East-West Labour Migration and the Swedish Cleaning Industry a Matter of Immigrant Competition?

TheMES Themes on Migration and Ethnic Studies occasional papers and reprints

42 (2015)

Annette Thörnquist East-West Labour Migration and the Swedish Cleaning Industry A matter of immigrant competition?

TheMES Themes on Migration and Ethnic Studies

All rights reserved. No part of this publication may be used or reproduced in any manner whatsoever without written permission except in the case of brief quotations in articles and reviews. © REMESO and the author 2015. TheMES No. 42/15 ISSN 1651-8306 Cover: Martin Ålund

Series Editor: Erik Berggren, +46 11 36 32 66, [email protected]

Orders: Eva Rehnholm Tel. +46-11-36 32 38 Fax. +46-11-36 30 04 E-mail: [email protected]

Address: Linköping University ISV-REMESO SE-601 74 Norrköping, Sweden

Webpage www.isv.liu.se/remeso The author acknowledges research funding support from the Swedish Research Council for Health, Working Life and Welfare (FORTE), Project Number: 2011-0338, ‘East-West labour migration, industrial relations and labour standards in a Swedish-Baltic context’. The author also warmly thanks the leader of this project, Professor Charles Woolfson, and the project member, Ph.D. Candidate Indre Genelyte, at the Institute for Research on Migration, Ethnicity and Society (REMESO), Linköping University. Many thanks also to Dr. Bjarke Refslund, researcher at the Department of Political Science, Aalborg University. Abstract The Swedish cleaning industry provides a good example of an ethnically segregated labour market. Almost 50 per cent of the workforce was born abroad, and an unknown number of undocumented immigrants are working informally in this labour market, especially in the metropolitan areas. Price pressure in public procurement of cleaning services and unfair competition generate market disturbances and labour abuse. In the light of these structural conditions, the aim of the present study is to discuss the presence of temporary labour migrants from Poland and the Baltic countries in the Swedish cleaning industry. The study shows, that since there is a shortage of jobs in this labour market, there is also a potential risk of competition between previous and new labour migrants, especially in large cities. Hence, when discussing labour migration to Sweden, or other destination countries, we need a broad social perspective that takes into account the impact of existing segregation and labour market segmentation, as well as the balance of power between the social partners and the labour market policy pursued by the governments. Thus, this study also reveals another important factor that most likely has an impact on the comparatively limited inflow of labour migrants from the newer EU countries to the Swedish cleaning industry, namely that the employers can easily recruit workers in Sweden participating in labour market projects, which means considerable wage subventions. In other words, employers can use a tax-funded reserve army of cheap labour to gain competitive advantage, for example in public procurement. This practice wors- ens the problems of price pressure, oversupply and competition by undercutting labour standards, which may generate further social segregation and labour market segmentation.

Key words: Swedish cleaning industry, temporary labour migration, newer EU countries, labour market segmentation, immigrant competition, tax-funded reserve army of labour.

Table of contents

I: Introduction 1 Aim and scope of the study 2 Theoretical approach 3 Methodological considerations 5 II: The structure of the Swedish cleaning industry 8 III: Trends in East-West labour migration 12 IV: Ways of approaching the Swedish labour market 16 Posted workers and temporary agency workers 16 Self-employed workers 19 Recruitment of labour from ‘third countries’ 21 V: East-West labour migration and the Swedish labour market 24 The buyers’ labour market 25 A segmented labour market 26 A reserve army of tax-financed cheap labour 27 VI: Concluding remarks 30 References 33

East-West Labour Migration and the Swedish Cleaning Industry

East-West Labour Migration and the Swedish Cleaning Industry A matter of immigrant competition? Annette Thörnquist

I: Introduction Traditionally, the vast majority of the workers in the cleaning industry are women. In Sweden, as well as in other Nordic countries, cleaning also provides a good example of an ethnically segregated labour market. Foreign born workers are highly over-represented, especially in the metropolitan areas where most immigrants have settled. These workers have mainly come from Southern Europe, South-East Asia and Africa (Bettio and Verashchagina, 2009; Gavanas 2010; Abbasian and Hellgren 2012). Since the enlargement of the EU in 2004 and 2007, there has also been an east-west labour migration within the EU/EEA region. In Norway and Denmark, most labour migrants from the newer EU member states who are working in the cleaning industry have come from Poland, but there is also an increasing inflow of workers from Lithuania, Romania and Bulgaria (Lisborg 2012; Friberg and Eldring 2013; Rose Tronstad and Andersson Joona 2013; Scmulyar Green and Spear 2014; Refslund 2014). In Finland, labour migrants in cleaning have mainly come from Russia and Estonia (Statistics in Finland 2013; Komulainen 2012; Ollus et al. 2013). When it comes to the Swedish cleaning industry, almost 50 per cent of the workers are foreign born people, compared to around 16 per cent of the population as a whole (Almega Serviceentreprenörerna 2014; SCB 2014a). However, we do not know much about labour migrants from the former Eastern Bloc who are working temporarily in the Swedish cleaning industry. Swedish studies on labour migrants in cleaning have mainly dealt with domestic cleaning (Gavanas 2010; Gavanas and Darin Mattson 2011; Calleman 2011; Shmulyar Green 2013). Even though domestic services have increased significantly in the 2000s, the main part of the multifaceted cleaning industry is office cleaning in public institutions and private companies, carried out mainly by contract companies. However, there is so far little research on temporary migrant workers in this sector. Previous studies have mainly concerned immigrants living in Sweden (e.g. Akhavan, Sharareh 2006; Abbasian and Bildt 2007; Abbasian and Hellgren 2012), or undocumented immigrants engaged informally in this and other low-wage sectors (Elmhorn 2008; Khosravi 2009; Björklund Larsen 2010)

1 Annette Thörnquist

Aim and scope of the study The aim of the present study is thus to discuss the presence of temporary labour migrants from Poland and the Baltic states in the Swedish cleaning industry, how workers have approached this labour market, and how trends in east-west labour migration to this sector can be explained. The discussion also includes posted workers, even though they should not be regarded as labour migrants (European Commission 2014). The term migrant workers/labour migrants has been defined in various ways in different contexts. The United Nations’ International Convention of Migrant Workers and its Committee aimed at protecting labour migrants and their families has a wide definition. A migrant worker is ‘a person who is to be engaged, is engaged or has been engaged in a remunerated activity in a state of which he or she is not a national’ (UN 2005: 4). The Swedish Committee for Labour Migration (Kommittén för arbetskraftsinvandring), which preceded a revision of the regulations on labour migration from outside the EU/EEA region in 2008, defined labour migrants as ‘migrants who move to Sweden for labour market reasons, and who are formally allowed to settle in Sweden’ (SOU 2006:87: 118)1. The problem with this definition, however, is that it mainly refers to labour migrants who have been staying in Sweden for 12 months or more, and who thus are included in the population register as ‘immigrants’(cf. Andersson Joona and Wadensjö 2011). There are also various definitions of ‘temporary migrant workers’ (Equality and Hu- man Rights Commission 2011). According to the International Organization for Migra- tion (2014), this term refers to ‘skilled, semi-skilled or untrained workers who remain in the destination country for definite periods as determined in a work contract with an individual worker or a service contract concluded with an enterprise’. ‘Temporary’ may implicate longer or shorter time-periods in a migrant’s life. However, this study focuses mainly on workers who are staying for less than a year. Temporary migrant workers in that sense, who are working in low-wage jobs, are among those who are most at risk of being abused as cheap labour. After World War II, the guest worker system became an established model for recruiting foreign labour in the Western world (Castles 1986; Schierup, Castles and Hansen 2006). In the early 1970s, Sweden began to pursue a more restrictive labour migration policy, while there was an increasing inflow of refugees in the following de cades. Since the mid-1990s, however, temporary labour migration has increased again, albeit within a new economic, political and demographic context. With the enlargement of the EU in 2004, transitional restrictions for east-west labour migration became a much-disputed issue in the older member states. Sweden was the only EU-15 country that allowed free access to its labour market, while the United Kingdom and Ireland introduced some minor restrictions (Arvidsson 2009; Gerdes and Wadensjö 2013).

1 The author’s translation.

2 East-West Labour Migration and the Swedish Cleaning Industry

Swedish research on labour migration from the newer EU countries in the former Eastern Bloc corresponds well with other European and Nordic studies (e.g. Kahan- ec and Zimmermann 2008, 2009; Lemos and Portes 2008; Galgóczi et al. 2009; Kerr 2011; Guild and Carrera 2012). In general, high-level economic studies have found little impact on the Swedish labour market and industrial relations model. The authors have emphasised that the inflow of workers has been comparatively low, especially in low-wage sectors. The main reasons have been higher demand for labour in other countries, such as in the United Kingdom and Ireland, and higher wages, for example in Germany, Denmark and Norway (Tamas and Müntz 2006; Andersson Joona and Wadensjö 2011; Gerdes and Wadensjö 2008; 2013; Engdahl 2013). It has also been maintained that many labour migrants from the newer EU states prefer to work in English or German speaking countries for linguistic reasons. Moreover, the problem of wage dumping has been regarded as limited as well. The overall explanation has been that the strong Swedish labour market regulations for equal rights cover all workers – including migrant workers (Ruhs 2012; Wadensjö 2012). Thus, the high union density (around 70 per cent), and a collective bargaining system covering almost 90 per cent of the labour market, make the Swedish labour market less ‘flexible’ in this respect. However, these studies have generally concerned labour migrants who have worked in Sweden for 12 months or more. Researchers who have studied the posting of workers and labour migration for shorter spells, on the other hand, have emphasised the risks for low-wage competition and severe abuse of migrant workers. In fact, they have shown that wage dumping is a problem also in industries with comparatively strong trade unions, such as in the construction and transport industries (e.g. Woolfson et al. 2010; Thörnquist 2011a; 2013a; Jonsson et al. 2014).

Theoretical approach Several economic studies on east-west labour migration within the enlarged EU have emphasised that in cases where negative effects on wages and working conditions can be found in the host countries, these effects generally appear in low-skilled service jobs, such as jobs in the cleaning industry, and in the hotel and restaurant sector (Lund- borg 2009; Somerville and Sumption 2009; Anderson and Ruhs 2010). Moreover, since foreign born workers are over-represented in these sectors, especially in the urban areas, the problems will most likely hit previous immigrants. As Somerville and Sumption (2009: 3) have put it:

This means that previous immigrants (foreign-born workers already present in the country) bear the brunt of competition from new immigrants. Most native workers are sheltered from immigrant competition, since they can work in jobs in which they have a comparative advantage over immigrants. But some natives might also lose out due to new immigration if they are unable to move into these “language-intensive” jobs (for example, due to poor communication skills).

3 Annette Thörnquist

Somerville and Sumption do not discuss in terms of segregation or labour market segmentation. Yet, the phenomenon they describe – competition between previous and new immigrants – can be related to the processes implied in these concepts. In the present study on east-west labour migration and how it concerns the Swedish cleaning industry, this relationship will be illustrated. One may say that the labour market in cleaning is horizontally segregated, and that it includes structures of labour market segmentation as well. The notion of horizontal segregation generally refers to gender segregation in the labour market, but it can also apply to segregation of other social- ly categorised groups, for example foreign born workers (Melkas and Anker 1997; Anker 1998; Bettio and Verashchagina, 2009). If the labour market is characterised by established sub-groups with different wages and working conditions with little social mobility between these groups, we may also talk about labour market segmentation (ibid.; Piore 1971; 1979; Wilkinson 1981; Deakin 2013). As regards the concept of immigrant competition, it often denotes competition between immigrants and native born workers. As the quotation above indicates, however, it can also refer to competition between previous and new immigrants (see also Lim 2002). This is how the concept will be used in this study as well. Thus, the question asked in this context is whether the social structure of the labour market may help to explain trends in temporary labour migration targeting the cleaning industry. This issue will also be discussed in the light of the labour market policy pursued in Sweden over the past ten years. The main purpose of this policy, which can be described as a workfare (work-for-welfare) policy, has been to increase the supply of labour in order to include people in the labour market, and thus reduce social security costs. Even though this policy was introduced in the 1990s (Kildal 2001; Junestav 2011), the centre-right alliance government in office during the period September 2007 to September 2014 strongly reinforced it. In accordance with this supply-side labour market policy, a large number of unemployed workers, many of whom have an immigrant background, have been engaged in various forms of labour market schemes. As will be shown in this report, they also constitute a reserve army of tax-financed cheap labour (see alsoThörn - quist 2011b, 2013b; 2014). The development of the Swedish cleaning industry in recent decades reflects the structural transformation within the public sector that has been related to the introduction of New Public Management policies (Hood 1995; Ferlie et al. 1996; Almqvist 2006; Hermann and Flecker 2012; Kamp et al. 2013). When the public sector was exposed to competition in the early 1990s, cleaning and facility services were among the first activities that were outsourced, and the use of private subcontractors has escalated ever since (Sundin and Rapp 2006; Rapp 2006; Calleman 2011). Moreover, so-called customer choice models have been introduced, for example in home-based care and services for elderly and disabled people in the municipalities. This means that the ‘customers’ can chose between the municipal provider and external providers authorised by the municipalities. In customer choice models, as well as in the system of outsourcing, the services are still tax-financed and prices are determined by political

4 East-West Labour Migration and the Swedish Cleaning Industry decisions. Thus, these markets can be regarded as a ‘quasi-markets’ (Kastberg 2005; le Grand 2007). An increasing problem in the Swedish cleaning industry in the 2000s has been a downward pressure of prices in public procurement, including also involvement of fraudulent actors in this process (Almega Serviceentreprenörerna 2014; cf. Jonsson et al. 2014). Unfair competition generates market disturbances, undercutting of wages and labour abuse. The present study also illustrates how this practice hit workers with a weak position in the labour market.

Methodological considerations This study draws primarily upon in-depth interviews with representatives for the main trade unions in the cleaning industry, the Swedish Building Maintenance Workers’ Union (Fastighetsanställdas förbund) and the Swedish Municipal Workers’ Union (Svenska Kommunalarbetareförbundet). Moreover, representatives for the main industry and the employer association, the Almega Service Contractors (Almega Serviceen- treprenörerna), and the sister organisation for domestic services, the Almega Home Service Companies (Almega Hemserviceföretagen), have been interviewed. Other important informants have been experts at the Swedish Trade Union Confederation, LO, officials at the Swedish Tax Agency (Skatteverket), the Work Environment Au- thority (Arbetsmiljöverket) and the Swedish Migration Board (Migrationsverket). The length of the interviews with the labour market parties was around 1–2 hours, while the interviews with the authorities generally were shorter and focused more on concrete facts. The names of the informants are not published, since the main interest has been in the opinions of the organisations they represent. In addition to the interviews, official statistics have been used, as well as public reports, reports from the labour market parties, and media material. The official Swedish statistics on temporary labour migration from other EU countries are limited though. As Sweden did not introduce transitional restrictions for workers from the newer member states, there is no register over work permits from the newer EU countries after 2004, for example. Moreover, there has been little coordination of statistical data between authorities in different EU countries, as well as between authorities in Swe- den. Since 2006, people from other EU/EEA countries have so-called resident right (uppehållsrätt), which means that they do not need a resident permit while working or living temporarily in Sweden. Those who intended to stay for a longer time-period than three months – for labour market reasons or for other reasons – were earlier required to register at the Swedish Migration Board. However, many person refused to register, or they were not aware of the regulations.2 In May 2014, the requirement to register after three months in Sweden was abolished.

2 Interview with official at the Swedish Migration Board (February 2014).

5 Annette Thörnquist

Many foreign workers who are working temporarily in Sweden fall outside the tax register as well. Posted workers in companies based in other countries can work in Sweden for 183 days during a 12 months period without being obliged to pay Swedish income tax. In principle, this time-limit applies to self-employed persons and companies without a permanent establishment in Sweden as well. However, there are also cross-national agreements for different industries, which means that entrepreneurs from other EU/EEA countries can stay for longer time-periods without paying Swedish tax and social security contributions. Moreover, foreign employers who post workers to Sweden, for example, can circumvent Swedish tax regulations by constant circulation of workers between the sending country and Sweden. In other words, the workers leave the country shortly before the tax-free time-period has expired. In many cases, they return again for another six-months period of work. Labour migrants often use this possibility to avoid being registered for Swedish tax as well (Thörnquist 2013a; Jonsson et al. 2014). When it comes to labour migration between the Nordic countries, the national agencies for statistical services cooperate, which means that our knowledge about Nordic labour migrants is comparatively good. However, we know far less about labour migrants from other EU/EEA countries who are working in Sweden for shorter spells than a year. This may explain why most Swedish studies on labour migration have concerned immigrants who have worked in Sweden for at least a year, and who are included in the national population register. The lack of reliable statistical data is a problem in many other EU member countries as well, which means that new registers have been introduced or discussed (e.g. Galgóczi et al. 2009, 2012; Refslund 2014). On the 1 of July 2013, an amendment to the Swedish Foreign Posting of Employ- ees Act (SFS 1999:678) came into force, which means that all foreign employers who post workers to Sweden for more than five days during a 12-month period are obliged notify their presence to the Swedish Work Environment Authority (SFS 2013:351). The companies must also have a contact person in Sweden, who can receive notifications from the authorities and provide documents that show that employment agreements and working conditions are in accordance with the EU Posted Workers Directive (96/71/ EC), as well as the Swedish Work Environment Act and some other Swedish minimum labour standards (European Commission 1996). However, the problem for the Work Environment Authority is how to control that all employers (including many small companies) do register.3 It should also be mentioned in this context that the Work Envi- ronment Authority has been vastly scaled down since 2007. If the employers refuse to report postings in time, the fines are relatively low as well (20,000 SEK). In addition, it can be a complicated matter for the trade unions to monitor the observance of the regulations in the companies. An investigation made within the LO unions showed that around 55 per cent of the local union branches found it more or less difficult to get in

3 Interview with official at the Work Environment Authority (February 2014).

6 East-West Labour Migration and the Swedish Cleaning Industry touch with the companies. For example, some companies lacked a Swedish workplace address (Jonsson and Larsson 2013). After this introduction, the report proceeds as follows. In the next section, the structure of the Swedish cleaning industry will be described. In section III, some general trends in labour migration from Poland and the Baltic States to Sweden since 2004 will be presented and discussed, while section IV concerns how temporary labour migrant workers approach the Swedish cleaning industry. The question asked in section V is how trends in temporary labour migration to the Swedish cleaning industry can be explained. In section VI, finally, the results are summed up and discussed.

7 Annette Thörnquist

II: The structure of the Swedish cleaning industry The Swedish cleaning industry includes around 2,100 limited companies with a yearly turnover of over of 27.5 billion SEK. In total, over 52,000 workers are employed in these companies. However, if we add those who are employed by the state, the municipalities and the county councils – including publicly owned enterprises – over 79,000 employees are working in the cleaning sector. The share of workers employed by public employers has been shrinking over the past decades. Since 2005, it has decreased with 17 percentage points. The main reason has been the interlinked processes of cut-backs in the public sector and increasing outsourcing and privatisation (Almega Serviceentreprenörerna 2014). Over the past ten years, many well-established contract companies in the cleaning industry have decided not to participate in public sector procurement due to an increasing downward pressure of prices, including involvement of non-serious actors.4 This problem concerns ‘general cleaning’ in particular. As a result, it has become more and more common to provide a diverse variety of facility services, including basic office cleaning as well as various forms of ‘specialised cleaning’, such as window cleaning, graffiti cleaning, carpet cleaning, and sometimes also catering and gardening.As a whole, however, profitability is low in this highly labour intensive industry where the labour costs represent the major part of the production costs (ibid.).5 Three multinational cleaning companies, one of which has over 8,000 employees in Sweden, have a very strong market position. Together they have around 15,000 employees, and their total turnover is over 11 billion SEK. In addition, the state-owned company Samhall AB, with a mandate to create work for persons with reduced working capacity due to functional impairment, is one of the largest actors in the cleaning industry with almost 6,000 employees in this sector. However, the dominating company structure in Sweden, as well as in other EU countries, is small and medium-large enterprises. Only 2.36 per cent of the 2,100 limited companies have more than 100 employees, and around 78 per cent have less than 11 employees. There are also around 5,000 registered sole traders working on their own account, or in limited partnership companies (Almega Serviceentreprenörerna 2014; cf. EFCI/FENI 2014). In the domestic service sector, approximately 90 per cent of the enterprises have less than 10 employees and the majority are self-employed persons (Almega Home Service Industry 2014).6 The large share of small enterprises in cleaning is partly due to the use of sub-contractors (ibid.). Moreover, tax reforms and other political incentives have been introduced to promote entrepreneurship, particularly among women and immigrants.

4 In 2013, 48.6 per cent of the members in the Almega Service Contractors participated to various extent in public procurements (Almega Serviceentreprenörerna 2014). 5 Interview with industrial political analyst at the Almega Service Contractors, Stockholm (March 2014). 6 Interview with negotiator at the Almega Home Service Industry, Stockholm (October 2014)

8 East-West Labour Migration and the Swedish Cleaning Industry

A reform on tax deduction for domestic services that came into force in 2007 – the so-called RUT reform (SFS 2007:346) – has contributed to a significant increase in the number of small enterprises, especially among foreign born women. In 2010, they represented 63 per cent of the newly established enterprises (Välfärd No. 1, 2012). In addition, the Act on System of Choice in the Public Sector, LOV (SFS 2008:962) that was introduced in January 2009 in order to promote the establishment of customer choice models in health care and home-based care and services for elderly and disabled people in municipalities, helps to explain the increase of small private cleaning and home service companies in recent years (Tillväxtverket 2012).7 A new F-tax (business tax) reform issued in 2009 has also made it easier to get in the position of a F-tax cer- tificate and set up an enterprise (Government bill 2008/09 No. 1).Among other things, the new regulations allow self-employed persons to have only one client, even if this client is their former employer. Thus, it it has become easier for employers to contract out work to former employees in order to downsize the organisational structure, and thus transfer costs and production risks to the workers. This practice undermines the notion of employee and blurs the border between regular employment and genuine self-employment. This notion defines the personal scope of labour law, and thus it is of central importance for workers’ rights and security in working life (Rönnmar 2004). Consequently, there is also an increasing risk of ‘false self-employment’, that is employment disguised as self-employment in order to circumvent collective agreements, labour laws, pay-roll tax and other employer duties implied in a regular contract of employment (Thörnquist 2011a; 2013a). Over the past ten years, the number of companies and employees in cleaning have increased, especially in the regions of Stockholm, Göteborg and Malmö. Moreover, the effects of the recent economic crisis have been modest in this industry as a whole (Almega Serviceentreprenörerna 2014; Välfärd No. 1, 2012). It should be emphasised though that there is no shortage of labour in the cleaning industry. As in other parts of the low-skilled service sector, such as in the hotel and restaurant industry, there is instead a shortage of jobs, and job-competition is high, albeit with regional variations (LO 2013; Hotell- och restaurangfacket 2012). This is a most important factor to take into account when discussing east-west labour migration to this industry. As was mentioned initially, almost half of the employees in the cleaning industry – more exactly 48 per cent – are foreign born workers, most of whom have come from other EU countries (outside the Nordic region) and Asia, but also from Africa and South America. The rate has increased significantly in the 2000s, and it is expected to increase further in the future. For young people, cleaning jobs have often been their first step into the labour market. As diagram 1 illustrates, native born people dominate among workers aged 15–25 years. However, while young workers often move on to

7 Since 1 January 2010, LOV is mandatory for the primary care in county councils and regions. For the municipalities, LOV is still a voluntary law.

9 Annette Thörnquist other jobs, foreign born workers remain in this sector to a considerable higher degree (cf. Urban 2013). Thus, they are highly over-represented among workers aged 26–45 years. The large share of Swedish workers in older age groups is partly due to demographic factors; the average age is lower among foreign born people. Moreover, in the public sector, the overall share of native born workers, as well as older workers, is higher.

Diagram 1: Native born and foreign born gainfully employed persons in the Swedish cleaning industry, by ages, 2012

Foreign born

Native born

Source: Välfärd, No. 1, 2012, Statistics in Sweden, SCB.

The overall share of women in the cleaning industry is 78 per cent. In the public sector, the share is no less than 89 per cent, compared to 59 per cent in private cleaning companies.8 The highest share of male workers, and workers with an immigrant background, can be found in the metropolitan areas (Almega Serviceentreprenörerna 2014). Traditionally, undeclared work and other forms of economic criminality have been common in the cleaning industry. According to the Swedish Tax Agency, this means around 2 billion SEK in unpaid taxes annually (Dagens Arena, 7 June 2011). Regular employment contracts are often used to disguise informal work, which illustrates the strong links between the informal and formal economy. For example, work has been passed over to undocumented migrants (Elmhorn 2008; Khosravi 2009; Björklund Larsen 2010; SOU 2010:63). Sub-contractors are also used to camouflage irregular- ities, such as tax evasion, informal work and various forms of labour abuse (RITA 2008). The Almega Service Contractors has introduced a system for certifying clean-

8 The figures refer to 2013.

10 East-West Labour Migration and the Swedish Cleaning Industry ing companies, among other things to promote fair competition. Today, over 150 of almost 1,000 member companies are certified, representing over 50 per cent of the employees. Thus, most certified companies are large actors. There is also an established cooperation between the labour market parties and the Tax Agency in order to curb economic criminality. Even though the problem has decreased, fraudulent actors still cause serious problems in the cleaning industry. Since cleaning is a strongly diversified industry, and cleaning is carried out in all sectors of society, including private homes, it is difficult for the organised social partners and the authorities to control this business.9 Workers in the cleaning industry are organised in several trade unions, and they negotiate with various employer organisations. This organisational structure compli- cates labour relations. Among other things, the wage level may vary between different unions, and sometimes the workers’ organisational residence can be unclear (Rapp 2006).10 Cleaners in contract companies are organised in the Swedish Building Main- tenance Workers’ Union while the Swedish Municipal Workers’ Union organises those who are employed in the municipal sector, including workers in the privatised parts of this sector, as well as cleaners in private homes. Those who are employed in the governmental sector are organised in the Union of Service and Communication Employees (Facket för Service och Kommunikation, Seko). The Hotel and Restaurant Workers’ Union (Hotell- och Restaurangfacket) organises hotel cleaners. In addition, cleaners in construction sites are generally organised in the Swedish Building Workers’ Union (Svenska Byggnadsarbetareförbundet, Byggnads). The main employer organisations in the private sector are the Almega Service Contractors, the Almega Home Service In- dustry and the Visita (in the hospitality industry). Since cleaning has been increasingly outsourced over the past decades, most trade unions negotiate with the Almega Ser- vice Contractors. The Swedish Association of Local Authorities and Regions, SALAR (Sveriges Kommuner och Landsting, SKL), represents the employers in municipalities and county councils, while all governmental bodies are organised in the National Agen- cy for Government Employers (Sveriges Arbetsgivarförbund). In contrast to the private sector, there are no industry or branch associations in the public sector.

9 Interview with industrial polical analyst at the Almega Service Contractors, Stockholm (March 2014). 10 Interview with union official at Swedish the Building MaintenanceWorkers’ Union, Region Tvärs, Västerås (February 2014).

11 Annette Thörnquist

III: Trends in East-West labour migration What then do we know about the general trends in labour migration from Poland and the Baltic States to Sweden since 2004? As was mentioned above, the official statistics include very little information about labour migration to Sweden for shorter spells than a year. Let us therefore take a look at the statistics over labour migrants from Poland and the Baltic states who have been working, or expressed an intention to work in Sweden for 12 months or more, and who thus are included in the population register as ‘immigrants’. As we can see in figure 2 and 3, the large majority of these immigrants are men coming from Poland, followed by women from Poland and men from Lithu- ania, while there are comparatively few registered labour immigrants from Latvia and Estonia. It should be mentioned though that all labour migrants who have been working in Sweden for more than a year are most likely not registered.11

Diagram 2: Annual immigration of Polish citizens to Sweden for labour market reasons, 2004–2013, (employees)

number

year

men women

Source: SCB/Statistics in Sweden (2014 b). Source : Statistiska centralbyrån (SCB)

11 Information by telephone from Statistics in Sweden, SCB, Department for Population Statistics (August 2014).

12 East-West Labour Migration and the Swedish Cleaning Industry

Diagram 3: Annual immigration of Estonian, Latvian and Lithuanian citizens to Swe- den for labour market reasons, 2004–2013, (employees)

number

year

Estonia, men Latvia, men Latvia, women Lithuania, men Lithuania, women

Estonia, women Source: SCB/Statistics in Sweden (2014b). Source : Statistiska centralbyrån (SCB)

As diagram 2 and 3 also illustrate, there are also some interesting differences between the countries regarding the patterns of labour migration. The number of registered Po lish labour migrants peaked in 2006, but decreased rapidly again in the following years. This trend continued during the economic crisis that started in 2008, while the number of labour migrants from the Baltic states increased, and even reached it highest level in the years after the crisis. The exceptions were women from Lithuania and Estonia. The diverging trends in labour migration from Poland and the Baltic states has mainly be explained by differences in the unemployment rates in these countries (Rose Tronstad and Andersson Joona 2012). In Poland, almost 20 per cent of the workforce was unemployed in 2004, while the Baltic States were hit harder by the financial crises (figure 4). The harsh austerity policy launched in the wake of the crisis meant that the share of the population at risk of poverty increased dramatically (Sommers and Woolfson 2014). Moreover, the recession in Sweden was comparatively short. The Swedish GDP declined by minus 5 per cent in 2009, but increased rapidly again by 6.6 per cent in 2010 and 3.9 per cent in 2011. The overall unemployment rate increased from 6 per cent before the crisis to around 9 per cent in 2009. In the following years, it has remained just below 8 per cent. The crisis in Sweden had probably little impact on the inflow of labour immigrants from the newer EU countries (Wadensjö 2012). Over the past few years, when the economic situation has improved in the former Eastern Bloc, the trend in labour migration to Sweden has turned downwards again.

13 Annette Thörnquist

Diagram 4: Unemployment rate in Poland, the Baltic states and Sweden, 2004–2013 (per cent of the workforce)

25.0

20.0

15.0

10.0

5.0

0.0 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 E stonia L atvia Lithuania Poland Sweden

Source: EUROSTAT, 2014.

Except for the wage gap within the enlarged EU, the main driving forces behind east- west labour migration from Poland and the Baltic states since 2004 have no doubt been high unemployment and deteriorating living conditions in wake of the financial crisis, and the subsequent austerity policy pursued by the governments, especially in the Baltic states (Kahanec and Zimmermann 2008, 2009; Galgóczi et al. 2012; Ollus et al. 2013; Sommers and Woolfson 2014). It should be emphasised though that labour migration from these countries includes a broad population, comprising also people who are not at risk of poverty, but who emigrate to make a better fortune in the west. In addition, workers are actively recruited, for example professionals in the health care sector (Andersson and Hammarstedt 2011; Andersson Joona and Wadensjö 2011; Wadensjö 2012; Sippola 2014). Labour migration from Poland and the Baltic states to Sweden should also be related to the migration flows from these countries to other EU/EEA member states. The main destination countries for migrant workers from Poland, Latvia and Lithuania have been Germany, the United Kingdom, Ireland, Spain, Denmark and Norway, while the main part of the Estonian labour migrants work in Finland (Asset Skills Research Group 2007; Platovona and Urso 2012; Quirico 2012). Despite the absence of transitional restrictions, the inflow of labour migrants in Sweden from Poland and the Baltic states, who have stayed for at least a year, has been comparatively modest regarded in a broader European context. On the other hand, many persons from the newer EU countries, especially people from Poland, Hungary and Estonia, were already living in Sweden before the enlargement of the EU in 2004 (Pedersen and Pytlikova 2008; Rose Tronstad and Andersson Joona 2013; Zelano et al. 2014).

14 East-West Labour Migration and the Swedish Cleaning Industry

The male dominance among labour migrants from Poland and Lithuania illustrated in diagram 2 and 3, may indicate that many labour migrants have worked in typically male jobs, such as building and construction jobs. Previous research have shown labour migrants from the newer EU countries who have worked in Sweden for at least 12 months are slightly over-represented in the construction sector. As a whole, however, they work in different parts of the labour market, and the share of migrant workers from these countries in low-paid jobs is lower in Sweden than in many other older EU countries (Wadensjö 2012: Wóicicka 2012; Gerdes and Wadensjö 2009; 2013). When it comes to labour migrants from the newer EU countries, who has been working in Sweden for less than a year, the series over registered residence rights from other EU/EEA countries may give some information about the general trends after 2004.12

Diagram 5: Number of registered residence permits/residence rights, employed and self-employed workers from EU/EES region (outside Nordic countries)*

Source: Ekonomifakta (2014).

* Until 2006, residence permits

As diagram 5 shows, the number of registered residence permits for labour market reasons for persons from the EU/EEA region (outside the Nordic countries) did not change much until the turn of the century 2000, when it became more common that workers from the candidate countries in Central and Eastern Europe applied for work permits to work temporarily in Sweden as employees, or they came to Sweden as self-employed workers (Migrationsverket 2015). The subsequent development reflects the increasing inflow of labour migrants from newer EU countries, especially from Poland and Lithuania. The number of registered rights of residence decreased during the crisis, but increased again after 2009 (Ekonomifakta 2014). However, it deserves to be said once again that many workers who have worked in Sweden for three months are not included in this register.13

12 Interview with official at the Swedish Migration Board, Norrköping (February 2014). 13 Information from statistical analyst at the Swedish Migration Board, Norrköping (March 2014).

15 Annette Thörnquist

IV: Ways of approaching the Swedish labour market Many temporary migrant workers from Poland and the Baltic States have approached the Swedish labour market on their own account, while other workers have been recruited or mediated in one way or another. According to the interviewed representatives for the trade unions, labour migrants from the former Eastern Bloc, who are working for shorter spells in the Swedish cleaning industry, are generally not organised, and the local unions branches mainly get in touch with them when they contact the unions on their own initiative, or if they work in companies where there are organised workers. The informants estimate that temporary labour migrants from Poland and the Baltic states are engaged mainly in small cleaning companies, or in small firms with a focus on domestic services. In many cases, the employers are immigrants, who recruit workers from their former home countries or other immigrants living in Sweden (see also Thörnquist 2013b; cf. Sorainen 2010).14 Workers in domestic cleaning are sometimes also employed by individuals in private households, or they are self-employed (Cal- leman 2011). According to a Swedish TV documentary on labour migration, in which several women from the newer EU countries were interviewed, the financial crisis, and the subsequent tough austerity policy, have forced many women to leave their children – the so-called EU-orphans – to work temporarily in private households in Sweden, especially in the regions of Stockholm, Gothenburg and Malmö, where the demand for domestic cleaning have increased significally the wake of the RUT reform (SVT, 1 September 2010; cf. Scmulyar Green 2013; Scmulyar Green and Spear 2014).

Posted workers and temporary agency workers Let us then take a closer look at the the recruitment of temporary labour from Poland and the Baltic states, and to what extent these practices may concern the Swedish cleaning industry. According to the Swedish Work Environment Authority’s register 38,251 posted workers worked in Sweden in 2014. Most of these workers were posted from Germany, Poland and Lithuania (Arbetsmiljöverket 2015). Workers who were posted from Poland and Lithuania were mainly engaged in the building and construction industry – the largest industry for posting as a whole (36.6 per cent) (ibid; cf. Dølvik 2008; Cremers 2011).15 As far as the register shows, there were no companies engaged in cleaning specifically. However, work in other sectors may also include cleaning jobs, for example in the building and construction industry. In addition, over 9 per cent of the posted workers who were registered in 2014 worked in unspecified

14 Interview with the second Vice President of the Swedish Building Maintenance Workers’ Union, Stockholm (December 2013); interview with union official and regional safety representative at Swedish the Building Maintenance Workers’ Union, Region Tvärs, Västerås (February 2014); interview with union official at the Swedish Municipality Workers’ Union, Stockholm (February 2014). 15 The sending country is not necessarily the same as the workers home country.

16 East-West Labour Migration and the Swedish Cleaning Industry sectors. Most of them were from Thailand, Poland and Germany (Arbetsmiljöverket 2015). As reported by the interviewed representatives for the labour market parties, however, there are few posted workers in the Swedish cleaning industry. For example, there are no posted workers within the Swedish Building Maintenance Workers’ Union’s bargaining areas for cleaning and janitorial services (see also Jonsson and Larson 2013). The main reason mentioned in the interviews was that cleaning companies generally are contracted for comparatively long periods, such as 2–4 years or more, and the contracts can be prolonged as well.16 Unlike in many other EU countries, office cleaning in Sweden is also carried out mainly during day-time, at least in the formal sector of the labour market. This means that there is little supply of night jobs, which generally are less attractive jobs for native workers.17 When it comes to domestic cleaning services, a common marketing argument is to provide continuity and consistency in service personnel. Moreover, many customers prefer workers who can communicate in Swedish.18 Thus, posted workers do not fit very well into this sector either. The mode of organising production in office cleaning may also help to explain the seemingly limited use of temporary work agencies in this industry as well. When it comes to the recruitment of labour from the newer EU states to the Swedish labour market in general, previous studies have shown that temporary employment agencies have played a minor role, compared to what is known from Norway and Denmark for example (Peterson 2012; 2013a; Andersson Joona et al. 2013). However, Swedish as well as foreign temporary work agencies have to some extent mediated labour from the newer EU countries to Swedish cleaning companies and private customers ever since 2004.19 One of these entrepreneurs, a native born Swede, who ran a temporary work agency based in Tallinn, provided cleaners to a price of 75 SEK per hour in 2005. His way of arguing was characteristic for many agencies at the time.

I do the Estonian workers a big favour by paying them 40 to 50 SEK per hour. This is twice, or even four times, more than they can expect to get at home. (Svenska Dagbladet, 29 April 2005)20

The arguments used to justify lower wages for migrant workers are mainly the same today. Temporary work agencies mediating workers from the newer EU countries in the

16 Interview with industrial political analyst at the Almega Service Contractors, Stockholm (February 2014); interview with union official and regional safety representative at the Swedish the Building Maintenance Workers’ Union, Region Tvärs, Västerås (February 2014); interview with union officials at the Swedish Mu- nicipality Workers’ Union, Stockholm (February 2014); interview with union official at the Seko, Stockholm (March 2014). 17 Interview with industrial political analyst at the Almega Service Contractors, Stockholm (February 2014). 18 Interview with negotiator at the Almega Home Service Industry, Stockholm (October 2014). 19 Interview with industrial polical analyst at the Almega Service Contractors, Stockholm (February 2014). 20 The author’s translation.

17 Annette Thörnquist cleaning industry operate generally in other industries as well, especially in the building and construction sector. For example, a much-disputed temporary work agency has provided Latvian labour in Sweden since 2004 to prices far below the market prices in construction, cleaning and a number of other industries. Over the years, the agency has had subsidiaries in Sweden as well as in Latvia, among them a company in Sweden for domestic services, including cleaning, washing, gardening and minor renovations. In 2012, the Swedish Tax Agency made a revision of the temporary work agency’s tax payments during the previous two years, which among other things showed that the taxes and social insurance contributions were paid in Latvia. Another strategy used to minimise the tax costs in Sweden was to pay only a small share of the workers’ remuneration in wages, based on low minimum wages in Latvia (around 15 SEK). Instead, the main part of the remuneration was tax-free allowances. The Tax Agency considered the company to be permanently established in Sweden, which meant that the owners were required to pay over 6 millions SEK in unpaid payroll taxes and preliminary income tax (Aftonbladet, 4 June 2009; Stoppafusket.nu 19 April, 2013). Another case that illustrates the involvement of Baltic temporary employment agencies in the Swedish cleaning industry is a company that started in 2005 with a focus on domestic cleaning services. In 2010, almost 100 Latvian cleaners (mainly women) were working in this company. The owner hired them from a temporary employment agency in Latvia. In this case too, the main part of the workers’ payment consisted of tax-free allowances. According to reports in the press, which among other things were based an analysis of the individual employment contract the workers had signed, the employment conditions were extremely insecure. For example, the workers lacked elementary social benefits, and they ran the risk of being ‘fined’ if they did not follow the regulations prescribed in the contract. Moreover, tight schedules and no organised breaks in the work meant high time pressure and stress. In 2011, the Latvian government introduced stricter tax regulations on allowances for people working abroad in order to avoid abuse of the system. According to an interviewed worker in this company, the workers were now expected to enter into self-employment to circumvent the new regulations. The owner of the Swedish cleaning company denied most of these statements, but admitted at the same time that the triangular employment relationship included problems for the company a well. Thus, the owner decided to employ the workers directly instead of hiring them (Expressen, 10 February 2011). The company is still one of the larger actors in domestic cleaning services in Sweden with almost 140 employees. Today, one of the main marketing arguments is that all workers are employed in Sweden. Temporary employment agencies based in the former Eastern Bloc provide labour directly to Swedish companies as well. They generally work in the quiet, approaching Swedish companies by e-mail and offering them labour to low prices. The main problem for the organised labour market parties is that less-serious contract companies may

18 East-West Labour Migration and the Swedish Cleaning Industry recruit labour mediated by these agencies, or by Swedish based agencies, to press down prices and thus win contracts.21 In January 2013, Sweden implemented the Directive 2008/104/EC on Temporary Agency Work. The Directive includes a statement on equal treatment, saying that ‘the basic working and employment conditions applicable to temporary agency workers should be at least those which would apply to such workers if they were recruited by the user undertaking to occupy the same job’ (European Commission, 2008: 2, 14§). However, new regulations often generate new loopholes. According to the Swedish Building Workers Union, an unintended effect of the Directive has been an increasing risk of false self-employment. The companies prefer to engage self-employed subcontractors instead of temporary agency workers in order to evade the regulations on equal treatment. Temporary work agencies have sometimes also started providing self-employed workers for the same reason (Byggnadsarbetaren, November 2012).22

Self-employed workers There are an unknown number of self-employed workers from Poland and the Bal- tic states working temporarily in the Swedish cleaning industry. In general very few self-employed workers from the newer EU countries are registered in Sweden (SCB 2014b).23 As was mentioned previously, many self-employed workers stay for shorter spells to avoid being registered under the Swedish tax regime. As was mentioned above, people from the former Eastern Bloc worked to some extent as self-employed in Sweden even before the enlargement of the EU in 2004. By the turn of the century 2000, several small recruitment agencies providing self-employed workers from the candidate countries in Central and Eastern Europe appeared in Swedish labour market. At the time, self-employment was sometimes used as a way of circumventing the requirement on work permits for workers from countries outside the EU, including the candidate countries. In these cases too, the agencies operated mainly in the construction sector, but also in cleaning, manufacturing, agriculture and forestry. Thus, recruitment agencies saw lucrative business opportunities in providing workers who were formally self-employed, but who according to the trade unions and the Swedish Tax Agency often worked as employees. In other words, they were considered to be falsely self-employed workers. Most of these agencies disappeared from the labour market after 2004 and 2007, when people from the newer EU countries could work as employees in Sweden without any requirement on work permits. In recent years, the incidences of false self-employment have increased again in east-west labour

21 Interview with industrial polical analyst at the Almega Service Contractors, Stockholm (February 2014; Au- gust 2014). 22 Interview with union official at the Swedish Building Workers’ Union, Stockholm (January 2015). 23 Interview with migration political analyst at the LO, Stockholm (February 2014).

19 Annette Thörnquist migration, mainly in the construction and road haulage industries (Thörnquist 2011a; 2013a). According to the interviews with the trade unions, there have been few established cases of false self-employment in the Swedish cleaning industry in recent decades (cf. Calleman 2011). On the other hand, the trade unions fear that many foreign as well as Swedish self-employed workers in cleaning may have an unclear employment status, since the law allows self-employed persons to have only one client.24 For example, in some municipalities, private companies in home-based elderly care and home services use self-employed workers instead of employees to lower labour costs. In general, it is difficult for the trade unions and the authorities to control the true employment status of self-employed migrant workers who are working temporarily in Sweden, and who are not registered in Sweden (Thörnquist 2013b). The trade unions estimate that many self-employed women from Poland and the Baltic states are working in Sweden as sole traders or in limited partnership companies.25 A quick review of small entrepreneurs who are advertising their cleaning services on the web confirms that many women from the newer EU countries, especially Polish women, are providing domestic cleaning, but sometimes also cleaning in offices and hotels. An interesting phenomenon is that the attribute ‘Polish’ seems to have become something of a brand name for these enterprises. The entrepreneurs often relate to the stereotype of the hard-working Polish worker to prove that they are careful and assiduous workers. Friberg (2012) has noticed a similar trend among employed Polish construction workers in Norway. This is an example from a small limited partnership company working in the Swedish cleaning sector:

We are a team of female Polish cleaners. We work hard, and we are very experienced in all kind of office cleaning and domestic cleaning services.We work very effectively and carefully, and we do indeed know how to take responsibility for our work.26

The workers’ strategy to highlight their Polish identity can also be related to the tough competition in the over-supplied cleaning industry. Many small cleaning and home service companies try to gain competitive advantages by marketing their specific niches. This is common, especially among service providers within the customer choice system (LOV), where there is no price competition (Thörnquist 2013b; 2014). When it comes to Latvian enterprises advertising cleaning services on the web, this strategy seems to be rare though. On the other hand, it is not uncommon that temporary work agencies,

24 Interview with the second Vice President of the Swedish Building Maintenance Workers’ Union, Stockholm (December 2013; August 2014); interview with union official and regional safety representative at the Swedish Building Maintenance Workers’ Union, Region Tvärs, Västerås (February 2014); interview with expert at the LO, Stockholm (February 2014). 25 Interview with migration political analyst at the LO, Stockholm (February 2014). 26 The author’s translation.

20 East-West Labour Migration and the Swedish Cleaning Industry and other labour market intermediaries, use the attribute ‘Latvian labour’ synonymous with ‘cheap labour’ implicitly in their advertisement. Hence, for small Latvian firms and Latvian self-employed workers, it can be complicated to use nationality as a cultur- al capital and market argument.

Recruitment of labour from ‘third countries’ When discussing temporary east-west labour migration within the EU and how it may concern the Swedish cleaning industry, it is also important to take into account the recruitment of workers from ‘third countries’, which refers to countries outside the EU/ EEA region. According to an OECD report (2011), Sweden has one of the most liberal labour migration policies within the organisation. The revision of the Swedish regulations on labour migration in 2008 meant that the employers can recruit workers from third countries even if there is no labour shortage at hand (LO 2012; 2013). Since all applications and approved work permits are registered at the Swedish Migration Board, we know much more about the inflow of labour migrants to Sweden from these countries than from the EU member states, outside the Nordic region. As the table below shows, the large majority of the third country workers have been recruited to the ‘green sector’ (agriculture, horticulture, forestry and fishery) as seasonal workers, most of whom have come from Asia, Africa and South America. The second largest sector is the information and technology industry with high-skilled workers recruited mainly from India. Cleaning is among the 6 most common industries. However, other sectors, such as assistants in kitchens and restaurants, may also include cleaning jobs (cf. Peterson 2013b). Most of the workers who have been recruited to the cleaning industry have come from Mongolia, Iraq, Bolivia and the Republic of Uzbeki- stan (LO 2013).

Table 1: Number of approved work permits for ‘third country’ workers 2009–2014 in the 6 most common industries (first time applications on a whole year basis)

Sector 2009 2010 2011 2012 2013 2014 Workers in agriculture, gardening, forestry and fishery 7,200 4,508 2,821 5,708 5,915 2,885 Computer experts 27 2,202 2,208 2,795 3,259 3,477 2,525 Personnel in restaurants, canteens and large-scale catering establishments 769 1,049 1,323 861 830 666 Engineers and technicians 541 525 630 558 415 424 Assistants in kitchen and restaurant 257 548 796 570 470 364 Cleaners 295 487 798 553 397 208

Source: Migrationsverket/Swedish Migration Board (2015).27

27 In 2014, 903 IT operators were registered separately due to a modification in the Swedish standard classifica- tion of occupations.

21 Annette Thörnquist

In cleaning, as well as in the restaurant sector, the work permits have generally been issued for 24 months. However, all permits can be prolonged for another 24 months, and the share of prolonged permits in cleaning increased rapidly from 2010 (LO 2013). The possibility to recruit labour from third countries for a comparatively long time-period has no doubt been an attractive option for the employers, especially in small companies. According to the LO unions, however, the main reason for recruiting workers to low-skilled service jobs has been to gain a cheap labour. As there has been a shortage of jobs in these sectors in recent years, the unions claim that the employers have preferred to replace Swedish workers with migrant workers (LO 2013; Hotell- och restaurangfacket 2012).28 The determining factor for immigrant competition is whether labour migrants are used as a complement or a substitute to native workers. Thus, the unions have pressed for stricter control of the system to prevent job competition and wage dumping (ibid.; Ds 2013:57). In addition, the OECD report mentioned above, recommended that Sweden should ‘reinforce mechanisms for ensuring employer compliance, in particular for occupations not generally considered to be in shortage’ (OECD 2011: 16). The new regulations introduced by the Migration Board in January 2012 have meant considerably stricter control of the reasons for recruiting workers. After reminders from the European Commission, Sweden also ratified the EU Directive 2009/52/EG on ‘minimum standards on sanctions and measures against employers of illegally staying third-country nationals’ in 2013 (European Commission 2009). This means that employers are obliged to notify the Swedish Tax Agency when employing third nationals, and control that the workers have valid resident permits and work permits. The Directive also apply to individuals in private households who employ workers directly for domestic services. As we can see in the table above, the number of approved work permits decreased more or less in most LO sectors in 2012 and 2013, especially in the cleaning industry and in restaurants, canteens and large-scale catering establishments. The exception was the green sector. According to the employer organisation Svenska Skogs- och Lantar- betsgivareförbundet, SLA, labour migrants are essential for forestry, agriculture and horticulture. The main argument is that it is difficult to recruit Swedish workers, despite high unemployment, especially among young people (in 2014, over 20 per cent) (SLA 2014).29 In 2014, the number of approved work permits in the LO industries continued to decrease, now including a dramatic drop in the green sector as well. This was partly due to the limited supply of berries. Since 2011, there are also special restrictions for the recruitment of berry pickers to curb abuse of work permits and human trafficking.30 According to the LO, the stricter regulations introduced in recent years to control the

28 Interview with the second Vice President of the Swedish Building Maintenance Workers’ Union, Stockholm (December 2013); interview with union official at the Swedish MunicipalityWorkers’ Union, Stockholm (January 2014); interview with the head of the Swedish Hotel and Restaurant Workers’ Union (March 2014). 29 Interview with the President of the SLA, Stockholm (March 2013). 30 Information by e-mail from the Swedish Migration Board (August 2013).

22 East-West Labour Migration and the Swedish Cleaning Industry recruitment of third country nationals, have ‘almost abolished’ the liberal reform from 2008 within the LO sector.31 On the other hand, undocumented migrants are still used as cheap labour in several sectors, especially in cleaning and in the hotel and restaurant industry (Elmhorn 2008; Björklund Larsen 2010; SOU 2010:63). Work in the shadowy parts of the labour market generally means extremely poor employment and working conditions. The workers are often engaged in long supply chains, where their identities easily can be concealed (RITA 2008; cf. Refslund 2014). In office cleaning, this practice has mainly been observed in sites where cleaning has been carried out at night-time and thus is less visible.32 As in many other western countries, undocumented migrants are also working in domestic cleaning (e.g. Gavanas 2010; Gavanas and Calleman 2013). According to Swedish Building Maintenance Workers’ Union, approximately 4,000 undocumented migrants were working in the Swedish cleaning industry in 2008, and around 50 per cent of them could be found in the Stockholm area (RITA 2008). The union estimates that the number has not declined in recent years.33 Consequently, ‘immigrant competition’ is tough in these areas for many reasons. This also means that those who have the weakest position in the labour market are even more at risk of being abused.

31 Statement made by LO’s migration political analyst at the conference ‘Dealing with low-wage labour migration’, Arena Idé/Friedrich Ebert Stiftung, Stockholm (11 November, 2014). 32 Interview with the second Vice President of the Swedish Building Maintenance Workers’ Union, Stockholm (December 2013); interview with union official at the Swedish MunicipalityWorkers’ Union, Stockholm (January 2014). 33 Interview with the second Vice President and work environment expert at the Swedish Building Maintenance Workers’ Union, Stockholm (December 2013).

23 Annette Thörnquist

V: East-West labour migration and the Swedish labour market As was mentioned in the introduction, previous studies have estimated that the overall inflow of labour immigrants from the newer EU countries to low-wage sectors in Swe- den has been low, compared to what is known from many other countries in Western and Northern Europe (e.g. Wadensjö 2012; Ruhs 2012; Rose Tronstad and Andersson Joona 2013). The interviews that were carried out in this study on temporary labour migration targeting the Swedish cleaning industry indicated the same trend. How then can this development be explained? Are there reasons other than those we know from high-level economic studies, such as the impact of the strong Swedish labour market regulations, strong trade unions and the tricky Swedish language? To begin with, something should be said about the conception of the Swedish labour market regime though. In the academic debate on labour migration from the newer EU states to Sweden, the impact of the strong Swedish labour market regulations has sometimes been a bit overestimated. Ruhs (2012: 120) provides a good example.

The requirement on equal rights in Sweden’s highly regulated labour market effectively meant that, from the employers’ view, migrant workers were as expensive as Swedish workers. This explains, to a considerable degree, why Sweden has experienced relatively low levels of labour immigration from EU8 nationals (just over 50,000 EU workers during 2005–2011). The insistence of equal labour rights in practice made Sweden’s policies towards admitting and employing EU8 workers much more restrictive than suggested by its formal decision to grant EU8 nationals immediate access to the labour market.

The aim of Swedish regulations laid down in labour laws and collective agreements is to safeguard equal rights for all workers in the labour market and thus forestall competition between workers. However, this does not necessarily mean that Swedish employers cannot use temporary labour from low-wage countries to reduce labour costs. The Swedish labour market regime is no guarantee against exploitation of workers in companies where neither workers nor employers are organised. Temporary labour migrants in low-skill sectors are among those who have the weakest position in the labour market, and who thus are most at risk of being abused. Swedish regulations as well as EU regulations on taxes and social security contributions for temporary work in another member state, include loopholes as well. As was mentioned earlier in this text, the tax-free time-period of 6 month can easily be ‘extended’ by constant circulation of workers between Sweden and the sending countries. Moreover, ‘shopping’ between different national systems for social security contributions within the EU has become more and more common to reduce labour costs, which the temporary employment agencies described above illustrate (see also Jonsson 2013). Another common strategy used by foreign subcontractors is to conclude Swedish collective agreements for their employees, but without following these agreements. In fact, the workers have to work substantial ‘off the record’ hours. In other words, the employers use ‘false’ or ‘double’ collective agreements in order to conceal the fact that they are practising wage-dump-

24 East-West Labour Migration and the Swedish Cleaning Industry ing to gain competitive advantages (Thörnquist 2013a). If the purpose of engaging subcontractors is to lower production costs, the main contractors may well have an interest in these practices as well, and there is no general obligation for main entrepreneurs to take formal responsibility for wages and working conditions among subcontractors.34 Over the past ten years, union power has weakened as well. The Laval conflict in 2004/2005 illustrated the fragility of the Swedish bargaining system within the framework of EU legislation. The European Court of Justice’s verdict limited the trade unions’ possibilities to take industrial action against foreign employers with posted workers in Sweden (Woolfson et al. 2010). In addition, the former centre-right alliance government introduced legislation in 2007 that changed the regulations on the trade union administered unemployment funds, which meant a significant increase of the individual fund members’ fees, at the same time as the possibility to deduct tax for such expenses was abolished. These regulations, which struck members in the LO unions in particular, resulted in a dramatic drop in trade union membership (Kjellberg 2011; 2013). Even though the coverage of collective agreements is still very high in Sweden regarded in an international context, there are also an increasing number of workers who are not covered, which will be illustrated further below (Kjellberg 2013; Arbetet, 17 October 2014). Hence, when discussing the impact of the ‘strong Swedish labour market regulations’ on trends and developments in labour migration from the newer EU states to Sweden, it is necessary to problematise this notion, and take into account the changes in the balance of power between labour and capital in the Swedish labour market in recent decades.

The buyers’ labour market In this study, it has been emphasised that east-west labour migration should be regarded in a broad social and organisational context. For example, the widespread practice of outsourcing cleaning and facility services, the length of the contracts and the limited supply of night jobs have been mentioned as important factors to consider. Further- more, third country workers have been recruited, and undocumented migrants are still common in this industry, especially in the Stockholm area. The presence of temporary labour migrants in an industry, should also be related to the supply of native labour. Since there are few job vacancies in the Swedish cleaning industry, and job competition is high, this labour market can be characterised as ‘the buyers’ market’. It has become difficult for young and unemployed people to get ordinary jobs in the low-skilled service sector, such as in cleaning and in hotel and restaurants (LO 2013; Hotell- och Restaurangfacket 2013). Moreover, union density is low in these industries. In 2013,

34 Since 1 April 2014, the Building Agreement in the construction sector includes a clause stating that the main entrepreneurs are obliged to control that their subcontractors pay wages and follow laws and regulations. This clause, which also applies to subcontracting chains, came into force 1 January 2015. However, the main entre- preneur is not economically responsible for the subcontractors.

25 Annette Thörnquist the overall union density in the Swedish Building Maintenance Workers Union was 48 per cent, and in the Hotel and Restaurant Workers’ Union 30 per cent (Kjellberg 2013). In general, however, union density among cleaners is lower. In a large district of the Swedish Building Maintenance Workers’ Union in Central Sweden, the interviewed trade union representative estimated that union density was as low as 20–30 per cent.35 In many small cleaning companies, especially in those with a focus on domestic services, the workers are neither organised nor covered by collective agreements. This problem has become more and more common in the privatised parts of the municipal welfare sector (Thörnquist 2013b; 2014). Today, over 50 per cent of the Swedish municipalities have introduced the Act on System of Choice, LOV, in home-based care and services for elderly and disabled people (SKL 2015). Since this system often is under-funded, and the providers have no impact on prices, they focus hard on reducing labour costs. In large cities, where the market for domestic services often is over-supplied, many small companies avoid collective agreements to press down labour costs and thus gain competitive advantage. In this case too, workers have sometimes signed individual employment contracts that include extremely insecure and poor working conditions. For example, employers have required the employees to use their own cars and bear all the costs of the transports between the customers. In some cases, the employees have even been forced to use their own cleaning equipments and cleanser (Thörnquist 2013b). In a regular contract of employment, the employer is obliged to cover all production costs. Thus, the employers transfer costs and risks to the employees. In other words, the workers have to pay for working. Even though the workers are employed, the employers treat them as self-employed whenever it suits. Consequent- ly, this practice too blurs the borders between employment and self-employment and undermines the notion of employee.

A segmented labour market Let us then return to the question on whether the existing structures of social segregation and segmentation in the cleaning industry may help to explain trends in labour migration. In this study, the Swedish cleaning industry has been characterised as a horizontally segregated labour market, where women and foreign born workers are highly over-represented. There are also structures of labour market segmentation in this industry. Immigrants and women generally work in basic office cleaning, and in domestic cleaning, while the more specialised and well-paid parts of facility and janitorial services are ‘sectors dominated by white Swedish men’.36 Moreover, there seems to be little cross-over capacity between these sectors.

35 Interview with union official at the Swedish Building MaintenanceWorkers’ Union, Region Tvärs, Västerås (February 2014). 36 Interview with the Second Vice President of the Swedish Building Maintenance Workers’ Union, Stockholm (August 2014).

26 East-West Labour Migration and the Swedish Cleaning Industry

According to the Swedish Municipal Workers’ Union, female cleaning workers have lower wages than male workers, and female workers and immigrants are more often employed on fixed-term and part-time contracts (Ölmebäck 2011; Kommunal 2012; cf. Wall 2014). Immigrant women can often be found among those who are employed in the least well-paid sectors of this industry as well (Abbasian and Bildt 2007; Gava- nas 2011; Gavanas and Darin Mattson 2013). On the other hand, female team-leaders and managing directors are far more common in the cleaning industry than in the industrial world as a whole (Almega Serviceentreprenörerna 2014). This is another aspect of horizontally segregated labour markets (Melkas and Anker 1997). With few job vacancies, and a significant risk of immigrant competition – especially in the metropolitan areas – it is less likely that labour migrants from the newer EU countries, who intend to work in Sweden for shorter spells, would approach this labour market to any great extent. The exception is the market for domestic cleaning services. However, this market tends to be highly over-supplied as well, at least in large cities (Thörnquist 2013b; 2014).

A reserve army of tax-financed cheap labour However, there is still another most important aspect of the Swedish labour market that should be taken into account when discussing recent trends in east-west labour migration in the low-skilled service sector. This aspect concerns the labour market policy pursued in Sweden over the past ten years. As was mentioned in the introduction, the main purpose of this policy, which also is a workfare policy, has been to increase the supply of labour. Many cleaning workers have been unemployed before they started working in this industry. In 2013, almost 54 per cent of the newly recruited workers in the Almega Service Contractors’ member companies were unemployed, compared to 44 per cent in 2011 and 30 per cent in 2009 and 2010 (Almega Serviceentreprenörerna 2014). In the Almega Home Service Industry, the share was no less than 75 per cent in 2013.37 This also means that a considerable number of the workers in cleaning have been recruited through the Swedish Public Employment Service (Arbetsförmedlingen), and many of these workers have been engaged in various forms of labour market projects. The Public Employment Service organises a vast variety of labour market schemes. In addition, the European Social Fund (ESF) is co-financing employment-related projects for social inclusion and increased labour supply.38 Since the unemployment rate is considerably higher for young people and foreign born people, they are often over-represented in labour market programmes as well. In addition, there are special schemes for immigrants. Thus, private as well as public employers who engage workers par-

37 Interview with negotiator at the Almega Home Service Industry, Stockholm (September 2014). 38 The Swedish ESF Council manages the use of the Social Fund and the Integration Fund in Sweden. The owners of the projects are private as well as public actors and NGOs.

27 Annette Thörnquist ticipating in labour market projects can expect to get considerable wage subventions from the state. Even the state owned company Samhall, which is something of a labour market project in itself, may use this possibility. The control of the wage subventions has generally been weak, and the system has often been over-used.39 In line with the labour market policy pursued in recent years, the main task of the Swedish Public Employment Service has been to fulfil its commission to increase the supply of labour (Thörnquist 2011b; 2013b). Hence, many employers use this opportunity to legally recruit cheap labour. The aim has often been to gain competitive advantage, for example in public procurement. At the same time, the purchasers main focus has long been on price rather than quality. According to a survey made within the Almega Service Contractors in 2013, 63 per cent of the member companies had participated in public procurements in which the price was as low as 200 SEK per hour. Moreover, almost 57 per cent of the companies had employees with some kind of wage subventions from the Public Employment Ser- vice (Almega Serviceentreprenörerna 2014). Many small companies in domestic cleaning, and many external providers of domestic services in municipalities – especially providers within the system of choice, LOV – systematically use this possibility to reduce labour costs as well. In general, the control of external providers has been weak, and the municipalities have often prioritised diversity over quality (Thörnquist 2013b; 2014). In fact, many cleaning companies have become dependent on wage subsidies to survive.

One of the worst examples we have seen was an employer who had 17 employees. Among these workers, no less than 15 had some kind of wage subsidies.40

Price pressure in public procurement, low remuneration rates and tight economic frames for providers within the system of customer choice, have contributed to the increasing use of wage subventions.41 As was mentioned earlier in this text, the aim of the supply-side labour market policy was to make workers less dependent on social subsidies and thus reduce social costs. In practice, however, this workfare policy seems also to have helped to transfer the problem of subsidy-dependence from the workers to the employers, not least those who have started small companies in the wake of the RUT and LOV reforms. Hence, the widespread use of wage subventions means further oversupply, increased price pressure and further market disturbances. This also means increasing job

39 According to a public investigation launched in 2013, the risk of incorrect disbursements to employers who employ workers in labour market programmes is mainly related to the complex regulations that govern the system of wage subventions (SOU 2014:16). 40 Interview with union representative and regional safety official at the Swedish Building MaintenanceWorkers’ Union, Region Tvärs, Västerås (February 2014) 41 Interview with union official at the Swedish Municipality Workers’ Union, Stockholm (January 2014); interview with industrial political analyst the Almega Home Service Industry, Stockholm (September 2014).

28 East-West Labour Migration and the Swedish Cleaning Industry competition in this segregated labour market, and thus potential risks for immigrant competition as well. The possibility to reduce labour costs and gain flexibility by using a tax-funded reserve army of cheap Swedish labour is a factor that most likely also has an impact on the Swedish employers’ interest in recruiting temporary labour migrants from the newer EU countries. Thus, in cleaning, low-wage competition is to a large extent generated within the Swedish labour market.

29 Annette Thörnquist

VI: Concluding remarks As we have seen in this study, the lack of reliable statistical data on temporary labour migration from other EU countries makes is difficult to estimate the presence of migrant workers in the Swedish cleaning industry, especially those who are staying for shorter spells. In previous research, the overall inflow of labour migrants from the newer EU member countries, who have stayed in Sweden for at least a year, has been regarded as comparatively low, particularly in low-wage sectors. As far as the interviews with labour market parties and authorities in this study have revealed, temporary labour migration from Poland and the Baltic states targeting the Swedish cleaning industry has been limited as well. In the academic debate, an important explanation behind this trend in east-west labour migration has been that the Swedish labour market is highly regulated and organised to guarantee equal rights for all workers, including migrant workers. Thus, from this point of view, the employers have little to gain from recruiting workers from low-wage countries in the former Eastern Bloc just to gain cheap labour (Wadensjö 2012; Ruhs 2012). In this study, however, it has been argued that we need to apply a critical perspective on this conception of the Swedish labour market regime, and take into account changes in the balance of power between labour and capital in recent decades. Even though the Swedish labour market is still strongly regulated, and has a considerably higher trade union density and collective bargaining coverage than most other EU member states, this labour market regime is no guarantee against exploitation of temporary migrant workers, for example workers who are recruited to companies where the labour market parties are not organised. Nor is it a guarantee against social dumping. In the wake of the Laval conflict, and its juridical aftermath, union power in Sweden has decreased, and it has become more difficult to prevent the undercutting of wages and labour standards. In addition, trade union density has declined within the LO since 2007. Among vulnerable groups of workers, such as young people and immigrants who are working temporarily in cleaning or in other low-skill jobs, union density is generally very low. Despite the introduction of national regulations as well as EU regulations aimed at preventing low-wage competition and abuse of foreign labour, it has often been difficult for unions and authorities to control the implementation and observance of these regulations in the companies. In addition, the regulations on taxes and social fees for employers and workers who are engaged temporarily in Sweden include loopholes that have been frequently used by foreign subcontractors and labour market intermediaries, but also by migrant workers themselves in order make a living. Thus, many temporary migrant workers and posted workers fall outside the system of social protection in Sweden, as well as in their home countries. If the main reason for using foreign subcontractors is to lower production costs, Swedish main contractors may well have an interest in these practices as well. Unless there is an agreement between the labour

30 East-West Labour Migration and the Swedish Cleaning Industry market parties in an industry, such as in the construction sector, there is no obligation for main contractors to control wages and working conditions in subcontracting chains. This study also underlined that when discussing trends and developments in labour migration to individual industries and sectors in a destination country, it is important to take into account how production and work are organised. Moreover, it is important to problematise the social structure of the labour market, as well as the impact of the labour market policies pursued by the governments. This broad social approach indicates that we need more case studies of countries and branches to understand trends and developments in east-west labour migration, as well as the effects this migration may have on labour markets and industrial relations in the destination countries. In previous research, Eldring and Schulten (2012) and Refslund (2014), for example, have emphasised this aspect as well. As most case studies have concerned the construction industry, we also need more studies of low-wage sectors, such as cleaning and agriculture (Refslund 2014). Moreover, the present study showed that the common practice of outsourcing office cleaning to companies contracted for 2–4 years or more, helps to explain why employers may have little interest in recruiting posted workers for shorter spells. There is also comparatively little supply of night jobs, which normally are less attractive jobs for native workers. The liberalised regulations in 2008 for the recruitment of labour from outside the EU/EEA region meant that employers could engage third country nationals for time-periods of 2 x 2 years. However, as the system was over-used in the cleaning industry, as well as in other parts of the low-skilled service sector, this option has been strongly reduced. So far, the stricter regulations for recruiting third country labour migrants that have been introduced in recent years to prevent low-wage competition and labour abuse, have been effective, at least in the view of the LO. The study also indicated that the presence of temporary work agencies providing cheap labour from the former Eastern Bloc seems to be have been more common in the Swedish cleaning industry than what is known from earlier research. The presence of temporary agencies and other labour market intermediaries providing cheap labour from the newer EU countries to Swedish cleaning companies and private customers have in fact been a controversial issue ever since the early 2000. Even though the Swedish Tax Agency and the labour market parties have increased their efforts to curb the widespread problem of various kinds of unfair competition in cleaning, the problem still exists and generates price pressure, market disturbances and undercutting of wages. The RUT and LOV reforms, and the new F-tax reform introduced during the period 2007–2009, have all contributed to a significant increase in the establishment of small cleaning companies, not least among foreign born women. The rising demand for domestic cleaning services, which has emerged in Sweden in the wake of the RUT reform, has also been a strong incentive for many female labour migrants from the former Eastern Bloc to work temporarily in Sweden as employees or as self-employed

31 Annette Thörnquist workers. However, this market tends to be highly over-supplied as well, at least in the metropolitan areas. When discussing the inflow of migrant workers from the newer EU countries to the Swedish cleaning industry, or to other industries, it is also important to consider the social structure of the Swedish labour market, including structures of segregation and segmentation based on class, gender and ethnicity. Since unemployment is high in this industry, and foreign born workers are highly over-represented, especially in general cleaning, there is also a risk of immigrant competition. In this case too, competition is most pronounced in the metropolitan areas where the share of foreign born workers is high and many temporary labour migrants and undocumented immigrants are working. Hence, it is reasonable to assume that the existing structures of segregation and labour market segmentation also have an impact on temporary labour migration from the newer EU countries in the Baltic region to the Swedish cleaning industry. However, this study also revealed another important factor, which has not been discussed in previous research on trends and developments in labour migration, namely that Swedish employers in the cleaning industry, as well as in several other low-skilled sectors, can easily recruit Swedish workers who are participating in various kinds of labour market schemes. In other words, they can use a tax-funded reserve army of cheap native labour, which also means considerable wage subventions. This option has most likely also had an impact on the employers’ interest in recruiting temporary migrant workers from the newer EU countries. Moreover, the intensive use of wage subventions has no doubt worsened the problems of price pressure, job shortage and low-wage competition in the over-supplied Swedish cleaning industry. This also implies a risk of further social segregation and labour market segmentation. Ultimately, the possibility to use workers with highly subsidised wages can be related to the supply-side labour market policy that has prioritised workfare over welfare.

32 East-West Labour Migration and the Swedish Cleaning Industry

References Abbasian, Saeid (2006) Deltidsarbete och deltidsarbetslöshet bland städare i Stock- holms län: En studie av integration och jämställdhet. Working Paper från HE- LA-projektet 2006:6. Stockholm: Arbetslivsinstitutet. Abbasian, Saeid and Carina Bildt (2007) Arbetsvillkor och anställningsförhållanden inom städbranschen i Stockholms län. Arbetsliv i omvandling 2007:8. Stockholm: Arbetslivsinstitutet. Abbasian, Saeid and Carina Hellgren (2012) ‘Working conditions for female and immigrant cleaners in Stockholm county: An intersectional approach’. Nordic Journal of Working Life Studies, Volume 2, Number 3, 161–181. Aftonbladet, 4 June 2009. ‘Ulvskogs attack: ”Människohandel”’. (Melén, Johanna). Akhavan, Sharareh (2006) The health and working conditions of female immigrants in Sweden. Stockholm: Karolinska Institutet. Almega Home Service Industry (2014) ‘Rör inte Rut’, https://www.almega.se/hemser- viceforetagen/rut, visited in September 2014. Almega Serviceentreprenörerna (2011) En renare städbransch: En sammanställning och analys för att få en tydligare bild av omfattningen av svartarbetet i städbran- schen. Stockholm: Almega. Almega Serviceentreprenörerna (2014) Branschrapport 2014. Fakta och utveckling för städ- och serviceentreprenadbranschen. Stockholm: ALMEGA. Almqvist, Roland (2006) New Public Management – om konkurrensutsättning, kon- trakt och kontroll. Stockholm: Liber AB. Alsos, Kristine and Line Eldring (2008) ‘Labour mobility and wage dumping: The case of Norway’. European Journal of Industrial Relations, 14 (4), pp. 441–459. Anderson, Bridgit and Martin Ruhs (2010) ‘Migrant Workers: Who Needs Them? A framework for the analysis of staff shortages, immigration, and public policy’, in Ruhs, Martin and Bridgit Anderson (eds.) Who Needs Migrant Workers? Labour shortages, immigration and public policy. Oxford: Oxford University Press, pp. 15–53. Andersson Joona, Pernilla, Stina Petersson and Eskil Wadensjö (2013) ‘Temporary agency work and labor migration to Sweden’, in Friberg, Jon Horgen and Line El- dring eds. (2013) Labour Migrants from Central and Eastern Europe in the Nordic countries: Patterns of migration, working conditions and recruitment practices. TemaNord 2013:570. Copenhagen: Nordic Council of Ministers, pp. 273–298. Andersson Joona, Pernilla and Eskil Wadensjö (2011) Rekrytering av utländsk arbets kraft: Invandrares arbetsmiljö och anknytning till arbetsmarknaden i Sverige. Rapport 2011:1. Stockholm: Stockholms universitets Linnécentrum för integrationsstudier (SULCIS). Andersson, Lina and Matz Hammarstedt (2011) ‘Inkomst och yrkesposition bland invandrare från de nya EU länderna’. Ekonomisk Debatt, Vol, 39 (2), pp. 50–58.

33 Annette Thörnquist

Anker, Richard (1998) Gender and Jobs: Sex Segregation of Occupations in the World. Geneva: International Labour Office. Arbetet, 17 October 2014. ‘Många företag är oorganiserade’, (Wreder, Johanna). Arbetet, 24 October 2014. ‘Bidragsjobben dumpar hela branscher’, (Pejer, Mats). Arbetsmiljöverket (2015) (Work Environment Authority) Helårsrapport 2014: Register för företag som utstationerar arbetstagare Sverige. Rapport 2015:10. Stockholm: Arbetsmiljöverket. Arvidsson, Monika (2009) ‘EU Labour Migration: Government and Social Partner Policies in Sweden’, in Galgóczi Béla, Janine Leschke and Andrew Watt (eds.) EU Labour Migration Since Enlargement: Trends, impacts and policies. Aldershot: Ashgates. Asset Skills Research Group (2007) The Role and Importance of Migrant Workers in the Cleaning Industry. Northampton: European Union, European Social Fund. Bettio, Francesca and Alina Verashchagina (2009) Gender segregation in the labour market: Root causes, implications and policy responses in the EU. European Com- mission Directorate-General for Employment, Social Affairs and Equal Opportuni- ties Unit G1. Luxembourg: Publications Office of the European Union. Björklund Larsen, Lotta (2010) Illegal yet Licit: Justifying Informal Purchases of Work in Contemporary Sweden. Diss. Stockholm: Department of Social Anthropology, Stockholm University. Byggnadsarbetaren, November 2012. ‘Mellanhänder som skor sig har ökat’, (Berg- gren, Jenny). Calleman, Catharina (2011) ‘Precarious Employment in Sweden? Care Work and Do- mestic Work in a Twilight Zone between Public Law and Private Law’, in Thörn- quist, Annette and Åsa-Karin Engstrand (eds.) Precarious Employment in Perspec- tive: Old and New Challenges to Working Conditions in Sweden. Work and Society, Vol. 70. Brussels: P.I.E. Peter Lang. pp. 131–162. Castles, Stephen (2006) ‘Guestworkers in Europe: A Resurrection?’ International Mi- gration Review, Volume 40, Issue 4, December 2006, pp. 741–766. Cremers, Jan (2011) ‘In search of cheap labour in Europe: Working and living conditions of posted workers’, CLR Studies 6. Brussels: European Institute for Construc- tion Labour Research. Dagens Arena, 7 June 2011 ‘Så motverkar vi svart städning’ (Milton, Jonas). Deakin, Simon (2013) Addressing labour market segmentation: The role of labour law, Working Paper No. 52. Geneva. Governance and Tripartism Department, Interna- tional Labour Office. de los Reyes Paulina (ed.) Inte bara jämställdhet – Intersektionella perspektiv på hinder och möjligheter i arbetslivet. SOU 2014:34. Stockholm: Arbetsmarknadsdepar- tementet (Labour Market Ministry).

34 East-West Labour Migration and the Swedish Cleaning Industry

Ds 2013:57 Åtgärder mot missbruk av reglerna för arbetskraftsinvandring. Stockholm: Justitiedepartementet. Dølvik, Jon Erik (2008) Mobility of Labour and Services across the Baltic Sea after the EU Enlargement: Trends and Consequences. Working Paper No. 161, Oslo: Insti- tute for Labour and Social Research. EFCI/FENI (2014) European Federation of the Cleaning Industry. http://www.feni.be. Ekonomifakta, ‘Fakta och statistik: Arbetskraftsinvandring’, ekonomifakta.se, visited in December 2014. Eldring, Line and Thorsten Schulten (2012) ‘Migrant Workers and Wage-Setting Insti- tutions: Experiences from Germany, Norway, Switzerland and United Kingdom’, in Galgóczi Béla, Janine Leschke and Andrew Watt (eds.) EU Labour Migration in Troubled Times: Trends, impacts and policies. Aldershot: Ashgates, pp. 235–260. Elmhorn, Camilla (2008) ‘Migrerande arbeterskor: Informellt arbete och papperslösa invandrare i Stockholm’, in Den kalla och den varma staden: Migration och stads- förändringar i Stockholm efter 1970. Stockholm: Stockholmia. Engdahl, Mattias (2013) International Mobility and the Labor Market. Diss. Uppsala: Department of Economics, Uppsala University, IFAU. Equality and Human Rights Commission (2011) Temporary and Circular Migration: Empirical evidence, eurrent policy practice and future options in EU Member States. SWEDEN. Brussels: European Commission, European Migration Network. European Commission (1996) ‘Directive 96/71/EC of the European Parliament and of the Council of 16 December 1996 concerning the posting of workers in the framework of the provision of services’. Official Journal of the European Union, I 18/1, 21/1 1997. European Commission (2008) ‘Directive 2008/104/EC of the European Parliament and of the Council of 19 November 2008 on temporary agency work’. Official Journal of the European Union, I. 327/9, 5/12 2008. European Commission (2009) ‘Directive 2009/52/EC of the European Parliament and of the Council of 18 June 2009 providing for minimum standards on sanctions and measures against employers of illegally staying third-country nationals’. Official Journal of the European Union, I 168/24, 30/6 2009. European Commission (2014) ‘Employment, Social Affairs & Inclusion: Posted Workers. What is Posting?’ http://ec.europa.eu/social/main.jsp?catId=47, visited in November 2014. Eurostat (2014) Unemployment rate by sex and age groups - annual average, % http://ec.europa.eu/eurostat/data/database, visited in December 2014. Expressen 10 February 2011. ‘Ett rent helvete: De städar med slavkontrakt’, (Josefsson, Jessica). Ferlie, Evan, Lynn Ashburner, Louise Fitzgerald and Andrew Pettigrew (1996) The New Public Management in Action. Oxford: Oxford University Press.

35 Annette Thörnquist

Friberg, Jon Horgen (2012) ‘Culture at work: Polish migrants in the ethnic division of labour on Norwegian construction sites’, Ethnic and Racial Studies, Volume 35, Issue 11, 201, pp. 1914 – 1933. Friberg, Jon Horgen and Line Eldring eds. (2013) Labour Migrants from Central and Eastern Europe in the Nordic Countries. Patterns of migration, working conditions and recruitment practices. TemaNord 2013:570. Copenhagen: Nordic Council of Ministers. Friberg, Jon Horgen, Jens Arnholtz, Line Eldring, Nana W Hansen and Frida Thorarins (2014) ‘Nordic labour market institutions and new migrant workers: Polish migrants in Oslo, Copenhagen and Reykjavik’. European Journal of Industrial Relations 20 (1), pp. 37–53. Galgóczi, Béla, Leschke Janine and Andrew Watt eds. (2009) EU Labour Migration since Enlargement: Trends, impacts and policies. Aldershot: Ashgates. Gavanas, Anna (2010) Who cleans the welfare state? Migration, informalization, social exclusion and domestic services in Stockholm. Stockholm: Institute for Future Studies. Gavanas, Anna and Catharina Calleman eds. (2013) Rena hem på smutsiga villkor? Hushållstjänster, migration och globalisering. Göteborg: Makadam. Gavanas, Anna and Alexander Darin Mattson (2011) Bland Rolexklockor och smutsiga trosor: Om skattereduktioner och segmentering på den svenska hushållstjänste- marknaden. Arbetsrapport 2011:9. Stockholm: Institutet för Framtidsstudier. Gerdes, Christer and Eskil Wadensjö (2013) Immigration to Sweden from the New EU Member States. Report 2013:5, Stockholm: Seips. Guild, Elspeth and Sergio Carrera (2012) Labour Migration and Unemployment: What can we learn from EU rules on the free movement of workers? CEPS Liberty and Security in Europe, February, http://www.ceps.eu. Hermann, Christopher and Jörg Flecker (2012) Privatization of Public Services: Im- pacts for Employment, Working Conditions and Service Quality in Europe. London: Routledge. Hood, Christopher (1995) ‘The New Public Management in the 1980s: Variations on a Theme’. Accounting, Organizations and Society, Vol. 20, No. 2/3, pp. 93–109. Hotell- och Restaurangfacket (2012) Till vilket pris som helst: En uppföljning av arbetskraftsmigranters villkor i hotell- och restaurangbranschen. Stockholm: Ho- tell- och Restaurangfacket. Hviid, Kirsten and Ann-Mari Flyvholm (2010) Udenlandsk arbejdskraft i Norden: Sammenfatning af landerapporter fra forprojekt til Nordisk Ministerråds globaliseringsprojekt. TemaNord 2010:588, København: Nordisk Ministerråd. International Organization for Migration (2014) ‘About migration/Migration terms’. http://www.iom.int/cms/en/sites/iom/home/about-migration/key-migration-terms-1. html, visited in October 2014.

36 East-West Labour Migration and the Swedish Cleaning Industry

Jonsson, Claes-Mikael (2013) Vinnare och förlorare – om konkurrens med arbetskrafts- kostnader inom EU. Stockholm: Landsorganisationen i Sverige. Jonsson, Claes-Mikael and Göran Larsson (2013) Gäst i verkligheten – om utstatio- nerad arbetskraft i praktiken. Facket och EU. Stockholm: Landsorganisationen i Sverige. Jonsson, Claes-Mikael, Thord Pettersson, Håkan Löfgren and Kristoffer Arvidsson (2014) When labour costs squeeze the price – an analysis of public sector invest- ments in infrastructure. Stockholm: Swedish Trade Union Confederation. Junestav, Malin (2011) ‘Promoting Employment or Employability? The Move from Active Labour Market Policy to Workfare’, in Thörnquist, Annette and Åsa-Karin Engstrand (eds.) Precarious Employment in Perspective: Old and New Challenges to Working Conditions in Sweden. Work and Society, Vol. 70. Brussels: P.I.E. Peter Lang, pp. 219–246. Kahanec, Martin and Klaus F. Zimmermann (2008) ‘Migration in an enlarged EU: A challenging solution?’ Paper prepared for the workshop. Five Years of an Enlarged EU – a positive-sum-game: Brussels, 13–14 November 2008. Kahanec, Martin and Klaus F. Zimmermann (2009) Migration in an Enlarged EU: A challenging solution? Economic Papers 363. Brussels: European Commission. Kamp, Annette, Klemsdal, Lars and Lena Gonäs (2013) ‘Working in the Public Sector: Introduction to the Thematic Issue’. Nordic Journal of Working Life Studies, Vol- ume 3, Number 2, 1–8. Kastberg, Gustaf (2005) Kundvalsmodeller: En studie av marknadsskapare och skapade marknader i kommuner och landsting. Göteborg: Förvaltningshögskolan, Göteborgs universitet. Kerr, Sari Pekkala and William R. Kerr (2011) Economic Impacts of Immigration: A Survey, NBER Working Paper No. 16736. Cambridge: National Bureau of Econo mic Research. Khosravi, Sahram (2009) Irreguljära immigranter i Sverige, ojämlikhet och utanför- skap. Socialförsäkringsrapport 2008:4. Stockholm: Försäkringskassan. Kildal, Nanna (2001) Workfare Tendencies in Scandinavian Welfare Policies. Geneva: International Labour Office. Kjellberg, Anders (2011) ‘Trade unions and collective agreements in a changing world’, in Thörnquist, Annette and Åsa-Karin Engstrand (eds.) Precarious Employ- ment in Perspective: Old and New Challenges to Working Conditions in Sweden. Work and Society, Vol. 70. Brussels: P.I.E. Peter Lang, pp. 47–100. Kjellberg, Anders (2013) Kollektivavtalens täckningsgrad samt organisationsgraden hos arbetsgivarförbund och fackförbund. Department of Sociology, Lund Uni- versity Studies in Social Policy, Industrial Relations, Working Life and Mobility. Research Reports 2013 (Updated 26 March 2014). Kommunal (2012). Vägval välfärd. Rapport. Stockholm: Svenska Kommunalarbe- tareförbundet.

37 Annette Thörnquist

Kommunal (2014) ‘Branscher och yrken: kök och städ’. http://www.kommunal.se/ Kommunal/Branscher-och-yrken/Kok-och-stad/. Visited in October 2014. Komulainen, Sirkka (2012) ‘Developing Comprehensive Mechanisms for Labour Immigration: MATTO Programme in Finland’, in Platonova, Anna and Giuliana Urso (eds.) Larbour Market Inclusion of the Less Skilled Migrants in the European Union. Brussels: International Organization for Migration IOM pp. 47–55. Le Grand, Julian (2007) The Other Invisible Hand: Delivering Public Services Through Choice and Competition. Princeton: Princeton University Press. Lemos, Sahra and Jonathan Portes (2008) New Labour? The Impact of Migration from Central and Eastern European Countries on the UK Labour Market. IZA DP No. 3756. Bonn: Forschungsinstitut zur Zukunft der Arbeit. Lim, Nelson (2002) Friends or Foes?: Immigrant Competition Hypothesis Revised. DRU 2913. Working Paper 2002:11. Rand Publication Cooperation. http://www.rand.org/content/dam/rand/pubs/drafts/2008/DRU2913.pdf. Lisborg, Anders (2012) Human Trafficking and Forced Labour in Denmark? Migration and working conditions for a group of migrant workers employed in the cleaning industry, in the green sector and as au pairs. Copenhagen: The Danish National Board of Social Services. LO (2012) Sökes: Kompetent, flexibel och billig arbetskraft – hur arbetsgivarnas krav ökar polariseringen på arbetsmarknaden. Delrapport 1: Arbetsmarknaden efter krisen. Stockholm: Landsorganisaionen i Sverige. LO (2013) Fusk och utnyttjande – om avregleringen av arbetskraftsinvandringen. Arbetsmarknad. Stockholm: Landsorganisationen i Sverige. Lundborg, Per (2009) ‘The Dimensions and Effects of EU Labour Migration in Swe- den’, in Galgóczi, Bela, Janine Leschke and Andrew Watt (eds.) EU Labour Migra- tion in Troubled Times: Skills Mismatch, Return and Policy Responses. Farnham: Ashgates, pp. 69–86. Melkas, Helinä and Richard Anker (1997) ‘Occupational segregation by sex in Nordic countries: An empirical investigation’. International Labour Review, 341–363. Migrationsverket (2015) (Swedish Migration Board) ‘Fakta och statistik: Beviljade uppehållstillstånd av arbetsmarknadsskäl’. http://www.migrationsverket.se/Om-Mi- grationsverket/Statistik/Oversikter-och-statistik-fran-tidigare-ar.html. OECD (2011) Recruiting Immigrant Workers, Sweden. Paris: OECD Publishing. Ollus, Natalia, Anniinan Jokine and Matti Joutsen eds. (2013) Exploitation of Migrant Workers in Finland, Sweden, Estonia and Lithuania: Uncovering the links between recruitment, irregular employment practices and labour trafficking. Publication Se- ries No. 75. Helsinki: EuropeanInstitute for Crime Prevention and Control, affiliated with the United Nations. Ollus, Natalia and Anniinan Jokine (2013) ‘We’ve got people lined up behind the door. Placing the Trafficking and Exploitation of Migrant Workers in Context in the Restaurant and Cleaning Sectors in Finland ’, in Ollus, Natalia, Anniinan Jokine and

38 East-West Labour Migration and the Swedish Cleaning Industry

Matti Joutsen (eds.) Exploitation of Migrant Workers in Finland, Sweden, Estonia and Lithuania: Uncovering the links between recruitment, irregular employment practices and labour trafficking.Publication Series No. 75 Helsinki: European Institute for Crime Prevention and Control, affiliated with the United Nations, pp. 31–55. Pedersen, Peder J. and Mariola Pytlikova (2008) EU Enlargement: Migration flows from Central and Eastern Europe into the Nordic countries – exploiting a natural experiment. Working paper 08:29. Aarhus: Department of Economics, University of Aarhus. Peterson, Stina (2012) Bemanningsföretagens roll i rekryteringen av utländsk arbetskraft. Rapport 2012:1. Stockholm: Stockholms universitets Linnécentrum för integrationsstudier (SULCIS). Peterson, Stina (2013a) Utstationerades arbetsförhållanden på svensk arbetsmarknad Rapport 2013:1. Stockholm: Stockholms universitets Linnécentrum för integrationsstudier (SULCIS). Peterson, Stina (2013b) Bemanningsanställdas arbetsvillkor på svensk arbetsmarknad: En del av två FAS-projekt. Två atypiska anställningsformer. Anställning i beman- ningsföretag och egenföretagande. Vad betyder de för utomeuropeiska invandrare? Working paper 2013:2. Stockholm: Sockholms universitets Linnécentrum för integrationsstudier (SULCIS). Piore, Michael J. (1971) ‘The Dual Labour Market: theory and implications’, in Gor- don David M (ed.) Problems in Political Economy. An Urban Perspective. Lexing- ton, Massachusetts D. C.: Heat. Piore, Michael J. (1979) Birds of Passage: Migrant Labor and Industrial Societies. Cambridge, UK: Cambridge University Press. Platonova, Anna and Guiliana Urso eds. (2012) Labour Market Inclusion of the Less Skilled Migrants in the European Union. Brussels: International Organization for Migration, IOM. Quirico, Monica (2012) Labour migration governance in contemporary Europe: The case of Sweden. Torino: Fieri Working Papers. Rapp, Gunilla (2006) ‘Möjligheter och hinder för avknoppare – Tioårigt perspektiv på privatiseringen av städverksamheten i två kommuner’, in Sundin, Elisabeth and Annette Thörnquist (eds.) När anställda blir företagare: Studier om företagande i ett strukturomvandlings- och arbetsmarknadsperspektiv. Arbetsliv i omvandling 2006:17. Stockholm: Arbetslivsinsitutet. Refslund, Bjarke (2014) ‘Intra-European labour migration and deteriorating employment relations in Danish Cleaning and Agriculture: Industrial relations under pressure from EU8/2 labour inflows? Economic and Industrial Democracy, vol On-line first, 10.1177/0143831X14550421. RITA (2008) RITA – Rättvis ingång till arbete. En rapport om papperslösa i städ- branschen. Projekt inom Fastighetsanställdas Förbund finansierat av Fastighets,

39 Annette Thörnquist

Näringsdepartementet och Europeiska flyktingfonden. Stockholm: Fastighetsan- ställdas förbund. Rose Tronstad, Kristian and Pernilla Andersson Joona (2013) ‘New patterns of migration from Central and Eastern Europe to the Nordic countries’, in Friberg Jon Horgen and Line Eldring (eds.) Labour migrants from Central and Eastern Europe in the Nordic countries Patterns of migration, working conditions and recruitment practices. TemaNord 2013:570. Copenhagen: Nordic Council of Ministers, pp. 19–42. Ruhs, Martin (2013) ‘EU Enlargement and Labour Immigration. The UK Experience’. European Liberal Forum. In: Zelano, Karin (ed.) Labour Migration: What’s in it for us? Brussels: European Liberal Forum, pp. 63–120. Ruhs, Martin and Bridgit Anderson eds. (2010) Who Needs Migrant Workers? Labour shortages, immigration and public policy. Oxford: Oxford University Press. Rönnmar, Maria (2004) ‘The Personal Scope of Labour Law and the Notion of Em- ployee in Sweden’, in The Mechanisms for Establishing and Changing Terms and Conditions of Employment: The Scope for Labor Law and the Notion of Employee, 2004 JILPT Comparative Labor Law Seminar, JILPT Report No. 1. Tokyo, 2004, pp. 159–165. SCB (2014a) Statistiska centralbyrån (Statistics in Sweden). ‘Befolkningsstatistik i sammandrag 1960–2013. http://www.scb.se/sv_/Hitta-statistik/Statistik-efter-amne/ Befolkning/Befolkningens-sammansattning/Befolkningsstatistik/25788/25795/ Helarsstatistik---Riket/26040/. Visited in November 2014. SCB (2014b) Statistiska centralbyrån (Statistics in Sweden), SCB. Invandring- ar (medborgare utom Norden) efter grund för bosättning, medborgarskap och kön. År 2004 – 2013’. http://www.statistikdatabasen.scb.se/pxweb/sv/ssd/ START__BE__BE0101__BE0101J/ImmiBosattMedb/chart/chartViewLine/?rx- id=6638b33c-e9b6-4827-a117-dd89b345a22b. Visited in January 2015. Visited in January 2015. Schierup, Carl-Ulrik Castles, Stephen and Peo Hansen eds. (2006). Migration, Citizen- ship, and the European Welfare State: A European Dilemma. Oxford Scholarship Online. Scmulyar Green, Oksana (2013) ‘För barnens skull. Kvinnliga migranter från Ös- teuropa i EU:s hushållstjänster’, in Gavanas Anna and Catharina Calleman (eds.) Rena hem på smutsiga villkor? Hushållstjänster, migration och globalisering. Stockholm: Makadam, pp. 161–187. Scmulyar Green, Oksana and Andrea Spear (2014) ‘Caugth in the vulnerability trap’, Baltic Worlds, Volume VII:4, pp. 27–28. SFS 1999:678 Lag om utstationering av arbetstagare. SFS 2008:962 Lag om valfrihetssystem. SFS 2013:351 Lag om ändring i lagen (1999:678) om utstationering av arbetstagare.

40 East-West Labour Migration and the Swedish Cleaning Industry

Sigvardsdotter, Erika (2012) Presenting the Absent: An account of undocumentedness in Sweden. Uppsala: Kulturgeografiska institutionen, Uppsala universitet. Sippola, Markku (2014) ‘Balancing between exit, voice, and loyalty: Labor market policy choices in Estonia’, in Sommers, Jeffrey and Charles Woolfson (eds.), The Contradictions of Austerity: The socio-economic costs of the neoliberal Baltic mo del. New York: Routledge, pp. 87–102. SKL (2015) Sveriges Kommuner och Landsting, ‘Valfrihetssystem i kommuner 2014 – Beslutsläget i införandet av LOV’. http://skl.se/, visited in January 2015. Somerville, Will and Madelene Sumption (2009) Immigration and the labour market: Theory, evidence and policy. Equality and Human Rights Commission. Published March 2009. ISBN 978 1 84206 100 8. Sommers, Jeffrey and Charles Woolfson eds. (2014). The Contradictions of Austerity: The socio-economic costs of the neoliberal Baltic model. London and New York: Routledge. Sorainen, Olli (2010) ‘Finsk landerapport’, in Hviid, Kirsten and Ann-Mari Flyvholm, Udenlandsk arbejdskraft i Norden: Sammenfatning af landerapporter fra forprojekt til Nordisk Ministerråds globaliseringsprojekt.TemaNord 2010:588, København: Nordisk Ministerråd, pp. 30–40. SOU 2006:87 Arbetskraftsinvandring till Sverige – förslag och konsekvenser. Stock- holm: Arbetsmarknadsdepartementet. SOU 2010:63 EU:s direktiv om sanktioner mot arbetsgivare. Betänkande av Sanktions- utredningen. Stockholm: Justitiedepartementet. SOU 2014:16. Det ska vara lätt att göra rätt: Åtgärder mot felaktiga utbetalningar inom den arbetsmarknadspolitiska verksamheten. Stockholm: Arbetsmarknadsde- partementet. Statistics in Finland (2013) Sysselsättning 2011. Yrke och socioekonomisk ställning. http://www.stat.fi/til/tyokay/2011/04/tyokay_2011_04_2013-11-06_tie_001_ sv.html. Visited in November 2014. Stoppafusket.nu 19 April, 2013 (Norberg, Anna-Lena). Sundin, Elisabeth and Annette Thörnquist (eds.) (2006) När anställda blir företagare: Studier om företagande i ett strukturomvandlings- och arbetsmarknadsperspektiv. Arbetsliv i omvandling 2006:17. Stockholm: Arbetslivsinsitutet. Sundin, Elisabeth and Gunilla Rapp (2006) Städerskorna som försvann. Arbetsliv i omvandling 2006:2. Stockholm: Arbetslivsinstitutet. SVT 1 (Swedish Television) 1 September 2010. Uppdrag Granskning. ‘RUT – mest för de rika’. Sydsvenska Dagbladet 29 April, 2005 ‘Billig personal hyrs ut i Tallinn’, www.sd.se (Magnusson, Erik).

41 Annette Thörnquist

Tamas, Kristoff and Reiner Müntz (2009) Labour Migrants Unbound? EU Enlarge- ment, Transitional Measures and Labour Market Effects. Stockholm: Institute for Future Studies. Thörnquist, Annette (2011a) ‘False Self-Employment: A Topical but Old Labour Mar- ket Problem’, in Thörnquist, Annette and Åsa-Karin Engstrand (eds.) Precarious Employment in Perspective: Old and New Challenges to Working Conditions in Sweden. Work and Society, Vol. 70. Brussels: P.I.E. Peter Lang, pp. 101–129. Thörnquist, Annette (2011b) I utkanten av arbetets marknad. ESF-projekt i ett ar- betsmarknads- och arbetsmiljöperspektiv. Vetenskaplig skriftserie från REMESO, Temagruppen integration i arbetslivet, TIA, 2011:3. Linköping: Linköpings universitet. Thörnquist, Annette (2013a). False (Bogus) Self-Employment in East-West Labour Migration: Recent trends in the Swedish construction and road haulage industries. Themes on Migration and Ethnic Studies, TheMES No. 41/13. Linköping: REME- SO, Linköping University. Thörnquist, Annette (2013b) Mångfaldens marknad och arbetets villkor: Om följder av kundval (LOV) i hemtjänsten. Arbetsliv i Omvandling nr. 2: 2013. Lund: Work, Technology and Society, WTS, Lunds universitet. Thörnquist, Annette (2014) ‘Mångfaldens retorik och arbetets praktik: Konkurrens- utsättning och jämställdhet i hemtjänsten’, in de los Reyes Paulina (ed.) Inte bara jämställdhet – Intersektionella perspektiv på hinder och möjligheter i arbetslivet. SOU 2014:34, Stockholm: Arbetsmarknadsdepartementet (Labour Market Minis- try), pp. 115–144. Thörnquist, Annette and Åsa-Karin Engstrand eds. (2011) Precarious Employment in Perspective: Old and New Challenges to Working Conditions in Sweden. Work and Society, Vol. 70. Brussels: P.I.E. Peter Lang. Tillväxtverket (2012). Kvinnors och mäns företagande 2012: Företagens villkor och verklighet. Fakta & Statistik. Stockholm: Tillväxtverket. UN (2005) The International Convention on Migrant Workers and its Committee. Fact Sheet No. 24 (Rev.1). New York and Geneva. Office of the United Nations High Commissioner of Human Rights. Urban, Susanne (2013) ‘Restaurant work experience as stepping stone to the rest of labour market: Entry to and exit from the restaurant sector among young people of immigrant backgrounds in Sweden’, Nordic Journal of Migration Research, Volume 3, Issue 2. Wadensjö, Eskil (2012) ‘EU Enlargement and Labour Immigration. The Swedish Experience’, Zelano, Karin (ed.) Labour migration: What’s in it for us? Brussels: European Liberal Forum, pp. 29–62. Wilkinson, Frank ed. (1981) The Dynamics of Labour Market Segmentation. London: AP.

42 East-West Labour Migration and the Swedish Cleaning Industry

Wóicicka, Aleksandra (2012) ‘EU Enlargement and Labour Immigration. The Polish Experience’, in Zelano, Karin (2012) Labour migration. What’s in it for us? Brus- sels: European Liberal Forum. Woolfson, Charles, Christer Thörnqvist and Jeffrey Sommer (2010) ‘The Swedish model and the future of labour standards after Laval’, Industrial Relations Journal, Vol. 41, No. 4, pp. 333–350. Välfärd No. 1. 2012. ‘Många nya företag i städbranschen’ (Håkansson, Maria), Statis- tiska centralbyrån, SCB. Zelano, Karin ed. (2013) Labour Migration: What’s in it for us? Brussels: European Liberal Forum. Zelano, Karin, Gregg Bucken-Knapp, Jonas Hinnfors and Andrea Spehar (2014) Map- ping and analysis of types of migration from the CEE countries. Country Report Sweden. Imagination Working Paper No. 13, 2014. Ölmebäck, Emma (2011) Ständig standby – en rapport om visstidsanställdas villkor. Stockholm: Kommunal.

43 The Swedish cleaning industry is an ethnically segregated labour market in which almost 50 per cent of the workforce was born abroad. Moreover, price pressure in public procurement of cleaning services and unfair competition generate market disturbances and labour abuse. Against this background, this study discusses the presence of temporary labour migrants from Poland and the Baltic states in this labour market. Since there is a shortage of jobs in cleaning, there is also a potential risk of ‘immigrant competition’. The author argues that when discussing labour migration, we need a broad social perspective that takes into account the impact of existing segregation and labour market segmentation in the destination country, as well as the balance of power between the social partners and the labour market policy pursued by the government. Among other things, the study reveals a factor that most likely also has an impact on east-west labour migration to the Swedish cleaning industry, namely that the employers can easily recruit workers in Sweden participating in labour market schemes, which means considerable wage subventions. Thus, employers can use a tax-funded reserve army of cheap labour to gain competitive advantage, for example in public procurement.

Dr. Annette Thörnquist is Associate Professor of History and working life researcher. In recent years, she has worked as a researcher at the Department of Economic His- tory, Stockholm University, at REMESO, Linköping University and at the Research Institute for Work, Technology and Culture (F.A.T.K.), University of Tübingen.

TheMES – Themes on Migration and Etnic Studies. Occasional papers and Reprints publi- ceras av REMESO, Institute for Research on Migration, Ethnicity and Society. REMESO är ett institut för forskning, forskarutbildning och grundutbildning på Institutionen för samhälls- och välfärdsstudier vid Linköpings universitet. TheMES är en en serie för working papers, återutgivning och parallellpublicering av vetenskapliga arbeten. www.ep.liu.se

REMESO publicerar också serien REMESO RAPPORT, vilken innehåller rapporter och redovisningar från samarbeten med, och expertuppdrag från, partners, myndigheter på internationell, nationell och lokal nivå, NGO:s och näringsliv. www.ep.liu.se

TheMES No. 42-2015 Series editor: Erik Berggren, ISSN 1651-8306 Tel. +46 11 36 32 66, E-mail: [email protected]