ARE LITIGATION AND COLLECTIVE BARGAINING COMPLEMENTS OR SUBSTITUTES FOR ACHIEVING GENDER EQUALITY? A STUDY OF THE BRITISH EQUAL PAY ACT

We present a socio-legal case study of the recent equal pay litigation wave in Britain, which saw an unprecedented increase in the number of claims, triggered in part by the entry of no-win, no-fee law firms into this part of the legal services market. Although the rise in litigation led to greater adversarialism in pay bargaining, litigation and collective bargaining mostly operated as complementary mechanisms in advancing an equality agenda. Litigation may be a more potent agent for social change than some recent analyses, which stress the limits of the law in the face of organisational pressures to canalise and diffuse human rights, have suggested.


Introduction
The passage of the Equal Pay Act 1970 was a landmark event in post-war British industrial relations, not simply for introducing a qualified right to pay equality between women and men but for breaching the principle that pay and conditions of employment should be determined by collective bargaining or other private contractual means, not by the state. Whilst the principle of legal support for the institutions of collective bargaining was well recognised at that point, and statutory minimum wages were set for sectors without effective voluntary arrangements, direct legislative modification of pay and conditions was regarded as exceptional. Since the passage of the Act, the relationship between equal pay legislation and collective bargaining has continued to be complex and contested. As the coverage of collective bargaining declined, litigation over pay equity issues sharply increased, in particular over the decade starting in 2003, but this is not a straightforward story of one form of regulation substituting for 382 S. Deakin et al. another. The rise in litigation was concentrated in areas of the economy where collective bargaining retained a presence, principally in the public sector. The intervention of new actors in the form of claimants' law firms, taking cases on a no-win, no-fee basis, changed the dynamics of the litigation process, with wider consequences for collective bargaining and for the effectiveness of legislation on gender equality.
In this article we review the causes and consequences of the rise in equal pay litigation in Britain during the 2000s. After an explanation of our methodological approach (Section 2), we assess the impact of changes in the legal framework which, on the one hand, opened up new types of legal claim and, on the other, provided the opportunity for claimants' law firms to enter a part of the market for legal services which had previously been dominated by public agencies and trade unions (Section 3). Next we examine litigation trends, assessing the importance of the leading cases for the development of equal pay law, and the effectiveness of litigation in providing redress for litigants (Section 4). We take a closer look at the litigation strategies of the law firms and unions, respectively, and at the responses of unions and employers to the rise in claims, drawing on interview material (Section 5). Finally, we offer an evaluation of the recent litigation wave, with the focus on its implications for the stability and efficacy of the collective bargaining system (Section 6). We argue that litigation and collective bargaining are best regarded as complements, in the sense that litigation is unlikely to be effective in advancing an equality agenda in the absence of well-functioning arrangements for collective wage determination. Conversely, collective bargaining 'in the shadow of the law' is likely to lead to more egalitarian and equitable outcomes than would be obtained from a purely voluntarist approach based on the autonomy of the wage determination process. More generally, the recent British experience suggests that litigation can be a potent mechanism for advancing social rights.

Empirical analysis of equality law: questions, methods and data
Empirical analysis of the operation of equality laws has cast light on the limits of the law as a means of implementing a human rights agenda for social reform. On one hand, there is the potential for civil and social rights to be embedded in organisational practice via human resource management (HRM) processes; on the other, these same processes can lead to a channelling and possibly a weakening of the human rights perspective which initially informed the legal change (Edelman et al., 2001;Dobbin, 2009;Greene, 2009, 2010). In their focus on HRM, these studies highlight one way by which legal rules, which are not self-enforcing, can be put into practice. At the same time, they leave open the possibility of other modes of implementation which are both more adversarial in character and more contentious in their results. The recent experience of British equal pay litigation poses the question of how an adversarial approach to enforcement, based on repeated and confrontational litigation, operating alongside the setting of pay and conditions through collective bargaining, compares to the process which Dobbin (2009, p 3), writing of US civil rights legislation, describes in terms of 'the personnel profession's compliance efforts [translating] the law into practice'.
The empirical study of equality law, as of any other area of labour law, requires a multi-level approach that first identifies the regulatory ambit of the relevant legal rules and their reception by social actors, prior to making an assessment of the impact of the law on behaviour and outcomes (Deakin, 2011). Thus analysis of the scope of the law and of its formal regulatory content is needed to establish its rationale and intended Equal pay litigation and collective bargaining 383 effects, but must be complemented by empirical research aimed at establishing the contours of the operation of the law at the level of litigation and enforcement, and its influence on actors' strategies.
In this article we present a socio-legal analysis of the operation of equal pay law, which combines a narrative account of the development of legal doctrine with descriptive statistics on litigation trends, case studies of leading decisions and material drawn from interviews with actors who played leading roles in the process by which the law played out in practice. To this end, we draw on a range of data sources. Data on litigation trends are drawn from annual records of employment tribunal statistics, published by the Employment Tribunal Service and its predecessors since the mid-1970s (ETS, various years). These statistical series provide data on the volume of claims and some general information on their outcomes (see Figures 1-4 later) but next to no detail on particular cases. For these, we constructed our own dataset of leading decisions based on press reports (local and national press and specialist publications on labour and equality law) and the published texts of legal judgments (law reports). We identified around a dozen leading cases, initially from legal decisions and references to disputes in the national press, and then researched references to them in the local press and specialist publications (see Table 1 and Section 4.3). Finally, we conducted interviews with principal actors in equal pay litigation in the key period covered by the study in the late 2000s to establish their approaches to litigation, their attitudes towards the law and their assessment of its impact. The interviewees did not form a random sample; on the contrary, they were approached because of their roles in the events that are described in the article, although for reasons of confidentiality, individuals and their organisations are not identified by name. They were HR managers in six local authorities, officials of five trade unions involved in equal pay claims, and a solicitor in a claimants' law firm focussing on equal pay litigation. Interviews were recorded and transcribed and coded manually using emergent categories (see Section 5).

The evolution of the legal framework governing equal pay in Britain
The Equal Pay Act 1970 was passed in the aftermath of strike action by women workers protesting about unequal terms and conditions. The most prominent of these disputes, involving the Ford Motor Company's Dagenham Plant, arose out of terms and conditions which had been collectively negotiated, as did a lengthy strike, shortly after the Act had come into force, at the Trico Folberth plant in Brentford in 1976 (Meehan, 1985). The principal remedy supplied by the Act took the form of a claim for equal pay before a labour court dealing with individual disputes (an 'industrial tribunal', later renamed 'employment tribunal'). The claimant was required to find a better paid 'comparator' of the opposite sex who worked in the same employment. Under the 1970 Act, the comparator had to be employed on a similar job ('like work') or on work which had been determined to be of equivalent value under a job evaluation scheme ('job rated as equivalent'). In addition, the Act provided for collective agreements to be amended through arbitration before a collective labour court (the Central Arbitration Committee) so as to bring them into line with the principle of equality (Deakin and Morris, 2012, para. 6.4).
The implementation of the Act was delayed for five years to allow for payment structures to be amended voluntarily. Prior to 1970 it was not unusual for the terms of collective agreements to make separate provision for female and male workers. This type of 'direct discrimination' (discrimination formally based on gender) was largely removed during the period when the Act was being phased in. Average hourly wages for female workers, which had been around 60% of average male wages in 1970, had risen to 70% of the male average by 1976, a change subsequently attributed to the implementation of the equal pay principle through sector-level collective bargaining (Zabalza and Tzannatos, 1985).
In a 1979 ruling, Hy-Mac, 'indirect discrimination' arising from the application of rules or practices which were formally gender-neutral but had the effect of disproportionately benefitting workers of one sex over the other was held to be outside the remit of the collective arbitration provisions of the Act, effectively nullifying them. The ruling was successfully challenged before the European Court of Justice in 1982, but the UK government, which was by now committed to a policy of labour market deregulation, implemented the Court's ruling in a minimalist fashion, replacing the power to re-align collective agreements according to the equality principle with a largely symbolic measure enabling a court to declare discriminatory agreements 'void'. After this the emphasis in litigation turned to individual claims, the scope for which was enhanced in 1983 by the introduction of the right to equal pay for work of 'equal value' (in US terms, 'comparable worth'), again following a ruling of the Court of Justice. The equal pay principle had been embodied in a provision of the fundamental text of European Community law, the Treaty of Rome, and this enabled the Court to give an expansive reading to the content of sex discrimination law in a series of rulings from the mid-1970s to the late 1990s. These decisions established, amongst other things, that individual claims could be brought with respect to 'indirect' sex discrimination, subject to an employer's defence of 'justification' or (as it later became known) 'proportionality', which the Court tended to interpret restrictively, thereby putting the onus on employers to provide non-discriminatory explanations for persistent inequalities (for further details on these legal changes, see Deakin and Morris, 2012, paras. 6.5, 6.80 and 6.106).
Notwithstanding the interventions of the Court of Justice, individual workers faced substantial obstacles to realising a successful equal pay claim. As it was no longer necessary to find a fellow worker employed on 'like work' or 'work rated as equivalent', comparisons could be made across occupational boundaries, although not beyond the scope of employment units or even, in some instances, across different workplaces, unless those workplaces were governed by common terms and conditions, such as those deriving from a single collective agreement or a series of linked agreements. For an equal value claim to succeed, the tribunal had to be satisfied that the jobs of the claimant and her comparator were equivalent by reference to a cluster of factors relating to, amongst other things, the nature of the work and the qualifications of the job holders. Even if a finding of equal value was made by a court-appointed expert, the employer could fall back on one of a number of 'genuine material factor' defences, which referred to factors such as labour scarcity ('market forces') and seniority, although the use of collective bargaining to set pay and conditions was not in itself a defence (see Deakin and Morris, 2012, paras 6.96-6.101). Partly because of the procedural complexity of claims, together with the scope for appeals on points of law and references to the Court of Justice for clarification on the meaning of European Union law, it was not unusual for the more high-profile equal value cases to take years to be resolved. The Enderby case, which involved comparing the pay of speech therapists and clinical psychologists in the National Health Service, was litigated for over a decade. Extended litigation of this type was mostly financed Equal pay litigation and collective bargaining 385 through funding from the government agency with responsibility for overseeing sex discrimination law, the Equal Opportunities Commission (EOC). 1 In their review of discrimination law, published in 2000, Hepple, Coussey and Chowdhury concluded that 'the equal value procedure has largely failed to deliver pay equity to women' (Hepple et al., 2000, p 97). Between 1976 and 1998, just over 12,300 equal pay claims had been registered with tribunals, of which only 20% were successful at a tribunal hearing or were settled in the claimant's favour. The success rate for equal value claims requiring a report from an independent expert was lower at around 15%. The vast majority of the equal value claims were against public sector employers, in particular British Coal (then a state-owned enterprise) and the National Health Service. In 1998, a total of £3.4 million had been paid out to around 1,000 successful female claimants. According to Hepple et al., 'although the settlements were significant for these women, they involved a long and expensive procedure at considerable public cost'. The study also doubted whether the equal value procedure had made any discernible impact on the gender pay gap (Hepple et al., 2000, 98).
Although the pace of legal change slowed after the mid-1990s, a further significant intervention of the Court of Justice occurred in 1998 when it ruled that the two-year limit on arrears in equal pay claims was contrary to the principle requiring effective remedies to be provided for breach of EU law rights (the Magorrian case). The effect of the Court's ruling, which was brought into effect in the UK in the early 2000s, was to substitute a six-year time period for arrears of wages, aligning equal pay law with the normal period of limitation of claims for breach of contract (Deakin and Morris, 2012, para. 6.103). This change in the law was triggered by cases arising from the exclusion of female part-time workers from occupational pension schemes, and its significance for collective bargaining was not immediately appreciated. It was to prove important in providing the context for the entry of no-win, no-fee law firms into equal pay litigation because it greatly enhanced the scale of the compensation that could be made available to claimants who could show that they had been the victims of historical discrimination. 2 A further change in the law, in this case governing the financing of civil litigation, helped create the conditions for the entry of no-win, no-fee law firms. English law has traditionally been hostile to contingent fees of the kind that have long been accepted in the USA, under which the successful parties pay their legal representatives a fee calculated as a percentage of the winnings, but a version of the contingent fee, known as the conditional fee, was introduced in 1990, and from 2000 it became possible to recover success fees from costs paid by the losing party. The regulations governing conditional fees were relaxed in 2005 in a way that further expanded the scope for their use. These changes were associated with a government policy of encouraging private financing of litigation to reduce use of public funds to support the costs of claims through the legal aid system. 3 Their application to the employment context was somewhat fortuitous 386 S. Deakin et al.
and was most likely unintended, since legal aid was not generally available at that time (or since) for employment tribunal claims. 4 4. The rise of equal pay litigation: main trends, sectoral context and leading cases Figure 1 shows long-term trends in equal pay litigation. In the first years of the equal pay jurisdiction, a few hundred applications a year on average were registered by industrial tribunals. By the early 1980s, as the scope for claims under the 'like work' and 'work rated as equivalent' headings was gradually exhausted, the number had fallen to no more than a few dozen. With the enactment of the new equal value jurisdiction from the mid-1980s, the volume of registrations returned to the levels of the 1970s. There was a further increase to the level of several thousand claims a year at the end of the 1990s, around the time that European law began to have a major impact on the rights of part-time workers excluded from occupational pension protection. The next major increase occurs in the mid-2000s and is associated with the entry of no-win, no-fee law firms and with the mass claims that they and others, including trade unions, launched against public sector employers in the local government and health service sectors. The volume of claims spikes in 2007-8 at over 60,000; by 2012-13 it had fallen back to around 20,000 a year. Figure 2 compares the number of equal pay claims with the total number of claims registered before employment tribunals across all relevant jurisdictions in the period from the late 1990s to the present. Although there has been a steady increase in the volume of total claims across this period, the trend in equal pay claims did not keep pace with the general increase until the late 1990s, and has generally followed a pattern of its own. Thus the rise in equal pay claims does not appear to be a simple function of a wider increase in claims across this period. The rise in overall claims is in part a result of the availability of new types of claim, in particular those relating to working time protection, which were introduced after 1998.

General litigation trends
Nor is there a straightforward relationship between the increase in the volume of equal pay claims and the decline in collective bargaining coverage over this period. The combined coverage of collective agreements and related forms of collective wage determination, including orders or wages councils and other statutory bodies, peaked at over 80% in 1979 and has been falling at a steady rate ever since. By the late 1990s it stood at just over 40% and in 2011 was down to below 25% (Ewing and Hendy, 2013). Equal pay litigation, however, did not begin a steep and sustained rate of increase until the mid-2000s, well after the decline in collective bargaining coverage had set in.
The decline in collective bargaining coverage and the rise of individualised employment litigation can be understood as related aspects of a larger, long-term shift in the nature of the British industrial relations system. There has been a move away from reliance on self-regulation by unions and management as the principal mechanism Equal pay litigation and collective bargaining 387 for determining pay and conditions and resolving disputes, towards a rights-centred and law-driven system (Deakin and Wilkinson, 2011). Had state policy continued to support multi-employer collective bargaining after 1979, it is likely that the collective arbitration mechanisms for aligning collective agreements with the equal pay principle would have been renewed in an effective form after the European Court judgment of 1982. If this step had been taken, implementation of pay equality through collective bargaining might have continued throughout the 1980s and beyond much as it had done in the 1970s, where it proved effective in narrowing the gender pay gap (Zabalza and Tzannatos, 1985). In this sense it could be argued that the decline in the effectiveness of collective agreements as a source of regulation, coupled with the ending of state support for the institutions of multi-employer bargaining, were necessary conditions for the later rise in equal pay litigation, even if the timing of that rise was contingent on a number of institutional factors, including those which prompted the increase in conditional fee litigation in the mid-2000s.
However, it cannot be assumed that as collective bargaining declined as a form of regulation, litigation on equal pay (or any other employment law jurisdiction, for that matter) was simply substituted for it. This is because the rise in equal pay claims has been focussed in those sectors of the economy where sector-level collective bargaining remained effective, particularly in the public sector. This trend, identified by Hepple et al. (2000) in the late 1990s, remained strong in the 2000s, as we shall now see.

Mass litigation in the local authority sector
4.2.1 Background: the implementation of single status in local government employment. The rise in equal pay claims which began in 2004-5 saw a near doubling of claims to reach over 8,000 a year, and peaked with over 60,000 claims in 2007-8. This increase was almost entirely located in just two sectors, local government and public health services, with the local authority sector leading the way. The catalyst for the rise in litigation in the local government litigation was a series of long-running disputes over the implementation of a national-level collective agreement, the 'single status agreement' or SSA. Between 2004 and 2008, the trade union Unison supported over 40,000 equal pay claims by its members, the large majority in relation to the SSA (Jaffe et al., 2008), and the GMB union supported around 30,000 (GMB, 2015). Claimants' law firms are reported to have taken a further 30,000 claims in this period (Gibson, 2013).
The catalyst for the SSA was equal pay litigation supported by trade unions in the first half of the 1990s, involving female canteen workers whose pay had been cut as a result of a compulsory competitive tendering exercise. The House of Lords (then the UK's highest appellate court) ruled in 1995 that the 'market forces' defence was not available to the claimant's employer simply on the grounds that unless wages were cut, the contract to supply the services in question would have been lost (the North Yorkshire case). The implication of this ruling and a related decision, Cleveland, when coupled with the application of the legal principle that employees' 'acquired rights' would be preserved upon a transfer of employment arising from outsourcing, was that the national agreement for local government employees would operate as a floor of rights, binding local authorities themselves where they kept services in house, as well as external contractors. The Cleveland case was eventually settled in the mid-2000s for sums reported to be between £6 million and £7 million.

S. Deakin et al.
The SSA, which was concluded in 1997, sought to streamline local authority payment structures and minimise the risk of future equal value claims by bringing together the terms and conditions of employment of manual workers (previously covered by an agreement known in England and Wales as the 'White Book') and those of administrative, professional, technical and clerical workers (APT&C, previously covered by the 'Purple Book'). The SSA put in place a national single spine pay structure (known as the 'Green Book'). 5 Under the SSA, each job was to be evaluated at local level using a locally agreed job evaluation scheme. An earlier job evaluation under the White Book had been carried out in 1987, but it had ignored the effect of bonus schemes, which had tended to benefit workers in male-dominated grades by, on average, up to 30% of basic pay. While these bonuses had at some earlier point been related to productivity, monitoring of  performance had generally ceased and, as later litigation made clear (see Section 4.3), they had come to be seen as forming part of basic pay.
The SSA was designed as a framework agreement which was to be implemented locally. Despite the significance of the issue of single status for potential equal pay claims, few implementation agreements were struck at local level during the first five years of the SSA's operation. Given the slow pace of implementation, a new nationallevel agreement, the National Joint Council Pay Implementation Agreement, was made in 2004, setting a deadline of April 2006 for job evaluations to be completed and a further one of March 2007 for the new payment system to be in place. However, only half of the almost 400 local authorities in England and Wales had fully implemented the agreement by this point.
Those local authorities that did implement the SSA within the first few years appear to have avoided the litigation and back pay issues that occurred at many other councils. Of the six local authorities we interviewed for this research between late 2008 and early 2010, only one had completed the process early, and this council had experienced no litigation or back pay issues. Four of the local authorities we spoke to had faced litigation or the threat of litigation from both trade unions and lawyers representing individual claimants. One authority had yet to start the job evaluation process at the time of our interviews.
Trade unions and employer representatives we interviewed put the long delay between the negotiation of the SSA and its implementation in many local authorities down to a number of factors. First, the process of implementation involved a detailed job evaluation of every post in each local authority and the parties needed to reach agreement about which evaluation tool they would use as well as the details of the process of evaluation. In large local authorities, evaluating every post was a fairly onerous procedure, and the process had to allow time for appeals. Thus, it could take several years. Many councils began their evaluations using paper-based systems, which were slow and cumbersome, and it was not until computerised systems became available that the process could be speeded up. Some councils used the opportunity provided by single status to deal with other, long-standing HRM issues and bring in a range of changes, so that negotiation over the implementation of single status became enmeshed with a wider reform agenda. Rather than the process of job evaluation being completed at a national level around job profiles, each local authority had to conduct its own job evaluation process. As one interviewee described it, every local authority had to 'reinvent the wheel' despite the fact there was much commonality across them amongst the roles their employees performed.
Second, pilot projects conducted in the first few years revealed that the implementation of single status was likely to have significant impacts for some employees' existing terms and conditions and for the financial stability of local authorities. For a significant proportion of workers (up to 40% in some local authorities) there were likely to be sizeable pay cuts, particularly once bonus systems were removed. As one HR manager told us, unions representing manual workers could see that 'their members were going to be hammered'. Pay cuts were something that many union officials had not factored into their negotiation positions, even though advice given to officials by Unison in 1997 and the GMB, TGWU and Unison combined in 1998 had indicated that this was a possibility and that union officials should negotiate pay protection arrangements for a 'reasonable period' for those negatively impacted by grading reviews (Unison, 1997: GMB et al., 1998.

S. Deakin et al.
For most of the officials interviewed as part of our research, the issue of pay cuts was something that they and their members struggled to come to terms with. As one noted, 'putting in front of people a proposal which results in the pay cut for anyone of any order is difficult . . . [they will say] I didn't join the trade union and pay you £10 a month [for] you to negotiate a pay cut'. Initially many union officials tried to negotiate for lifetime protection for any employee whose pay was to be reduced. Many local authorities, on the other hand, aimed to negotiate settlements in which pay increases for the 'winners' would be offset as far as possible by pay decreases for the 'losers', thereby minimising the overall increase in the expenditure on wages and salaries. Thus in many local authorities the negotiations were slow and fractious, and in some cases they simply broke down. As one union official noted, the unions had assumed that the SSA 'would lead to local authorities putting serious resources' aside for its implementation, but this 'largely didn't happen'. A number of employers we interviewed reported deliberately holding back on negotiating further once the pilot studies revealed the extent of potential change for them, because of fear of significant industrial unrest and the impact on employee morale. In addition, some local authorities reported deliberately delaying the process so as to allow others to go first: 'it was let's sit and watch them make some mistakes and learn from them'. Unions also reported being 'fobbed off' in their attempts to engage employers in negotiations. The feeling was that some local government administrations had other priorities in the political cycle and adopted something of an 'ostrich position' around equal pay and single status.

Litigation arising from the local government SSA.
The most prominent no-win, no-fee law firm to operate in the equal pay sector during the 2000s was Stefan Cross Solicitors, a Newcastle-based firm which was established in 2003. Its founder, Stefan Cross, had previously worked as a specialist in personal injury litigation for the leading claimants' law firm Thompsons, and had been a trade union activist since his teens. When acting for the unions in the late 1990s and early 2000s, he had taken a prominent role in the Cleveland litigation. Stefan Cross Solicitors accepted its first cases in March 2003 and a year later was processing 100 claims. By 2008 it had 30,000 cases on its books. 6 Claims brought in relation to the implementation of the SSA were of four main types (see McLaughlin, 2014). First, claimants sued for back pay in relation to past inequalities which the non-implementation or inadequate implementation of the SSA had failed to deal with. These most often took the form of claims arising from the payment of unjustified bonuses to workers in male-dominated grades. This type of claim was described to us by a claimants' lawyer in the following terms: pay differentials in [a particular local authority] are so huge that they are using [the argument] 'we're looking after the women' and actually it's a cloak for we don't want huge pay cuts for our men. We've got a position in [this council], a grade five refuse driver, who on a national pay scale is £12,000, gets £26,500, he gets paid more than social workers and newly qualified solicitors in [this council]. So they've got a 125% effective bonus arrangement.
A second type of claim arose from 'pay protection schemes' under which the pay of more highly paid workers was maintained at an artificially high level for a certain period after the implementation of the SSA, thereby perpetuating historical differences Equal pay litigation and collective bargaining 391 that favoured the male-dominated grades. A third legal strategy was to challenge the job evaluation schemes put in place at local level as part of the process of implementing the SSA, where it could be shown that they had a tendency to discriminate (see Wright, 2011;Gilbert, 2012). Finally, challenges were mounted to 'job enrichment schemes' under which job descriptions were redefined in an allegedly discriminatory manner. This was described in the following manner by the claimants' lawyer we interviewed: The other brilliant device they've invented is to change the jobs altogether. Under the White Book scheme you have four grades of cook and four grades of gardener. Under the Green Book almost universally they have one grade of gardener and they call them something else, they'll call them environmental operative, and they will score the job of the environmental operative on a theoretical basis and then they will apply a whole load of jobs to this new grade of environmental operative. . . . So the grade 1 gardener, the grade 2 gardener, disappear altogether and they end up being an environmental operative. Now what they don't do, they don't do that for kitchen staff. They don't create a new grade of kitchen operative, or catering operative, they stay and get evaluated job by job and hey presto they're still at the bottom. In fact what's happened in a lot of areas is that they've actually created a desert of no male jobs in the lower grades.
A further type of claim, which was rarely pursued but gave rise to more controversy than any other, took the form of actions against trade unions which had negotiated agreements that were indirectly discriminatory (Deakin and Morris, 2012, para 6.100). Unions could be sued under the Sex Discrimination Act 1975 for committing discriminatory acts. In the Allen v. GMB litigation, the courts held that this provision could be invoked to supply compensation to workers who had suffered losses as a result of the union's involvement in concluding a collective agreement that preserved discriminatory practices or sought to compromise or qualify claims for historical inequalities. The union argued as its defence that its bargaining strategy was justified by a number of considerations, including the need to balance the needs of different workforce groups amongst its membership. After a number of appeals this argument was rejected, on the basis, amongst others, that officials had acted against the interests of the union's female members by concealing information from them and pressuring them to settle claims prematurely. The Allen litigation prompted employers in subsequent cases to seek contribution from unions in respect of liabilities arising from discriminatory agreements, and to put forward union involvement as the basis for a genuine material factor (GMF) defence. These tactics were largely unsuccessful, with the courts eventually coming round to the view that responsibility for discriminatory payment structures would in most situations lie with the employer. The Allen case came to be seen as somewhat exceptional, but only after a period of several years during which unions faced a novel and open-ended litigation risk.
The claims made in relation to the SSA were at least as complex to bring as the types of claim that had featured in the litigation wave of the 1980s and 1990s. Establishing equal value in the first place could require either an independent report or lengthy deliberations by the tribunal, and employers would seek to rely on the GMF defence to justify pay protection and job enrichment schemes. Legal uncertainties arising from the novel nature of some of the claims meant that obtaining a clear ruling on the scope of the GMF defence could delay consideration of the merits of a claim by months or years.
The rapid rise in cases registered did not translate into immediate litigation success. Figures 3 and 4 show trends in registrations and disposals of equal pay cases in the period from 2007, when cases brought following the initial rise in mass claims began to be decided. Disposals have generally run well below the level of registrations, and Of those that were withdrawn (43% in 2011-12 and 50% in 2012-13), a substantial proportion are likely to have resulted in a payment of some kind being made to claimants. Official statistical series do not indicate whether withdrawn applications led to a settlement, but it is likely that many of them did. Table 1 sets out in synoptic form the features of some of the most important equal pay cases decided in this period, most of which arose under the SSA. As the table shows, by contrast to the disposal figures just mentioned, claimants in these leading cases were generally successful in the legal arguments that they put forward and in obtaining compensation. Some of these claims were continuing at the time of writing (January 2015), as further appeals were pursued or additional legal points raised.    procedures. Further complications arose from the possibility of duplicate claims, from arguments that claimants had mis-described their jobs, and dispute over the comparison of employees in non-teaching roles in community schools to employees in other council establishments. These issues were appealed from the initial tribunal determination, first to the Employment Appeal Tribunal (EAT) in 2010, and then to the Court of Appeal in 2012. Meanwhile, separate appeals were being heard on procedural issues arising from the constitution of the initial tribunal hearing which dealt with the GMF defence and eventually on the validity of that defence itself. The employment tribunal finally ruled in the claimants' favour, finding that the bonus scheme operated by the council was not transparent, with the result that the GMF defence failed.
In tandem with this litigation, the council was also facing claims in the regular courts (the County and High Courts) from claimants who were out of time to bring their cases before an employment tribunal. These claims were eventually appealed to the Supreme Court, which ruled in 2012 that the normal six-year limitation period for breach of contract actions would apply to claims brought before the County or High Court, thereby circumventing the six-month time limit for claims before tribunals (the Abdulla case).
In an attempt to avoid future equal pay liabilities, the council then sought to change the terms and conditions of its entire workforce. Employees who did not accept the new, non-discriminatory terms were dismissed and re-engaged. This led to two additional forms of claims: claims by various unions for compensation in relation to the council's failure to consult over the dismissals, and unfair dismissal claims by individuals who had had their terms and conditions forcibly changed. The claims by the unions were resolved while the majority of the unfair dismissal claims were withdrawn; the remaining claims were lost by the individuals with costs awarded against them. The employer's attempt to force through changes to terms and conditions led to industrial action by the Fleet and Waste teams. Between 20 December 2010 and 12 January 2011, approximately 450 refuse loaders undertook discontinuous industrial action involving half-day strikes and working to rule, at least in part by way of a response to the ending of bonuses.
In its Annual Audit letter 2011-12 published in November 2012 after the Supreme Court decision in Abdulla, the council recorded that its accounts from 2006-7 to 2011-12 reflected combined actual and potential equal pay settlements totalling £757 million. These sums included 'second-generation' claims that were anticipated following the success of the initial cases. The Audit Letter noted that 'the affordability of this presents a major challenge to the Council.' In January 2014, following further legal defeats, the Council indicated that it might have to sell a number of major assets, including an international conference centre, to meet its liabilities.

Case study 2: Sunderland City
Council. The first claims against Sunderland City Council were lodged in 2004. By the end of the litigation there were around 1,186 claimants, although the precise numbers fluctuated throughout. The first generic claim form was settled in 2007 and the first substantive tribunal hearings commenced in 2008, on the issue of whether the council had a GMF defence. These hearings entailed 30 days of evidence after which the claimants had succeeded across the board and the council's defence was held to be a sham. During 2008 there were also three separate EAT hearings dealing with more procedural aspects of the litigation in addition to an appeal on the GMF ruling, which was then further appealed to the Court of Appeal (the Brennan case).

S. Deakin et al.
During 2009 and 2010 the tribunal dealt with the separate issue of the validity of the job evaluation scheme adopted by the employer under the SSA. One hundred twenty-five days of litigation ensued, during which around 30,000 pages of documents were disclosed by the council to the claimants. Final argument on this aspect of the case was heard over three days in January 2011 with approximately 500 pages of submissions from each side. In February 2012 the claimants' challenge to the job evaluation scheme was upheld.
There were further EAT hearings during this time. On 16 and 17 November 2011 the issue of a claim in contribution and disclosure of settlement agreements with the unions was considered, and on 1 February 2012 the EAT overturned a tribunal decision striking out various claims in which employees sought to rely on different comparators to those identified in internal grievance hearings.
All the claims by those instructing Stefan Cross Solicitors were settled by around 2012. Thus for many claimants, the process of obtaining compensation for the pay disparities had taken eight years. The case went to the EAT on six occasions, as well as giving rise to a hearing on a complex point of law before the Court of Appeal.

Attitudes towards negotiation, litigation and fairness
We now draw on our interview material to provide further information on the motivations of the principal actors, their views of the litigation and negotiation strategies they and their counterparties were pursuing, and their perceptions of the fairness of outcomes.

Litigation versus negotiation?
Opinions differ on whether the intervention of no-win, no-fee law firms changed the approach of trade unions to dealing with local authority employers, to the extent of inducing them to use litigation to a greater extent. The advice given by Unison in 1997 on implementing single status had argued that 'negotiation is better than litigation', but it also noted that 'where employers delay unnecessarily, it should be made clear that they are opening themselves to equal pay challenges' (Unison, 1997, p 6). However, litigation was seen as a matter of last resort. As one regional union official explained to us, in relation to employer intransigence: You can only put a certain amount of pressure on, because there was no compunction on the employers to move forward quickly on this, there was absolutely no reason in the world at this stage why they should move forward. Could we compel them to do so in the end? All we could do is talk, correspond, shout a little bit, stamp our little feet . . . we couldn't say right, we're going on strike because you won't talk to us, it just wouldn't happen, it just was impossible.
The explanation given to us for the unions' reluctance to initiate litigation was twofold. First, 'litigation won't deliver equal pay. [Only] negotiation will deliver equal pay'. Achieving equal pay was seen as a matter for joint regulation, with unions and employers together conducting the job evaluation process and implementing the resulting pay structure. According to this argument, litigation could deliver back pay, but not equal pay. Some argued that the focus on back pay that arose from litigation initiated by claimants' law firms distracted employers from considering the wider picture. As one national union official argued: 'we are not going to see that the intervention of no win, no fee lawyers has led to any significant narrowing of the gender pay gap. . . . Single status will have had a much greater impact'. Additionally, unions accused claimant law firms of 'parasitic' behaviour in acting on the back of completed phases of the job Equal pay litigation and collective bargaining 397 evaluation scheme conducted by employers and unions. This enabled them to take claims on the basis of 'work rated equivalent' rather than the more complex 'equal value' claims: 'no win, no fee lawyers have not been keen to pursue equal value claims because of the uncertainty, because it's not easy to predict what the outcome is going to be, and because of the length of time those cases have taken'.
Second, unions argued that litigation increased employer intransigence, pointing to cases in which employers had chosen to fight them at every stage, dragging out the process over years. One official pointed to the Enderby case, which had taken 14 years to be completed in the 1990s, and to litigation against Cumbria council, which involved 3,000 female workers in over eight years of litigation in the 2000s. They also argued that many of the local authority equal pay cases which were being litigated at the time of this interview (2010) were far from clear-cut, and that even when claims were, in their view, clearly justified, some employers continued to fight them.
Third, the union officials we spoke to rejected the claim (widely circulated in the British media at the time of the Allen decision) that they had been slow to act prior to the intervention of Stefan Cross Solicitors. One union official told us that there had been internal debate in his union in the early 2000s about whether to litigate and 'the view was that we need to reach negotiated settlement . . . to deliver equal pay'. Another union official accepted that procrastination and inertia could explain the first few years of inaction, and that in the early 2000s more use should have been made of the threat of litigation. However, this respondent suggested that after 2003, independently of the entry of no-win, no-fee law firms, unions had begun to make greater use of the threat of litigation and had issued equal pay questionnaires, designed to force employers to disclose information on payment structures with a view to bringing a claim, to get employers to engage in serious negotiation in cases where they had previously refused to do so.
Union officials also regarded some of the litigation that they were engaged in as unhelpful. They saw themselves as having been forced into litigation in the middle of jointly conducted job evaluations to prevent their members being enticed away by potentially generous back pay settlements from no-win, no-fee law firms. Some employer respondents also referred to HR resources being diverted away from the job evaluation process to address litigation issues.
The decision of the Court of Appeal in favour of the claim against the union in the Allen litigation was seen as having had a fundamental impact on the position of the different players in the equal pay negotiation and litigation context. The tribunal held that whilst it was legitimate for the union to attempt to balance a range of conflicting interests, the means used had not been proportionate. The tribunal found that there had been a practice 'which involved persuading those who had [equal pay] claims to take less than they were worth [or] . . . not to bring them at all'. The tribunal noted 'how much care, effort and discussion went into pay protection and how little went into back pay'. The tribunal considered that the female claimants had been placed 'in a position where they were in fear that, if they pressed for more, it might lead to job losses and their being seen as traitors by their colleagues'. The tribunal was also critical of the union for accepting at face value the employer's 'plea of poverty'.
The union respondents we spoke to argued that the GMB had agreed to a low back pay settlement to deal with past pay inequalities not to prioritise pay protection but to avoid potential job losses through outsourcing and service cuts. They suggested that in the wake of the Allen ruling, unions had become hesitant to endorse agreements with employers over back pay and had advised their members to reject offers that they would have otherwise considered reasonable. Union respondents also pointed to 398 S. Deakin et al.
additional negotiation costs arising from Allen; officials reported taking advice from lawyers at every stage of the negotiation process and to be referring agreements to national level to be checked. This was a view confirmed by an employer respondent who said, 'irrespective of the ins and outs of that case [Allen], the trade unions are worried about doing a deal with the employer'. Union officials argued that the overall effect of Allen would be to slow down the implementation of single status. They felt uncomfortable more generally with levels of litigation which ran counter to the traditional focus on achieving collective solutions: If we are saying that an individual doesn't like the collective settlement and then exerts their individual right over the collective good and that's allowed to continue, then there is no such thing . . . as collective agreements and trade unions cease to have a purpose to exist and I am going to retrain as a lawyer.
By contrast, the law firm representative we spoke to questioned the effectiveness of the union approach to getting employers to engage in collective bargaining. He argued that unions tended to litigate strategically, that is, on a selective basis, and often with ulterior motives: You have to ask yourself the question well why did they pursue cases in only 12 out of the 400 local authorities when the pay structures at least under the White Book and the Purple Book were national, they were national pay scales and national agreements. If there's a problem in St Helens there's also a problem in Manchester, and if there's a problem in Cleveland there's a problem in Warwickshire, and they pursued the Warwickshire case but don't pursue any of the other cases. And the answer is because there's always an underlying dispute and unless they have another dispute they won't pursue any other issues. That's where the opportunity arose for [nowin, no-fee law firms which] knew they had all these potential issues not being pursued.
Second he viewed unions as excessively conservative in their approach to pay bargaining, as favouring the status quo and hence being unwilling to countenance altering radical changes to existing pay structures: They will do absolutely everything to ensure that the status quo is maintained and that is their negotiating strategy, that's the way they've approached single status, that's the way they've approached everything to do with this. Although the documents and the policy says it's designed to deal with equality issues, the reality is that . . . if [they] can't maintain the status quo [they'd] rather not deal with it, otherwise what's the explanation for [the unions' line] which is we will negotiate rather than litigate.
Third he rejected the claim that litigation undermined collective bargaining, arguing that the fact that somebody succeeded in pursuing the cases was the log jam to get the single status agreement in place, which is ironic when people now claim that pursuing equal pay cases is a hindrance to collective bargaining when in fact the truth is it's exactly the opposite, equal pay cases are a catalyst in proper equal pay bargaining, and in fact the refusal of the unions to actually pursue these cases is why 11 years after they made the agreement they still haven't done it, because you'll find you've put in the claims and suddenly the employers start acting . . . a lot faster than they were acting before. So it was the combination.
Relatedly, he argued that unions would only litigate thanks to the competition provided by independent law firms: Well they continue to essentially act on our coat tails in the vast majority of cases. There still isn't in most areas an initiative, otherwise you'd find [it], if you just look at the [employment tribunal Equal pay litigation and collective bargaining 399 statistics], just look at the ACAS stats, and look at the areas where we act and where we don't act, and see how many equal pay cases are submitted in the south west, how many equal pay cases in the south, how many equal pay cases in London. If the unions have got a national strategy it shouldn't make any difference, they should have the same strategy everywhere.

Equal pay or fair pay?
Back pay had not been a part of settlements before the mid-2000s when no-win, nofee law firms had first become involved in equal pay litigation. Since claims for six years' back pay were not possible until around the same time, it is difficult to establish a straightforward causal link between the entry of no-win, no-fee law firms, and the growing incidence of back pay claims. Nevertheless, management representatives we spoke to said that back pay had not been part of early settlements following the replacement of the white book by the SSA. When back pay did emerge as a negotiation issue, it was not related to the assessed value of a potential claim but was a 'compensatory sum that [claimants] would accept in return for giving up their right to make a claim', according to one of our interviewees. Amounts cited to us were between £8,000 and £10,000 in agreements struck between 2003 and 2005. Following the Allen ruling, they tended to be calculated as a percentage of an assessed claim and were in the vicinity of £30-40,000 on average, with some claims as high as £100,000, according to media reports. For low-paid women these amounts were 'life changing', according to a union official we spoke to. This official cited a member close to retirement who told her she had 'put central heating in [her] home for the first time ever'. He added that 'without the union, that money would never have been on the table'.
In contrast, the claimant's lawyer we spoke to argued that unions had largely ignored their low-paid female members in negotiations, some of whom were 'working full-time in the most difficult circumstances for £12,000 a year . . . [which] is virtually poverty wages', and had put most of their energy into protecting male workers who were going to see their pay cut as a result of SSA. He cited the case of a local authority where a new pay structure was implemented in 2003-2004 but women got not a single penny in back pay as a result . . . [while] the men got five years, 100% pay protection. . . . We don't come onto the scene until late 2007 [and] within three months of us coming on the scene they're paying out £15 million in back pay to the women. . . . And the unions say 'well we've been negotiating'.
In a different local authority where we interviewed, an HR manager told us that an agreement covering the former White Book manual workers had involved no back pay and six years' pay protection for the mostly male employees. Workers in the male-dominated manual grades had been put at the top of their new grades while the employees in the mostly female grades who had gained a pay rise were put at the bottom of the incremental scale. This council had agreed to look at enriching the jobs of those in receipt of pay protection to limit the impact on them of the ending of the period of pay protection as this 'was a big issue for the unions at the time'.
Unions accepted that they had tried to minimise the impact on the 'losers' under single status but argued that this was an expected part of their role. 'It's traditional in most authorities that have a reorganisation, that a person gets lifetime protection; a person leaves, a new person comes in and gets that job' at the lower rate, we were told. A GMB official in the Allen tribunal had argued that 'for a union 400 S. Deakin et al.
to agree a result, without fighting tooth and nail to prevent anybody from being a loser to a greater extent than is absolutely necessary, would be likely to drive all union members away'. Transitional arrangements had also been included in the SSA itself, though union guidance had highlighted that this could only be for a limited period and 'not applied in a way which perpetuates gender discrimination' (Unison, 1997, p 7). Moreover, the combined advice given to officials by the GMB, TGWU and Unison had stated that single status would result in existing relativities being challenged and that officials should 'be aware that not all members may welcome changes in existing hierarchies (the pecking order) and the effect on differentials that may result' (GMB et al., 1998, p 14).
However, interviews with some union officials suggest that in certain cases at least, unions were negotiating to protect losers on a longer term basis through job enrichment schemes. One official accepted that 'this is going to look a bit obvious if we are only modernising and improving and enriching those services where they were predominantly male before', and hence 'we want every job looked at . . . [but] we would seek for them to prioritise those areas because that's where our members are going to be most adversely affected'. The same official later acknowledged that enriching jobs 'limited the ability of people to make claims'. In the Brennan litigation involving Sunderland City Council, a job evaluation scheme failed where it was found by the EAT to have been implemented in a 'cynical' manner so as to limit the impact on the predominantly male employees.
Unions nevertheless argued that they had to balance the interests of different groups of workers in striking an agreement as well as considering any potential future impact of an agreement, such as job losses, outsourcing or cuts in public services (on this point, see also Thornley, 2006;Colling, 2012;Guillaume, in press). Some employers we interviewed argued that their unions had acted 'responsibly' in negotiations by considering limited council budgets and trying to 'future proof . . . services'. An HR manager suggested that whilst the SSA had resulted in significant pay rises for low-paid women and back pay compensation that 'might well be deserved', this could increase pressures to outsource jobs: 'they might gain in the short term [but] in the long term they could find themselves out of a job' as a result of competition from private sector providers. However, the extent to which claims of a lack of employer resources were justified has been questioned in the light of evidence of central government's reluctance to finance in full the implementation of the equality agenda set out in the SSA (Conley, 2013).

Assessment and conclusion: equality law as an agent of social change
The advent of mass litigation over equal pay has proved enormously controversial in Britain over the course of the past decade, putting into 'stark relief the potential conflict between individuals' legal rights and collective attempts to agree viable, affordable guarantees of long-term pay equality for all employees in large, financially constrained public organizations' (Dickens, 2007, p 483). As we have seen, critics of claimants' law firms accused them of undermining collective bargaining and putting the financial stability of local government and even of some unions at risk. Predictions of union bankruptcies have not been realised, however, and whilst the full impact of equal pay claims on local government finances has yet to be seen, it is unclear how far litigation can be linked to cuts in services or asset sales. Whether the intervention of claimants' law Equal pay litigation and collective bargaining 401 firms led to a shift in union strategy is hard to judge. Some union officials we spoke to maintained that they were moving to a more adversarial stance towards employers and making greater use of litigation before the entry of no-win, no-law firms after 2003. At the same time, comments of some union respondents, to the effect that litigation was sometimes initiated after 2003 in circumstances where it was harmful to the process of reaching an agreed settlement, suggests that the arrival of a new actor, prepared to take a more confrontational attitude towards both employers and the unions themselves, had led the unions to change their position, even if this was not in a way that union officials always felt happy with.
If there is evidence that litigation had, as our law firm interviewee suggested, acted as a catalyst in encouraging unions to pursue back pay and other claims on behalf of their female members, there is also evidence that the law firms' strategies were dependent on the framework of collective pay determination that the unions were at the forefront of maintaining. Without formal pay structures of the kind associated with collective bargaining, claimants' law firms would not have had the opportunity to bring claims of the kind they were pursuing. Without collective bargaining, it is doubtful that pay arrangements would have been sufficiently transparent for litigation to be launched.
The history of equal pay legislation in Britain illustrates the intertwining of collective bargaining and litigation strategies at every stage. As far back as the early 1970s, the Act's objective of eliminating discriminatory pay structures was implemented through changes to sector-level collective agreements which were made in the 'shadow of the law', in anticipation of the Act coming into force in 1975. At a later point, in the late 1990s and early 2000s, the litigation in the North Yorkshire and Cleveland cases was a major catalyst for the SSA, and the implementation of the SSA, in turn, triggered the next wave of litigation in the 2000s. The dependence of litigation strategies on the framework provided by collective bargaining is illustrated by the dearth of claims brought in the private sector during this period, 7 which is also a result of the fragmentation of employment in the private sector, associated with outsourcing. Asked why claimants' law firms focussed their activities on the public sector, the lawyer we interviewed responded as follows: The private sector is much more difficult because the gender segregation is not within organisations generally, it's between organisations. . . . So the restrictions on the equal pay legislation [are] such that if you have an employer who is only doing one job done by women, they have no prospects because they've got no comparator that they can make valid comparisons [with]. So the irony is the more privatisation there is the more difficult it is actually. . . . Privatisation is actually one of the best ways of preventing equal pay claims, either get rid of the comparator or get rid of the claim driver.
If litigation has essentially been complementary to collective bargaining in advancing an equality agenda, the recent British experience illustrates the role that legal actors can play in stimulating institutional and organisational change. In context, in the USA, with a weak state and decentralised court system, the HRM profession emerged as a significant actor in the implementation of civil rights legislation, developing 'private