What is driving rates of social policy preliminary references to the CJEU? Evidence from the United Kingdom and France

Abstract

Preliminary references to the Court of Justice for the European Union are unevenly distributed across the EU, creating differing access to justice for European citizens. This study presents case studies of the UK and France, exploring factors affecting rates of social policy preliminary references from 1996–2009. The UK had a rate twice that of France. What accounts for this difference? Analysis of documentary evidence and 25 expert interviews help to explain the differing rates. Themes were related to policy, structural factors and the agency of actors. In the UK, policy themes are the free movement of persons and the ‘Right to Reside’ test. Legal aid and legal NGOs help individuals access the Court and drive test case strategies. In France, a high degree of dualisation in the welfare state creates an insider/outsider dynamic. Coupled with the resistance of courts and a lack of comparable actors to drive preliminary references, this contributes to a lower rate of references.

Keywords:

• CJEU
• social policy
• preliminary references
• UK
• France
• courts

Introduction

The Court of Justice for the European Union1 (‘Court’) has been an important driver of EU integration. The legal order of the European Union is an interlinked network of European and national courts, all effectively functioning as European courts. Through this system, most of the costs of enforcing EU laws are shifted onto the legal systems of the Member States. Enforcement within the Member States happens largely through the actions of litigants and national courts, essentially outsourcing governance (Kelemen 2008).

The reach of the Court is limited to the cases that are brought before it. Its caseload is comprised of enforcement actions brought by the Commission and preliminary references from the national courts of the Member States. National courts thus function as ‘agenda-setters’ in the European legal system (Chalmers, et al. 2010, p. 158). However, preliminary references are unevenly distributed across the Member States of the EU, and some Member States' preliminary references are concentrated in particular policy areas.

How do we account for the different usage of the European legal process across the Member States? This paper presents part of a larger project completed for the author's doctoral thesis, which sought to explain variation in rates of social policy preliminary references2 across the member states of the EU-15 from 1996–2009. The study began with a quantitative analysis, which was followed by a Qualitative Comparative Analysis (Ragin 1987, 1994). Finally, two case studies were conducted in the UK and France. These case studies helped to unpack the complexities of how the preliminary references process functions in the Member States.

Explaining Variation in Rates of Preliminary References

Potential explanations for variation in national rates of preliminary references have been much theorised (See, e.g. Alter 1996, 2000, 2008, Burley and Mattli 1993, Carrubba and Murrah 2005, Cichowski 2007, Golub 1996, Mattli and Slaughter 1998, Stone Sweet and Brunell 1998, Stone Sweet 2004, Vink, et al. 2009, Weiler 1994, Wind et al. 2009). Because most prior studies have not been specific to an issue area, the explanations are largely general and relate to the legal system as a whole. Factors considered have included what Alter calls the ‘friendliness’ of a legal system to litigation (2000); judicial review (Alter 1996, Mattli and Slaughter 1998); the effects of exposure to the EU (Stone Sweet and Brunell 1998); ‘containment’ by the preferences of political and judicial actors (Conant 2002); and the power of organised interest groups (Golub 1996, Cichowski 2007, Alter 2008, Slepcevic 2009). Although these more general factors play an important role in determining rates of preliminary references, they do not address the effects of policy misfit, which has been identified as an important factor in the Europeanisation of domestic policy (Börzel 1999, Risse, Cowles and Caporaso 2001). The more general factors also do not explain why references from some Member States are concentrated in particular issue areas.

By focusing on preliminary references in the area of social policy, it was possible to consider the effects of underlying policy fit on rates of references. Social policy is an important area for EU law. The Court has been a driver in the increasing Europeanisation of social policy. It has chipped away at Member State control over who has the right to receive services under the national welfare state, where those services may be consumed and who can provide those services (Leibfried and Pierson 1995). The role of the Court is essential to a governance-based account of the development of the European Union social dimension, especially in terms of regulatory policy-making (Majone 1993, 1995).

The findings of the quantitative and QCA elements of this study support the importance of national legal culture, judicial activism from lower level courts (Leibfried and Pierson 2000; Scharpf 2010); and compliance with the EU (Falkner, et al. 2005; Falkner and Treib 2008). In terms of policy misfit, Bismarckian welfare states3 had higher rates of social policy preliminary references. This reinforces the findings of Martinsen (2005a) in the her study of cases in the area of social security for migrant workers. The percentage of the population that belongs to a union (‘union density’) was also important, confirming the importance of the organizing power of interest groups (Alter 2008; Cichowski 2007; Golub 1996; Martinsen 2005b; Slepcevic 2009). The case studies presented in this paper were used to inform and explain these findings.

Methods

The theoretical argument of this study is that the different usage of the preliminary reference procedure arises from the institutional structures of the national legal system and of the welfare state, as well as because of the agency of actors within the national system. France and the UK were chosen through purposive sampling because they provide the best insight into the research question. They are alike enough to be comparable but have differing rates of referral. The UK has a rate of social policy preliminary references twice that of France. The UK and France are comparably sized and have similar long histories with the EU. However, they differ in key ways in their legal systems and welfare states.

Evidence was collected via documentary analysis and 25 expert interviews. A starting point for the documentary analysis was the cases that comprise the preliminary references sample. There were 27 relevant preliminary references from France, and 69 from the UK in the period. Additionally, searches were conducted in national media, legal and academic databases in order to identify relevant documents and news items.

Then, 25 semi-structured expert interviews were conducted with lawyers, interest group officials, court officials and legal experts familiar with the history of preliminary references in the Member States. 8 were conducted in France, 11 in the UK, 3 with officials at the Court of Justice, and 3 with officials associated with the Commission in Brussels.4 Participants were offered anonymity. Those who are named in this article explicitly consented to be named. The circumstances of those participants who want to remain anonymous have been masked to ensure that anonymity.

The resulting data were hand-coded for themes and sub-themes. The themes fall into the three types of explanatory factors. First, there are themes related to the underlying policy structure of the Member State. Second, there are themes related to structural factors of the national legal system. Third, there are themes related to the actions of individuals and groups.

Results for the United Kingdom

Policy Structure Themes

Free Movement of Persons/ European Citizenship

The UK is classified as a low-spending Beveridgean welfare state (Bonoli 1997). It provides low-level, often means-tested, support. Apart from meeting any means test requirements, eligibility to receive benefits is based upon UK residence.

The UK was also one of only three established EU Member States that permitted nationals of the new accession states to move to the UK for work after 2004. It was at that point that the net inflows and outflows of EU migrants into the UK made a large shift well into a positive balance, after years of being at or near stasis.

Figure 1 Net Migration from the EU to the UK, 1991–2010 Source: Oxford Migration Observatory 2012. www.migrationobservatory.ox.ac.uk

These elements lead to the first UK policy theme, free movement of workers/persons and European citizenship. 21 of the 69 cases from the UK were in the area of free movement of workers, European citizenship or both. Which migrants are able to use European rights and to what benefits they are entitled have been very active questions. These have been really significant references from the UK, which have shaped the EU citizenship debate. These references contributed to the development of the jurisprudence gradually extending the free movement of persons. The right to free movement has been interpreted to include derived rights for third country nationals (TCNs) based upon their relationship to EU nationals. In the process, tiers of rights have developed according to citizenship status, with the broadest rights for nationals of the pre-2004 EU Member States, followed by nationals of the A85/A26 Member States that joined the EU in the 2004 and 2007 accessions, and then third TCNs.

This tiering is significant for determining welfare claims. Whether and to what extent migrants are able to access a national welfare state has large financial implications. Paul Eden, who co-represented the plaintiff in the landmark Collins case concerning access to Job Seekers Allowance (JSA) for unemployed EU migrants said:

I was as aware as everyone of the potential financial dimensions to this case. The courts did not want to put the financial burden on the state. If everyone coming to the country for the first time had the right to seek JSA that has obviously a huge potential cost. There were Home Office lawyers sitting there in the phalanx of government lawyers, monitoring the implications for asylum and immigration (Interview, 15/09/11).

Mr Collins had been refused the JSA on grounds that he was not habitually resident in the UK. In the Collins case, the Court found that the UK's requirement of a genuine link with the labour market constituted indirect discrimination, because UK nationals will be able to fulfil the requirements more easily. However, it was justified because the JSA had been designed to address unemployment for those living long term in the UK.

The Child Poverty Action Group (CPAG) is an advocacy organisation using litigation as part of a broader strategy to end child poverty. In addition to EU migrants to the UK, their strategy has also benefitted UK migrants to the rest of the EU. CPAG has been involved in cases about UK nationals' right to export their benefits from the UK when they move within the EU. Sarah Clarke, a solicitor at CPAG, said:

That's become quite a big issue since the European Court of Justice made a decision that one of the UK's major disability benefits and the carer's allowance … are sickness benefits, not special non-contributory benefits. There are different categories of benefits that are basically treated differently in European law and special non-contributory benefits, if you move within the EU you can't take them with you, but sickness benefits you can take with you (Interview, 02/09/11).

The EU rights to free movement were grounded in the free movement of workers. This strong nexus between the right to residency and economic self-sufficiency limits who is able to access these European rights. As Clarke noted that:

I think it's striking that the people who tend to lose out under European law, the people who don't get their rights recognised, tend to be women, and often women with children, because they fall out of the movement of workers, which is all based upon worker status. (Interview, 02/09/11).

Right to Reside

The first sub-theme involves the Right to Reside Test. It was created in May 2004, in response to concerns about the EU enlargement (House of Commons Library 2011). It applies to applicants for most benefits. Under the Right to Reside Test, other EU nationals have the right to reside so long as they remain ‘qualified’ persons or are the family member of a qualified person. According to UK immigration regulations7, a ‘qualified person’ is an European Economic Area (EEA) national who is in the UK and is: a jobseeker, a worker, a self-employed person, a self-sufficient person, or a student. Who qualifies for these sub-categories has been the subject of much litigation.

The European Commission commenced formal infringement proceedings against the UK in 2010 over the application of the Right to Reside Test.8 On 29^th September 2011, the European Commission announced that it had sent the UK a reasoned opinion that the Right to Reside Test is in contravention of EU law. By requiring non-UK citizens who are EEA nationals to pass the Right to Reside Test, the UK indirectly discriminates against those EU citizens.9 Chris Grayling, the Minister for Employment, has said that the UK has formally rejected the Commission's opinion, ‘in the strongest possible manner.’10

The Supreme Court previously had considered the test in March 2011 in the Patmalniece case11, rejecting the argument that the test directly discriminated against EU nationals, but agreeing that it constituted indirect discrimination. However, the Supreme Court found that the indirect discrimination was justified as a proportionate means of achieving a legitimate aim, namely protecting the public purse.

CPAG run a welfare rights advice hotline. Two-thirds of all calls to the hotline are for assistance with the Right to Reside Test. It is a convoluted area and many are ill informed. One welfare rights adviser said that:

Most clients only know as much as the DWP/HMRC/Local Authority tells them. This information is usually dispensed by frontline staff who only have a cursory understanding of EEA nationals' rights … In particular, we find that clients are often advised to claim income support … and are refused because, by doing so, they become economically inactive … and lose their ‘right to reside'’ (Interview, 07/09/11).

A further complication is that the Member States may discriminate against newer Member State nationals for up to seven years after the date of accession. Although the UK allowed nationals from the newer Member States to relocate to the UK for work, a Workers Registration Scheme (WRS) was instituted for A8 and A2 nationals, requiring them to be in registered paid employment for 12 months before they were eligible for benefits. The WRS ended in May 2011 for A8 nationals, but was extended until December 2013 for A2 nationals.

Rights of family members of UK or EU nationals

The next sub-theme is the derived rights of family members of UK or EU nationals. Some of the most important European citizenship cases have recently arisen from the UK in this sub-theme, including the Baumbast 12 case. In Baumbast, the Court found that a German immigrant to the UK was entitled to remain because his children were in education, even though he effectively had ceased to be economically active by becoming employed outside the EU. More recently, in the Teixeira and Ibrahim 13 cases, the Court found that subsequently TCNs may claim a right of residence as the primary carer of an EU child in education in the Member State even though they are not economically active. CPAG has been granted a preliminary reference in the Punakova 14 case, about whether previously self-employed A8 nationals have Baumbast-type rights of residence. This is illustrative of the process by which the freedom of movement has been extended via derived rights.

There is confusion about who qualifies as a family member. A welfare benefits adviser described a case where the client, a Dutch national living with her daughter, was refused Pension Credit despite having a right to reside as an extended family member of her daughter who was a worker. (Interview, 07/09/11).

The process by which the freedom of movement for persons has been extended through derived rights has created uncertainty as to the limit of the right. As different claimants present, their specific factual situations have had to be addressed via litigation. This has resulted in increased preliminary references in the area of free movement of persons.

Anti-discrimination

The second major policy structure theme is anti-discrimination cases. This reflects the fact that there was a misfit between the UK's policies in this area and Europe in some ways. Most of the cases address matters at the edges of rights and protections. Although later than other Member States, notably France, anti-discrimination had been extensively legislated in the UK. This earlier UK anti-discrimination was consolidated into the Equality Act 2010, which closely follows the EU Equal Treatment Directives15.

There were 19 discrimination cases in the UK sample. Seventeen of those concerned equal treatment of men and women. The majority of these cases were brought on behalf of women. Two were brought on behalf of both men and women, and two were on behalf of transsexuals.16 Despite the protections afforded in UK law, examples of discriminatory treatment are still found. Christine Boch, who was involved with the Brown 17 case on pregnancy discrimination, said that at the Citizens' Advice Bureau, ‘We do still have loads of people who are the victims of the most blatant form of, for example, pregnancy discrimination.’ (Interview, 23/09/11). In her experience, it happens primarily but not exclusively in two sectors – catering and cleaning.

There was also an age discrimination test case from the UK, the Age Concern case, challenging the UK's default retirement age. The Court found that a default retirement age could be justified if it was a proportionate means of achieving a legitimate aim. However, the aims of the strategic litigation went beyond the immediate result. Andrew Harrop, formerly of Age Concern, said that, ‘We wanted to get a tight interpretation of 'objective justification,’ in order to show that the prohibition of age discrimination could only be disregarded in special circumstances' (Interview, 07/09/11). Moreover, although the initial result was a setback, the legal challenge worked in the long run:

It was an interesting example of the usefulness of litigation in a political lobbying context. The ECJ opinion was somewhat helpful in that it left some room for favourable interpretation, but it guided the High Court to dismiss the case. However, over the course of the case the issue had garnered huge interest and entered the public debate. It turned around stakeholder views … We lost the case in the High Court in the autumn of 2009, but going into the 2010 General Election we had all three parties committed to ending the practice, against what remained pretty robust lobbying from the CBI (Interview, 07/09/11).

Finally, in the Coleman 18 case the Court agreed that discrimination against an employee because she is the carer for a disabled person is in violation of the Equal Treatment Framework Directive 2000/78/EC (‘Framework Directive’). Although associative discrimination was not covered on the face of the Disability Discrimination Act 1995 (DDA), the claimant argued that the DDA should be construed in light of the Framework Directive. Beyond the immediate result, the Coleman case was very significant for UK anti-discrimination law. Prior to the case, associative discrimination had not been included in the draft legislation for the Equality Act 2010. After the decision, the government announced that it would ‘extend the prohibition against associative and perceptive direct discrimination and harassment to other strands and areas where this does not currently apply.’19

Structural Elements

The next set of themes relates to the elements of the UK legal system that make it friendly or unfriendly to litigation to enforce rights. The relatively smooth functioning of the judicial system for administering EU law related to social policy was affirmed by this study. Some of these elements have benefitted references, some were barriers.

Structural Benefits

Acceptance of courts

Participants reported that UK courts are generally accepting of EU rights and courts are open to preliminary references. Liz Barratt, a solicitor, said that, ‘UK courts are very receptive to preliminary references. On immigration issues, the courts are very prepared to sort of knock it off to Europe’ (Interview, 15/09/11). There are differences by courts. Sarah Clarke reported that, ‘The Upper Tribunal [Social Security] has become quite good about referring … the Court of Appeal is probably a bit more hostile’ (Interview, 02/09/11). Liz Barratt said:

Once you get to the higher courts then the judges and barristers are more familiar with EU law. Lower courts may be nervous about sending a reference off, compared to the Supreme Court. I know people who have asked at the tribunal and been bounced up and only got their reference higher up (Interview, 15/09/11).

Victoria Phillips, solicitor for the Stringer 20 and British Airways v. Williams 21 cases, said that in her experience it is easier to get a reference from a higher court. She did note a few cases that had gone directly to the Court from the tribunal stage, ‘but I think it's a bold employment judge who's prepared to go for a reference (Interview, 16/09/11).

Some reported that there used to be more resistance to EU law in UK courts. In Advocate General Eleanor Sharpston's experience as a barrister before 2006, ‘when I was in practice, resisting a reference on behalf of the government was relatively easy for experienced counsel’ (Interview, 08/09/11). She said that:

It used to be almost a dirty word in English, you know, if you've got nothing else you run the Euro Defence… Obviously, if you were running an EU law point then it was because you had nothing better to say. In effect of course, that was a real caricature (Interview, 08/09/11).

Employment and Social Security Tribunals

The second structural benefit in the UK legal system is the tribunal system of first instance courts that litigants can access with few barriers. The tribunals for employment and social security cases were cited as important to promote social policy cases. These systems handle an incredible volume of cases. There were 557,100 cases in the Employment tribunal and Social Security and child support Tribunals in 2011–12.22 In the Preston 23 case, the Court noted that after its earlier judgments about the rights of part time employees to join pensions systems, ‘some 60,000 part-time workers in the United Kingdom in both the public and the private sector commenced proceedings before industrial tribunals’ (para. 17).24

Although plaintiffs do not need a lawyer to access the tribunals, the tribunals will also sometimes act to ensure that plaintiffs have better representation as the cases move up through the system on appeal. Sarah Clarke of CPAG said that, ‘Sometimes the Upper Tribunal refers cases to us, if the claimant isn't represented and it's a complicated issue’ (Interview, 02/09/11). Paul Eden, the co-counsel in Collins, which rose up through the social security tribunal system, discussed how he suddenly found himself out of his depth when the Commissioner decided to send a preliminary reference:

Neither the government nor our side said ‘wouldn't it be a good idea to go to the ECJ.’ We weren't lawyers at that level … It was an entirely different matter to go off to the ECJ, rather than the Commissioner in London (Interview, 15/09/11).

However, help came from the tribunal, which put them in touch with potential pro bono legal support (Interview, 15/09/11).

Legal Aid

A Judge at the Court stated that ‘of course you need finance,’ in order for cases to be brought (Interview, 08/09/11). The availability of legal aid was mentioned often as a benefit to preliminary references.

Legal aid only applies in some types of cases. There is no legal aid in the employment tribunals or for third party interventions. Legal aid is only available to persons, so organisations have to find funding for strategic litigation. There is some limited legal aid available at the Court for indigent litigants. Paul Eden said that, ‘I should say that the Court was tremendously helpful…. They were very good about the legal aid process’ (Interview, 15/09/11). He described support to understand the processes of the Court, as well as financial support.

Threats to legal aid were frequently mentioned as a potential problem. Sarah Clarke said:

There's a network of advice workers and advice agencies in the UK, so there are Citizens' Advice Bureaux, Law Centres, local authorities might have welfare rights units, and there are independent advice agencies as well, all of whom might refer cases to us. … But we don't know quite what will happen about that, because the government is proposing to take legal aid away for social welfare cases, which would devastate the advice sector. I would think it would be catastrophic. I think legal aid is quite an important part of how the entitlement system works in the UK (Interview, 02/09/11).

The proposal mentioned by Clarke at the time of this interview has materialised in the Legal Aid, Sentencing and Punishment of Offenders Act 2012 (LASPO), approved in April after 14 defeats in the House of Lords. The legislation makes deep cuts to the provision of legal aid services for social welfare from April 2013. It will eliminate advice provision to approximately 500,000 individuals annually by the Government's estimates. By one legal aid group's calculation, legal aid income to non-profit agencies will be reduced by 92% (Smith 2011). These cuts will be implemented over two years. As legal aid was cited frequently as important for the advancement of preliminary references in the area of social welfare, these cuts have major implications for the ability to enforce EU social rights.

Structural Barriers

Money is the biggest identified structural barrier. There were two different issues. First, it is expensive to bring a test case. Victoria Phillips identified the availability of finance as crucial, listing the sources of funding for some key UK preliminary references:

You'd have to have some big funding. It's interesting to look at those cases that have been referred. My colleague… was involved in the Preston 25 litigation. That was … all trade union-funded. We had another colleague who had the UK BECTU 26 reference. Now that went straight from the High Court to the ECJ, and again that was trade union funded. … EHRC or the EOC have funded a few cases, but I can't think of anything in the employment or industrial relations world that has gone without funding. Coleman 27 was EOC-funded (Interview, 16/09/11).

However, she pointed out that it would be considerably less expensive if the reference were made by the lower courts to the Court without having to appeal a case up to the higher courts, ‘it's not costly - other than the costs of travel - it's not costly to make an application to the ECJ…. There's less documentation, there's less fuss.’ (Interview, 16/09/11). Nicola Smith, a solicitor in Scotland, said, ‘Money is another big barrier. You're always relying on someone to do it pro bono’ (Interview, 16/09/11).

The second element of funding barriers is the possibility of a large award of costs against the losing party. In some cases lawyers have managed to agree to a protective costs order, essentially agreeing that the litigation is in the public interest and therefore the costs will not be passed along to the losing party. However, barring that, the prospect of bearing all of the costs for an expensive action is a deterrent. Andrew Harrop, formerly of Age Concern, said:

The main challenge for us as a litigant was that the government refused to agree to a protective costs agreement. All along we were more afraid of an award of costs against us than we were of our legal fees, because costs would be so unpredictable. At one point we had two QCs against us. (Interview, 07/09/11).

Nicola Smith also identified potential costs as a barrier, particularly in light of the sorts of cases that result in preliminary references. ‘There's always a substantial risk of losing and getting stuck with all the costs. In these cases there's always a genuine disagreement as to the area of law. It could go either way, usually’ (Interview, 14/09/11). Roger Smith of JUSTICE has identified the fact that costs are not awarded in Employment Tribunals as a benefit to strategic litigation arising from the system (2003). However, as legal aid is not available in the employment tribunals, this may balance out this benefit.

Actors

NGOs, the Equality and Human Rights Commission (EHRC), its predecessor commissions and unions are crucial actors to advance preliminary references. They help to overcome the barriers to social policy preliminary references.

Test case strategy

For these actors, support for strategic litigation, or test cases, is part of an organised plan. CPAG's strategy is well-developed and they openly solicit on their website for the fact patterns for test cases that they're seeking. Sarah Clarke said that the test case strategy was one part of the organisation's broader efforts to lobby for an end to child poverty in the UK (Interview, 02/09/11).

Some legal NGOs run training sessions or advice lines for client advisers on European rights. Nicola Smith collaborated with other disability rights NGOs in other Member States to write Legal Strategy: A Good Practice Guide for Lawyers:

We had identified that there were very few cases on people with learning disabilities in the Member States involved and none at the European Court. We looked at the types of cases that it would be useful to have in the employment field. We also considered how NGOs could help with access to justice because it is difficult for this particular group to access courts, more so than for other groups. (Interview 14/09/11).

Paul Eden said that both sides in test case litigation are looking for favourable cases. ‘The government is looking for their test cases and CPAG is actively seeking their test cases. Both sides are looking for the best facts they can’ (Interview, 15/09/11).

Organisations often have larger purposes behind test cases. In addition to the legal result desired, they seek publicity for a cause or to exert political pressure. One participant said of the Age Concern case:

It was primarily there to change the world, and to change the situation with regard to retirement, but actually it had another very powerful benefit, to show Age Concern and Heyday as champions of older people, particularly of the baby boomers, who were coming up to an age where they could be booted out just because of their age' (Interview, 02/09/11).

Several types of groups emerged as important. NGO support for litigation can come from either specialist legal NGOs that regularly take on strategic litigation or from general NGOs that provide support for an occasional strategic case. Within the legal NGOs category there are also community law centres, which occasionally take test case strategies. Unions also provide support for test cases. Victoria Phillips, who has had two union-funded preliminary references, said ‘This is all entirely strategic litigation’ (Interview, 16/09/11). Finally, the EHRC and its predecessor the EOC have provided support for cases.

The impact of these actors who have experience with the system is very important for preliminary references, especially considering the frequent inequity between the experience levels of the government and plaintiffs. Having the support of a pressure group can level the playing field. Paul Eden noted that, ‘There's the socio-legal theory about one shot versus repeat players Brian [Collins] was a classic one-shot player. CPAG and the Government were repeat players’ (Interview, 15/09/11).28

Results for France

Generally speaking, there was less data available in France. Although media and academic coverage of the Court grew in France over the period studied, it was still much lower than in the United Kingdom. There were fewer potential interviewees, and they tended to have had isolated experiences with the Court. The cases and expert interviews did not reveal equivalent legal NGOs with test cases strategies in the time period studied.29

Policy Structure themes

Dualization of the welfare state

The first policy theme for France is that France's strongly dualized welfare state does not generate circumstances conducive to social policy preliminary references. France has a ‘dual social protection system.’ Its primary component, the Sécurité sociale, is a social insurance system financed through employment-related contributions and providing benefits linked to contributions. A secondary component, the Solidarité nationale, is a non-contributory scheme ‘generally designed to cater for those who have been unable to build up an adequate contribution record (Bonoli 2000, p. 123–124). France's social insurance system had the largest proportion of financing from employer and employee contributions in Europe (Bonoli and Palier 2001). The social insurance funds that manage these contributions are relatively independent from the state. Through the principle of ‘management by interested parties,’ unions, employers and mutualist representatives were granted joint management (Lallement 2008, p. 54). Unions are involved in managing these funds and as such have ‘a de facto veto power against welfare state reforms’ (Bonoli and Palier 2001, p. 339.) ‘The trade unions act as the representatives and defenders of the systems. They defend both the interests of the salaried population and their own interests (Palier 2008, p. 111). The unions have become integrated into institutions that have a ‘broader constituency’ than just union members, and the unions increasingly must intermediate between the short-term interests of their membership and the broader general interests (Lallement 2008, p. 61).

The French welfare system is generous. Hall notes that, ‘at 53 percent of GDP, France's public expenditure has reached Scandinavian levels, although the redistributive impact of its tax and transfer systems is more meagre’ (2008, p. 8). The system is regressive, and most social benefits go to the richest half of society (Smith 2004).

The contrast between the traditional Bismarckian system of social insurance, and the newer system of basic social protection is increasingly leading to welfare ‘dualization’ and ‘two distinct worlds of welfare’ (Palier 2010, p. 96). The national solidarity scheme consists of healthcare, family benefits, and policies against social exclusion (Palier 2010, p. 96). Eligibility is based upon citizenship, benefits are either universal or means-tested, and the funding is through taxation (Palier 2010). About one-third of the French population does not participate in the ‘normal’ labour market and social insurance arrangements, and therefore is covered by this second world. (Palier 2010).

In 2001 Palier noted the increasing distance between the social insurance schemes, which were becoming less solidaristic, and the emerging state-financed health care, family policy and social minima policies, which were increasingly subject to means tests and activation principles geared toward getting recipients into employment. Smith, in considering why it has been so difficult to reform the French welfare state, answered that, ‘a key reason is that so many comfortable people resist change’ (2004, p. 2). This may perhaps also be revealing of why France has so few social policy preliminary references.

Reverse Discrimination

France already had laws providing for gender equality, at least at the level of grand principles, before the European legal order began. It was at the insistence of France that Article 119 (now Article 157 TFEU) on Equal Pay was inserted into the Treaty of Rome (Cichowski 2007). However, the EU has been important in ensuring a more strict application of rights. In an interview, the President of the former French equality agency, the Haute autorité de lutte contre les discriminations et pour l'égalité (HALDE) said that they had quickly noted that, ‘There were not enough complaints from women on the grounds of gender and pregnancy.’ (Interview with Equinet Europe 2012). The second policy theme identified is in line with this observation, as reverse discrimination cases were important in the period studied. None of the five cases in the area of equal treatment of men and women in France was brought on behalf of female plaintiffs. Every case involved men challenging more protective treatment for women.

The two most important cases were the Griesmar 30 and Mouflin 31 cases, with judgements released in close succession at the end of 2001. In Griesmar, a male magistrate challenged provisions in the scheme that granted service credits to female civil servants for each child but not to male civil servants. In Mouflin, Mr Mouflin wished to retire and claim his pension early in order to care for his wife, who had a terminal illness. However, the pensions scheme only allowed female civil servants to retire to care for an invalid spouse. In these cases the Court confirmed that pensions provided under the French civil service retirement scheme fell within the scope of Article 157 (ex Articles 141 and 119).

After Griesmar there was more public awareness of the Court of Justice and EU rights in France (Interviews, 22/08/11 and 24/08/11). There was a proliferation of newspaper and other media coverage of Griesmar, in particular, and responses within France to the Court's decision. They triggered a series of online commentaries, with participants in the discussions discussing potential strategies to enforce their rights or to challenge discriminatory policies.

France continues to have gender discrimination issues for women. In May 2012 the Conseil Constitutionnel repealed the new law on sexual harassment, forcing the cancellation of all ongoing prosecutions under the prior law. There were demonstrations in the streets of Paris. It will likely be a delay of several months before new legislation can be adopted. Furthermore, speaking of gender discrimination, one Court official said:

Particularly in France, it was very difficult to make French judges understand what indirect discrimination is. Because they have a very formal understanding of equality in France, and indirect discrimination is a much more substantive concept. So it really took years to make them understand and to get the concept into the writings (Interview, 08/09/11).

Opening up the French Civil Service/ Protected Occupations

The third policy theme is challenges to the closed French systems for the civil service and French protectionism for certain occupations. Access to the French civil service is tightly controlled and regimented through national examinations and the elite French universities.

Five of the cases were related to efforts to open up the French civil service, or other protected occupations, to nationals of other Member States who have credentials from their home Member State. One such case was that of Isabel Burbaud32, a Portuguese national, who wished to become a public hospital administrator in France based upon her qualifications in Portugal. The Court held that she could not be required to pass the public competitive examination for this position in the French civil service. In another case, Josep Peñarroja Fa challenged a decision denying him enrolment in the court registries of translators. He believes that the restriction is related to nationality.33 He successfully brought a test action, although at the time of interview he was still awaiting the response of the French authorities to the Court's decision in his favour. He noted that one likely response was for the French authorities to now say that the register is full (Interview, 25/09/11). The controlled French system for its civil service and professions conflicts with the opening logic of European integration.

Structural Themes

Resistance of Judges and Courts

Resistance of French judges and courts to preliminary references specifically, or European rights and law generally was the first theme related to the French legal structure. Every lawyer reported resistance to preliminary references by French courts as a barrier. A Court official said that, ‘I think what you need at the national level is of course judges who are willing to start a preliminary reference procedure. What is generally well known is that if they can avoid it, they often prefer to avoid it’ (Interview, 08/09/11). Another respondent said that it was ‘absolutely not’ his experience that French courts were open to the idea of European law or European rights (Interview, 25/09/11). Jean-Edouard Robiou de Pont, the counsel in the Wood 34 case, said that in his experience it was ‘very rare’ for a French court to agree to send a preliminary reference that had been requested of it.

Respondents also commented on a resistance to European law among French judges and courts. Mr. Robiou de Pont said, ‘I think that French judges think that a lawyer only uses the European rights if he has nothing serious to say’ (Interview, 23/08/11). Another participant said, ‘The judges do not like European law’ (Interview, 24/08/11). A variant of a common response was, ‘They wish to keep cases in France’ (Interview, 23/08/11). Philippe Derouin, a lawyer who has brought three preliminary references to the Court, including one in which he was the named party,35 noted that, ‘I think that the Paris administrative court, at the lower level and the court of appeals, pride themselves on never having referred a case to the ECJ’ (Interview, 24/08/11). One lawyer remembered that this was not a new trend, ‘even in the old days the Conseil d'État refused to acknowledge that there were EU competencies for a long time’ (Interview, 22/08/11).

Mr Derouin noted that there are two routes to the Court: to file a case and hope for a referral, or to make a complaint to the Commission. If the Commission takes it seriously enough it will start an infringement procedure (Articles 258–260 TFEU). However:

In France, in the areas that I know better, there are almost as many infringement procedures as there are referrals. Whereas in countries like the UK, Germany or the Netherlands, there are three times as many referrals as there are infringement procedures (Interview, 24/08/11).

This ratio suggests that there is much more resistance in the French system to preliminary references than in these other Member States.

Mr Derouin described a lengthy legal process for preliminary reference in which he was the named plaintiff, at the end of which, ‘because this had been lasting for ten years, we could convince the tribunal de sociale sécurité that there really was an issue’ (Interview, 24/08/11). However, the tribunal did not make a referral directly to the Court. It instead referred the case to the Cour de Cassation, which also did not refer the case, although it noted that there was an EU question. The Cour de Cassation referred the case back to the tribunal, which finally sent the preliminary reference to the Court (Interview, 24/08/11).

Wilful misinterpretation by courts

This resistance to EU law also manifests in the way that French courts interpret and apply EU law. Participants observed many instances of French courts interpreting EU directives in opposition to Court precedent. Mr Derouin observed that:

One of the ways of the French court to resist is to disobey themselves, and the administrative courts tend to decide in favour of the administration. They tend to give a neutralising interpretation to European law or the European legislation or to the Treaties (Interview, 24/08/11).

Another interviewee said that:

The French judges do what they want with the text. There are famous cases of the French courts, where the judges say that it is obvious what we must do based upon European law. But if you look at what they have said, it is nothing like the law. It is very different (Interview, 23/08/11).

Improvement over time

In the second major structural theme, the interviewees indicated that the situation is improving. Several respondents said that the French legal system had become more open to European law and rights over the past decade. Mr Derouin attributed this change to the leadership of the Conseil d'État, stating that it ‘showed [lower courts] the way’ (Interview, 24/08/11). Seven of the 23 cases in this sample were referred by the Conseil d'État. Another participant said that:

Certainly the Griesmar case has triggered some changes in the litigation in France. And I feel like there is a lot more understanding of French law and EU law in France … I would say that in the last ten years it has changed a lot (Interview, 22/08/11).

Actors

Lack of union or NGO support for litigation

In France there were very few of the actors that were identified in the UK case study. France has the lowest union density, or percentage of the population that belongs to a union, in Europe. Moreover, in the broader study union density was serving as a proxy for having interest groups willing to intervene to enable cases to get to the Court to enforce rights. Because French unions are so heavily interlinked with the management of French social insurance funds, unions are an unlikely source of support to challenge the system. Most of the French cases challenging social insurance provisions were brought against the social insurance funds, meaning that the interests of the unions were likely to be on the opposing side to the plaintiffs. There is very limited assistance available to individuals wishing to access the French legal system. This makes it difficult to overcome the substantial barriers in the French system.

There was briefly an agency that could intervene on behalf of individuals in anti-discrimination matters. In May 2005 the French equality agency HALDE came into being. However, its staff was ‘ridiculously small’ and by aggregating all types of discrimination it ‘dilutes each issue’ (Guiraudon 2008, p. 146). In 2006 its powers were expanded and it gained a statutory ability to be a party to cases. Louis Schweitzer, President of the HALDE, said that the number of complaints received by the agency had grown from 1,400 in 2005 to 10,549 in 2009 (Interview with Equinet Europe 18/11/11). The largest group of these complaints was in the area of discrimination based upon origin. However, 70% of the claims processed by HALDE in 2009 proved to be outside of its areas of competence. In the interview, Mr. Schweitzer discussed a problem of interest:

The HALDE considered that the fact that children's access to family benefits was attached to the regularity of their right of residence was discriminatory. The HALDE considered it was contrary to articles 8 and 14 of the European Convention on Human Rights as well as article 3 of the International Convention on Children's Rights. Many courts echo the HALDE's analysis and cancel the payment refusals. However, despite the repeated recommendations of the HALDE, at this very stage, the government will not amend the law on this issue despite it being clearly contrary to the international commitments of our country. This illustrates the limits of our powers (Interview with Equinet Europe 18/11/11).

It is striking that the HALDE was so powerless to effect change in one of the major issue areas that has generated most of the UK case law – being legally resident for benefits purposes. On 1^st May 2011 the HALDE was dissolved and its functions were incorporated into another, more general, organisation, the Défenseur des droits (DDD). The structure and remit of the organisation is still developing.

Without legal NGOs or unions or an equality agency to intervene as ‘repeat players’ (Galanter 1974) to help overcome the structural barriers on plaintiffs' behalf, individuals become crucial to the success of claims. However, without the benefit of being an elite or a very determined lawyer, individual plaintiffs seem to have difficulties accessing the system.

Importance of individuals or elites

A key theme in the category of actors is the necessity of having individuals who will push for change through litigation. Without these protagonists it is unlikely that preliminary references will happen. This was highlighted by a Judge at the Court, who said, ‘For certain important changes, you need people who pursue that aim with much litigation and force. If those people are not there you can forget about it, I think’ (Interview, 08/09/11). One French lawyer said that most people would not consider taking a test case, as ‘That requires really serious people’ (Interview, 22/09/11). As an example of the determination required, Philippe Derouin had tried at least once or twice a year for ten years to obtain a preliminary reference in a tax area before he was successful (Interview, 24/08/11). Jean-Edouard Robiou de Pont, the counsel in the Wood 36 case, described how ‘both courts in town refused to pose the question to the CJE37,’ but he persisted until ‘one very brave judge’ sent the preliminary reference (Interview, 23/08/11).

The specialist knowledge required to be able to bring a preliminary reference to the Court can be a barrier:

For a long time there were two different types of lawyers, there were the national lawyers and there were the European lawyers … It takes a certain type of lawyer to have the ambition to take a European case … to bring a test case, and also to have the two types of knowledge together – European and national law – it is quite rare, this type of person. So in France it is not unusual that there were not these types of cases, these types of people (Interview, 22/08/11).

Another interviewee said, ‘I think a lot depends on the involvement of the lawyers and how far they are familiar with EU law’ (Interview, 22/08/11). As an example of a protagonist, several interviewees mentioned Hélène Masse-Dessin38, the attorney in a number of gender equity preliminary references as a driving force behind legal strategies. One interviewee described her as, ‘certainly looking at legal strategies that might be useful to generate preliminary questions’ (Interview, 15/09/11). Ms Masse-Dessin appeared for the party requesting the reference in three matters in this sample.

The key individual can be a lawyer, but it can also be the party bringing the suit, who is determined to push the action. In six cases the party requesting the preliminary reference represented him or herself in written or oral proceedings before the Court. When the stature of the Court is considered, it seems extraordinary that so many matters should have been argued by the parties themselves.

A related sub-theme is the importance of elites. Many of the parties to the actions that generated the preliminary references were elites by virtue of professional status, who presumably have better access to resources or knowledge. Josep Peñarroja Fa, who was the plaintiff in a matter that resulted in a preliminary reference, is the President of the Association of Sworn Translators. He is a lawyer and a translator from French to Spanish who passed a competitive examination in his home Member State, Spain. Having heard of difficulties for his colleagues with ‘free circulation in France,’ he applied to be admitted to the register of court translators in France for both the Cour d'Appel de Paris and the Cour de Cassation. Both applications were rejected. Subsequently he brought a case to challenge policies of the French courts as a restriction on the free movement of services. As there were insufficient funds for a lawyer, he brought the cases himself and represented himself at the Court. He said of the cases, ‘legally speaking it was me against France, but really it was all the Spanish sworn translators represented by my association against the French authorities’ (Interview, 24/09/11). Philippe Derouin is a partner at major multi-national law firm in Paris. He was the party to a case about the proper social security contributions when an individual has income in more than one Member State of the EU. The Griesmar case had a French magistrate as the plaintiff. Another case, Olympique Lyonnais SASP v Olivier Bernard and Newcastle UFC 39, was brought on behalf of a professional football player, with support of his Premier League club, over the breach of a training contract with his previous club. Clearly there were resources available to him that would not be available to the average citizen with an employment dispute.

Discussion and Conclusion

This article has explored factors affecting national rates of preliminary references in the UK and France. The results clarify how policy structures can influence the rates of preliminary references and the areas in which they arise. The larger study indicated that a Bismarckian welfare state contributed to a higher rate of social policy preliminary references, making the UK anomalous. The themes from the policy structure category in the UK case study offer some insight. Beveridgean welfare states can offer other points where European cases arise. As eligibility for benefits in Beveridgean systems is primarily based upon residence establishing a nexus to the state, who is ‘resident’ for the purposes of benefits becomes a crucial and, in the UK, highly litigated issue. Also, the fact that the UK became a destination state for migration for many nationals of new accession Member States after 2004 contributed to the rise in these cases. Because the UK's equal protection legislation lagged behind that of Europe in some areas, there also were potential points to challenge the policy structure in these areas. This policy misfit created opportunities for challenges via preliminary references.

France has a Bismarckian welfare state that is very generous to insiders, who are seldom motivated to challenge the system. The high degree of dualization in the system creates a strong insider/outsider dynamic. Without help, outsiders will have difficulty accessing legal systems to enforce their rights. In France there was little help of this nature to outsiders.

Even in the newer areas of French social policy, such as social protection, anti-discrimination and disability, there are few cases. France does have cases in the area of reverse discrimination and opening up the civil service and protected professions. France had policies about equal pay before the Treaty of Rome. The areas in which French law was out of sync with EU law in the period studied concerned the disadvantage that French policies benefitting women caused to men. Anti-discrimination cases were a major category of cases from France, and all of the cases during the period studied were brought by or on behalf of men.

The French system is characterised by being very closed for its civil service and regulated professions. These restrictions put France in conflict with EU law on free movement of persons and the right to establishment. Therefore, cases related to opening up the civil service and the regulated professions are an important policy theme.

In the second category, there are themes related to structural factors in the Member State legal system. These factors can present barriers to social policy preliminary references or they can benefit them. In the UK, although some structural themes are barriers, there are beneficial structural factors that helped to offset those barriers. The acceptance of courts is a broad theme. Courts have become accustomed to greater oversight over the policy process. The cost of litigation, both the fees to bring legislation and the threat of costs shifting onto the losing party are barriers. Finally, the employment and social security tribunal systems and legal aid are benefits to preliminary references, and help to overcome some of the barriers. However, the LASPO may radically shift the balance between structural barriers and benefits, as it eliminates legal aid for most social welfare cases, with cuts beginning in April 2013.

In contrast, in France, although there was reported improvement over time, the resistance of courts and judges to European law and European rights is a barrier to references. Wilful misinterpretation of EU law by French courts also creates barriers to references.

The third category of themes relate to the actions of individuals and groups. These actors, whether they are plaintiffs, supporting groups or judges in the national system, affect rates of preliminary references. They can act to overcome structural barriers in the system.

In the UK organised test case strategies are backed by different types of organisations. These organised actions can be more strategic and advance the case law by soliciting particular cases needed to clarify points of law. This arises from a culture of activism in the NGO sector. The efforts are backed by NGOs, unions, the EHRC and its predecessor commissions. These actors benefit from legal aid and provide advice through benefits advisers and assist plaintiffs to access the tribunal systems and courts.

In France, the major theme in the category of actors is the importance of individuals. A few determined lawyers have accounted for a number of the French preliminary references. Individuals are also important as plaintiffs in the actions. These individual plaintiffs are often elites themselves, who presumably have better access to information and resources to overcome the substantial barriers in the system.

In both France and the United Kingdom, the themes related to policy structures affect the types of cases that arise. These are the opportunities in the systems for strategic actions. In both Member States, cases arose that were the most significant to that national policy context. The interlinked nature of the EU legal system will then ensure that these established precedents affect every EU Member State. This has implications for the development of EU social welfare law. The Court's justification for decisions based upon the circumstances of a particular case, or Member State, is then applied in a different policy context, with sometimes haphazard results.40

Acknowledgements

I am grateful for the research participants, who gave so generously of their time. I'd also like to thank colleagues and reviewers for their helpful suggestions to improve this article.

Notes

 1. The Court was previously known as the European Court of Justice (ECJ), and some interview respondents refer to it in this way.

 2. Social policy preliminary references were defined as those in the areas of: free movement of persons; European citizenship; social provisions; social security for migrant workers, and free movement of services, but only those services that are social policy-related, including health and education.

 3. Bismarckian social security systems are primarily financed through social insurance contributions by employers and employees. They are focused on horizontal redistribution across the recipient's lifespan, rather than vertical redistribution across society. They are earnings-related and the right to receive benefits is tied to work or its equivalent.

 4. Some interviewees fell into more than one category. For example, a Judge at the Court might also be a lawyer in France or the United Kingdom. In those instances, the individual was interviewed about all of his or her relevant experiences.

 5. Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia.

 6. Bulgaria and Romania.

 7. The Immigration (European Economic Area) Regulations 2006, SI 2006/1003, reg 6(1)(a)-(e).

 8. European Parliament Committee on Petitions, Notice to Members: Subject: Petition 1119/2009 by Piotr Kalisz (Polish) on the British Authorities' refusal of his application for unemployment benefit (‘Jobseeker's Allowance’), CM∖829426EN.doc, PE448.691, 2 September 2010.

 9. European Commission press release IP/11/1118, 29 September 2011.

10. Hansard HC Deb 28 November 2011, vol 536 col 667.

11. Patmalniece (FC) (Appellant) v Secretary of State for Work and Pensions (Respondent) [2011] UKSC 11

12. Case C-413/99, Baumbast and R. v Secretary of State for the Home Department, [2002] ECR I-7091.

13. Case C-480/08, Teixeira v Lambeth LBC and Secretary of State [2010] ECR I-01107 and Case C-310/08, Harrow LBC v Ibrahim and Secretary of State, [2010] ECR I-01065.

14. Case C-148/11, Secretary of State for Work and Pensions v Margita Punakova.

15. Council Directive 2000/43/EC implementing the principle of equal treatment between persons irrespective of racial or ethnic origin [2000] OJ L180/22; Council Directive 2000/78/EC establishing a general framework for equal treatment in employment and occupation [2000] OJ L303/16; European Parliament and Council Directive 2006/54/EC on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation [2006] OJ L204/23.

16. There have been prior anti-discrimination cases brought on behalf of men, notably Case C-262/88, Barber v Guardian Royal Exchange Assurance Group [1990] ECR I-1889, on pensionable ages for men and women.

17. Case C-394/96, Mary Brown v Rentokil Ltd [1998] ECR I-4185.

18. Case C-303/06, Coleman v Attridge Law and Steve Law [2008] ECR I-5603.

19. Hansard HC Deb 02 April 2009, vol 490, col 88WS.

20. Case C-520/06, Stringer and others v HMRC [2009] ECR I-179.

21. Case C-155/10, British Airways v Williams [2011].

22. Ministry of Justice, HM Courts & Tribunals Service, Annual Tribunals Statistics, 2011–12.

23. Case 78/98, Shirley Preston and Others v Wolverhampton Healthcare NHS Trust and Others [2000] ECR I-3201.

24. The industrial tribunals were the predecessors to the employment tribunals.

25. Case 78/98, Shirley Preston and Others v Wolverhampton Healthcare NHS Trust and Others [2000] ECR I-3201.

26. Case 173/99, The Queen v Secretary of State for Trade and Industry, ex parte Broadcasting, Entertainment, Cinematographic and Theatre Union (BECTU) [2001] ECR I-4881.

27. Case C-303/06, Coleman v Attridge Law and Steve Law [2008] ECR I-5603.

28. See Galanter 1974.

29. Although it was not a preliminary reference, the French NGO Group d'information et de soutien des immigrés (GISTI) was involved in triggering the European Parliament to challenge several provisions of the Directive on the Right to Family Reunification (2003/86/EC) in the Court in Case C-540/03, European Parliament v Council of the European Union [2006] ECR I-05769.

30. Case C-366/99, Joseph Griesmar v Ministre de l'Economie, des Finances et de l'Industrie et Ministre de la Fonction publique, de la Réforme de l'Etat et de la Décentralisation [2001] ECR I-9383.

31. Case 206/00, Henri Mouflin v Recteur de l'académie de Reims [2001] ECR I-10201.

32. Case C-285/01, Burbaud v Ministère de l'Emploi et de la Solidarité [2003] ECR I-821.

33. Joined Cases C-372/09 and C-373/09, Josep Peñarroja Fa [2011] ECR I0000.

34. Case C-164/07, James Wood v Fonds de garantie des victimes des actes de terrorisme et d'autres infractions [2008] ECR I-4143. Mr Wood, a UK national, had resided in France for twenty years with his French partner. One of their children was killed while on holiday outside the EU. French law provides for a payment to French family members in such a situation. Although other family members received payment, it was denied to Mr Wood because he was not French. The Court held that this constituted discrimination based upon nationality.

35. Case C-133/06, Philippe Derouin v Urssaf de Paris - Région parisienne [2008] ECR I-1853.

36. Case C-164/07, James Wood v Fonds de garantie des victimes des actes de terrorisme et d'autres infractions [2008] ECR I-4143.

37. The French acronym for the prior name of the Court.

38. Unfortunately, Ms Masse-Dessin's caseload did not permit her to be interviewed at the time that she was approached.

39. Case C-325/08, Olympique Lyonnais SASP v Olivier Bernard and Newcastle UFC [2010] ECR I-2177.

40. See O'Brien (2009) on the ‘monocular’ approach whereby the Court precedents around migrant workers have developed.