What will Happen to Race Equality Policy on the Brexit Archipelago? Multi-Level Governance, ‘Sunk Costs’ and the ‘Mischief of Faction’

Abstract

This article considers how one of the ‘archipelago of contradictions’ raised by Brexit is the prospect of unconventional policy change, in so far as it includes – amongst other options – ‘returning’ to prior conventions that were scaled up from the UK to the EU, and then returned to the UK through EU directives. To explore this, the paper divides UK equality legislation into three types: (a) that which was created in the UK (b) that which flows from membership of the European Union and (c) that which reflects an outgrowth of the two. The translation of this into social policy has typically taken a patchwork approach, including a discursive public function which addresses the rights of distinct groups as well as their modes of interaction. The scope and scale of existing equality approaches have therefore become central to the kinds of social and political citizenship achieved by Black and Minority Ethnic (BAME) Britons. While the dangers of Brexit rhetoric are apparent to see, we do not yet know how withdrawal from the EU revises (a), (b) or (c). The article makes a tentative attempt to shed light on these entanglements by focusing on public policies enacted to pursue race equality in particular.

Introduction

Reflecting critically on the scope and reach of Britain's Parliament, the eighteenth-century libertarian Jean-Louis de Lolme complained that ‘Parliament can do everything but make a woman a man and a man a woman’ (1911[1775]: 970). With both surgical and non-surgical gender reassignments now protected in the Equalities Act 2010, de Lolme, were he alive today, might have cause to revise his view. From the perspective of contemporary policy making, this anecdote is illustrative: most obviously because it highlights how equality agendas have evolved and transformed notions of elementary rights (EHRC, 2016; Cowan and Calder, 2013). Perhaps more obviously pertinent for this article, and during a period in which UK equality policy has been informed by a series of European Union directives (especially those flowing from the Treaty of Amsterdam, 1999), the anecdote invites us to register the multiple levels beyond Parliamentary sovereignty through which this has come to pass.

Interesting in and of its own right, this is an especially urgent task in the era of Brexit. If we are to grasp the implications of the referendum outcome for UK equality policies, the article argues that we shall need to track what we might call unconventional policy change. It is unconventional in so far as some UK equality frameworks had been ‘uploaded’ to the EU level before being returned to the UK through EU directives to all member states. This also provided a catalyst for other changes to UK approaches. The main question this raises – i.e., where does withdrawing from the European Union leave us – might be broken down into three parts: (1) what are the conceptual approaches best suited to a policy analysis of this topic – specifically: which analytical tools might yield the most insights? (2) does Brexit mean that the UK will ‘return‘ to prior conventions that were scaled up from the UK to the European Commission? (3) given that these conventions in turn have revised UK approaches, can EU approaches be unstitched from UK approaches without altering what has been established in both?

The article will take up these general questions with the particular example of Race Equality; something that has a distinctive UK policy character that is traceable across multiple levels. This is grounded in an approach through which post-war migrants who arrived as Citizens of the United Kingdom and Commonwealth (CUKC), and subsequent British-born generations, have been recognised as ethnic and racial minorities requiring state support to overcome distinctive barriers in their exercise of citizenship. While not unproblematic, it is a markedly different method to that of European neighbours with comparable colonial histories. For example, France pursued a robustly assimilationist strategy in which equality was understood as uniformity and, until the beginning of this millennium, Germany had a ‘returnist’ approach in which labour migrants were guest-workers (gastarbeiter) expected to return to their country of origin (Meer and Modood, 2011). In the UK, in contrast, under the remit of several Race Relations Acts which approached equal opportunity as equal access, the state sought to proactively integrate minorities into the labour market and other key spheres of British society (e.g., education, health and political participation).

Indeed, it is now over 40 years since the introduction of a third Race Relations Act (1976) cemented a state sponsorship of race equality by consolidating earlier, weaker legislative instruments (RRA 1965, 1968). Alongside a remit spanning public and private institutions, recognition of indirect discrimination and the imposition of a statutory public duty to promote good ‘race relations’, it also created the Commission for Racial Equality (CRE) to assist individual complainants and monitor the implementation of the Act. Such developments, too, reflected ongoing community based anti-racist struggles. There are many examples but perhaps one that stands out is the Lawrence family campaign into the improper investigation of the racist murder of Stephen Lawrence, and attendant findings of institutional police racism (MacPherson, 1999). Some scope for redress therefore, against racially structured barriers to participation, represents one characteristic in the British approach to citizenship, and serves as the cornerstone of an unwritten ‘British multiculturalism’ (Meer et al., 2015). Another way of putting this is to say that the public function of anti-discrimination policy has been integral in cultivating a ‘web of beliefs’ that characterise certain political traditions (Bevir and Rhodes, 2003), and from which several things have flowed: pluralised national identities are possibly the most obvious (Uberoi and Modood, 2012). It is essential therefore to understand that race equality in the UK has always had a discursive character that goes beyond public policy and administration, not least because it tips into debates about identity, belonging and community formation (Meer, 2015).

Since the referendum result, both governmental and non-governmental organisations have recorded a spike in racially motivated activity (National Police Chiefs Council, 2016; Institute for Race Relations, 2016). Official police figures report a 49 per cent rise in incidents to 1,863 in the last week in July 2016 when compared with the previous year (the following week saw a record 58 per cent increase to 17,870) (Travis, 2016). Some of these incidents have been documented in real time and shared via social media, illustrating how racialised antipathy toward BAME groups, as well as white Europeans, often invokes rhetoric on national identity and belonging.

Commentators and researchers have rightly moved to consider whether this pattern of activity may be isolated to the referendum, or if it forms part of a wider trajectory with underlying dynamics that the referendum amplified (Young, 2016; Devine, 2016). This is necessary and valuable work, but what remains overlooked is the prospective status of race equality apparatus as a matter of public policy. Since we are only just coming to terms with the overwhelming ‘archipelago of contradictions’ (Shaw, 2016: 104) raised by Brexit, this oversight is one amongst many. Archipelagos of course can be found isolated in large amounts of water or neighbouring a large land mass. Borrowing from Shaw, we might figuratively use the designation to grasp some of the policy challenges and possible contradictions of the UK as a ‘Brexit Archipelago’. In this landscape it is paramount not to lose sight of the fact that race equality in the UK is a question of public administration, and civil and criminal law, as much as it is a matter of public discourse and everyday attitudes. A key task therefore is to consider what this race equality will look like in Brexit Britain.

Outline and methodology

To this end the article proceeds in three stages. The first sets out the conceptual relationship between the policy process and multi-level governance, and how developments in race equality might be understood through these frames. The second stage tries to divide race equality legislation into three types that function across multiple policy levels: (a) that which was created in the UK, (b) that which flows from membership of the European Union, and (c) that which is an outgrowth of the two. These concerns have been explored through a mixed-method case-study analysis. Most closely associated with the work of the Chicago School, case-study research has been marked by periods of intense use and disuse throughout modern social science inquiry (cf. Feagin et al., 1991). Its under-use is somewhat surprising given it is very hospitable to ‘how’, ‘what’ and ‘why’ questions. The multiple sources of evidence used in this article are derived from projects having a common concern with race equality at EU, UK and Scottish levels, and accord with those found in Yin's (1994) typology of (i) policy documents and government bills, (ii) archival records, and (iii) semi-structured interviews. The latter includes interview data with stakeholders involved in EU and UK equalities legislation, utilised to address gaps in the public policy literature. This data is derived from research supported by the European Commission ¹ , the Economic and Social Research Council ² and the Royal Society of Edinburgh (RSE) ³ . Cumulatively, of the more than 60 interviews available, 15 are directly and indirectly used here in order to address gaps in the policy literature. The third stage of the article considers the prospective options in the light of Brexit, and the discussion in stages one and two. This shows how, during its membership of the EU, Britain has typically pioneered race equality legislation. This means that the status of existing settlements is subject to their own dynamics, too, in a way that should not be attributed to Brexit alone.

Part 1: The policy process and multi-level governance

To the extent that there is a prevailing account of what the study of policy process should incorporate, we might say this seeks to explore a series of distinct but interrelated ‘stages’ (Hogwood and Gunn, 1984). While the precise configuration and description of these tend to reflect the wider theoretical stance that is adopted, they typically comprise: problem identification, agenda setting, consideration of potential actions, implementation of agreed action, and evaluation. When linked together this is what is deemed to comprise a ‘policy cycle’ (Goodin et al., 2006). Somewhat deontological, these frames are best when read as non-linear ideal types, but which nonetheless envelope a significant breadth and depth in approaches to policy analysis (cf. Smith and Kattikireddi, 2013). What is worth registering are the ways it is increasingly common to see the use of ‘synthetic approaches’ (Ayres and Marsh, 2013: 645) that employ these stages to bridge across a number of theoretical positions. Recognising that policymaking is a complicated and dynamic process, it is difficult at the outset to employ ‘one model’ that spans path dependencies and flux, agents and coalitions. It is easier to state that most policy process literature has been less interested in charting the mechanics of the policy cycle across multiple levels. Perhaps this is due to a prevailing tendency of methodological nationalism across the social sciences, something which can fix our aperture at state levels. One of the arguments of this article is that, if we are to grasp the fate of race equality policy in light of Brexit, we need to bring in some of the literatures better suited to understanding policy making and remaking between the UK and EU. For, despite what is claimed by Brexiteers, the EU has never been a ‘blueprint for a workable system of government’ (Marks and Hooghe, 2004: 17) but, instead, reflects ‘an increasingly complex pattern of policy-making’ between member states. Therefore, to provide a set of conceptual linkages across the policy process and the EU, in order to understand the prospective status of race equality post-Brexit, it would be useful to turn to some of the literatures on multi-level governance (hereafter MLG).

This is not a simple task. The literatures on MLG are wide-ranging and reflect competing traditions of thought. To avoid getting entrenched in important but lengthy debates, and in the tradition of Wittgenstein, the article approaches the meaning of concepts as something derived from their use. This is not the same as ignoring the provenance of concepts; being sensitive to this is important in understanding how such meanings have come to be been forged. Bache and Flinders (2004: 2–3), for example, trace MLG to ‘a new wave of thinking about the EU as a political system rather than [as] a process of integration. . .that followed swiftly from the accelerated deepening of the integration process in the mid to late 1980s’. The concept therefore closely tracks the dynamic formation of EU and attendant political settlements. This is a view supported by Awesti (2007: 3) for whom MLG emerged ‘as a response to the state-centric, intergovernmentalist theory of the EU which dominated studies throughout the so-called period of ‘eurosclerosis’ ⁴ from the late 1960s.’ Jordan (2001: 196) perhaps put it most succinctly when he characterised MLG as part of an intellectual effort ‘to investigate the various parts rather than the whole of the EU’.

The ‘parts’ here describe a variety of actors in the policy process across public, private and civil society but in ways that are not limited to state boundaries. Here governance is as important as multiple levels, specifically in the view that European societies have grown so complex, dynamic and differentiated that no single policy approach commands hierarchical control (Sabel and Zeitlin, 2012). An understanding of governance thus needs to grasp the role of networks that blur boundaries between state and civil society, and which rely on organisations and NGOs at various levels of consultation and partnership (O'Toole et al., 2016). If we take the European Commission Code of conduct on the partnership principle for Cohesion Policy 2014–2020 (European Commission, 2012: 3), both features are included and framed in the following terms: ‘Multi-level governance means coordinated action by the European Union, the Member States and local and regional authorities, based on partnership and aimed at drawing up and implementing EU policies’. The challenge for policy analysis then is not only to ‘grasp the dynamics of scale’ (Davies, 2013: 497) but, given the complexity of the policy process, to consider why some approaches are successful in this environment and others are not.

This question is integral to understanding the EU Race Equality Directive (RED) that emerged from the Treaty of Amsterdam (1999) and which came into force in 2000. As elaborated below, the answer partly relies on the idea of ‘policy windows’, informed by Kingdon's (1995[1984]) view that an opportunity for policy change opens when a number of ‘streams’ coalesce. In the case of the RED directive, and perhaps reflecting what McGoey (2012) understood as the strategic role of ignorance, the ‘EU was proceeding from a low base in this area’ (Geddes and Guiraudon, 2007: 127) in a manner that undoubtedly elevated UK expertise.

It is worth registering that the EU's legal competences in the social policy field across the board are otherwise quite minimal and, in this respect, RED is unusual. What made the incorporation of RED easier were at least two conditions for ‘uploading’. Namely, that British approaches were actively taken up by EC policy makers, and RED amounted to ‘a relatively costless measure that mapped well with an already existing national framework’ established in the UK practice on race equality (Geddes and Guiraudon, 2007: 129). In order to discuss and address the policy analysis challenge I set out earlier, we first need to understand the constituent features of UK and EU race equality policy. This is presented in part two below. Before we move to this, and in an effort to caution against seeing policy processes as uni-directional, it's important to register that what came back from the EU following this ‘uploading’ was co-terminous with a decade of policy ferment in UK anti-discrimination, something that led to novel outcomes because it tried to address the intersecting nature of inequalities.

One of the specific characteristics of this policy ferment, discussed below, is that it was fostered by a broadly conceived intersectional approach, of the kind ultimately enshrined within The Equality Act 2010. This Act was deemed ‘a major landmark in the long struggle for equal rights’ (Hepple, 2011: 1), and considered to be an advance on other paths to domesticating European legislation elsewhere in the EU (O'Brien, 2013). Precisely how much of this is a debt owed to MLG policy processes?

Part 2: Something old, something new, something borrowed. . .

Let's begin with legislation. The first thing to say is that it is a curious feature of British citizenship that its possession has never conferred a formal right to non-discrimination, not least because the UK has no ‘written’ constitution as found in many liberal democracies. What has been amassed instead is a body of legislation that is overseen by the judiciary and intermediate organisations, and which protects both citizens and non-citizens from discrimination on specific ‘grounds’. As I outline below we might divide this legislation into three kinds:

1. (a) that which was created in the UK

2. (b) that which flows from membership of the European Union

3. (c) that which is a combination between the two.

The key question is to what extent withdrawal from the EU revises a, b, or c? This is elaborated by examining the interaction of specific British approaches and generic EC directives. The article then considers to what broader approaches these directives may be tied politically, as well as legally, with respect to human rights discourses, the creation of the Equality and Human Rights Commission (EHRC) and the overarching Equalities Act 2010. Let's begin with (a).

(a) Endogenous equalities

Perhaps some of the most powerful anti-discrimination instruments enacted in Britain can be found in the statutory torts of unlawful discrimination found in both the 1975 Sex Discrimination Act and the 1976 Race Relations Act, which prohibit direct and indirect forms of discrimination, and impose statutory duties of care. For example, section 1(1)(b) of the 1976 Race Relations Act is modelled on the 1975 Sex Discrimination Act. It results, on the one hand, from a practical concern to enact tested legislation. On the other hand it was a shrewd political manoeuvre by the then Home Secretary (Merlyn Rees) to find cross-party support for race relations legislation from a variety of quarters (precisely because they had already supported sex discrimination legislation).

Section 71(1) of the 1976 RRA (as amended primarily in 2000 and 2003), in particular, required all public authorities to adopt a general duty to promote race equality, and sought to eliminate racial discrimination, ensure equality of opportunity and promote good ‘race relations’ (through such things as outreach work and diversity awareness training). There are also specific duties enabled by the legislation, such as the implementation of a written policy on race equality, perhaps as part of an overall policy; an assessment of the impact of new and current policies on minority ethnic staff, students and other service users; the monitoring of recruitment and progression of minority ethnic staff and students; and the monitoring of grievance, disciplinary, appraisal, staff development and termination procedures by ethnicity. The Secretary of State is also empowered to impose specific duties on key listed public authorities. Broadly, these selected authorities must publish a Race Equalities Scheme and meet specific employment duties (the scheme is effectively a strategy and action plan).

It is notable that these measures have developed in a way that places a specific emphasis on managing group relations. In doing so, Britain has perhaps borrowed something from an American approach to tackling racism. If this is so, then it has also gone further in focusing on how society can achieve fair treatment for different groups, something that reaches beyond how these groups might blend into society. This means that British anti-discrimination frameworks tried to ‘address the rights of distinct groups as well as their modes of interaction, and so are not merely concerned with the rights of individuals’ (Rudiger, 2007: 46).

These anti-discrimination frameworks comprise a significant body of legislation that applies to all ethnic and racial minority groups categorised according to race, colour, nationality (including citizenship) or ethnic or national origins. To illustrate how different it was to the approaches of EU neighbours we might register its dynamic quality. For example, some religious minority groups, especially Sikhs ⁵ and Jews ⁶ , have in case law cumulatively established precedents in the application of this legislation to protect them, though in a way that has not been extended to all religious minority groups (Meer, 2015). What is nonetheless important to note is that while ethnic and racial categories vary across the EU, a stand-out feature that according to Simon (2005: 14) has been taken up is the idea of indirect discrimination. This is something most fully pioneered in UK approaches, and characterised as ‘a source of one of the key trends in matters of non-discrimination and promotion of equality’ (ibid.). It is to this that we now turn.

(b) European Equalities

European Commission (EC) directives derived from Article 13 of the 1999 Treaty of Amsterdam came into force in the first part of the millennium. Under a Labour government these directives were co-terminous with important changes in established legal responses to racial and religious anti-discrimination measures in Britain. Prior to this, across the EU, there had been no consistent level of protection against racial discrimination (and little legislation that has consistently protected people from discrimination that takes place on grounds of religion, disability, age or sexual orientation ⁷ ). In recognition of this, the EC introduced the following Article:

Without prejudice to the other provisions of this Treaty and within the limits of the powers conferred by it upon the Community, the Council, acting unanimously on a proposal from the Commission and after consulting the European Parliament, may take appropriate action to combat discrimination based on sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation. (Office Journal of European Communities C325/33: 11 and Council Directive 2000/78/EC)

This Article was enacted through the issuing of two directives. By 2014 it was reported that both directives had ‘been transposed into national laws in all 28 Member States and the conformity of all those laws with the Directives had been [satisfactorily] checked by the Commission‘ (EC, 2014). The important point for the purposes of our discussion is that it also meant that ‘in effect, the British framework has been “uploaded” to EU level’ (Geddes and Guiraudon, 2008: 129). What this claim doesn't sufficiently answer is why? An interesting account here might be garnered from Kohler-Koch (1996: 362–3), and concerns the logic behind other examples of ‘uploading’ to the EU level, and specifically about shifting policy problems from the national to the European agenda to avoid public pressure. In his words:

[M]ember states themselves. . .[have] considered joint problem-solving to be more attractive than preserving their national autonomy. As a consequence, governments may accept a further transfer of authority to the [European] Community to increase, at least indirectly, their problem-solving capacity (ibid).

There is a clear rational choice reading at work here. Namely, policy problem solving is mutually beneficial, but first and foremost in the obvious self-interest of national governments. It is worth problematising this by registering the role of coalitions and lobbying too. Geddes and Guiraudon (2007: 134), for example, persuasively show that while ‘NGOs played no role in the negotiation of the directive and were not at the negotiating table. . . the kinds of ideas mobilised by the SLG [Starting Line Group] were present and did have influence on the legislation because they had been fed into the Commission policy’. The Anglo-Dutch-led Starting Line Group (SLG) was a network of more than 250 NGOs, which actively sought the inclusion of anti-discrimination articles within the Amsterdam Treaty (Amiya-Nakada, 2007). Carles (2008), too, points to ‘results from a social mobilisation process and from policies elaborated at international and European level’ by civil society actors. In this respect it could be argued that, in these policy developments, certain ‘policy coalitions’ (Sabatier and Jenkins-Smith, 1993) were operative. For example, a strong coalition of stakeholders and advocates across equality strands was successful in ensuring that a unification of equalities legislation would neither risk regression nor levelling downward. Geddes and Guiraudon (2007: 126) put it as follows:

The ways in which these acquired some purchase at EU level and were then broadly reflected in the resultant legislation. . .tells us quite a lot about the constitution of the EU as a political field and the kinds of political and social capital that are privileged within it.

The implication is that the EU had spaces which ‘tended to privilege networks of expertise’ (ibid. 131) which propelled not only the British but also some variation of the Dutch model, both of which are ‘linked to a network of actors including NGOs and academic activists with good links to European institutions, particularly the Commission and the Parliament’ (Geddes and Guiraudon, 2008: 133). Perhaps the key point here is that scholarship on multi-level governance has tended to be taken up in ways that emphasise the ‘multi-level’, but underappreciate the ‘governance’ feature of the relationship. Or perhaps, more precisely, to think of governance in terms of the distribution or diffusion of ‘governing’ arrangements, rather than the ways policy-making processes are also subject to governance dynamics. This is not necessarily a flaw in the design of the concept, but an observation on the limited ways in which it has been used to think about the alternating and diffuse means through which policy can be made. For although never present at actual negotiating tables, the involvement of the SLG illustrates the way in which tested practices from British and Dutch contexts could be marshalled and mobilised to influence ‘the content of legislation because they had been fed into the Commission policy development process’ (Geddes and Guiraudon, 2008: 133). One participant in the process, an expert in discrimination law who monitored the impact of the directive, provides an interesting contextual insight:

[T]he Race Directive was approved in Europe very quickly – this was when Haider had been elected in Austria and there were too many racist incidents across Europe so no member state would want to be seen to be voting against an anti-racism measure. For the next directive there was much more politicking going on. The Catholic Church used its influence so protection for religious organisations is particularly good. (Interview with author)

The RED required member states to make discrimination on grounds of racial or ethnic origin unlawful in employment and training. Unlike the Employment Directive, it went further in requiring member states to provide protection against discrimination in non-employment areas, such as education, access to social welfare, and the provision of goods and services. In many respects it mirrored the approach of the RRA in promoting proactive initiatives for combating discrimination in member states. RED required member states to establish bodies as an institutional support for equal treatment provisions, but it is arguable whether it endorsed proactive measures to promote equality within institutions. For example, the positive action clause in the directives, which was phrased as an exception rather than as an explicit means to achieve equal treatment, ‘offers an insufficient basis for such an approach [because] the Directives remain focused upon individual litigation against specific acts of discrimination once they have occurred’ (Rudiger, 2007: 49).

This raises the question as to what contribution these directives could make where they did not increase the levels of protection that were already available in Britain. According to a British MEP closely involved in scrutinising the legislation:

[T]he Race Equality Directive was not a huge advance because we already had a fairly comprehensive Race Relations Act. But it did improve it in the area of burden of proof and it also showed the UK that this was the right way to go and stopped any kind of regression or going backwards in terms of race equality legislation. The key areas which are still to be fully developed in the UK are disability and age, and in both these areas the Employment Directive has been helpful in pushing the UK further than we were going, as well as religion of course (Interview with author).

In this respect the impact was mainly political in shoring up a particular approach. In practical terms it moved the burden of proof away from the claimant onto the organisation or party against whom a charge of discrimination is made. Both these reasons were cited as assets by UK Race Equality Network (2016) during the EU referendum.

The fact that the UK has to comply with this directive was an important reason why the current government could not and did not attempt to weaken our race discrimination laws when it was considering how to implement the Red Tape Challenge and cut down on laws and regulations that affect business. Additionally, it is important that UK nationals regardless of race are protected when they work in other member states (UK REN, 2016).

Despite marking an advance on what had gone before in a number of EU countries, other weaknesses of the Race Equality Directive quickly became apparent. One was that it stopped short of requiring substantive positive action, another is that it relied on legal complaints rather than effective proactive rule-making within organisations. As Hand (2008: 599) observed, the directive did not ‘expressly permit the adoption of “measures providing specific advantages” but merely the adoption of “specific measures”’. Perhaps a more nebulous criticism is that while at an EU level it advised a dialogue on racism, it did not monitor this or propose an EU-wide response to it. In many respects then, although the RED marked an advance elsewhere in Europe, it was not of its own accord an advance in the UK. A better way to see it is as a catalyst, especially when coupled to the impact of a human rights agenda, something to which we turn next.

(c) Catalysts for change?

The late John Urry (2004: 5) once noted that ‘relationships between variables can be non-linear with abrupt switches occurring, so the same “cause” can, in specific circumstances, produce different effects’. Perhaps this is one way to think about our third type of equalities legislation. An example of type (c) is the 1998 Human Rights Act, which brought into domestic law key provisions of the European Convention on Human Rights (ECHR). One of the contributors to the legislation put it as follows:

The way to understand the introduction of the Human Rights Act is in the context of a debate about the Bill of Rights rather than as specifically about equality and discrimination although that was certainly part of it. [. . .] However, it's not particularly helpful at dealing with all the details of discrimination in everyday life, which is why you also need specific equalities legislation . . . for example, the Human Rights Act cannot be used directly against private bodies, although all law must comply with it, while specific discrimination legislation directly impacts on the private sphere. (Interview with author)

The aforementioned Bill of Rights is an unsettled topic that is expected to return to the Brexit archipelago. What form this might take is unclear, but the Bill of Rights Commission (2012) under the Coalition Government (2010–2015) actually sought to build on the European Convention of Human Rights (EHRC). It was also notable that the difference between a concern with rights and equality quickly became apparent in the way that the Human Rights Act promotes a more individualistic approach. Since it considers the majority of people in need of protection from some form of discrimination, it perhaps risks de-emphasising specific experiences of historically disadvantaged minority groups. The implication for prospective policy making in this area is that uniform rights for individual citizens could take precedence over recognising the situation of diverse and disadvantaged groups in society.

This might facilitate a shift from a group-based approach to a focus on individual rights. While such a move might assist the principled operation of human rights legislation in promoting, for example, the right to religious freedom, it may be less sensitive to promoting specific anti-discrimination measures. So something modelled on a Human Rights approach could protect the right to practise religion in accordance with religious beliefs, as is exemplified by provisions including Article 9 of the ECHR. This, however, is different to the remit of anti-discrimination measures. The latter would be concerned with how discrimination against religious minority groups picks out individuals on the basis of discernible characteristics, and attributes to them an alleged group tendency, or emphasises those features that are used to stigmatise or that reflect pejorative or negative assumptions based on the individual's real or perceived membership of that group.

This raises the question, then, as to whether an increasing focus on the former risks ignoring how different minority groups are disadvantaged in different ways, and moves the emphasis away from a more specific recognition of diversity. Rudiger (2007: 52) argued that this was observable at the EU policy-making level where a new focus on human rights informed the change from the European Centre on Monitoring Racism and Xenophobia into the Fundamental Rights Agency. It is to these issues that we now turn through examples of recent changes to longstanding British approaches that have historically recognised diversity in their promotion of equality.

According to one respondent, it was widely understood that ‘another review of discrimination law was unlikely to happen again for a long while’ and that this presented the opportunity ‘as a bare minimum to harmonise some quite disparate pieces of legislation’ (interview with author). There is a convoluted history as to what this review initiated: before the 2010 Act, the Equality Bill (released in May 2008 and sent for consultation in July) combined all UK equality enactments so as to provide comparable protections across each equality ‘ground’. The Bill itself relied on the legislative framework of prior UK legislation, with amendments, found in the Sex Discrimination Act (1975), Race Relations Act (1976) and Disability Discrimination Act (1995). Those explicitly mentioned in the previous 2006 Equality Act (which created the EHRC) included age; disability; gender; proposed, commenced or completed gender reassignment; race; religion or belief; and sexual orientation. This Act was presented as a blend of traditional non-discrimination obligations, substantive equality goals around equal participation, and statutory duties to promote respect for diversity, human dignity and human rights. It was particularly noteworthy because it was the first occasion on which equality and diversity were expressly tied together (see especially s.8(1) and (2) of the 2006 Equality Act).

The QC and legal scholar Bob Hepple (2010) pointed to four key milestones in the resulting 2010 Act. These included (i) an integrated perspective of equality law that is promoted by a harmonized (single) Equality body, namely the Equality and Human Rights Commission (EHRC); (ii) consistency in definitions of discrimination, harassment and victimization across different ‘grounds’ i.e., identity categories of age, sex, ethnicity, religion, disability, sexuality and human rights; (iii) the extension of ‘positive duties’ which proactively promoted equality in public authorities across all characteristics (initially including socio-economic inequality); and (iv) expanding the application of how ‘positive action’ is practiced (not to be confused with ‘positive discrimination’). In his view therefore it is a ‘transformative Act’ (Hepple, 2014: 1).

Despite all of this, it is striking that in the final 2010 Equality Act there is only a partial commitment to intersectionality in so far as principle 12 allows multiple discrimination but only on two grounds, each of which have to be claims of direct discrimination (rather than harassment or indirect discrimination). According to Hepple (2010: 16) this reflected a concession to the business lobby ‘who opposed any provision on multiple discrimination’ on the ground that it would become “unduly burdensome” to business’.

Part 3: The mischief of faction

In the previous two sections we have discussed the challenges of how to understand the prospective status of UK race equality in the context of Brexit. To do this the article has turned to MLG literatures, and then distinguished between the provenance of different legal and policy materials that make up the patchwork of existing UK race equality policy. The conceptual task is to connect the policy change literatures to the MLG literatures. It is clear that an adequate understanding of public policy developments in this area must incorporate an understanding of the multiple levels across which race equality in the UK has been forged. Some of this is augmented by an understanding of the ways in which ‘advocacy coalitions’ proved effective in propelling some change over others (Sabatier and Jenkins-Smith, 1993). Some of it has reflected an EU level ‘window’ thrown open by a specific ‘policy constellation’ that formed part of the Treaty of Amsterdam (Rudiger, 2007: 41). It has been argued that, while multi-level governance has tended to be taken up in ways that emphasise the ‘multi-level’, it has underappreciated the ‘governance’ feature of the relationship, especially the ways policy-making process is also subject to governance dynamics. The unanticipated consequences of this have certainly been profound because they proved a catalyst for changes to existing approaches to race equality in the UK, especially by tying together a number of different grounds in addition to race i.e., driving home an intersectional agenda through the 2010 Equalities Act and the creation of a single Equality and Human Rights Commission (EHRC).

In this respect MLG literatures are useful in helping to gauge the character of these developments, and especially reading off their dynamic and multi-directional flow e.g., not just from the UK to the EU but vice versa. There is also further possible explanatory currency in taking seriously the MLG assumption that ‘adaptation to EC institutions and policies drastically increases the cost of exit from existing arrangements for member states’ (Pierson, 1996: 144–145). With this in mind it is difficult – although not impossible - to imagine that a post-Brexit administration would actively unpick the domestication of the equality components of the Treaty of Amsterdam (1999), especially since the UK was already compliant with race equality provisions. In other words, the Equality Act (2010) is not merely about being compliant with EU directives, but reflects an endogenous trajectory in incorporating the existing race equality provisions discussed in part two. This means that the ‘sunk costs’, as Awesti (2007: 17) calls them, and which extend to wider patterns of societal level adaptation, are likely to be sufficiently ‘sticky’ regardless of the ‘current preferences of political leaders’ (ibid) – or what James Maddison memorably termed the ‘mischief of faction’.

This should most certainly not be taken to imply that Brexit will not diminish race equality in the UK. The myriad of Brexit uncertainties are such that they might be modelled through a Johari window model (e.g., that which is known to be known and that which is known to be unknown, etc) (Luft, 1969). In terms of the politics, if the ‘policy image’ (Jones, 1994) of race equality comes to be associated with an EU imposition, rather than an endogenous creation, it becomes much more contested by those seeking to uncouple as much as they can from the EU. Put in other terms, if race equality becomes part of the ‘collateral rhetoric’ (O'Brien, 2013: 490) of Brexit, rather than a longstanding – though unsettled and incomplete – British conversation, then the symbolic politics of this would do great harm. Equally, in the area of trade agreements, there may well be economic incentives to reduce equality protections that would lead to either a direct or indirect discriminatory outcome for BAME groups. This is all speculative of course, but there are obvious conditions which might cultivate such a move. Mulder (2016) offers the following scenario concerning gender equality in particular but the lesson is salutary and registers with race too:

[I]t is not difficult to see how a UK government facing turbulent economic times and having put UK businesses in the position of having to compete with the rest of the world outside of a trading block, would feel like there is little choice other than to reduce the level of protection and return to the ineffective and formalistic legal framework . . .

Trading off race equality in the name of efficiency and competitiveness is therefore a very real concern, but to some extent such a scenario would mark an accelerated roll back on commitments which are already precarious. For example, the Coalition Government (2010–15) significantly undermined features of the Equality Act in the name of deregulation and competitiveness, summarised in O'Brien's (2013: 486) complaint, that ‘innovations in the structure of equality law forged by a generation's experience of its application are in danger of being brushed aside with scarcely a political murmur’. The point being that UK governments already have the capacity to roll back from race equality commitments, and have shown the political will to do so, independent of other developments.

The Equality Act 2010 was possibly the final piece of substantive legislation introduced by the Labour government (1997–2010), and the Conservative-Liberal Coalition inherited it before the legislation had bedded in, and so were charged immediately with delivering it. Perhaps illustrating a direction of travel, the then Home Secretary, now Prime Minister May, announced a review of the public sector equality duty in the government's ‘red tape reduction challenge’. The report of the Independent Steering Group on the Public Sector Duty struck an ominous tone, especially their description of ‘useless bureaucratic practices which do nothing for equality’, concluding that ‘Equality is too important to be tied up in red tape. Let's cut it out’ (Government Equalities Office, 2013: 7). As O'Brien (2013: 486) notes, the review itself offered a ‘reminder of the need for constant vigilance if hard won institutional gains are not to be squandered’. A reminder too, if needed, of the capacity of UK governments to set an agenda of regression on race equality commitments within the existing parameters of multi-level governance arrangements.

There may however be other possibilities, too. For example, even though Equalities are broadly reserved to the UK Parliament, a certain kind of drift in MLG has meant that Scotland is simultaneously the same and different. Schedule 5 of the Scotland Act 1998 (c46) basically incorporated the functions of the third Race-Relations Act (1976). In May 2012 however, the Scottish government placed specific duties on public authorities, also known as the Scottish Specific Duties, requiring a listed authority to publish a mainstreaming report on the progress it has made in integrating the three needs of the General Equality Duty (GED) to: (i) eliminate unlawful discrimination, harassment and victimization; (ii) advance equality of opportunity, and (iii) foster good relations. This was followed in 2016 with a new Race Equality Framework which marked a further contrast with the discontinuation of statutory equality impact assessments in England, and possibly marks a divergence from understanding race equality instruments as an administrative burden.

Conclusion

During the consultation on harmonising different equality bodies and different equality legislation, a concern that was repeatedly voiced drew attention to the risk of rolling back previous race equality achievements through other means. In particular, there was a fear that, in reshaping the sector so radically, and even if there was no immediate ‘dilution’ and settlements were ‘levelled up’ across different grounds, it would undermine existing settlements. This is because, when a distinct race equality commission was no longer able to publically agitate for race equality, and when legislation became streamlined, a less favourable political administration in more cash-strapped times would encounter less resistance. In their study Craig and O'Neil (2013) showed precisely these developments. They noted, for example, that the budget of the harmonised EHRC was reduced by the coalition administration to the equivalent of less than one of its constituent bodies (from £70m when it started in 2007 to £17m presently). They also highlighted how statutory equality impact assessments have been discontinued. In addition, Ware (2013: 8) notes how the single Equality Act had the effect of diluting funding to communities themselves.

In the best traditions of race-equality, however, the response to these developments needs to stem from an anti-racism that mobilises and agitates across a range of sectors, taking in the arts and education in addition to political and policy arenas. This reading is consistent with a longstanding view that anti-racism is an unsettled, incomplete and on-going pursuit. The theme running throughout this article is that, during its membership of the EU, Britain has typically pioneered race equality legislation. This means that the status of existing settlements is subject to their own dynamics too, in a way that should not be attributed to Brexit alone.