Labor Migration Database

Up-to-date knowledge in the field of internationalization of the labor market in Europe. Find scientific publications (provided with an abstract) and search for similar publications via keywords.

This database is the result of the multidisciplinary literature review and data collection by the INT-AR project – a project researching the internationalization of the European labor market, as well as transboundary labor recruitment. The database is composed of a collection of scientific publications with short summaries listing their main insights for policymakers and researchers. In addition, all sources have been tagged allowing the user to easily find comparable resources within the database.

Go to the database 

Stay informed of

  • the latest publications;
  • additions to the Database;
  • developments related to labor migration
Sign up to our newsletter

links to other comparable websites or databases

The Migration Observatory An Oxford University database with research, analysis and policy papers
The Migration Policy Institute A US database with factsheets, research, analysis and policy papers
IZA World of Labor A German database with articles on migration, mainly from an economic perspective
Migration Policy Centre A migration research center associated with the European University in Florence


This series of accompanying papers provides insights into different aspects of labor migration, including who can be considered a labor migrant and the macro-economic effects of labor migration. The papers have been peer reviewed and are periodically updated to reflect current developments.

1. Who is a labour migrant?

Download this paper

In the public debate different definitions in which one can comprehend who counts as a migrant are used. Moreover, also in the regulatory frame that applies for migrants, in statistics on the stocks and flows of migrant workers, in analysis of labour mobility and cross-border recruitment, in data sources and research the definitions vary. As a result, clearly assessing the size of the migration, the impact for the (Dutch) labour market and the evolution over time is not easy. Different, sometimes conflicting definitions pose challenges for consistent and coherent policymaking. Applied to the narrower term of ‘labour migrant’ the same confusion exits, although it seems logical to separate this category from asylum migration and knowledge migrants. Assessment of the impact of today’s and future labour migration asks for more unified definitions.

In INT-AR Paper 1, some examples of the notion of a ‘labour migrant’ are treated.

The greatest common denominator in the cited definitions seems to be that labour migrants are workers that come to the Netherlands in search for temporary work. The question is how this way of defining matches with the right of free movement and the liberty to look for a job in another member state that is enshrined in EU-treaties for all EU-citizens (with of course an increased potential after the EU-enlargement in 2004). The European Institutions tend to apply a different definition that seems to be at odds with the Dutch ‘mainstream’ terminology. In the EU, labour migration is defined as ‘Cross-border movement of people for the purpose of employment in a foreign country’, but in the special migration EU-glossary, there is only explicit reference to non-EU workers.[i]

EU-citizens who use the right of free movement are not labelled labour migrants; labour migrants are workers from 3rd countries who leave their country or region in order to work in the EU, often in search of a better life. For EU-citizens, the free movement prevails, according to Directive 2004/38/EC of the European Parliament and of the Council on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States. According to this Directive EU-citizens have the right to come to the Netherlands, as long as they fulfil certain obligations, just as Dutch citizens are welcome to work, live and study in another EU Member State. As such, the free movement that can take the form of labour mobility is an established right for all EU-citizens and, in principle, subject to minor restrictions.


2a. Social Security and free movement: European coordination

Download this paper


Since the establishment of the European Economic Community (EEC), the free movement of citizens of the Member States has been a key theme in the European cooperation. The removal of barriers for the free movement of citizens belonged to the basic principles of the 1957 Rome Treaty. This Treaty forms the basis for the EEC and contains several provisions that guarantee the free movement (Treaty of Rome, 1957, Articles 48-51). The Treaty gives European citizens the right to go to another EEC member state to seek employment and to work in all EEC member states. Three INT-AR papers deal with the social security that is applicable in case of free movement of workers. Paper 2a treats the general principles of the social security in case of cross-border work. Paper 2b looks after loopholes and question marks in the legislation and the link with free service provision. Paper 2c assesses the notion of welfare tourism.

The coordination of social security

Although the actual arrangement and application of the social security system does not belong to the competence of the European Institutions, the coordination of national social security schemes became one of the first and crucial regulated fields of cooperation between Member States. In order to prevent that EU-citizens who work and reside temporarily in another Member State lose their social security rights the Member States considered it necessary to govern the coordination of their social security systems. These arrangements and the fine-tuning of the different national systems in cross-border situations know a long history. Since 1958, the Council has concluded regulations for the social security coordination of mobile and migrant workers. The coordination was (and is) based on the principle that persons moving within the EU are subject to the social security scheme of only one EU member state. The regulations have to guarantee the equal treatment and to counteract discrimination.

In the course of the years, this legislation has been revised, based on jurisprudence and practical experience. The European Council modified the corresponding Regulation number 3, concluded in 1958, 14 times. The subsequent and completely revised Regulation 1408/71 was likewise modified 39 times. These modifications had to do with adaptations to changes in national legislations and with improvements based on case law stemming from proceedings of the Court of Justice of the European Union. The readjustment of the coordination rules in 2004 took the form of an overall modernisation exercise, leading to Regulation 883/2004 (also called the Basic regulation), with the additional aim to simplify the procedures. At the same time, the legislator wanted to get rid of several exemptions. The necessary implementing Regulation 987/2009 was concluded in spring 2009. The modified rules came into effect from 1 May 2010. The basic principle – only one legislation applies – stayed upright. The last revision dealt with several important question marks. The coordination mechanism had to be made resistant to practices that attempt to circumvent the mandatory contribution of social security payments, for instance by using foreign letterbox companies. Crucial issues were how to get a grip on cross-border recruitment and the free movement of workers that pursue activities in multiple countries, like in the international transport and the inland waterways, but also in the construction or infrastructure sectors. After the conclusion, the procedures and working methods to be followed in case of cross-border work have figured several times on the agenda of the Administrative Committee for the coordination of the social security systems. This Committee, with representatives from all Member States, has the tasks to monitor all questions and problems of administrative and interpretative nature and to facilitate the uniform application of EU-legislation (based on Title IV of Regulation 883/2004).

The main principles of the current rules

Regulation 1408/71, concluded in 1971, and the complete revision with the Council Regulations concluded in 2004 and 2009 guarantee workers who are EU-citizens the right to social security benefits, irrespective of the place of work or residence. Workers who decide to settle down in another EU country have, in principle, the right to be treated as if they were citizens of that country. The objective was and is a reciprocal coordination, not the harmonisation of the social security systems in the Member States, in order to regulate cross-border situations. The general rule for the social security in the case of labour migration is the application of the country where the work is performed principle (the Lex loci laboris). Workers have to be insured and will build up and acquire social security rights in the country where the work is pursued. The 2004/2009 revision enhanced the right of equal treatment. Whilst this principle was originally applied only for persons who at the same time had decided to work and live in another country, the condition to reside there is no longer an absolute must. A second modification broadens the scope. Next to workers, self-employed, civil servants, students and pensioners, with their family and relatives, non-active persons are inserted in the application of the coordination rules. The scope of application no longer contains an exhaustive enumeration. A third change is the simplification of several rules with regard to unemployment: an unemployed who decides to move to another Member State in order to find a job, retains during a certain period the right of unemployment benefit payments in the country of origin.

The functioning of these rules

In addition to the coordination rules for the social security, the terms as formulated in Directive 2004/38/EC with regard to the residence rights of EU citizens that want to live in another Member State apply. EU citizens have the right to live on the territory of a Member State for a maximum period of three months without further conditions or formalities to fulfil, except the obligatory possession of a valid identity card or passport. EU citizens who want to stay for a longer period in another Member State either must be employed or self-employed in the host country or dispose of sufficient resources for themselves and their family members to ensure that they do not lay claims on the social assistance system of the host country during the stay. During the first three months of residence, the host country is not obliged to confer entitlement to social assistance. As long as EU citizens and their family members not become an unreasonable burden on the social assistance system of the host country, they cannot be expelled. If they are an employee or self-employed, expulsion is under no circumstances legitimate. Every EU citizen residing lawfully for a continuous period of five years in the territory of the host country, acquire a permanent residence in that country. Until the expiry of this period, expulsion is possible if a person does not comply with the applicable requirements.

Important case law

The Court of Justice of the European Union has clarified the right of access to (parts of) the social security. In a few cases, the central question was whether economically inactive EU citizens residing in another Member State are entitled to claim rights to social benefits, which apply to the citizens of that country. The coordination regulations seem to suggest these rights, although Directive 2004/38/EC does not impose this obligation. In November 2013, the Court ruled in the so-called Dano-case. A woman of Romanian origin, a long-time resident in Germany, had never worked in Germany and was not a job seeker. Her appeal on social assistance for job seekers was dismissed. The case was brought before the European Court. The Court confirmed, according to the EU-legislation regulating the free movement (Directive 2004/38), that the host country is not obliged to grant social assistance during the first three months of residence. Is that period exceeded (but still less than 5 years), providing one’s own subsistence becomes a condition for the right of residence. If someone does not meet these requirements, no entitlement to social benefits exists. There is in general only entitlement for benefits for which contributions are paid (e.g. unemployment benefits). The ruling is important in light of the discussion on social security and welfare tourism (elaborated in paper 2c).

Jan Cremers (Tilburg University)

2b. Social Security and free movement: Legal Loopholes

Download this paper


The Treaty of Rome contains several provisions to ensure the free movement (Treaty of Rome in 1957, Article 48-51). Free movement gives citizens the right to seek work and to go to work in another Member State. Three INT-AR papers will outline the main policies associated with this movement. Paper 2a was devoted to the social security in cross-border work and the consistency with European rules on the residence right. Paper 2b exemplifies some loopholes in the rules and the relationship with cross-border services, illustrated by a case of a letterbox company. Paper 2c discusses the so-called welfare tourism.

Some legal loopholes

An exception, formulated in the coordination of social security Regulations, to the country of employment principle applies to posted workers who work temporarily in another country on behalf of their employer in the frame of the freedom to provide services. A person who is posted for a period shorter than 24 months remains subject to the social security system of the country where he/she is usually employed. The administrative work and costs necessary to integrate workers who work for a short time in another country in completely different social security systems do not outweigh the benefits. Besides, it is hard to derive rights in such short periods. However, the relationship between posting and social security has appeared susceptible to various ways of problematic use.

The Basic Regulation 883/2004 defines the maximum period of posting at 24 months.[i] After these 24 months, in principle the rules of the country where the work is performed apply, unless an agreement is reached within the meaning of Article 16 paragraph 1 of the Regulation.[ii] This period of 24 months is not mentioned explicitly in the Posting of Workers Directive 1996/71/EC. Earlier research revealed two important problems.

  • Posting of workers in the frame of cross-border service provision is used as an instrument for transnational recruitment whereby workers are recruited from countries with low social security costs. This occurs in the construction and agriculture, and meanwhile in other sectors with strong competition on labour costs.
  • Secondly, the creation can be detected of fictitious foreign companies that recruit workers who then are posted into several countries. Examples can be found in international road transport and inland waterway transport.

Cross-border provision of services

The deficient control on compliance with the social security payments and with existing employment conditions, combined with low risks of detection create plenty of space for unscrupulous companies that circumvent through various misleading ways the rules and regulations. Falsification of the so-called A-1 forms (a standardized statement that proves the social security registration of any country), the use of the posting formula to foreign workers who have been living in the country of employment and job mediation by fake agencies or letterbox structures are practices that have come to light in controls. A report drafted by the labour inspectorate and the most engaged social partners in 9 countries revealed the complicated control on the legality of the posting. The fact that the country where the work is pursued has only limited powers when it comes to checking the legitimacy of posted work and the employment relationship between the posted workers and the posting company in the country of origin impedes effective control. Limited power resources and poor functioning sanctions, partly because of a lack of international cooperation in gathering information and dealing with fraud, hamper the effective exercise of checks on lawfulness. Besides, a host country has few competences to control to what extent a company is legitimate; cooperation of the host country is crucial.[iii]

Free establishment and the use of letterbox companies – a case

In recent years several transport companies (in the Benelux countries) received the offer to transfer the workforce to intermediate companies (located in, for instance Cyprus) and to hire the staff through these intermediate service suppliers. The ‘companies’ publish glossy websites in various languages and from different addresses in Europe. With reference to the changes in the coordination of social security as a result of Regulations 883/2004 and 987/2009 (applicable since May 2010), the intermediates offer to act as employers for the workforce. The original employer of the truck drivers would become the ‘client’ and only receive an invoice for supplying services, whilst the truck drivers would continue to work for the original employer. Truckers who work for this client (their former employer) earn the minimum wage (taxed for income in the home country) and all other wage components and bonuses (for overtime and irregular work, for instance) are paid undeclared (neither in the home country of the worker, nor in the country of establishment of the letterbox). The social security payments are based on the minimum wage and the former employer is ‘freed’ from additional payments (such as mandatory employer contributions in the home country or sectoral charges for pensions or vocational training). The recruitment of workers through these letterbox companies is presented as a ‘perfectly legal’ method to make substantial savings on the labour costs of drivers. The introduction of a labour contract under a foreign legal frame (even though workers do not live there and never will get to that country) creates this opportunity. In short, the use of foreign intermediaries is lucrative because of the lower social security contributions. The used method to circumvent the mandatory social security contributions refers to the Basic Regulation on the coordination of the social security. It often takes place through the provision of A1 (previously E101) forms in the intermediate’s country of registration. This standardised EU-form declares where a person is registered for the social security and in which country contributions are paid. The issuing of the form has some built-in control mechanisms. The competence to figure out in which country the government must provide an A-1 document lies first in the hands of the competent body in the country of residence of the driver. Besides, the coordination rules impose conditions on the genuine character of a company: it must be a company with local operations and output in the country of establishment. In some cases, it turned out that uninsured drivers drove around with foreign A-1 documents, which were issued without the competent body involved. Paychecks made it clear that social security payments were not duly made in the country of establishment, since only the basic wage and not the total wage premium was paid.

In one case, brought up by the Dutch transport union FNV, it took 16 months (and a change of government) to convince the social security authorities to enforce Dutch social security and working conditions. The case was clear from the beginning and transmitted to the Dutch authorities in March 2012, but partly because of the fear that the authorities of the workers’ country of residence would be accused of ‘creating barriers to the free provision of services’, the competent social security institution – Sociale Verzekeringsbank (SVB) – was reluctant to act immediately. The final statement issued by SVB in October 2013 was clear, however: based on Regulations 883/2004 and 987/2009, truck drivers living in the Netherlands and on the payroll of one of the intermediates had to stay in the social security system of their country of residence – the Netherlands. The case had other far-reaching consequences. The SVB stated that the labour relation was still intact with the client (the original employer) ‘as the activities were still pursued under the subordination and authority of the client’. The SVB concluded that the intermediate firm could not prove that it was a real undertaking with activities in the country of registered office. In summary, the link between the driver and the intermediate was qualified as artificial and created with the sole aim of circumventing social security obligations in the country of residence.

Jan Cremers (Tilburg University)

[i] Article 12.1 Regulation 883/2004: A person who pursues an activity as an employed person in a Member State on behalf of an employer which normally carries out its activitie s there and who is posted by that employer to another Member State to perform wo rk on that employer's behalf shall continue to be subject to the legislation of the first Member State, provided that the anticipate d duration of such work does not exceed twenty-four months and that he is not sent to replace another person.

[ii] Article 16.1 Two or more Member States, the competent authorities of these Member States or the bodies designated by these authorities may by common agreement provide for exceptions to Articles 11 to 15 in the interest of certain persons or categories of persons.


2c. Social Security and free movement: Welfare tourism?

Download this paper

Welfare tourism debated

In this third paper on social security and free movement, we look at the scale and size of benefits in case of migration. At the end, we make a preliminary balance. In 2013, the Dutch Government notified the European Commission that the free movement of EU citizens posed the risk of migration motivated by the search for social security payments. The responsible state secretary spoke of a major problem with EU citizens applying for assistance without having worked in the Netherlands. His evidence came from a fraud case by Bulgarian citizens who had received some local benefits. In total, he claimed a number of 4200 EU citizens seeking payment of social benefits in the Netherlands. The government wanted to put an end to this and asked the European Commission to come up with a solution. The Commission in turn argued that it was a non-issue; the Netherlands scored well below the average in terms of the number of resident Europeans on benefits. A year later, the European Court of Justice judged in a case of a Romanian woman who moved to Germany and applied for social assistance there (C-333/13 - Dano). The Court stated that the right to free movement does not automatically imply the right to social assistance: an EU citizen who resides more than three months in another Member State must have sufficient resources in that country to provide for his or her livelihood. According to the Court, EU countries have the right to refuse welfare benefits (special, non-contributory cash benefits) to nationals of other Member States, who have come to their country especially for those benefits.[i]

In another case (Alimanovic, in September 2015), the Court ruled that equal treatment with regard to the right to social assistance depends on the legality of stay in the country. Remarkable was that the person had worked for 11 months (in Germany) and then had enjoyed a lawful period of six months of unemployment benefits. The German social insurance agencies decided after those six months no longer to provide benefits; the Court approved this policy. The Court referred to Article 7.3.c of the EU Directive 2004/38. This article states that EU citizens who become involuntarily unemployed within the first twelve months of their stay in another country retain their status as legitimate job seeker in the country and are entitled to equal treatment in the social security for a period of at least 6 months. After a period of six months, the host Member State has the right to refuse access to non-contributory forms of social assistance.

Who is entitled?

Entitlement to benefits differs depending on the type of benefit. Relevant is the distinction between contributory benefits (such as unemployment benefits and work-related pensions) and social assistance. Non-Dutch EU citizens cannot simply settle in the Netherlands and then apply for social assistance. One condition for the application of the free movement and the right to establish in another EU country is that a person is able to provide the own subsistence and living costs. An EU citizen should, after a stay of three months, either be employed or self-employed, or have sufficient means of subsistence.[ii] People who have worked in the Netherlands and have contributed to the social security are consequently insured for their social costs and, if they fulfil all other conditions, have the right to receive, for instance, unemployment benefits. Workers from another EU Member State, who have contributed and have worked at least 26 weeks in a reference of 36 weeks, derive the right to receive unemployment benefits during 3 months. This period is extended if the worker has worked for at least 4 years in a period of 5 years.

Available data

Few detailed figures are available on the composition of the group of benefit recipients in the Netherlands. The overall number of benefit recipients rose to 490 000 people (in August 2015, including state pension recipients with additional benefits), after a decline that began since 1998 and brought back the total to 370,000 in mid-2007. Changes in the conditions for receiving benefits make the numbers not fully comparable; however, the figure provides a proxy idea. It is worthwhile, independent of the question how many people are included, to analyse who receives benefits and whether there have been important changes over the years. The Dutch statistical office CBS records EU-citizens in the category western allochthones. A large part of this category origins (according to the CBS-definitions) from a German or Indonesian background. The size of citizens coming from Central and East European countries (CEE MOE) has increased in the last 10 years (to 177,000 per 1 January 2015). There were 108,000 citizens with a Polish background. The total of residents coming from Bulgaria and Romania in 2013 was around 4% of all EU-residents (or 0.23% of the total population). The share of western benefit receivers in the overall figure of benefit receivers was stable; in absolute size the overall total increased in recent years to 54,500 persons (in August 2015, a decrease started from spring 2015). In total, 4,160 citizens from CEE-countries that belong to the labour force (15-65 year old) received (by the end of December 2013) social protection benefits. The use of social protection by CEE-migrants is slightly higher than the use by the autochthones (2.8% compared to 2.3% per 1-1-2014).

However, the share of unemployment benefits of CEE-residents was substantially lower (in August 2014) with 2.4%, compared to 6.9% for Dutch workers. The percentage of non-Dutch unemployed is increasing.[iii]

The 75,000 temporary workers from CEE-countries, calculated in the so-called Migrantenmonitor, in general have no entrance to the social protection or unemployment benefits. If these workers are included in the calculations of the benefit use, the share of CEE-workers having benefits falls sharply.[iv]

The total of non-western benefit receivers increased steadily to 241,000 persons (in August 2015). However, figures on the background of this category are less reliable.

To resume

The statistical office CBS has published data on benefit fraud that reveal a serious decrease of fraud in the period 2002-2008. The available data provide no evidence for the existence of large-scale welfare tourism related to social protection benefits. Most migration and mobility related research in Europe confirms that migrants are not interested in the country with the best social benefits. The general rule is that the presence of jobs is the most predictive parameter for an increase of labour migration and mobility. Research and literature provide a certain pattern for the impact of this type of labour migration on the welfare state, and particularly on the social security. Labour migrants in the lowest segment of the labour market perform very often work for which it is difficult to find sufficient labour supply in the receiving country. Partly, there is the effect of substitution, whereby other vulnerable groups are pushed out of the labour market. There are only few incentives for employers, in case of enough supply of cheap labour, to invest in these more vulnerable workers. Besides, the import of cheap labour can have a downward effect on the wages of the lowest pay scales. The use of labour migrants can conserve certain types of low-paid jobs (and the related production of goods and services) that normally would have disappeared. This all seems not very rational, as it could block innovation, but, for instance, it still keeps the Dutch agro- and horticulture profitable. In other sectors that are inland or site specific, like domestic work and caretaking, the influx of migrants guarantees the necessary and growing services that are closely related with the future of the welfare state.

The Dutch labour market is not very generous with regard to the offer of stable jobs. Many migrant workers work in precarious and flexible jobs with short time contracts, as such they belong to the invisible and unrepresented part of the labour market. The cross-border recruitment of workers has facilitated the expansion of flexible labour relations. Labour is becoming more and more a commodity, with work performed under a contract for the provision of services, by a person who is no longer considered a stakeholder. Thus, the impact of cross-border mobility of workers on the flexibilisation of the (Dutch) labour market is continuously expanding. If the result is less contributions to the social security schemes, in the end the impact on the financial sustainability of the welfare state will increase.

Jan Cremers (Tilburg University)



[iii] UWV Kennisverslag 2015-1, april 2015, Kenniscentrum UWV

[iv] Ontleend aan (bezocht op 12-1-2014).

3. Labour migration and substitution

Download this paper

Jan Cremers (Tilburg University)

What is substitution on the labour market?

Substitution appears if finding a job by one person results in the job loss (or non-obtaining of a job) of another person. The notion has been defined in different manners. Labour market experts qualify a situation where unemployed work below their skill level a form of conjunctural or cyclical substitution: in period of economic downturn and increasing unemployment the supply of labour is larger than the available employment. Employers have the opportunity in such situations to come up with higher demands and recruit high-skilled persons that will substitute lower skilled workers. Although labour migration regularly leads to a situation where there is a mismatch between the (high) skill level of the recruited foreign workers and the offered jobs, this phenomenon is not conceived as cyclical substitution. Substitution related to labour migration (widely understood, both as labour mobility of EU-citizens and labour migration by 3rd country workers) is most often used as a more elegant wording of ‘they take our jobs’. In the search for the reduction of labour costs employees are recruited who demand less and are flexible employable. There is substitution if the decrease of jobs for the local residents is an evincible consequence of the entering of labour migrants. The Dutch social-economic council (SER) distinguishes in its labour migration advice between direct substitution (job losses as a result of a job taken by a migrant) and indirect substitution (inland supply has less changes as a result of the existence of foreign supply). The SER-advice concludes that most available studies (at macro level) indicate low substitution effects, even in times of crisis. According to the SER, the outlook can be different if certain occupations or branches are analysed.

Research of substitution before the economic crisis

In one SEO-project dedicated to the economic effects, for the period 1999-2005, there was no evidence of large-scale substitution or displacement of residents after the arrival of long-term staying labour migrants from CEE-countries. In a follow-up, the effect of the presence on the market of labour migrants after 2005 has been analysed. According to the researchers, the influx of long-term residing labour migrants from CEE-countries in the period 1999-2008 had no significant net effect on the overall total of jobs for Dutch employees. It seemed that the labour market easily could handle this additional supply of workers. The researchers found positive effects in growth segments of the market: the supply of foreign workers supplemented the inland supply. Negative effects were registered in declining parts of the market: labour migrants competed with domestic supply. The effects in both parts of the market were so minor that the outcome in the end was zero. Though the supply of foreign workers doubled in the period 2005-2008, the labour market was not characterised by a substitution of Dutch workers.

The results correspond with the outcome of British studies for the period 1997-2005, a period that labour migration to the UK increased substantially. These studies hardly found any substitution of employment; but there was a downward pressure on low wages (and a positive effect for higher wage groups of workers). In the UK, this resulted, first and foremost, in a downward pressure on the wages of migrants already present on the market. Labour migration had no significant effect on the employment perspective of British workers. The Migration Advisory Committee (MAC) of the British government came up with another link between pay and the inland supply. In short, this governmental advisory body claimed that an improvement of the lowest wages in some sectors could serve as a strong incentive for the enlisting of British workers. Economic analysis in Norway presented similar outcomes: the supply of migrant labour creates a downward pressure on the lowest wage scales. There is an indirect substitution effect. However, in times of economic booming these effects are not perceptible because society at large profits of the presence of migrant labour.

Substitution during the economic crisis

Research on substitution at the Dutch labour market, commissioned by the ministry of labour, led (at the end of 2014) to the conclusion that workers in some sectors were exposed to substitution as a result of the influx of migrant labour from CEE-countries, as a result of unfair, distorted competition. It is cheaper for employers to hire staff from abroad through the use of fake agencies and middlemen. The main risks groups that suffer from substitution are the low-skilled, youngsters and migrants that are already present. However, migrants that enter the Dutch market particularly take up jobs and tasks that are neither aspired nor accepted by Dutch unemployed.

Most recent studies don’t provide evidence of a link between the presence of migrant labour from CEE-countries and the domestic employment during the crisis. Researchers expect migrant labour to be mobile and to disappear as soon as the employment outlook deteriorates; an effect that is strengthened by a restrictive access to social security benefit in host countries. In the study, commissioned by the Dutch labour ministry, research institute SEO found out that labour migration from Western European countries decreased, however, migration from Eastern Europe did not change. In agriculture, manufacturing, construction, retail and transport the total employment of Dutch workers decreased with digit percentages, combined with an increase of foreign labour. The jobs most often shifted from direct labour into temporary and flexible contracts. In these sectors, the search for cheap labour was combined with an ongoing process of flexibilization. The cost reduction was achieved through the recruitment of labour migrants who accept lower wages and high flexibility. This type of substitution mainly took place in labour-intensive jobs with standardised work, with a calculation and a competition based on price, not on quality, and in situations where language skills are neither essential, nor important. Labour migration has a distributional effect on certain categories of workers. These effects are both negative and positive. Some high-risk groups that have to compete with migrants are confronted with negative effects. Other categories hardly have to compete and encounter positive effects.

Recent findings on labour market effects

In a majority of analyses substitution is assessed from the perspective of job losses and direct and indirect employment effects. Another aspect is often neglected, that is the effect on the flexibilization of the labour market. The recruitment of migrant labour is almost exclusively through flexible, temporary and short-term labour contracts, thus further facilitating the externalisation and flexibilization of the labour market. Research in France, dedicated to the impact of immigration on the wages and employment of similarly skilled native workers, confirms this assumption from another angle. The author decomposes the population of native workers by job contract to investigate how job contract (and therefore wage rigidity) can protect the labour market outcomes of natives from immigration. He concludes that immigration has no wage impact on the population of native workers under indefinite-term contracts, whereas it reduces the wages of the native population under fixed-term contracts. This asymmetric wage impact indicates that the wages of workers under fixed-term contracts can be manipulated by firms to adjust labour cost. Since fixed-term contracts have a short duration and they terminate at no cost, they provide firms with a scope for wage adjustment: firms hire the new employees at lower wages than those who terminated their fixed-term contracts. The labour turnover among workers with indefinite-term contracts is lower and they tend to be more protected. As a result, firms are unable to lower the wage of the native population covered by indefinite-term contracts. But, firms use the exit of native workers under indefinite-term contracts to replace them by migrants. While indefinite-term contracts exert a protective effect on wages, an increase in the number of migrants decreases the employment level of native workers under indefinite-term contracts. If this is combined with a lowering of the working conditions and pay, the increase of migrant labour will negatively influence the share of direct labour on the labour market. The growth of migrant labour thus threatens the stable form of labour relations of workers in direct labour. There is no loss of employment as such, but a slow and advancing reduction of secured jobs and stable labour relations.

January 2017, a pdf of the second, revised version can be downloaded.

4. Labour market frictions and migration

Download this paper

Research on labour market shortages

This paper makes an assessment of labour market frictions and shortages and of the proposed policy to tackle the frictions, based on national and international studies. The used definition of labour market frictions is based on existing international notions that were also used by the Dutch socio-economic council (SER) in its advice on labour migration: shortage or lack of labour can be both in absolute terms (no labour supply available) or in relative terms (no labour available for the offered pay and working conditions).[i] The SER expects that the Netherlands, in line with the rest of the EU, will be confronted with shortages of (young) high-skilled workers and as a result will have to recruit 'knowledge'-migrants. In the SER reasoning it is necessary to develop a strong and active labour market policy, based on existing and future shortages, including initiatives that aim to improve the use of (potential) qualifications of migrant workers that are already residing in the Netherlands.

The UK and Norway have an experience of more than a decade with longitudinal metering of labour shortages. This experience is synthesised, followed by a section on the Dutch approach. Most observers conclude that the demand for well-skilled personnel and the expected shortages in technically skilled and social occupations and, to a lesser extent, in occupations in care-taking and education might lead to serious labour market frictions. At the same time, several sources note that is not an easy task to estimate the future needs on the labour market and the related shortages, which can be expected. The future needs correlate with the functioning of the labour market and with technological developments that have a direct influence on the future job content and the necessary size of the workforce. The technological development has an important impact on: production processes and tasks that can be split in parts and relocated to specialised plants spread over more countries, leading to an international economic division of labour, based on off-shoring, outsourcing or re-shoring. A lack of labour could lead to a transfer of (parts of) the production to other regions or countries, to substitution of labour by capital or to the import of necessary products.

Comparative European research

In a study, commissioned by the Employment and Social Affairs committee of the European Parliament (in 2015), an overview is provided of the expected labour shortages, looking at their nature, causes and size.[ii] Quantitative shortages arise from a shrinking labour force, a decrease of the labour participation, a growing demand and a geographical disparity in the population structure. Qualitative shortages become manifest in a mismatch between the demand and the available qualification level, in preferences of employees that do not match with the offered jobs and in a poor image of the available functions combined with inadequate information provided. Migration can contribute to a decrease of shortages in receiving countries, but it can lead, at the same time, to quantitative shortages in the sending countries. For instance, the migration of medical and care professionals form the CEE-countries has created serious shortages in hospitals and healthcare institutions in different CEE-countries. According to the study, there are no substantial quantitative shortages at aggregate EU-level. Shortages are mainly of a qualitative nature in relation to educational deficits and insufficient labour market connection. In contrast to other countries the regional disparity in the Netherlands is relatively low. In the Netherlands mainly qualitative shortages are expected, with a growing demand for specialists in manufacturing, the building and energy sector, corporate services and the IT-sector. The Netherlands scores high on the ranking with regard to shortages in some technical occupations and specific IT-functions, and there is more demand than supply for medical specialists.

A report of the European Commission treats the barriers in the employers’ search for qualified labour.[iii] According to a survey (in 2013) 40% of the interviewed employers was confronted with problems in the recruitment of skilled labour. Around half of the cases belonged in the category of real qualification shortages; one third of the cases was related to a remuneration that was too low. Unattractive working conditions, combined with atypical working time, poor career perspectives and a lack of training facilities, explain why it is often so difficult to find the right person, especially if no permanent job is in sight. The report signals for the Netherlands that there is no upward pressure on wages, because there is still enough skilled labour available on the labour market. The attentiveness to the development of talent (and a solid HRM-policy) might explain why labour shortages are still very limited.

Research on labour market shortages in the UK and Norway

The UK government installed in 2007 a permanent advisory institution for migration, the Migration Advisory Committee (MAC). The MAC is an independent, non-governmental public consultative body that advises the government in different policy areas that are related to migration issues. The institute, composed by 5 economists, a chairperson and representatives of home affairs and employment and social affairs monitors in particular three items: the impact of labour migration, the limitation of the migration that is based on the socalled Tiers 2 system – the visa system in the UK for high qualified labour migrants – and shortages on the labour market related to qualification needs. Since 2008 the MAC publishes an overview of labour market shortages (the ‘shortage occupation list’) and examines and assesses notified shortages on the basis of a particular own methodology. The list determines to a large extent the migration policy of the British government for workers coming from third countries.[iv]

Recent advises treat the applied income thresholds for labour migrants and the relationship with the actual work that is performed by migrants on the British labour market. In the spring of 2015 the list that eases the entrance to the labour market in case of serious frictions was revised. In the assessment the MAC made use of several existing sources, such as the Labour Force Survey, data and statistics on working time and earnings, research on qualification demands and unemployment data. The MAC defined over the years 12 indicators that are applied in the assessment. These are indicators related to the perception of employers, to the earning structure, the (available) labour volume and other relevant labour market data (such as the duration of existing demand and the relationship between vacancies and unemployment).

The Norwegian NAV business survey, commissioned by the government, examines the labour demand, per region and in sectors and occupations. The survey includes, since 2007, the recruitment of foreign labour. In some editions of the NAV-survey (2007 to 2013) the demand for qualified labour migrants form outside Norway is analysed. Labour shortages peaked in 2007 with, in total, 132,400 workers. After the start of the financial-economic crisis the labour demand was decreasing; in 2008 shortages lowered to 88,700 workers, the year after the figures halved to 45,000 workers. Due to a housing and property bubble the demand increased again in 2011. After these years, a permanent loss of jobs with a serious effect on the overall demand started in the building trades and the oil exploration (in 2012 the overall shortages were calculated at 36,700 workers, in 2013 this decreased to 32,200 persons and in 2014 to 26,000). In the spring of 2015 the survey showed a further decrease of the shortages to 20,800 workers. Due to a modification of the survey the data on the period 2007-2011 and 2012-2013 are not completely compatible. However, it is clear that the sectoral distribution has substantially changed. The first measuring pinpoints agro- and horticulture and hospitality as the sectors with increased demand for foreign labour. Later on most labour migrants are recruited in extracting industries, including oil exploration, and hospitality. Although the overall demand is weakening in 2013, the demand for migrant labour stays upright in some industries. In the long run this is the case in corporate services, the IT and communication sector, in caretaking and social services. Serious shortages remain in healthcare occupations, in services and in parts of the building trades (though the large demand has substantially decreased). The demand in different segments of manufacturing was rather constant. The 2015 survey indicated a modest growth perspective; employers expected a stabilisation of the overall employment, with vanishing frictions in the extracting and oil industries (including the processing industries).

Migration and labour market frictions – the Dutch analysis

The analysis and mapping of labour market frictions in the Netherlands are undertaken by several institutes. The UWV (Employee Insurance Agency) makes use of various instruments (for instance sectoral analyses) that describe frictions and surpluses per sector. The ROA (Research centre for Education and Labour market) produces regular overviews of current and future developments of the Dutch labour market. The CBS (Statistics Netherlands) publishes the Labour Force Survey and other statistics.

The UWV produced an overview in May 2014 with labour and shortages. The overview tries to map as clear as possible the occupational shortages. At the level of medium-skilled workers the institute found indications of shortages in operational technical functions and technical staff; at high-skilled and scientific level the indicators of shortages went in other directions, i.e. specific IT-functions (programmers), specific occupations in education and specific financial occupations (tax specialists and actuaries). In caretaking, it was indicated that only in special functions, for instance regional/mobile caretakers and specialists in geriatric medicine, shortages were signalled. On the occupational side shortages were found in the metal and electronical manufacturing and in other parts of manufacturing, in specialised corporate services and in IT.[v]

The ROA expects serious frictions and tensions between demand and supply on the labour market in the near future in the IT and technical segments. In the last updated research (2015), the ROA forecasts a growing demand for technical occupations over the whole skill range in technics: starting at the mediumskilled level to the academic educated. The austerity programs in the healthcare and social sectors have important consequences for the prognosticated employment growth in the sector; as a consequence the forecasted growth will be substantially lower, resulting in less frictions.[vi]

According to the statistical office CBS, labour migration can be one of the mechanisms that enter into force in case of labour shortages. With a regression analysis the office examined whether changes in the recent past of the size of the labour force have had an influence on the labour migration into the Netherlands. No significant relation was found between labour migration and supply of labour. In periods that the domestic labour supply accelerated there was no decrease of labour migrants’ influx. In years of shortages the influx did not increase. The CBS does not exclude that part of the migration was the result of shortages of domestic supply of labour in specific sectors, but, at aggregate level no clear link exists between the overall labour supply and the total influx of labour migrants. This is the main reason why the office has decided not to use the ageing of the population and the demographical shrinkage as predictors of the (future) labour migration.[vii]

A report of the European Migration Network (EMN) focuses on the need of third country workers and disregards the mobility of EU-citizens. EMN concludes that the labour market shortages are to be found in technical occupations, the IT-sector and certain specific labour market niches, at medium-skilled, high-skilled and academic level. The analysis of labour shortages has no important impact on the migration policy, with the exception of the upper level of the labour market, where labour migration is functioning as a means to strengthen the economy. For the rest, recruitment of labour migrants from third countries is not utilised to cover identified shortages; the initiative lies mostly in the hands of the (individual) employer. EMN notes that a shrinking and ageing labour population in all EU-countries will lead to a decreased potential of labour mobility in the EU. Fewer EU-migrants will be available and the competition with other EU-countries that want to attract migrants will intensify. Implicitly, EMN advocates a pro-active stance for a migration policy that is no longer considered a subordinated part of the actual labour market policy.[viii]

[i] SER, 2014, Arbeidsmigratie, Advies 14/09







[viii] 20b_netherlands_labour_shortages_nl_version.pdf

5. Provision of services and posted workers

Download this paper


This paper is based on a master thesis.[1] It summarises several aspects of the regulation of posted workers in the frame of the free provision of services. The downloadable extended paper describes:

  • the EU-context of the free provision of services;
  • the position of posted workers in relation to social security, tax and labour law;
  • posting in construction with a comparison of pay for posted workers in Belgium and the Netherlands;
  • a (first) assessment of enforcement measures and monitoring tools.


The free provision of services underpins the right for an undertaking, established in one Member State (hereafter, Sending State), to provide services in the territory of another Member State (hereafter, Host State) with its own workers. This concept, referred to as posting, raises the question what labour legislation is applicable to the working conditions of the worker during a posting period. Under the current version of the Posted Workers Directive 96/71/EC (hereinafter, PWD), Host States may impose a nucleus of labour conditions to posted workers. In 2014, the Enforcement Directive 2014/67/EU (hereinafter, ED) was introduced in order to ensure the more uniform implementation of the provisions of the PWD, with respect to the monitoring and enforcement of labour conditions of posted workers. While most Member States transposed the ED only recently into their domestic legislation, the European Commission proposed, in March 2016, a reform of the posting rules by applying the ´remuneration´ rather than the ´minimum rates of pay´ of the Host State in order to promote the ‘principle of equal pay for equal work at the same place’.

EU context

Posting of workers encompasses a wide range of issues in the frame of one of the economic freedoms of the Internal Market - the free provision of services. Contrarily to what the free movement of workers implies, posted workers do not gain access to the labour market of the Host State. By way of derogation from the lex loci laboris principle, posted workers remain subject to the legislation of the Sending State. Given the variety of labour systems and labour costs between Member States, this can result in an unfair competition climate regarding labour costs between local and foreign service providers. In this context, the PWD was created in order to protect the posted worker with a ‘hard core’ of provisions which apply in the Host State.

Posting of workers within the EU from the perspective of social security law, tax and labour law

The discussion about the revision of the PWD illustrates the conflicting stakes between the free provision of services on the one hand and the protection of workers’ rights within a climate of fair competition on the other hand. The imposition of domestic labour standards on foreign service providers touches upon a sensitive area as it conditions and restricts (indirectly) the free provision of services. This section explores the legal framework on posting at the EU level from three angles: social security law, tax law, and labour law.

Posting and social security law

Already before the adoption of the PWD, the concept of posting was introduced in 1971 in Regulation 1408/71 (now Regulation 883/2004) on the coordination of social security in the case of free movement of persons. Based on the principle of equal treatment, the starting point of the Coordination Regulation is the lex loci laboris principle: workers are subject to the social security system where the work is performed. An exception is made for workers being subject to posting for a period of less than 24 months. Workers remain subject to the social security system of the Sending State. The employer who posts workers to the territory of another Member State, has to carry out its activities normally in the Sending State. Article 14(2) of the related Implementation Regulation 987/2009 describes in more detail that the employer has to perform substantial activities, other than purely internal management activities, in the Member State in which it is established. As a consequence, the legislation of the Member State of establishment is not automatically applicable to the posted workers. Therefore, businesses cannot establish themselves in a non-genuine way  in low costs jurisdictions, while performing substantial activities in the territory of another Member State by structurally ‘sending’ workers from the low social security cost jurisdiction to the high social security jurisdiction.

Posting and tax law

In the EU Single Market, taxation remains almost utterly within the sovereignty of the Member States; each country is entitled to determine the connecting factor arising tax liability within its jurisdiction. In order to achieve a coordination of tax law regarding cross-border situations, Member States rely on the agreement of bilateral tax treaties, which are meant to avoid double taxation and double non-taxation. Within the EU, most of those bilateral tax treaties are based on the OECD Model Tax Convention. Similar to the lex loci laboris principle, earnings are taxed in the Country where the employment is performed. However, wages remain subject to taxation in the Contracting State (the Sending State) under two conditions. When the employee is present in the Host State for no longer than 183 days in 12 months, and the employer has no physical presence in the country where the employment is performed. It is quite remarkable that under tax law, posted workers remain subject to the Sending State for the first 183 days, while for social security law, the posted workers remain subject to the Sending State for a period of 24 months.

Posting and labour law

As regards to the compensation of the worker, under the current EU framework, the protection is equal to the ‘minimum rate of pay’ as laid down within the general applicable collective agreements or the statutory minimum wage. In some critical rulings, the CJEU interpreted the imposition of labour conditions beyond the minimal protection of the PWD, as being non-compatible with the free provision of services. With the recent approval of the PWD revision by the Council, characterised by an equal remuneration for posted workers, some might argue that this stigma in favour of the free provision of services has been abandoned. According to this, the question raises is to what extent a level playing field has been created between local and foreign service providers under the proposed revision of the PWD.

Posting in construction and a comparison on posted workers’ pay in Belgium and the Netherlands

The thesis describes how the construction sector in Western Member States has been confronted with a relatively high inflow of posted workers from CEE Member States. This can be explained by the labour costs driven model in the construction sector: posting within the European construction sector is triggered by differences in labour costs. Furthermore the thesis provides a comparison of the wages in Belgium and the Netherlands. In both countries, the gross wages as laid down within collective agreements are significantly higher than the statutory minimum wages. Therefore, it can be concluded that general applicable collective agreements at the sectoral level play an important role for Host States to create a level playing field. However, foreign service providers in construction are likely to have a competitive advantage based on labour costs due to differences in the amount of social security contributions and taxation. The proposed revision of the PWD is not going to improve this wage gap. Therefore, in order to create a full level playing field, one might argue that there is a need to adopt the EU legislation on social security in line with the lex loci laboris principle.

The need for adequate monitoring tools and effective enforcement mechanisms

At the same time, even if a perfect level playing field is not achieved, in order to minimise the disruptive effect of regulatory completion, Member States shall work altogether on matters dealing with enforcement and monitoring in order to create a climate of fair competition. Illegal practices of posting in the construction sector have been identified, such as the undercutting and circumvention of minimum rates of pay, fake posting by means of rotational and permanent posting, bogus self-employment, and the use of letterbox firms. Those illegal practices of social dumping within the construction sector underline the need for adequate monitoring and enforcement. Recently, the ED has been introduced in order to improve the monitoring and enforcement of the PWD. In the comparison, it appeared that Belgium can be praised for its registration system, LIMOSA, and for the active role of the Labour Inspectorate in monitoring the working conditions of posted workers as it is able to format its findings directly to the public prosecutor.

[1] Robert van der Vlies is LLM & MSc Candidate at Tilburg University. The extended paper was originally written as a Master Thesis, in the frame of the Master course Labour Law and Employment Relations

6. Applicable labour law in case of posting of workers

Download this paper (Dutch only)

The draft proposal to amend the Posting of Workers Directive assessed from the private international law perspective

Author: Veerle Van DEn Eeckhout (University of Antwerp)

On 8 March 2016, the European Commission published a proposal for a targeted revision of the Posting of Workers Directive. So far, the message in the media has been that this proposal must be seen as a step in the direction of the application of the principle ‘equal pay for equal work at the same workplace’. In the 6th INT-AR paper (download only available in Dutch) a specific aspect of the proposal is treated, notably the addition of a new article 2bis in the Posting Directive. This article deals with the aspect of long-term posting, where the duration of posting exceeds 24 months. In such cases, according to the proposal, the labour law conditions of the host Member States will have to be applied, where this is favourable to the posted worker.

The 6th INT-AR paper provides an in-depth analysis from a technical legislative perspective and examines the legal procedure that the European Commission wants to apply. According to the author, the extent to which the proposal can or cannot be fitted into and connected to developments in private international law, depends on how the ‘protection principle’ of the employee as the weaker party will further take shape in private international law.

7. Macro-economic effects of labour migration

Download this paper

Introduction: what are the relevant effects of labour migration?

In this paper we summarise the macroeconomic outcomes of labour migration as found in the relevant economics literature. The focus is predominantly on the (macro) economic effects for the ‘destination’ country. The economic effects of interest are on growth, on labour market outcomes and on the use of and contributions to the social security system. In surveying the literature it has proven difficult to make a clear distinction between the effects in general and the effects of labour, asylum, family or student migration in particular. The economics literature with respect to migration is indebted to the work of George Borjas and Barry Chiswick on the effects of migration to the US. However, migration patterns for European countries differ substantially from ‘traditional’ immigration countries such as the US and Canada. Therefore, we specifically include research on European countries in this paper.

Growth effects

Theoretically and partly in practice, immigration means additional labour supply, which could result in economic growth in the long run. In the short run however, due to decreasing returns to scale, mainstream economists expect a lower growth rate from population growth, thus also from immigration (Dolado et al, 1994). The exact size of these effects is, among other things, determined by the degree of substitution that can take place between migrant workers and the local workforce and the level of (physical or human) capital the migrants bring to the host country (see e.g. Kemnitz, 2001). In a local economy characterised by labour shortages, immigration could boost economic growth, when the new entrants have the required qualifications. On the other hand, in a local economy characterised by a relatively large labour reserve, immigration only leads to more competition for scarce jobs, potentially at the detriment of the labour market perspective of local workers with comparable qualifications. Economic growth is then not a likely result.

Not much empirical evidence is found for (positive or negative) effects on economic growth in the literature. Besides, most of the empirical literature focuses on the growth effect of migration in the country of origin; with some evidence that labour migration has positive effects on economic growth in the country of origin (e.g. Dos Santos & Postel-Vinay, 2003). In studies with a focus on the host country the causality of the effect runs the other way: economic growth leads to immigration and not the other way round (Morley, 2006; Islam et al, 2012) but the effects of migration on economic growth in the receiving country are at best ambiguous.

(Un)employment effects and wages

The effects of labour immigration on the labour market are related to the impact on economic growth. For labour market effects a comparison between the relative levels of human capital of local and migrant workers is relevant, next to what extent migrants are complements or substitutes to the local labour force. The theoretical literature does not establish definitive conclusions on labour market outcomes because of the heterogeneity of features of migration and the specificity of the host country (Danette & Fromentin, 2013). Furthermore, related to labour market outcomes, it’s crucial to think about distributional effects: who gains and who loses from immigration, and this depends critically on the skill mix of migrants relative to local workers (Dustmann, 2008). Also, theoretical predictions of the impact of immigration on the wages of local workers depend upon the assumptions used. The most important assumptions are whether the host economy is open or closed to international trade and the degree of substitutability between migrants and local workers (Friedberg & Hunt, 1995). Another important question in the migration literature is about assimilation: the extent to which labour market success becomes comparable with that of local workers as migrants spend time in their new country (Kerr & Kerr, 2011, Bart et al, 2004, referring to Chiswick, 1978, Borjas, 1999 and Bauer, Lofstrom and Zimmerman, 2000).

In a study for 18 OECD countries no significant long-term impact was found on unemployment, but it was found that ‘immigration may have a temporary impact on natives’ unemployment, depending upon the policy framework. In particular, a temporary increase in unemployment may be observed in a context of stringent anticompetitive product market regulation, or of high replacement rates of unemployment benefits’ (Jean & Jimenez, 2011). A panel study including 14 OECD countries concluded that an increase of migrants is likely to decrease wages in the destination countries in the long term although this impact seems to reverse in the case of Anglo-Saxon countries (Damette & Fromentin, 2013). Moreover, there is no evidence of adverse effects on unemployment in the long run, as yet outlined by Gross (2004) in the Canada case and Ghatak and Moore’s (2007) study of European countries. The findings of Damette & Fromentin (2013) are largely consistent with recent studies that used different methodologies (Ortega and Verdugo (2009) for France and Card (2009) for the United States and in particular with Okkerse’s conclusion (2008): the probability that migrants increase unemployment is low in the short term and zero in the long term.

Despite institutional differences, the results in Damette & Fromentin (2013) confirm that migration flows have no substantive adverse impact on the labour market of developed countries. Immigration is not the cause of higher unemployment rates in developed countries since the nineteen eighties. With respect to wage assimilation, Lehmer & Ludsteck (2015) conclude on the basis of German data that ‘the raw wage gap of migrants compared to native Germans decreases by 14 log percentage points in the first eight years.’ Furthermore they conclude that there are large differences between different immigrant groups.

Social security effects

Related labour migration effects on social security, the economics literature distinguishes between ‘pension systems’ on the one hand and social benefit systems for risks such as unemployment, sickness, disability and social assistance on the other hand. In the policy debate this is the distinction between ‘we need migrants to increase the workforce to pay for pensions’ and ‘migrants just come to our country to claim benefits’. Given the effects on labour market outcomes, the effects for both these claims are likely to be limited. Besides, migrants cannot ‘claim benefits’ and ‘steal our jobs’ at the same time. Nor is it possible that they are a solution for labour market shortages and a burden to the social security system at the same time.

Research on migration and (the financial sustainability of) pension systems suggests that the effects on the contribution rates of the social security system are quite small (Felderer, 1994; Börsch-Supan, 2000, based on German data).

With respect to the question whether migrants disproportional use social security benefits and/or welfare programs the primary source is the work by Borjas (1999). According to him, an important economic issue in the historical debate on immigration policy in the United States is whether migrants find their way in the welfare system. He argues that in the US, migrants receive a disproportionately large share of the welfare benefits distributed. Nannestad (2007) states that the effect on welfare state dependency depends on three factors: (i) the age and gender composition of the immigrant group, (ii) the migrants' degree of labour market absorption, and (iii) the entitlement structure of the welfare system. Since migrants tend to be less well-integrated into the labour markets of welfare states offering generous benefit levels, and since these welfare states also tend to have quite lenient eligibility criteria for many welfare programs, migrants should be expected to be a heavier burden on welfare programs, especially unemployment benefits and social transfers, in the more than in the less generous welfare states.

Boeri (2010) finds no evidence that legal migrants, notably skilled migrants, are net recipients of transfers from the state. However, there is evidence of ‘residual dependency’ on non-contributory transfers and self-selection of unskilled migrants in the countries with the most generous welfare states. In a more recent study Barrett & Maitre (2012) find very little evidence that migrants are more likely to receive welfare payments when all payments are considered together. Neither do they find evidence of higher rates of poverty among migrants. It seems fair to conclude that immigration is neither the solution for social security problems (demographic problems for old age pensions) nor, on aggregate level, a major problem for welfare state arrangements (unemployment benefits, social assistance, etc.).

Summary and conclusions

All in all, it seems fair to state that immigration has a relatively small and mild impact on the economies of ‘receiving’ countries. Very little evidence suggests that migration is detrimental to economic success, whereas economic success seems to elicit immigration. Translated to labour market outcomes this means that immigration has limited effects there also: migrants themselves do experience higher unemployment incidence, but immigration itself does not influence the unemployment level in a country a great deal. There is some evidence however that immigration causes higher levels of unemployment (and lower wages) for ‘native’ workers with similar levels of human capital as migrant workers. For the effects of immigration equity (inequality between higher and lower skilled, between immigrants and ‘natives’) seems to be a bigger issue than efficiency (macro levels of employment and unemployment).

Finally, the findings on labour market outcomes translate into rather limited effects on social security, either positive or negative. Current levels of immigration are not sufficient to counter the demographic problems that some European countries have with their old age pensions. And there is very little evidence that migrants are substantially overrepresented in welfare arrangements such as unemployment benefits and social assistance schemes. However, in the INT-AR paper number 2, we concluded from another angle that the cross-border mobility of workers has a serious impact on the flexibilisation of the (Dutch) labour market especially through temporary work agencies. If the result is less contributions to the social security schemes, the impact on the financial sustainability of the welfare state will increase in the long run.

8. Posting of workers in practice

Download this paper

Paper 8 focusses on several aspects of the enforcement experiences of compliance offices and labour inspections in the area of workers, posted in the frame of the free provision of services:

  • Experiences of labour inspections;
  • Fraud practices and mechanisms in posting;
  • Cooperation across institutions and social partners;
  • Redress, effective enforcement and prevention of fraud.

The content is based on European and national research and cooperation projects, initiated by the labour inspectorate, social partners and compliance offices. In order to do justice to earlier research, the paper starts with a short retrospect of findings. The second section summarizes the transnational experiences of labour inspectors in compliance activities. The third section provides a summarized analysis of fraud practices and mechanism in the Netherlands, based on an assessment of the work of the Dutch labour inspectorate in the period 2014-2015. The last section reflects on the problems of redress and follow-up, once the inspection services have examined breaches and fraudulent practices.

It can be concluded that monitoring of posting rules is difficult and that enforcement lacks strong sanctioning. Problematic for all stakeholders in a compliance campaign is the lack of effective sanctions. Fines are weak in an extra-territorial context and most countries have no specific posting-related enforcement instruments. The fact that the tackling of artificial arrangements and of fraudulent cross-border labour recruitment very often comes too late or that these practices can pop up repeatedly, leads to serious frustrations. Competence to deregister establishments lies outside the competences of the inspectorate and social fraud is still not seen as a major offense that could argument a European-wide sanctioning. Therefore, a more horizontal transnational cooperation right across all relevant policy areas is of the utmost importance. Inspectors often trace firms that are active in several countries, using the same methods, whilst their presence in the country of registration is only symbolic. This asks for cooperation, in the control and compliance activities, but also in the enactment and implementation of sanctions.  

A key factor for the achievement of results is concerted action of compliance and enforcement authorities. Special attention should be given to dubious subcontracting practices. Social partners report in several studies the appearance of artificial legal entities, like letterbox companies, that are established for the purpose of subcontracting work to one or more countries. The workers most often work under the direct supervision of the user undertaking, thus creating a situation of bogus subcontracting or illicit provision of manpower. Such artificial arrangements should be properly sanctioned. Effective measures are needed in order to promote genuine operations and prevent abuses. Fake entities should be refused the entrance to the market (such as withdrawal of licenses and certificates or the exclusion from public procurement bids). Sanctions need to have an EU-wide effect in order to avoid non-genuine actors starting all over again in other constituencies. Competences to decide on and to control compliance with the regulatory framework of pay, working conditions, as enshrined in collective agreements and labour legislation, should be more allocated to the country of employment. This asks for a reestablishment of the ‘lex loci laboris’ principle. Free movement of workers will only stay upright if this free movement takes place grounded on the principle of equal treatment in the territory where work is carried out. The competence to check the reliability of documents, which underpin the cross-border activity and, if necessary, to withdraw these documents, must become an EU-competence that can be performed by competence authorities in both the sending and the receiving country.

9. Cross-border recruitment and artificial arrangements

Download this paper

Artificial legal corporate entities and the recruitment of labour

This paper summarises the state of the art with regard to the scientific knowledge on ‘letterbox companies’. This multidisciplinary paper informs about the (possible) relationship between letterbox companies and practices that seem ‘perfectly legal’, but can be, not only from a moral point of view, dubious or unlawful.

Different perspectives – different definitions

Over the years specialists from several disciplines have examined and analysed the notion of letterbox companies. This resulted in different definitions depending on the perspective that was chosen. The OECD provided one of the most commonly used definitions in its Model Tax Convention 2014, whereby a letterbox company is a paper company, shell company or money box company, i.e. a company which has compiled only with the bare essentials for organisation and registration in a particular country, whilst the actual commercial activities are carried out in another country. This definition refers to a business that establishes its domicile in a tax friendly country with just a mailing address while conducting its commercial activities in other countries for purposes of minimising its tax liability. The European Commission similarly stated that ‘letterbox subsidiaries’ are artificial arrangements established in countries solely to qualify for a softer tax regime and cut their bill. The linkage with the free provision of services led to a definition that went beyond taxation. In the debates, critics of the uncontrolled mobility of national service providers referred to the creation of letterbox companies offering services at low prices, which would be able to operate from their registered offices across the whole territory of the EU. Interestingly, the European Commission referred in 2013 to legislative loopholes with a definition that said ‘letterbox companies are companies which have been set up with the purpose of benefitting from legislative loopholes while not themselves providing any service to clients’. In a recent study, commissioned by the ETUC, the definition focuses on a range of issues: a letterbox company is defined as a business that establishes its domicile in a given Member State while conducting its (substantial) activities in other Member States for purposes of circumventing or evading applicable legal obligations (lower taxes, wages, labour standards and social security contributions).

Mobility and freedom of establishment

In general terms, setting up a letterbox company in a foreign constituency is facilitated by the freedom of establishment, as enshrined in Article 49 TFEU. Besides, companies benefit from the internal market principles that guarantee both the right of establishment and the freedom to provide services, cf. Articles 54 and 62 TFEU. Corporate entities are creatures of national law and the rules for setting up companies vary significantly among Member States. Based on the freedom of establishment and the free provision of services these national entities are free to move around and get market access elsewhere in Europe. Thus, there is a real possibility to create artificial arrangements for the purpose of evading statutory and other obligations. The European Parliament asked to put an end to the letterbox companies by ensuring that businesses registered in a Member States are genuine and active ones. The EP suggested to tackle the phenomenon by the creation of an EU-wide list of enterprises, including letterbox companies, responsible for serious breaches of labour and social legislation. However, national laws determine the way registers are organised and the legal value of entries. The EU Council of Ministers so far refused to work towards a central business register.

The notion of the genuine undertaking

There is neither a unified integral and horizontal definition throughout the EU acquis, nor a comparable definition across Member States, nor a definition in identical terms in the different policy areas at national and European level. The most important existing notions are:

- in the field of taxation, the European Commission tabled changes in the Parent Subsidiary Directive, including a general anti-abuse rule, but this was restricted to tax transfers. The Directive says that Member States’ tax administrations, when assessing whether an arrangement or a series of arrangements are abusive, should undertake an objective analysis of all relevant facts and circumstances; an arrangement or a series of arrangements shall be regarded as not genuine to the extent that they are not put into place for valid commercial reasons which reflect economic reality.

- the Regulations for the coordination of the social security (883/2004 and 2009/987) provide criteria for the assessment of the genuine character of an undertaking.

- the Regulation for the international transport (1071/2009) formulates criteria on access to the sector, with provisions to eliminate letterbox firms; for instance Article 3 prescribes that undertakings engaged in the occupation of road transport operator shall have an effective and stable establishment in a Member State; and Article 5 provides a list of conditions relating to the requirement of establishment.

- the Enforcement Directive (2014/67/EU) for the posting of workers gives some guidelines in Article 4.2.

Case law on artificial arrangements

Settled CJEU case-law with regard to domestic measures targeting letterbox companies established in other Member States says that measures that are likely ‘to limit their freedom of establishment or their freedom to provide services’ are not per se incompatible with EU law, but have to be justified by overriding reasons of public interest. And restrictions must be appropriate to attain the objective pursued and cannot go beyond what is necessary. The overriding reasons of public interests, considered by the CJEU, are mostly limited to the prevention of abusive tax practices, for instance in case of ‘wholly artificial arrangements intended to escape the domestic tax normally payable’. According to the CJEU, the finding that there is a wholly artificial arrangement must be based on objective factors which are ascertainable by third parties with regard, in particular, to the extent to which the controlled foreign company physically exists in terms of premises, staff and equipment. Moreover, the CJEU has ruled that a host state may not refuse recognition of the legal capacity of a company incorporated under the law of another Member State, even if the company does not pursue any economic activity in the latter state. In the end, the CJEU requires not only the elimination of all discrimination on grounds of nationality against providers of services established in another Member State, but also ‘the abolition of any restriction, even if it applies to national providers of services’. This policy has eased the possibility for letterbox companies to be created through artificial arrangements in order to circumvent national mandatory rules and obligations.

The impact of artificial arrangements

By the late 1980s, first indications of the practice of bypassing the applicable rules through the use of foreign labour-only subcontractors led to questions about the possible relationship between cross-border labour recruitment and artificial arrangements. The free provision of services by foreign entities resulted in exemption from the host land social security legislation, questionable practices in the field of income and corporate tax and the watering down of national labour standards, mandatory pay and working conditions. The regulatory framework related to the phenomenon of artificial arrangements is stretched over various areas, with incoherent, contradictory or even conflicting rules in company, labour and contract law, internal market regulations, tax rulings and social security legislation. Competences to verify the genuine character of the activities are fragmented and spread over different national institutions. The dispersion and fragmentation of the competence to control and enforce make it difficult to monitor and combat abusive practices and the patchwork of regulations, combined with a lack of enhanced and straightforward cooperation beyond the limits of every separate policy area or discipline, hinders effective actions of inspections and enforcement services. Moreover, legal complexity and loopholes hamper effective application of the law and therefore favour unreliable actors. Unnecessary frictions between these areas of law should be avoided where prevention and combating letterbox companies require a consistent enforcement frame.

The signalled loopholes paved the way for firms and agencies that can be easily, and at low cost, established as a legal entity in a foreign constituency, disappear across the border, go bankrupt and start all over again. It also led to an advisory industry of incubators that can explain how ‘perfectly legal’ the course of action is. It is imperative to strengthen the legal framework and to repair loopholes and inconsistencies in a horizontal and coherent way. This in fact, asks for an impact assessment across a large part of the internal market acquis, not only with the aim to protect workers, but also in the interest of genuine economic actors and customers.

The Project

The Database Labor Migration was created as part of the multidisciplinary INT-AR research project, which, in close collaboration with policymakers, employers, employment agencies and unions, maps the consequences of today’s increasingly internationalized European labor market and transboundary labor recruitment for the Dutch economy, its labor market and social security system.

Project Team

The project team consists of:


  • Internationalizing the labor market
  • Cross-border job recruitment


Kenniscentrum Arbeidsmigranten

The Center of Expertise Labor Migrants (in Dutch) gathers, combines and enriches knowledge and expertise on the position of migrant workers on the labour market, in the economy and in society.

Instituut GAK

Instituut Gak (in Dutch) aims to contribute to the quality of social security and the labor market in the Netherlands by investing in social projects, research and chairs.