norwegian refugees who were not able to provide for themselves or needed medical care. Similar collaborations, with arrangements concerning the division of labour, were established between the state and other legations and refugee offices. This was most clearly the case with the Danes, with whom collaboration developed along the lines of the Norwegian model in autumn 1943. Legations thus came to have considerable influence over ‘their’ refugees, while the Swedish state was relieved of its financial responsibilities and found in the refugee offices a partner for solving problems the state had not been prepared for. It was a solution that the Swedish state as well as the legations benefited from. ⁵²

This system, however, was found to be wanting when tens of thousands of Baltic refugees sought protection in Sweden during autumn 1944. While many Balts could be put to work like other refugees, the Baltic national associations lacked the economic and political resources of the legations. Nor could the mass of Baltic refugees be supported by the refugee committees which were already under considerable pressure. The problem, therefore, was what was to be done with those who could not support themselves.

The arrival of Baltic refugees forced the Swedish government, for the first time, to take full responsibility for a large number of refugees. ⁵³ It was a formidable task, but the Swedish government had acquired considerable knowledge and previous experience of this problem in collaboration with the Nordic legations. Drawing on this experience, and showing creative flexibility in the acceptance that everything rested on the shoulders of the state, the Swedish authorities developed a practice that fundamentally altered Swedish refugee policy. Henceforth there was no doubt that it was the Swedish state – and not refugee offices, committees or other participants – which was responsible for the refugees. The solutions, practices and policies employed to cope with the Balts came to form the model for Swedish postwar refugee policy, in which the state was the principal actor.

As late as the 1930s, such a position would have seemed inconceivable. But state responsibility had increased gradually since then. Already in 1939, parliament decided to earmark 500,000 SEK (approximately 11,800,000 SEK or $1,800,000 as its current value) for the subsistence and vocational training of refugees. The decision also led to the formation of Nämnden för statens flyktingshjälp (NSF, the Board of the Government's Refugee Relief) to administer the funds. The instructions for the NSF made it clear that its essential task was that of distributing state support to the voluntary committees for refugee aid. This was almost certainly the first state-financed aid agency in Europe, and it marked the first tentative steps in the direction of the Swedish state taking economic and social responsibility for refugees. Voluntary organizations, however, still had the main responsibility for refugee aid. In order to ensure that the state's responsibility did not grow so large that it undermined the voluntary basis of Swedish refugee aid, the NSF made sure to divide its favours cautiously and sparingly.

The NSF was a short-lived affair, and on 1 October 1941, it was replaced with Statens flyktingsnämnd (SFN, The Government Board of Refugee Relief) which had a wider mission. SFN was not only responsible for handling state subsidies to the committees, but was also, among other duties, to provide direct support to foreigners who did not receive assistance through the committees or the legations. To begin with, the role of the SFN was modest: most refugees in need of help were supported by committees or legations. For the comparatively few refugees who lacked other support the SFN came to function as a provider of poor relief, on more or less the same terms as for Swedish citizens in need. But the workload increased drastically with the numerous Baltic refugees made up of a cross-section of the entire population of the region: the young, old, sick, healthy, well-educated and poor were included among them. Confronted with this influx of Baltic refugees who for different reasons could not provide for themselves and who lacked an organization to support them, the Swedish state was suddenly forced to take the lead. The SFN arranged educational and employment schemes, often together with SAK. In addition, it provided financial and social support for the unemployed, sick and handicapped, and guaranteed medical care and medicine – activities that had previously been much more limited since most refugees were Nordic and thus provided for by their respective legation. Furthermore, some new functions which had not been appropriate previously were assigned to the SFN. Among these were aid for mothers and special loans for Baltic families who had found a place to live but lacked household furniture. Meanwhile the SFN continued to provide financial support to the refugee committees, but what had been a major field of activity when the NSF was founded in 1939 by the end of the war was a negligible part of an enterprise which had increased dramatically from autumn 1944.

After the Second World War most refugees returned to their home countries, but the Baltic refugees and the non-Nordic ex-concentration camp prisoners remained. ⁵⁴ Both of these groups were the responsibility of the SFN, and for a transitional period until 1950 it continued its work. When Sweden returned to conditions of peace, the provisional solutions for refugee administration also came up for revision. From 1946 until the 1954 Aliens Act a number of changes in legislation and administrative practices were undertaken with the purpose of ensuring that foreign citizens were encompassed by the laws, obligations and rights that pertained to Swedish citizens. The result was that foreign citizens, refugees and immigrants alike, on paper at least, became eligible to more or less the same rights as Swedes in terms of economic and social protection. Nor was there any doubt that the state was the main actor as far as these issues were concerned. The voluntary organizations, for example, were wound down.

As part of this process of transition, it became increasingly clear that refugee issues needed a ‘comprehensive elucidation’, and that policy should not to be dictated solely by, for instance, the needs of labour or national security concerns. Establishing the ‘refugee expert’ was central to this ‘comprehensive elucidation’. Previous research on the welfare state has shown how the development of a Swedish wartime administration was a prerequisite for the postwar expansion of the welfare state. During the war, the state's capacities swelled, the bureaucracy learned to react flexibly and a cadre of professionalized civil servants opened up new possibilities of policy making in the 1940s. ⁵⁵ The same patterns are visible in the emerging apparatus surrounding the handling of refugees, which also displayed a flexible character and an ability to produce a set of competent professionals – in this case ‘refugee experts’. This indicates that the bureaucratic structure itself played an important role for the relative integration of foreign citizens into the Swedish welfare state.

The process of professionalizing the way refugee issues were handled took place step by step as officials at SAK and the SFN, for instance, gradually came to feel that they had mastered not only issues of employment and refugee support, but that they had a more comprehensive knowledge of the refugees, which made them uniquely positioned to deal with the issues involved. The tendency towards professionalization was accentuated from 1946 onwards, when the provisional refugee authority organization was replaced with a more permanent administration. The establishment of ‘refugee experts’ and the emergence of the view that the refugee policy should take many aspects into consideration had a clear impact on the 1954 Aliens Act. ⁵⁶

Henceforth, Sweden admitted more or less everybody who sought asylum in the country and the Swedish state guaranteed their rights. But after the backwash of the war had receded, relatively few found their way to this northern outpost. Between 1950 and 1967 only 24,000 refugees arrived in Sweden, primarily from Eastern Europe, of which some 7000 were Hungarians who had escaped during the 1956 Revolution. ⁵⁷ Instead the most important reasons why foreign citizens emigrated to Sweden was to be reunited with relatives or to seek employment. ⁵⁸

It is possible to argue that labour immigration already began during the war. In winter 1944, there are examples of Nordic citizens applying from their home countries for work in Sweden. This practice continued after peace had been restored. It is also fair to say that the refugees had been perceived as a significant resource during the war, not the least because they could be controlled to a greater extent than Swedish workers could. In this way, the refugees paved the way for labour immigration, at least in the eyes of the authorities.

The economic recession that was expected to follow the Second World War never set in, and instead the postwar years saw sustained economic growth. In Sweden this led to a significant labour shortage, something that proved how important the refugees had been for the Swedish labour market during the war. This was particularly evident in important economic sectors such as forestry and agriculture. For this reason, a recruitment of Nordic labour had already begun in summer and autumn 1945 in order to replace the refugees that were returning home. Skilled labourers for the manufacturing industries were most in demand, but there was also a need for unskilled manpower. Recruitment was mainly collective, meaning that Swedish employment organizations made deals with their Nordic counterparts. A healthy economy and labour market also attracted many individual Nordics to Sweden. An agreement in 1954 formally established a common Nordic labour market, but in practice the Swedish labour market had already been open to Nordic citizens since the second half of the 1940s and they remained the largest group of immigrants up to the 1960s. ⁵⁹

Toward the end of the 1940s, workers were also recruited from Hungary, Italy and Austria. Overall, this recruitment was set up according to the system already being employed in relation to citizens from the Nordic countries: employment organizations struck a deal with a central counterpart in the immigrant's home country and this allowed a process of explicit selection, as it was the Swedish participants (authorities and employers) that decided who was to be employed. ⁶⁰ Initially there was a distinct element of compulsion, as the foreign worker pledged to stay within a certain sector, with a specific employer, or in a limited geographical area. As the need for foreign labour increased, the regulations were relaxed and increasing numbers of non-Nordic immigrants came to Sweden outside the framework of collective labour recruitment.

In the postwar period the influential Swedish trade union movement showed a close interest in the rights of foreign labourers. Historians have interpreted this in two ways. On the one hand there are those who claim that this was an expression of genuine solidarity; while others have suggested that the primary motivation was to protect the social conditions of (Swedish) workers by making sure that their working conditions were not undermined by the recruitment of ‘cheaper’ labour. ⁶¹ Regardless of which interpretation one chooses to follow, the practical result was that foreign workers were generally afforded the same economic and social rights as Swedish workers. Successive politicians likewise maintained that foreign labour should be treated the same way as Swedish workers, and that they should have the same conditions concerning health insurance and unemployment benefits, for instance. ⁶² While these rights looked good on paper, it should be emphasized that in practice immigrants faced considerable obstacles to getting ahead within Swedish society and that well-educated foreign labourers, for example, often had to do jobs they were overqualified for.

During the 1950s and 1960s the Nordic (above all Finnish) group continued to be the largest category of immigrants, but those from Yugoslavia, Italy and Greece were also well-represented. Non-European immigration, however, was practically non-existent before the early 1970s. ⁶³

Recent research has suggested that Swedish refugee and immigration policy has been influenced by ethnic considerations – what I have elsewhere termed the ‘Nordic prerogative’. ⁶⁴ The Nordic prerogative is associated with older ideas about the Nordic or Scandinavian ‘people's’ geographic, cultural, economic and political affinity. ‘Scandinavianism’ was an ideology and a movement during the middle of the 1800s which sought to unite the Scandinavian people, even within a common kingdom. At the beginning of the 1900s the movement's ideas lost ground, but they were replaced by ‘Nordism’ which focused on more practical economic and political cooperation. After the First World War there were tentative attempts to formalize Nordic cooperation. But it was only after the Second World War that this took off with the formation of the Nordic Council in 1953. ⁶⁵ It is in this historical context that the ‘Nordic prerogative’ and its impact on refugee policy should be understood.

Neutrality and ‘Nordism’ were two important elements of Swedish foreign policy during the war. ⁶⁶ In the policy statement of the coalition government, these components are present in the following key sentences:

The coalition government wants to fulfill the declared and documented Swedish ambition to create trust in and respect for our people's will to remain autonomous and neutral. This ambition also includes cooperation with the other non-aligned neutral states … and the preservation and expansion of Nordic collaboration.’ ⁶⁷

This Nordic intention came out plainly in the political debates, and there was to be wide support for the notion that it was in principle self-evident that any Swedish commitments primarily applied to the Nordic countries. As a consequence, the Swedish responsibility for refugees was primarily perceived as applying to Nordic citizens. This became clear particularly in the way the different groups of refugees were discussed when they were admitted. Put briefly: Sweden at the time recognized that it had a political responsibility for refugees primarily insofar as these were ethnically Nordic. The reception of ethnic Danes and Norwegians was seen as natural, and the smaller group of so called ‘Estonia Swedes’ (a Swedish-speaking population living in the Baltic region which offered historical evidence of Swedish influences in that area) were seen as Swedes who had returned to their ‘mother country’. ⁶⁸ Other groups, in contrast, were seen as more problematic. The Balts were one such group; Jews – even Danish Jews – were another. This became evident not least in the debate about labour market issues in which these groups caused unease. There were accusations to the effect that these groups robbed Swedish citizens of jobs. Jewish refugees were subjected to traditional antisemitic portrayals and were accused of ruining Swedish business morale or of being governed by indolence. The Balts in particular were accused of being disloyal, of undermining working conditions, undercutting wages and refusing to join trade unions. ⁶⁹ Other concerns included the issue of what would happen when the war was over and the Balts (and the Jews) refused to leave. In labour-shortage Sweden, however, it was more commonly claimed that the foreigner did not want to work and thus took advantage of the nation's generosity. Both ideas existed simultaneously. The Balts were also encumbered with other ethnically-tinged notions, particularly that many of them were Nazis and/or war criminals. ⁷⁰ In the public debate, the Balts were not spoken of as people of a ‘sister nation’, and as a consequence it was possible to apply an ideological epithet generically to the whole ethnic group. For Finns suspected of being Nazis, or for Norwegian quislings for that matter, there was a superior principle that left little room to make generalizing ideological accusations. Finnish wartime collaboration was not cause enough to accuse all Finns in general of being Nazis and war criminals. ‘A brother in need remains a brother, even if he commits … tremendous stupidities,’ said Social Democrat Rickard Sandler, thus suggesting the salience of the ‘Nordic prerogative’. ⁷¹

Nordic refugees were better off in practical terms as well. Until spring 1943, a residence permit as well as a work permit was required for all refugee groups and all professions, but after that point the work permit for forestry and agriculture was abolished. However, during the autumn of 1943 the work permit requirement was scrapped for all other employment sectors too, but only for Nordic refugees (a residence permit was still required). One year later, the work permit requirement was abandoned for Balts, but this was clearly a pragmatic measure: this group of refugees was large, lacked a refugee office to support it and was dependent upon the Swedish state for economic assistance. Hence it was important that as many as possible found employment. It was invariably easier for Nordic refugees to find employment outside the physically-intensive sectors of forestry and agriculture, and also to enrol in vocational training and higher education. They owed these benefits partly to the discursive ‘Nordic prerogative’, but also to the activity of the refugee offices. But it would be difficult to claim that authorities like SAK and the SFN differentiated between categories of refugees explicitly in their rhetoric. Ethnic or national stereotypes are conspicuous by their absence in the surviving documents of these authorities, which do not even contain traces of antisemitism, which is otherwise documented elsewhere at the time. ⁷²

Even so, the nationality of the refugees did affect their prospects after the war, and thus to what extent they had control over their lives. Aspects of security policy, intersected with notions about ethnic traits, affected the conditions of the refugees. Soviet Russian refugees were monitored more closely than other categories and did not receive as generous residence permits and work permits as others – even when compared to others who were formally Soviet citizens, such as Ingrians and Balts. Over this period, it became less common to categorize refugees by nationality and more common to look at individual cases, but Russian refugees were still treated differently. ⁷³ This special treatment is evident also in that in cases of non-Nordic labour recruitment, when guarantees had to be provided, stating that the immigrant would be tied to a specific place or employer. ⁷⁴ Nordic citizens, by contrast, were almost immediately given the opportunity to lead a life in Sweden on conditions similar to those of Swedish citizens. This circumstance raises two important points: even if postwar authorities actively sought to equate ‘immigrants’ and ‘refugees’ with Swedish citizens in most respects, they still singled them out partly on grounds related to national security; and secondly, in practice nationality and ethnicity remained factors that affected what chances the refugee or immigrant had in Sweden.

Insofar as Nordic labour immigration was preferred, the postwar period makes it evident that the ‘Nordic prerogative’ remained in force. Non-Nordic immigration policy during the 1940s and 1950s was consequently characterized by an ethnic hierarchy which prioritized some groups over others. That Swedish immigration policy was restricted toward the end of the 1960s can partly be explained by a decreasing demand for labour. However, in addition to these labour market concerns, it should also be observed that in this period the previously dominant Nordic immigration decreased and was replaced by immigrants from, for instance, Yugoslavia and Greece, who were seen as less close to ‘Swedish culture’ and thus regarded as more of a problem. The trend during the 1970s to the 2000s points in the same direction. ⁷⁵ Thus one can reasonably say that Swedish immigration policy has been guided not only by welfare state ambitions, labour policy, economic and humanitarian considerations, or by concerns for the security of the state, but that it was also a product of salient perceptions of ethnicity.

The foregoing discussion suggests that a number of changes took place in different fields related to refugee and migration between the 1930s and 1950s, and that it is necessary to think in terms of a continuous process of change during the 1940s, rather than a specific ‘turning point’ around 1943, as previous research has suggested. Yet the fact remains: whereas the interwar period was characterized by xenophobia, by notions of the different value of different ‘races’, or by ‘Sweden for the Swedes,’ something occurred during the 1940s which made it possible for the legislation of 1954 to equate the ‘foreigner’ with the Swedish citizen and to allow for what was – practically speaking – free immigration to Sweden until the end of the 1960s. There are several possible explanations for this change in Swedish refugee and immigration policy during and after the Second World War. ⁷⁶

Firstly, the emerging Swedish welfare state clearly made use of experience and methods that were already tried and tested when handling the refugees. Training courses, state-instigated labour market adjustments and direct financial support are some examples of measures designed for Swedes but applied to Nordic refugees. These procedures were then utilized when dealing with refugees of other nationalities. Special arrangements for refugees were rare or non-existent. The decision to include the ‘foreigners’ in the Swedish system was thus made at an early stage, when Norwegian refugees already constituted an absolute majority of the refugees. This suggests that the ‘Nordic prerogative’ played a role in the process, and sanctioned the inclusion of other refugee groups too.

Secondly, when the Swedish state was faced with the challenge of taking full responsibility for tens of thousands of Baltic refugees this led to a new way of approaching how to take care of refugees and other migrants. Despite the problems, the authorities found that an expertise for dealing with a large-scale refugee influx had evolved. Step by step, state administration had learnt how to handle the practice of refugee admission, and had reorganized itself in order to better tend to the needs of the refugees. In addition, the state had come to realize how much refugees contributed to the labour market. In short, it was evident that Sweden was capable of admitting refugees.

Thirdly, and directly related to the above, during much of the Second World War and well into the 1960s Sweden had a constant demand for labour. The social and legal aspects of incorporating working refugees and ‘foreigners’ into the labour market was dealt with by granting them the same rights as Swedish workers to participate in state insurance schemes covering unemployment and health care, which was also of importance for the condition of ‘foreigners’ who were unable to provide for themselves.

Fourthly, the state administration was professionalized and the state's overall influence over the private life of individuals increased. A general administrative development in which the state official became an ‘expert’ in a growing state administration thus made its mark also in the formulation and administration of refugee and immigration issues. In this sense, the development of refugee policies was part and parcel of the structural and ideological overall development of the Swedish welfare state.

Finally, and more broadly, the internationalization of the refugee issues in the postwar period had consequences for the way in which legislation concerning ‘foreigners’ was handled on a national level. In the Swedish case, the 1954 Aliens Act bore clear traces of the influence of the 1951 UN Refugee Convention, and is one important reason why – at least formally – the refugees came to be included in the same social citizenship that has been associated with the emerging welfare state.