The literature recognises that, despite a lingering or nominal commitment to internationalist ideology, labour movements and individual trade unions are embedded in their particular national contexts and charged, as they see it, primarily with representing the interests of their national membership (McShane 2004; Penninx and Roosblad 2000; Ebbinghaus and Visser 1999). Also, as outlined earlier, trade unions have generally seen their interests as being best served by restrictions on immigrant labour largely because, as Castles and Kosack (1973) have pointed out, a surplus of workers on which employers can draw tends to weaken the position of trade unions and concomitantly have a depressing effect on wages. These challenges were faced by the trade unions across Western Europe after World War Two as labour migration developed and foreign workers arrived into national labour markets in large numbers. The challenges continue to the present day as migration evolves and changes.
In considering the nature of immigration over this period and the trade union response in the selected countries, I have deployed what Penninx and Roosblad (2000) identified as ‘the three main dilemmas’ that unions had to face when confronted with the issue of migrant labour (See Figure 3 below). First, was the question of whether to cooperate with employers and the state in the employment of migrant labour or to resist? Secondly, once migrant workers had arrived, whether to include them fully or exclude them. Thirdly, if following a line of inclusion, whether to adopt a policy of equal treatment or one of special measures for this new category of union member. I have attempted, in each case, to examine the initial attitude of trade unions towards immigrants, the extent to which they are included, if at all, in unions and their structures, the development of union policies on immigrant workers and the role of unions in the wider struggle against discrimination and racism. The focus is on labour migration and migrant workers and not on immigration for political asylum or refuge.
Figure 3: Analytical Framework, Europe
As presented in the analytical framework, the four possible explanatory factors that influence union responses to migration are set out. The first of these variables is the character of the immigration which may help to explain differences in attitude. Historical ties may play a role as in unions may be more open to migrants who have automatic legal status by virtue of colonial links. Equally they may accept migrants who share characteristics with the indigenous workers, cultural, physical, religious, while being less open to those that do not (Wrench, 2004; Penninx and Roosblad 2000). The second is the economic and labour market conditions that pertain in the society in that it seems reasonable to assume that trade union resistance to immigration will be greater in times of economic downturn whereas in times of economic growth and full employment, unions may be more amenable to co-operation. A third variable is the national politico-legal context in which trade unions are based in that national identity and historical path-dependency are integral to the development of trade union orientation (Taylor et al., 2011). The fourth and final variable is that of the industrial relations context and the position of trade unions in the power structure of the society, in other words, the model of trade unionism that applies as discussed above. Marino and Roosblad (2008) hypothesised that because corporatist unions are not dependent on rank and file mobilisation to affirm their position in the policy-making arena, they are likely to expend less effort on recruiting minority workers such as migrants than are more oppositional unions. In contrast, unions with a weak institutional position within an industrial relations system, such as in the Mediterranean model, would have a greater focus on membership levels and diversity as a way of justifying their role and consequently put more effort into organising and recruiting migrant workers.
3.3.1. Initial Immigration: co-operation or resistance
While there was a great deal of convergence in trade union responses to immigration within the nine countries, there was also a significant level of divergence. In the first instance all trade unions were confronted by the dilemma of their avowed commitment to international solidarity versus representing the interests of their existing membership. The concern that the importation of foreign labour would undermine trade union bargaining and have a depressing effect on wages and employment standards was shared by all, irrespective of their position within national power structures. In all cases immigration began at a time of economic prosperity within the individual states. These were periods of full employment and increasing labour shortages. This was a significant factor in initial trade union responses. However, there were variations in responses within the different states.
Trade unions in Austria and Sweden were initially resistant to immigration while those in Germany, the Netherlands and Britain took a more inclusive and open approach (Penninx and Roosblad 2000). With regard to those countries which experienced immigration later, unions in Spain and Ireland welcomed migrant workers from the start but did not necessarily actively organise them while Italian unions were initially ambivalent. However, in almost all cases unions supported, and sought, restrictions with regard to countries of origin, numbers, sectors, etc. Italian and Spanish unions were the exceptions as they did not support restrictions at any point (Kahmann 2002; Calavita 1999).
The acceptance of immigration by unions in Germany, the Netherlands, Sweden, Britain, Italy, Spain and Ireland was, in most cases, a conditional acceptance. German unions sought and attained rights but always on the basis that the employment would be temporary. The Dutch unions fully accepted ex colonial migrants and were positive towards the ‘guestworkers’ who arrived in the 1960s, and advocated equal treatment. The Swedish trade unions were initially resistant and thereafter insisted on restrictions though they did also insist on equal wages and employment conditions for migrant workers. In the case of Britain, though the trade unions were hostile to the general recruitment of migrant workers immediately after World War Two, like their Dutch counterparts, they did not oppose the arrival of Commonwealth migrants, as they were not regarded as labour migrants but as UK citizens with equal rights. But, unlike the Dutch, UK unions had no involvement in decision making in this regard.
When immigration to Italy first began in the 1970s the trade unions took a ‘wait and see’ approach but did not really engage with the issue until the 1980s. Spanish unions on the other hand seemed to accept immigration from its beginnings in the mid-1980s and reached out to migrant workers, irrespective of their status. Irish trade unions did not resist immigration when it began in the 1990s and they adopted a rights based approach, seeking equal rights and entitlements for migrant workers.
In Austria the trade unions resisted attempts to introduce immigration and thereafter looked for very restrictive measures of control. Through their strong position in economic decision-making, they sought strict quotas and lesser employment and social rights for migrant workers. According to Gächter they used this influence to pursue a policy of ‘protecting indigenous workers from immigrants’ (2000: 67). French unions, which had little influence over government policy, also favoured a restrictive immigration policy but, in contrast to Austrian unions, they considered migrant workers ‘an important section of the working class’ who should be organised (Lloyd 2000: 117).
Even for those unions who actively resisted the import of foreign labour initially it became increasingly obvious that the maintenance of such a position was unsustainable. In that post-war era most Western European countries experienced labour requirements, which far exceeded what could be filled by indigenous workers. And so the import of additional labour was seen as being necessary to re-build those countries and to sustain continuous economic growth. It is against this background that many unions modified their initial position of resistance towards labour migration into a position of acceptance coupled with a demand that it not diminish wages or employment standards and that migrants not represent a cheaper alternative to indigenous workers. In most countries unions succeeded with this demand, largely through a variety of forms of legislative provision (Wrench 2000; Castles and Kosack 1973). These provisions ranged from the severely restrictive and nationalistic in the case of Austria through the less restrictive in the cases of Germany and the Netherlands to the somewhat more liberal in terms of migrant workers’ rights in the cases of Britain and France and, to a greater extent still, Sweden.
Attitudes among trade unions became much more polarised as the recession of the 1970s arrived. There is evidence of some trade unions becoming resistant to labour migration as increased unemployment became a feature of many Western European countries. The world-wide economic downturn led directly to calls for and, in most cases the introduction of, restrictions on immigration. Penninx and Roosblad argued that an “alliance between governments and trade unions in favour of restrictive immigration policies since the mid-1970s seems to be a natural one” (2000: 189) and indeed, as a minimum, no national trade union movement opposed the introduction of restrictive policies and many actively endorsed them. German, Dutch, British and French trade unions all supported government initiated restrictions while in the cases of both Austria and Sweden, it was at the instigation of the unions that restrictions were introduced. In most cases this coincided with government policy. This was also a period when both governments and trade unions began to recognise the more permanent nature of immigration and trade union policies began to become more nuanced with calls for recruitment restrictions combined with the introduction of integration measures for immigrants already resident. However, there were exceptions as some unions recognised that the movement of people is an inextricable part of the ‘global age’ and increasingly viewed restrictive migration policies as neither desirable nor feasible (Haus 2002; Avci and McDonald 2000; Watts 1998).
So we can distinguish a number of different types of reactions on the part of trade unions to early immigration and to the question of whether to co-operate or resist. In their analysis of such responses, Penninx & Roosblad (2000) identified a four-way breakdown. Firstly, there is co-operation from a position of strength which would apply in the cases of Sweden, Germany, the Netherlands and, to a lesser extent, Ireland. In these countries, unions were part of the socio-economic decision-making process, giving them, for the most part, a direct input into decisions on migration with which they co-operated. The case of Ireland was somewhat different in that, though unions were social partners, they were not participants in the decision making around immigration. Austrian unions represent a second type of reaction, that being resistance from a position of strength in that they used their strong position within the socio-economic decision-making process to oppose immigration and to limit migrant workers’ rights within the country. The reaction of the British unions to that initial immigration was one of co-operation from a position of weakness in terms of their lack of a role in the national decision-making process but they did not oppose immigration, at least not initially. This too was the case with the Italian and Spanish unions who would not have been consulted by government but who did not oppose immigration and were supportive of immigrant workers when they arrived in the country. The situation in France represented another variant, that of resistance as an opposition movement. French unions, who were outside the decision making process, were critical of the Government’s lack of regulation of migration and looked for greater controls.
3.3.2. Migrants: inclusion or exclusion
The nature of the dilemma for unions changed after the mid-1970s. In the direct sense trade unions were relieved of the original one – that of co-operation or resistance - in so far as most governments were now implementing very restrictive policies relating to immigration due to the international economic downturn, and trade unions only had to endorse such policies.
The 1980s were characterised by greater divergence between unions and governments on immigration policy. In most cases trade unions opposed increasingly restrictive measures being introduced by national governments as being incompatible with protection of immigrant workers rights. Only in Austria did unions continue to call for restrictions and fully support government policies of short-termism and exclusion. Elsewhere, there was a growing recognition that trade unions interests are not necessarily best served by restrictive immigration policies. Particularly in countries such as Italy and Spain, which were in the early stages of immigration, and which had a large informal economy, union officials increasingly viewed restrictive immigration policies as counter-productive with the possibility of such policies channelling even more migrants into the informal economy (Watts, 1998). They feared the possibility of this having negative consequences not only for the vulnerable migrant workers, but also for indigenous workers as the growth of the informal economy undermines established labour standards.
In some countries trade unions became one of the first institutions in the host society into which migrant workers could integrate but other trade union movements pursued the integration of migrant workers less urgently (Wrench, 2000). While there may have been organisational and structural issues involved in this, the issue of racism was also a factor whereby racial prejudice was frequently seen within trade unions in their attitudes to organisation (Wrench 2000; Castles & Kosack 1973). Munck observes that “there is a history often overlooked in the annals of the official trade union movement which tends to airbrush out the sexism, racism and xenophobia which forms an integral element of most labour movements” (2008: 12). However, as it became clear that labour migration was now a permanent feature of Western Europe and as the concept of the guestworker faded and migrant workers became embedded in their host societies, trade unions became increasingly open to the recruitment and active representation of them, regardless of race (Penninx & Roosblad 2000).
In general, having accepted immigration as a phenomenon, trade unions went for a policy of inclusion whereby they accepted legally resident migrant workers as part of the national workforce who should be included and organised. In fact, in the case of Sweden, union membership was largely a requirement of employment. So, on an ideological level, trade unions were open to including migrant workers in their ranks. However, when examining the situation across the nine countries, it is clear that in many cases this was a passive openness to inclusion which is not, of itself, sufficient to constitute real inclusiveness.
Over time German unions increased their efforts to organise migrant workers and to integrate them into the workplace (Kühne 2000). This was facilitated by the reform of the Works Constitution Act in 1972, which accorded the right to immigrants to be elected on to works councils. Penninx and Roosblad (2000) argued that German trade unions were a major institutional force for integration, in the absence of adequate government policy. Dutch trade unions also took a positive attitude to organising and representing migrant workers from their arrival in the early 1960s, recognising the dual benefit of this approach in improving the conditions of migrant workers while minimising the possibility of unfair competition from them (Roosblad 2000). Irish unions also adopted a policy of inclusion from the start of labour migration to Ireland in the 1990s, taking a similar position to that of the Dutch (ICTU 2005).
While French unions operated a policy of inclusion in relation to immigrants there was, and is, a tension within French unions between particularism and universalism with equality of treatment seen as being a fundamental tenet of French trade unionism (Lloyd 2000). As mentioned above, in the Swedish case union membership was largely a condition of employment but unions also insisted on equal pay and working conditions for migrant workers. Thus while many immigrants may have worked in low paid jobs they were paid union wages and had access to full social welfare benefits ((Knocke 2000). Though British unions had accepted ex-colonial migration, it was only from the late 1970s that the unions began to actively tackle issues of racism and discrimination and to put specific structures and policies in place targeted at black and minority ethnic members (Krings 2009b; Wrench 2004). Austrian unions were reluctant to pursue inclusion measure and, to quote Gachter, “… in deed, if not in word (they) pursued a strictly discriminatory policy against immigrants settled in the country” (2000:84) Latterly this has changed somewhat and Austrian trade unions have put in place some measures to encourage greater inclusion of migrant workers (Fulton 2003).
Italian unions had little engagement with migrant workers initially as they were largely employed in sectors and jobs with little or no unionisation and also they settled to a large extent in central and Southern Italy where traditionally the labour movement had been weak. However, this changed over time and from the 1980s Italian unions were actively involved in the recruitment and representation of migrant workers, most markedly, on issues outside of the workplace such as regularisation, housing, language etc. (Basso, 2006; Calavita 1999). Spanish trade unions provided support to migrant workers from their first arrival in that country and, like their Italian counterparts, much of that support was provided outside of the workplace and on non-labour related issues (Kahmann 2002). In both cases they set up a range of services to assist immigrants, in many cases, beyond the usual/typical remit of trade unions.
Membership density is commonly seen as the obvious measure of inclusion but it is not an entirely reliable indicator. As can be seen here overall membership density varies across the nine countries studied. While that relates to some extent to international economic change, it is primarily dependent on national institutional factors, most particularly the position of the union in socio-economic decision making, whether the union operates as a service model and whether there is obligatory membership. Secondly, the data on immigrant membership is not comprehensive as some unions have an ideological objection to collecting such information while others have difficulty in compiling it and others still categorise migrants and ethnic minorities under the same headings (Visser, 2006; Fulton, 2003).
Sweden has high overall union membership levels and equally high levels among immigrants, there being a compulsory element to membership. Germany’s density is lower but again with equal levels among immigrants. Information on the case in France is indicative of a similar pattern. In Italy, however, where immigration came late, membership is somewhat higher among immigrants than among the indigenous workforce. This is partly explained by the fact that traditionally Italian trade unions do not differentiate between legal and illegal status and they also operated a very pro-active approach to recruitment. In Britain, Austria, the Netherlands, Switzerland, Spain and Ireland union density is lower among migrant workers than among indigenous workers. In so far as it can be assessed the general situation seems to be that overall rates of union membership vary according to national circumstances, there is variation between the level of membership within the indigenous workforce and the immigrant workforce but the variation is not huge and can probably be accounted for by national structural and organisational factors (OECD 2014; Visser 2006; Penninx and Roosblad 2000).
Involvement in trade union structures is possibly a stronger indicator of unions’ commitment to active inclusion and while there is some variation, the picture that emerges indicates that representation of migrant workers at decision-making level within unions throughout the nine countries is poor. To start with, in the case of Spain those who are in the country illegally are precluded from joining a union in the first place. In Britain immigrants appear to feature at all decision-making levels but it is important to note that, in the case of Britain ethnic minorities and migrants are counted as one category so that the picture presented represents the combination of both migrant and ethnic minority representation. In Ireland too immigrants feature at all decision-making levels but, at national executive level, only in three unions and the numbers generally are very low proportionately.
In Italy and Germany migrants are represented at a variety of decision-making levels though not at executive committee level. In the Netherlands representation is very low with migrant workers represented only at union delegation levels. Marino and Roosblad (2008) would argue that this is unsurprising as there was very little priority given to migrant worker recruitment at local level. In the case of Sweden, representation of migrants is thought to be low but the Swedish unions could not provide definitive figures. No information was available from the French trade unions (ETUC 2011; Fulton 2003).
Another measure of union commitment to inclusion is how actively it pursues its recruitment and internal integration policies. It is noteworthy that while representation of migrant workers at decision-making levels within trade unions is low, in most cases trade union bodies have had specific strategies and measures in place both to increase it and to increase migrant membership generally. Almost all trade union confederations and many of their member unions have designated staff with specific responsibility for migrant issues, many produce literature in a variety of languages and provide targeted training for migrant members. Unions in Britain, Germany, Sweden, Italy and Spain are particularly strong in this regard. Austrian unions have some inclusion measures in place while the Dutch trade union confederations have no strategies in place to increase migrant representation other than generalist anti-discrimination ones. The Irish trade union confederation has measures in place to increase migrant membership with some unions developing strategies specifically targeted at improving representation at decision-making levels. French unions have actively, and quite successfully, recruited migrant workers from the beginning and while they do have some measures in place to increase general minority participation at decision making levels, they do not monitor their effectiveness as any such monitoring is seen as being in contravention of the universalist principle.
Finally, the issues of illegal immigration and irregular employment have been of increasing concern for unions as the numbers employed in irregular situations throughout Europe grow. This has led to real fears of a negative impact on labour standards and wage rates. Unions took, and continue to take, different approaches in different national contexts with regard to inclusion of such workers. While many unions have supported government initiatives at regularisation, they have also generally either supported or been silent with regard to restrictive legislation to control illegal immigration. Some, such as those in Germany and the Netherlands support government measures to combat irregular migration but also provide services to workers in irregular situations once they are in the country. Austrian unions seek severe restrictions on irregular migration and do not provide any union supports to those in such a situation. In the case of both Britain and Sweden, unions do not have a stated policy on irregular migration and they provide union services to members only. Irish unions do not have stated policies either but, individual unions provide informal support to individuals in irregular situations. The majority of unions in France and Italy do not support restrictive measures and offer a full range of services to all workers, including those in irregular situations. Spanish unions do not support restrictions and they too offer supports to workers in irregular situations but, as pointed out earlier, they are legally prohibited from recruiting them (Krings 2007; Hamann 1998).
3.3.3. Migrants: equal or special treatment
In most cases the question of equal versus special treatment of migrant workers has had to be confronted by trade unions within the selected countries at some point in time. Generally unions are somewhat suspicious of special policies for certain groups as they aim to organise workers regardless of nationality or ethnic belonging. When it became apparent that many migrant workers were ‘here to stay’, most trade unions, over time, adopted some special policies in recognition of the particular circumstances that the former have to face. Penninx and Roosblad observe from their study that the historical evolution seems to be that ‘in the course of time but at different points in time’ most national trade union organisations come around to the view that the specific situation and characteristics of migrant workers require special attention and policies’ (2000: 198). These policies can range from anti-racism and discrimination, to the provision of specific targeted forms of assistance and advice to special committees and sections (Visentini 2011; ETUC 2003, 2011; Wrench 2004).
There is an inevitable overlap between many of the inclusion measures outlined in the previous section and special treatment, in that, in order to be inclusive many unions put in place special measures such as designated staff, specific training courses, producing literature in a range of languages, special committees etc. In almost all cases unions have put in place some or all of these measures at some time. German unions have measures such as special services and have also increasingly placed immigration issues on their national negotiation agenda. Austrian unions came very late to these policies and, while they do provide some targeted information and support services to immigrant members and have supported and initiated anti-racism and anti-discrimination training, it is all still within a context of the unions seeing their primary function as the protection of the indigenous Austrian worker (Fulton 2003; Gachter 2000). The Dutch unions have tended to frame policies of inclusion largely in anti-discrimination terms and Roosblad (2000) would claim that the rhetoric on the issue is not necessarily matched by the reality. While specialist committees and policies were developed by French trade unions in the early days of immigration9, they lean generally more towards an equality, rather than a separatist, approach as outlined earlier with a focus on anti-discrimination and anti-racism. Many do have designated staff with particular responsibility for migrant issues, and the confederations do have specific policies to increase migrant participation at activist and executive level (Fulton 2003; Lloyd 2000) Like their French counterparts, Swedish unions established specialist committees and sections in the early days of immigration but by the 1990s had come to see this organisational model as problematic and a contributing factor in the marginalisation of both migrant issues and of the migrants themselves. They then moved towards a more inclusive model with immigrant issues integrated into general union business (Fulton 2003; Knocke 2000). Through this model they run targeted recruitment and awareness raising campaigns and also provide internal anti-discrimination training.
British unions have been committed to the application of special measures since the 1970s which saw the growth of racism and right-wing extremism in the UK, including within the ranks of the trade union movement. They have well established and long-standing anti-racism and anti-discrimination policies and practices such as self-organisation structures for black and ethnic minority members, ethnic monitoring and positive action measures (Wrench 2000, 2004). Italian trade unions too adopted a particularist approach to the issue of migrant workers. They moved away from traditional forms of unionism and adopted a social movement type approach, meeting with migrants outside the workplace, providing a wide range of supports to all, irrespective of legal status or union membership, “it was outside the workplace and on issues not strictly connected with labour negotiations that the first mass contact between immigrant populations and trade unions took place” (Basso 2006: 2). Largely due to internal pressure from union members, Italian unions were forced to re-evaluate this approach in the mid-1990s and revert to a more traditional approach whereby they represented immigrants as workers primarily. However, they continued to provide specialist services and most trade unions have succeeded in including immigrants at all levels of the organisation, with a union rate equal to, and in some cases, higher than that of native Italian workers (Basso 2006; Fulton 2003).
Spanish unions have also introduced specialist measures to support, recruit and develop migrant workers, chief among them are the designated Immigration Offices and the network of Information Centres for Foreign Workers (CITEs and Centros) established by the two union confederations. In 2010 there were close to 100 centres throughout Spain providing a broad range of support and information services to both immigrant union members and non-members. Despite the level of services the fact that immigrants in irregular situations cannot be union members under Spanish law has an obvious knock on effect on both migrant union membership and on migrant worker involvement at decision making-levels (ETUC 2011; Fulton 2007; Kahmann 2002). Despite a general policy commitment to an inclusive and universalist approach, increasingly Irish unions have introduced special measures, such as language support, designated migrant worker organisers, targeted advice and information and active organisation through links with migrant representative and support groups.
Perhaps the most significant example of special treatment manifests itself in collective bargaining situations where unions negotiate issues of specific interest or relevance to migrant workers. An ETUC study of 2003 found that a number of union confederations include such issues in their negotiations (Fulton 2003). In the cases of France and Sweden quite broad equality of opportunity, anti- harassment and anti-discrimination agreements were sought. But many also negotiated around issues of equality of access to training, promotion, workplace benefits and recognition of qualifications. In Germany, Italy, the Netherlands, Spain and Britain issues pertinent to the cultural characteristics of specific migrant groups were raised in these situations. Unions in these cases have negotiated agreements around religious practices and holiday arrangements. These types of negotiations are not a feature of either the Austrian landscape where there is no specific outreach to migrant workers or the Irish one, where the representative focus is generally on migrants in the context of their employment circumstances e.g. low-paid, irregular. Though this evidence is not necessarily definitive, it does indicate that, in most cases, unions prefer to take an egalitarian approach to negotiation whereby in seeking special measures for the minority it is in order to give them equal opportunities with the majority.
Unions in all of the countries studied increasingly engage in co-operative initiatives with NGOs who work with and represent migrants. These include initiatives such as anti-racism and anti-discrimination campaigns; awareness raising campaigns around exploitative practices; Government lobbying on legislative issues; and joint support initiatives.
3.3.4. Trade union responses: convergence and divergence
When we look at the response to migration of trade union federations and their member unions across the nine countries selected we see a level of convergence emerging with initially hostile responses evolving, in most cases, into much more inclusive approaches over time. However, there are significant variations to be observed and so we look to the influence of the four explanatory variables the character of the immigration, the economic and labour market conditions, the politico-legal context and the industrial relations context (see Figure 3 above).
With regard to character of the immigration, it is seen that historical ties can play a role in union responses, such as in the Dutch and British cases where the unions were initially more open to ex-colonial migrants who had automatic legal status by virtue of their colonial links. It is also the case that societies are more likely to accept migrants who share characteristics, cultural, physical, religious, with the indigenous workers (Penninx and Roosblad 2000). However, consideration of the trade unions across the nine countries does not indicate that either the character of the immigration or of the immigrants themselves are crucial factors in the union response to migration but seem to be more a function of the socio economic environment at the time. Thus as Penninx and Roosblad point out, it seems to be the ‘characteristics which have been allocated to immigrants by the receiving society’ that is the important variable (2000: 201). These include issues such as their legal status (as in employed on work permits, or being undocumented); the frequently insecure and informal nature of their work (in generally non-unionised workplaces and their social exclusion).
As could be expected, economic and labour market conditions clearly played a role across all of the nine countries considered in the responses of trade unions to migration through its different phases. Unions were generally more open to immigration at times of economic growth and full employment as in the early days of post-war migration when there was an ever-increasing demand for labour or again during the economic boom of the 1990s. Union resistance was greater in times of economic downturn as can be seen from the almost universal calls for, and support of, restrictions on migration with the international economic downturn of the 1970s. This supports the view of Avci and McDonald that “in liberal industrialised countries, times of economic recession and high unemployment create pressures for restrictive immigration legislation, proposals which will be supported by trade unions as a means of safeguarding their interests” (2000: 191). There has been much less evidence of this response from union movements during the current recession with almost all having moved to a point where they identify migrant workers, and even those in irregular situations, as an important constituency with the focus on organisation and inclusion. This shift in perspective is widely recognised in the literature with trade unions increasingly realising the inevitability of immigration and recognising that “restrictive immigration policies are unrealistic and futile” (Watts 1998: 657). It is also the case that all of the countries now have quite restrictive policies in place on migration, removing possible internal pressure on unions from their membership. However, in countries such as Austria, France, the Netherlands, Sweden and Britain economic downturns led to the emergence, and in many cases electoral success, of the far-right anti-immigration movement.
The actions of trade unions, as actors in national contexts, are likely influenced by politico-legal factors peculiar to their individual countries. In almost all of the countries examined in this study the dominant discourse, terminology and policy development since the 1970s have centred round restriction of immigration, this despite the very evident labour market needs. While trade union movements have generally challenged this discourse they also cannot move too far ahead of their national memberships which, in many cases, are opposed to migrant labour. Wrench sees these factors as being especially significant, “ofparticular importance is the quality of public discourse on immigrants,ethnic minorities, and multiculturalism, in particular that of politicalopinion leaders, and the balance between ‘conflict’ and ‘consensus’ inpolitical life, also reflected in the relationship between trade unions,employers, and government” (2004: 8). Penninx and Roosblad too identified these as the ‘most important explanatory factors’ and point out that immigration and the presence of ‘foreigners’ was perceived differently in the light of different historical experience and regulation by different instruments (2000: 206). This is probably the most likely factor to explain the contrasting attitudes of German and Austrian unions to immigration with the Austrian unions’ stance of rejection of an open immigration policy going hand in hand with a discourse focused on the protection of the native workforce. It is also a significant factor in explaining the resistance of French unions to the introduction of specific migrant focused policies as they are strongly committed to the republican values of French society with their focus on equality and universality.
It is evident from the literature generally and from an examination of the situation in the nine countries specifically, that the industrial relations system and, most particularly, the level of union institutional embeddedness, are critical factors in determining the overall behaviour of trade unions, including their specific response to migration. In fact the pattern that emerges indicates that the particular model of trade union is a less important determining factor than is the broader socio-economic position of the trade unions (e.g. corporatist vs non-corporatist or strong vs weak institutional embeddedness). Thus, for the most part, strongly embedded unions such as those in Germany, Sweden the Netherlands and, latterly, Ireland co-operated with governments and employers, somewhat reluctantly, in the opening up of economies to ‘temporary’ immigration but were not pro-active initially in organising such workers. On the other hand weakly institutionally embedded unions in Britain, Italy and Spain did not oppose immigration but did engage directly and pro-actively with migrants and their issues from an early stage. This supports the thesis that where trade unions’ institutional position is strong, they focus on building institutional partnerships with less incentive to prove their strength through recruitment and mobilisation but where their institutional position is weak, with little influence on policy-making, they are more likely to be pro-active in recruitment and organisation (Baccarro et al. 2003; Frege and Kelly 2003; Ebbinghaus and Visser 1999). However, while this factor can explain the individual union movements’ approach to migrant worker organisation, it does not explain the differences in union response to migration. Nor does it explain the responses of the union movements in France and Austria who both sought restrictions on migration. The other factor here discussed by Gorodzeisky and Richards (2013) is the role of what they call ‘organisational security’. They argue that a single dominant union confederation (as, for example, in Austria, Britain and Ireland), holding a monopoly position, enjoys greater organisational security and therefore may have fewer incentives to organise migrant workers. This is in contrast the situation where there are competing union confederations which may be more willing to recruit migrant workers as a means of gaining organisational strength at their rivals’ expense.
And so, as has been argued by Penninx and Roosblad (2000), there is no clear-cut mono-causal explanation for individual trade union movement responses and no one factor on its own can explain the complexity and multi-faceted nature of national trade union movement responses. It is only possible to draw valid inference through consideration of all of the variables. When taken separately they have limited explanatory power.