This paper argues, first, for UK equality law to be liberalised to allow organisations more scope for conduct that is deliberately calculated to benefit social groups whose members have been systematically denied, or disadvantaged in their enjoyment of, social, economic and other goods. Secondly, drawing especially on the evolution of UK and EU equality law, I contend that the shape and implementation of law reform should follow a model in which law guides and informs organisational experimentation with such initiatives.
My account of positive action is conceptual, aiming to capture the necessary components of this behaviour. So positive action is taken to be activity designed to improve the position, in terms of the distribution of benefits or dis-benefits, of a given social group or sub-group (or of several such groups), on the basis that its members suffer systematic disadvantage in that regard. Such steps might take many forms and, in particular, might be more or less influenced by legal mechanisms related to equality. It also follows that, while positive action includes positive discrimination in the sense of goods being allocated (or detriments imposed) wholly or in part on the basis of identity, it covers much else besides.
It is important to be precise on this point because the phrase positive discrimination is often now loosely deployed only to mean practices that allocate substantial benefits and do so solely on the basis of identity. On this basis positive discrimination (malign) is contrasted with positive action (benign), frequently with the aim of rescuing the latter from disapprobation that it encounters in contemporary debate.Footnote 1 Yet such characterisations of positive discrimination are misleading in referring only to one type, and an extreme and unlikely one at that (whether in the UK, the EU, the US or anywhere else). This rhetorical move therefore elides difficult conceptual questions about positive action that need to be tackled while, ironically, ceding too much to simplistic arguments that decision-making should ignore identity.
In his 1986 article about the employment context, McCrudden identified five activities that “positive action might include, not in the sense of what is legally permissible, but in the sense of how the term appears to be used in common parlance”.Footnote 2 These were: first, “eradicating discrimination”; secondly, “facially neutral but purposefully inclusionary policies” by which employers sought to increase the proportion they employed from excluded or under-represented groups; thirdly “outreach programmes” in the sense of targeted advertisement, encouragement and training to attract qualified candidates from under-represented groups; fourthly, “preferential treatment in employment”, by which was meant under-represented groups being favoured in, for example, hiring, promotion or redundancy decisions; and, finally, “redefining merit”, through group membership being included in the qualifications for a given job.
This gives useful insight into what was then perceived as positive action. But there are reasons to use a more abstract definition here. First, it is notable that the last two categories, and arguably also “outreach programmes”, indicate that positive action was then accepted to include various types of positive discrimination, when usages now may have altered. In part building on this observation, a conceptual approach recognises that we do not know all the ways that contemporary organisations either do already, or might in future, purposefully devise measures to redress identity related inequality, disadvantage or exclusion. This starting point therefore enables a more complete analysis of the relationship, whether constraining or enabling, between law and organisational initiatives of this kind. This shift also encompasses positive action in the delivery of goods and services and in the performance of other non-employment organisational functions.Footnote 3 While the arguments point to a need for ongoing research about organisational practice, this emerges from the discussion rather than the empirical position being taken as prior.
Secondly, conceptual analysis surfaces difficulties with certain kinds of activity being automatically characterised as positive action. It is of interest that McCrudden reported his first category, “eradicating discrimination”, as the only type that was then at all well known in Britain, noting that it was permitted by the three then anti-discrimination statutes and was a legal duty under the Sex Discrimination Act (“SDA”) 1975 and Race Relations Act (“RRA”) 1976.Footnote 4 Yet if decades of anti-discrimination law have taught us anything, it is that eradicating unlawful discrimination may not always defensibly be characterised as redressing systemic inequalities. It all depends on the underlying rules. As the following analysis exposes, some anti-discrimination norms are supportive of existing patterns of distribution. Basing my enquiry on what organisations perceive as positive action (probably influenced by law) would therefore run the risk of assuming that which ought to be in question. Besides, it is perfectly possible that the empirical position has changed here also. Some organisations are perhaps now even likely to have a detailed awareness of when enforcement of anti-discrimination norms is an impediment to distributional change.
This subject-matter is not, however, to be conflated with the evolution, in theory or in practice, of positive duties to promote equality. It is true that innovation with such duties in the UK adds to the urgency of re-thinking the law on positive action. Moreover to the extent that recent UK experience is reflective of a more general trend, what is said will have resonance elsewhere. But the focus remains the UK law on positive action. In particular, the essay is as concerned with private as with public organisations, when UK duties to promote equality apply directly only to public authorities and to the exercise of public functions by others.Footnote 5
I should finally say a preliminary word on timings. UK equality law has recently entered the last chapter of a long overhaul, with introduction to Parliament in April 2009 of the long awaited Equality Bill.Footnote 6 My argument encompasses the changeability inherent in these developments. First, it is rooted in analyses of UK and EU equality law over time, seeking to capture lessons from experience before they are lost. Secondly, there is a great deal of uncertainty about the ultimate outcome of the current legislative process because of the potential for a change of government at the forthcoming election. The analysis therefore leads to brief commentary on the existing proposals, but should additionally inform evaluation of future alterations.
The article is in two parts. The first sets out specific grounds for UK law to make more space for positive action lawfully to be taken. The second sketches the model that change should follow, significantly flowing from the earlier arguments. The attempt throughout is to keep the focus on how law translates into organisational settings, thereby to illuminate what is needed to maximise the chances of useful innovation.
I: The Case for UK Law to Expand the Scope for Lawful Positive Action
A. The Persistence of Systematic Inequality
The first argument for change derives from growing realisation that systematic inequalities have persisted in the UK, indeed in some cases deepened, notwithstanding legal and other interventions to reduce them. The evidence base for this observation is considerable, and important advances are being made in explaining what has been found out.Footnote 7 For present purposes what matters is that it has become clear that existing interventions in pursuit of equality, legal or otherwise, have neither prevented nor effectively withstood this trend.
The Blair government's decision to commission an independent Equalities Review in fact partly grew out of acknowledgement that existing policy action in pursuit of equality, including the law, was not working as had been hoped. The terms of reference specifically adverted to the failure of anti-discrimination law.Footnote 8 The implications of emergent learning about persistent inequality has also now been taken further, through open recognition that social mobility has in significant respects been stagnant for decades and policy deliberation about what should be done about this.Footnote 9
There is, therefore, a general case for innovation, for going “back to the drawing board” in trying to craft effective tools to promote equality. But a specific case follows for organisations to have more freedom than at present to take positive action. The wide-ranging, local and problem-centred endeavour of organisations, private and public, is a vital potential resource for tackling group-based disadvantage to which these entities contribute or over which they exercise control. Failures to date suggest that regulation aimed at combating inequality needs to be better adapted to harnessing this organisational capacity.
B. Flaws and Contradictions in UK Equality Law
1. The Tradition of Excessive and Incoherent Constraints on Positive Action
The arguments in this section apply to the “traditional” strand of UK equality law and hence not to the law on either disability or age. The components identified are traditional in the sense that they date from the enactment of the SDA 1975Footnote 10 and the RRA 1976.Footnote 11 Legislation rooted in this model owes much to the US approach to combating inequality through law, which was immensely influential on the UK law-makers of the 1970s. At the same time, a similar strain runs through the EU law acquis. In spite of being long-standing, moreover, this model has contemporary power at UK and EU level. Crucially, there is a marked tendency for innovations to be added on without what is already there being re-cast. The 2000 EU DirectivesFootnote 12 therefore retained the elements I analyze below as their starting point. Recent UK legislation proscribing discrimination on grounds of religion or belief, and of sexual orientation, in part implementing EU law, then incorporated this template.Footnote 13 And the only aspect that is not intended essentially to survive enactment of a single Equality Act is the cramped approach to positive action. This backdrop not only makes the case for change but conditions the form that it should take.
UK equality law on this model has (and for now this continues), first, prohibited and discouraged measures targeted at correcting systemic inequality that are both perfectly sensible and that do no harm sufficient to justify legal proscription. Secondly, the conceptual intolerance of positive action has given rise to a notion of indirect discrimination that incoherently restricts what organisations may do.
The major technical devices that have combined to produce these effects are fivefold. The first structural feature involves giving anti-discrimination rights equally. There are largely equivalent protections, in particular,Footnote 14 for men and women, for people of different races, colours, ethnic origins, national origins and nationalities, for the holders of a wide range of beliefs, non-beliefs and religionsFootnote 15 and for bisexual, homosexual and “straight” people.Footnote 16
The second structural point is that the wrong of direct discrimination puts a potent equal treatment requirement at the heart of the complex of rules. This requires that people of the various protected identities who are in the same, or broadly similar, relevant circumstances to their counterparts, be treated the same, unless differential treatment is for reasons unrelated to the protected identity in question. This would more usually be put negatively to say that the law requires that people should not, because of one of the protected identities, be treated less favourably than equally or similarly situated others. The central message, therefore, is that it is wrong to take account of these aspects of identity in decision-making; they should instead be ignored.
Combining these two, it is clear that conduct that deliberately assists members of a group protected by anti-discrimination norms of this kind, even if it is to redress systematic inequalities, is at great risk of being unlawful direct discrimination against members of identity groups with symmetrical anti-discrimination rights. Assistance to women or to members of minority groups (whether racial, religious or based on sexual orientation) is likely to breach the anti-discrimination rights, respectively, of men or of other traditionally (or relatively) dominant social groups. Such behaviour can only be rescued from unlawfulness if it either falls into a gap in the coverage of the legislationFootnote 17 or comes within the exceptions that have been carved out of the prohibition on unlawful discrimination. Yet the third important design feature (and the only one currently intended for change) is that the positive action gaps and exceptions generally permit very little.Footnote 18 Beyond this, many of the existing positive action provisions are so bafflingly drafted that it is hard to tell quite what they do allow.
Much current provision for lawful positive action relates to employment,Footnote 19 with arguably the most significant instance being that the SDA 1975 and EPA 1970 save measures related to pregnancy and childbirth from unlawfulness.Footnote 20 But there are other positive action provisions that are perplexing in various ways. First, the comparatively clear provisions on targeted training and encouragement to apply for work do not permit much, with lawfulness anyway dependent on confusing pre-conditions.Footnote 21 Secondly, section 35 of the RRA 1976, mirrored in the law on religion or belief and on sexual orientation,Footnote 22 protects “any act done in affording persons of a particular racial group access to facilities or services to meet the special needs of persons of that group in regard to their education, training or welfare, or any ancillary benefits.” This wording is especially inscrutable, particularly as to whether it may be relied on in employment dealings.Footnote 23 Finally, a narrow reading of all positive action provisions is required by general principles of interpretation because these are exceptions to the equal treatment prohibition central to this strand of anti-discrimination law.Footnote 24
The fourth feature that is decisive to understanding the existing limits on positive action is that the prohibitions on discrimination apply to a very a wide range of actions. This may sound banal but has great practical consequences, especially where there is a complex, ongoing interaction. It is therefore highly relevant for the employment sphere and for any repeat relationship to which these rules apply. It means it is not only the award of a significant benefit, like a job or a promotion, or the imposition of a major detriment, like dismissal, that is regulated, but a very broad range of decisions and encounters. In the work context, for example, these include task allocation, the physical work environment, supervision, training, mentoring and appraisal. In fact, where there is a continuing relationship the opportunities for breaching anti-discrimination rights are legion.
An illustration helps to show the degree of constraint. On the one hand, if an organisation took steps to assist, for example women, with the aim of increasing the rate of applications for senior jobs, while maintaining scrupulously even-handed final selection procedures, it would be vulnerable to claims from men about the steps leading up to the actual selection. The measures taken would only be legally defensible if they could be shoe-horned into a positive action category. Also, the existence of positive action measures (even if lawful in themselves) may support a claim by an unsuccessful male candidate that the ultimate decision was direct sex discrimination. This would be the equivalent of attempts by women over the years to demonstrate how myriad workplace norms and practices not only favour men in themselves, but by affecting decision-making that looks superficially impartial. Ironically, where positive action measures are in place, being deliberate, it may be easier for claimants from traditionally (or relatively) dominant groups to make good this kind of assertion.
This demonstrates that the current limits on positive action go too far in prohibiting and discouraging targeted measures by organisations to address systematic group-based disadvantage or exclusion. It is an obviously good thing for organisations that perceive negative group-based trends to look for non-accidental underlying causes and to remedy any that are found, always provided others are not disadvantaged in, nor excluded from, obtaining the goods, or avoiding the detriments, in question. No harm is thereby done to members of other identity groups who are not affected by the same group-based factors. If there are other protected groups for which analogous considerations obtain, this makes a case, not for targeted action of this kind to be rendered unlawful, but for that situation to be catered for.
The fifth important design element, and the source of incoherent limits on positive action, is the notion of indirect discrimination. The wrong of indirect discrimination signifies that treating equally or similarly situated people the same will not always escape the strictures of this strand of equality law. It evidences some concern to root out unjustifiable practices that do not breach the equal treatment principle on their face, but that systematically disadvantage members of protected groups. It matters, however, that this time there is a justification “defence” built into the law on indirect discrimination. The implication is that unequal treatment on egregious grounds offends the ideal of equality in a way that equal treatment with a disproportionate adverse impact does not. To put this another way, the law on indirect discrimination does not inscribe certain conduct with innate wrongfulness, rather imposing a burden of justification where a particular outcome is produced. Of particular moment for this discussion is that a finding of indirect discrimination does not itself mandate differential treatment to remedy the imbalance. Instead a way must almost always be found to treat similarly situated members of protected groups alike, that either does not have the impugned consequences (nor disproportionately exclude another protected group) or that can be objectively justified.
So the message to be extracted for a would-be law-abider is complicated. It is clear that, while ignoring many aspects of identity in the way people are treated, an eye should be kept out for neutral practices that unjustifiably exclude members of protected groups. To this point the concept of indirect discrimination appears to encourage organisations to engage in reflective self-appraisal to uncover practices that systematically disadvantage or exclude protected groups.
But it is here that the incoherence becomes manifest.Footnote 25 If such an excluding practice is discovered, say that excludes women from recruitment or promotion, the law is extraordinarily constraining as to how this may be rectified, remaining resolutely opposed to steps being taken deliberately to assist the excluded group. To start with, say an organisation purposefully designed new, even-handed procedures rationally to alter the systematic gender effect it had discovered, this would very likely itself amount to direct discrimination against men. The act of designing the new system in this way would itself treat members of one protected identity group (women) more favourably than equally and similarly situated others from symmetrically protected groups (men), and on prohibited grounds (being for sex based reasons). Given that it would not fall within any existing category of lawful positive action, the fact that the changes were defensible on non-identity grounds would cut no ice because there is no justification defence to a claim for direct discrimination. It may be easier to see this point if you imagine the new systems were designed to benefit men (which, ironically, so many systems, indeed the whole world of work, arguably were, but as a matter of history).
Even if this pitfall were avoided, if the new system in fact systematically reduced the success rate of men as was intended, the organisation would be vulnerable to a further indirect discrimination claim. It would then find it could face a major legal battle to justify what it had done, because a prima facie indirectly discriminatory practice can probably not be justified by considerations related to the protected identity in question, whether sex, race, religion or belief, or sexual orientation. The old definition of indirect discrimination contained in the SDA 1975 and the RRA 1976, which continues for now to apply in relation to colour and nationality discrimination and to some instances of sex discrimination,Footnote 26 specified that objective justification of the challenged rule etc must be irrespective of the sex or race of the person to whom it was applied. The recent, EU-inspired wording has generally not mentioned this element. Instead, for the most part the requirement is that the practice be shown to be “a proportionate means of achieving a legitimate aim” (and the proposal now is for general harmonisation on this version).Footnote 27
The open legal question is whether UK anti-discrimination law still prohibits justifications that relate to the prohibited ground in issue. The new EU wording arguably supports a change.Footnote 28 There are also references in the EU positive action cases to selection criteria which, while gender neutral, were designed with women in mind (for example positively to recognise relevant experience outside the workplace).Footnote 29 While a little hard to interpret, these suggest that such measures may be lawful positive action under EU law if they are transparent and amenable to review for arbitrariness. It follows that an employer should have a potential justification defence at EU level to an indirect discrimination challenge to such measures.Footnote 30 The view is widely held, however, that outside the age and disability field, a prima facie indirectly discriminatory practice can only be justified by factors unrelated to the relevant protected identity.Footnote 31 At best there will be uncertainty on this issue until the point is decided by the ECJ.Footnote 32
Our putative organisation would therefore have to live with the risk that it was prohibited from justifying its new systems on the ground that they addressed an existing gender inequality. The weird implication would be that the organisation should have an independent rationale for its new practice that ignored the context that caused the old system to be questioned. Our self-reflective organisation that found an indirectly discriminatory practice, therefore, would find out that existing law not only stringently limited the positive action it could take to reverse this, but may require it to ignore the gender context in devising a new system that treated everyone the same.
The over-arching point is that anti-discrimination rules on this model conceptualise positive action as permissible only exceptionally or peripherally. The underlying premise is that an ideal of equality should be enacted that prioritises treating similarly situated individuals the same. This translates to a powerful emphasis on removing identity considerations from decision-making. It follows ineluctably that differential treatment in pursuit of equality is wrongful, to be justified only rarely as an exception to the equal treatment, identity-neutral ideal.
2. Contradictions about Positive Action: Disability Discrimination, Age Discrimination and Positive Duties
Yet the strand of anti-discrimination law analyzed above is in sharp contrast to the model underpinning UK law on disability discrimination since this was first legislated in 1995. Recent age legislation is also divergent in significant respects. The essential point is that this legal terrain is far more hospitable to positive action, with the consequence that organisations are being given contradictory messages by these different bodies of equality law.
To elaborate, far from constraining positive action, UK disability discrimination law was constructed to be permissive of this activity and, indeed, frequently to require it. The main technical devices that now ensure this are twofold, contrasting with the first two structural features considered above. The protection given is asymmetrical in that only people with disabilities are given the full range of rights.Footnote 33 This means that assistance to a person with a disability generally cannot give rise to a plausible claim of discrimination from someone without. It is true that the existing law on direct discrimination here militates towards people with analogous (but different) disabilities being treated equally well, and that following the ECJ decision in Coleman v. Attridge Footnote 34 this effect will extend to people whose association with a disabled person has analogous effects.Footnote 35 But a person unaffected by disability will not have a claim arising from assistance to someone who is.
Important obligations are also imposed on organisations to accommodate the needs of people with disabilities. As Baroness Hale of Richmond explained in Archibald v. Fife Council, the consequence is that: “It is common ground that the Act entails a measure of positive discrimination, in the sense that employers are required to take steps to help disabled people which they are not required to take for others.”Footnote 36
In the age field, the usual primacy is not given to equal treatment despite protection being symmetrical. Technically this results from there being a justification defence not only to a claim of indirect discrimination, but also to one of direct discrimination. Provided it was a proportionate means of achieving a legitimate aim, there will be no unlawfulness in treating someone less favourably because of age than an equally or similarly situated other.
This on its own is strongly suggestive that the age discrimination legislation is more permissive of positive action than the traditional strand. In addition, there are good reasons for thinking age-based positive action arguments are potential justifications for direct and indirect discrimination. Article 6(1) of the underlying Directive includes examples of differential treatment of this kind that is potentially justifiableFootnote 37 and this stance is reflected in early ECJFootnote 38 and TribunalFootnote 39 decisions.
In addition to the possibility of justifying direct age discrimination and the usual provisions on positive action, the Age Regulations also make further distinctive exceptions to the equal treatment principle. These should not necessarily be regarded as providing for positive action. If anything they add to the difficulty of discerning the “rule” from which the various derogations are exceptions. But they expand the scope for lawfully differentiating between different age groups, from which it follows that the positive action subset of such differentiation is more tolerated than elsewhere.
Disability discrimination law therefore directly challenges the contention that equality is best pursued through identity being ignored. Rather it suggests that where identity arbitrarily worsens a person's life chances, this should, within appropriate limits, be adjusted for. While more difficult to interpret, the law on age discrimination is also disruptive to the UK equality law tradition in suggesting that it is frequently justified to take account of identity in the name of equality.
The upshot is that in crucial respects UK equality law has taken to speaking with a forked tongue. In some places it conveys that the core requirement of the ideal of equality is that group identity be ignored. At others it gives an equally strong message that the same notion requires careful account of the impact of group identity.
These contrasts are bewildering to those with a deep understanding of the various rules. How much more confusing are they for people required to translate equality norms into organisational settings? The potential negative consequences, particularly for institutions engaging in positive action, are encapsulated by the 2006 experience of Avon and Somerset Constabulary. Complaints about a pioneering positive action campaign led to the force being advised the scheme may have contravened the SDA 1975 and the RRA 1976.Footnote 40 Chief Constable Colin Port said on 7 March 2006:
Last summer we tried to do something different. Through use of positive action and using the model of the Disability Discrimination Act we intended to recruit officers that more accurately reflect the diverse communities we serve. It was not and has never been our intention to discriminate against anyone who applies for a position with Avon and Somerset Constabulary.Footnote 41
It appears that the conceptual tension in equality law directly resulted in this organisation misconceiving what was legally permissible in pursuit of equality.Footnote 42
But further contradiction now arises from the expansion in positive duties on UK public authorities to promote equality, widely seen as exemplifying a broader regulatory trend through which equality considerations are inserted into organisational policy-making. The evolution has been encapsulated as being away from individual, retrospective, fault and litigation based approaches that impose negative restraints on potential discriminators, towards “proactive models”. These are said to be different in placing systematic obligations on “policy makers and implementers, service providers or employers” to diagnose and address institutional and structural causes of inequality with no need for fault to be found or individual complaints to be pursued.Footnote 43
Many important issues about these developments are tangential to this argument.Footnote 44 But it is essential to working out what the UK law on positive action should be that this regulatory innovation has been built on top of existing legal rules. This has resulted in measures being poorly articulated with what is already there and in further inconsistent messages being given to the subjects of the various legal regimes. Yet the potential for conflict between the old and the new tends to be muted, or at least to stay below the surface, in contemporary discussions of equality law.Footnote 45 This is despite emphasis on the theoretical discontinuity of the new methodologies with the old. It is striking that this should be so given that theoretical mismatches typically signpost problems in practice.
The positive duties placed on UK public authorities with regard to disability, gender and raceFootnote 46 exemplify this phenomenon. In the case of the race and sex duties, calls were issued to some organisations actively to pursue equality without anything being done to ease the limits on what they were entitled positively to do. Bolting on the positive duties to the old rules meant that organisations have been simultaneously required to avoid individual liability firmly rooted in the idea that identity should be ignored and actively to combat systemic group-based inequality within their spheres of influence. These two sets of commands are not necessarily irreconcilable at a deep philosophical level. But translating them into workable workplace guides for the lawful and effective pursuit of equality presents something of a challenge. Adherence to one or other set of instructions is likely to be compromised, as reported in the Equalities Review:
We were made aware… that many organisations which were keen to be proactive in making more opportunities available to both employees and clients were often frustrated by the rigidities within the equalities laws. None of the people who raised this concern were apologists for “positive discrimination”; their problem was that in order to carry out their principal functions properly they risked contravening the existing equality laws.Footnote 47
The disability duty did not fall into this trap precisely because the underlying individual anti-discrimination rights are so different. That made it perfectly consistent to require public authorities to have due regard in carrying out their functions, amongst other things, both to “the need to eliminate discrimination that is unlawful under this Act” and to “the need to take steps to take account of disabled persons' disabilities even where that involves treating disabled persons more favourably than other persons”. An equivalent combination in the gender and race duties would have been tautological, since meeting the latter duty would, very often, have required an act of unlawful race or sex discrimination.
C. Tension between Law and Organisational Thinking on Positive Action
A further argument for liberalisation derives from tension between UK equality law and organisational trends that militate, at least in theory, towards positive action. Aside from augmenting contradiction in the influences on organisations, this mismatch adds to the risk of sensible, defensible measures to combat systemic inequality being de-railed by individual claims. Moreover it is significant that this argument applies independently from that regarding positive duties to promote equality, since “diversity thinking” has broad influence in the private as well as the public sector.Footnote 48
The dislocation here derives from the popularity of an organisational approach to equality that emphasises the value of diversity. The ideal of equality underlying management theorising about diversity is one in which everyone, irrespective of identity, is able to participate in all aspects of life according to his or her capacities. It follows that more is required than the passive avoidance either of identity-based unequal treatment or of unjustified excluding practices.Footnote 49 Identity is emphatically not to be ignored. Rather organisations are called upon actively to adapt themselves to the needs of different individuals, whether employees, clients or other “stakeholders”.Footnote 50
It is important, however, not to over-emphasise the utopian aspects of this vision. These are undermined by strong emphasis on the pursuit of equality being aligned with business or organisational concerns.Footnote 51 The premise is that the case for any particular kind of diversity is, at least in part, dependent on the organisational benefits it would bring. It follows too that freedom is implicitly given to organisations to determine what version of diversity they should value and promote.
The diversity approach is also individualistic since the contributions of everyone are to be valued. This accounts for the tendency to represent diversity strategies as implicating everyone.Footnote 52 But this element in the thinking gives rise to theoretical and practical tensions. Aside from failing to take seriously demonstrable relationships between group identity and life chances, it does not sufficiently allow for practical (including legal) pressures to focus initiatives on particular groups. It is unsurprising therefore that diversity strategising in practice frequently concentrates on increasing the participation and attainment of particular identity groups.Footnote 53
Ambivalence about equality law also naturally follows from these premises. First, the dynamic pursuit of workplace equality is depicted as most compellingly justified by organisational imperatives. Secondly, it is implicit that law's intervention may be counter-productive given its capacity to distort organisations' self-referential setting of priorities. Equivocal attitudes to law by those espousing this perspective are therefore to be expected. A typical example, including in its circularity, was contained in the finding of the European Commission's 2005 research, The Business Case for Diversity: Good Practices in the Workplace, that: “For many companies legal compliance is a crucial reason for adopting equality and diversity policies and practices. The majority of companies however stress that it is not a driver for implementation, but the desired outcome of their policies.”Footnote 54 There is imperfect empirical information about how law actually motivates and directs employers' pursuit of equality or diversity. Still, we know enough to be sceptical of organisations' protestations that law is irrelevant.Footnote 55 Even so, ambiguity in organisational narratives about law is instructive in itself.Footnote 56
The question, then, is where positive action fits in. The point at this stage is that the idea that difference should actively be accommodated inevitably leads to support for positive action. It is true that the individualism associated with diversity thinking logically tempers this effect, with positive discrimination in major decisions a contradiction too far of the imperative to value everyone as an individual as well as support for positive action being inclusive. Nonetheless, targeted initiatives remain the obvious means by which demonstrable progress may be made in valuing difference.Footnote 57
I shall return to the empirical question of whether organisations are in fact engaging in more positive action (let alone whether they are doing so because of the diversity management trend). But the further scope for confusion created by the tension between legal constraints on positive action, and that diversity strategising encourages such activity, is sufficient in itself to support the case for change.
There are also prominent examples of the diversity package of ideas leading to positive action initiatives,Footnote 58 and to ones that are legally dubious. This evidences that initiatives inspired by diversity theorising, including worthwhile and justifiable ones, risk being undermined by equality law's current excessive strictures on positive action.
For example, the Government has been active in developing a diversity strategy for civil servants. The Cabinet Office launched a typical diversity plan in November 2005. The over-arching goals both emphasised the intrinsic value of inclusivity and characterised achieving this as critical to organisational aims.Footnote 59 The role of individuals was stressed,Footnote 60 while this did not inhibit group-based targets and priorities being identified. Prominence was in fact given to numerical goals for women in the Senior Civil Service, as well as members of ethnic minorities and people with disabilities and for the number of top management posts to be filled by women.Footnote 61
The active, systematic aspects of the diversity perspective were also present. The Foreword by the Cabinet Secretary and Head of the Home Civil Service, Sir Gus Donnell, illustrated this: “This plan commits to 10 key areas and it particularly targets accelerated progress towards a more visible and diverse Senior Civil Service. This is a signal that this plan is ambitious and intended to achieve broad and deep cultural change.” The 10 points covered setting “stretching, challenging and ambitious internal departmental Senior Civil Service targets”. These were also to be reflected in diversity targets for “feeder grades” and “a robust system of measurement and evaluation”. Permanent Secretaries would be held to account in performance appraisals for delivering diversity targets and for ensuring positive action programmes for junior staff (below the “feeder grades”).
Action was to be taken on recruitment, with particular emphasis on positive action to “access the widest and most diverse talent pool” and ensuring that recruitment policies and practices “support[ed] a truly level playing field for all”. The plan specified that: “Where we use search consultants or recruitment advertising agencies, we will incorporate positive action criteria and targets on search and attraction of diverse candidates into awarding, managing and renewing of contracts.” In addition: “Identification and accelerated development of talented people from under-represented groups in the feeder grades to the SCS is a key element of departmental diversity target action plans. We must also ensure that people in under-represented groups below the feeder grades receive targeted and appropriate development to enable them to progress into the feeder grades and beyond.”
As the account of existing law presages, however, these proposals set alarm bells ringing. First, they are at serious risk of going beyond the permitted categories of lawful positive action. Secondly, they might well support a direct discrimination challenge to a final selection decision from a candidate belonging to a group at which positive action was not targeted.
This is a strategy document so it is not generally known how far it was acted on. Publication of the follow-up strategy in July 2008, however, showed the Civil Service continuing in the same vein.Footnote 62 This from the introduction is a classic statement in the diversity “genre”:
This is not about favours and special interests; everyone is part of one group or another. It is about the performance and ability of the Civil Service to deliver. Diversity matters to all of us in the Civil Service if we are to understand our customers, develop evidence-based policies that take their different needs into account and deliver high-quality services to everyone who needs them, in an increasingly personalised way. We will only succeed if we value our people, invest in their development and create an inclusive culture, giving a lead to other employers as we do so.Footnote 63
This document also gave prominence to various examples of departments implementing the 10 Point Plan, with particular show-casing of positive action measures including ones straining at the boundaries of lawfulness.
The DCLG programme, “Pairing for Performance” was described as starting in December 2007 with the aim of developing talented people from ethnic minority communities. Relatively junior staff from the relevant background had been paired with a member of the DCLG Board, as follows:
The programme gives the buddies an insight into the role of leaders and a chance to develop skills in observation, behavioural analysis and feedback. The buddies are supported by their line managers and the human resources team as well as through intensive skills workshops and regular coaching sessions.
Board members have identified opportunities for the buddies, including day visits, attendance at Board meetings and meetings with Ministers.Footnote 64
Another example referred to was an initiative by the Department for International Development (“DFID”) to increase the representation of women in senior and middle management. Named “Crossing Thresholds”, it provided mentoring for “feeder grade” women managers. The 12-month programme covered career planning workshops, peer group support networks, coaching and mentoring. Thirty six women across DFID had been through the programme, of which 30% had moved to the next level and 11% into the Senior Civil Service. The Department planned extension of the scheme to staff from minority ethnic backgrounds.Footnote 65
Whatever the overall relation between diversity rhetoric and reality, therefore, these examples exhibit organisations being steered, not only towards positive action, but to legally questionable measures (on the current state of the law). This adds to the contradictory influences to which organisations are subject, while increasing legal risks for them. It only adds piquancy that the illustrations are provided by the Government responsible for the legal contradiction and for maintaining till now such stringent limits on lawful positive action.
II: Equality, Experimentation and Positive Action
The arguments in Part I build an urgent case for UK equality law to relax the constraints on positive action to the extent that EU law permits this. This need has now repeatedly been acknowledged, by the Equalities Review,Footnote 66 the Discrimination Law ReviewFootnote 67 and by the Government in its proposals for the Equality Bill.Footnote 68 In this section I draw on the arguments supporting change, and on the current state of EU law, to theorise the optimal legal response. This gives rise to an initial critique of the Equality Bill and will facilitate ongoing evaluation of what is done.
A. Facing up to Uncertainty
The first point is that it matters that reform falls to be undertaken in a context, in a sense, of disappointment and failure. It should elicit great humility that decades of legal and other measures to combat group-based inequality have co-existed with such unfairness enduring and even becoming worse. This has no doubt partly been determined by social, economic and political forces that national governments have little power to alter, indeed that they reflect and embody. Nonetheless any new intervention should be rooted in awareness that in critical ways we are ignorant of how society can alter itself to ensure members of different social groups have more equal life chances.Footnote 69
This argument holds despite there being extensive, probably inevitable, contestation about what a condition of perfect equality would consist in (indeed about whether it is possible to articulate any fully coherent conception)Footnote 70 and hence about what steps are justifiable to move closer to this state. The point is that even if there was agreement, there would remain a fount of ignorance about how effectively to use law or anything else to act on this unlikely consensus.
B. Facing up to Contradiction
The second argument is that there is a need for change to engage directly with the impact of the enduring strain in UK equality law that steers organisations away from attending to group identity in decision-making, and especially away from using targeted measures to counteract group-based disadvantage. The Equality Bill certainly does not change the default position, outside the age and disability context, that identity should be ignored. Indeed it could not alter that starting point without bringing UK law into conflict with EU law. Accordingly, much as the Bill proposes greater scope for lawful positive action, it will continue in practice to be challenging for organisations to reconcile law's requirements that identity be ignored with its indications that it is permissible, even desirable, in certain situations to resile from that. And the fact that public and private organisations have been subject to such divergent influences in recent years reinforces the need for them to be given substantial help with untangling the newly configured messages from within and outside the law.
This would also take seriously empirical evidence that a wholesale readiness to engage in positive action should not be assumed, notwithstanding examples of this practice and that it has arguably been encouraged by aspects of the law and by the trend in diversity strategising. The 2004 Workplace Employment Relations Survey evidenced the need for caution. It confirmed an increase in policies on equality issues, in that 73 per cent of workplaces had a formal written equal opportunities policy, or a policy on managing diversity, compared to 64 per cent in 1998. Moreover, this was due to a private sector increase from 55 per cent in 1998 to 68 per cent in 2004. Further analysis also confirmed that there had been behavioural change in the private sector,Footnote 71 which I hypothesise reflected adoption of the diversity template. Intriguingly, however, there was a decline in the public sector in the use of procedures to attract women returning to work after having children (from 20 per cent in 1998 to 12 per cent in 2004), while the proportions in the private sector remained unchanged.Footnote 72 There was also a decline in the public sector of the incidence of procedures to attract applications from ethnic minorities or older workers (older workers – 12 per cent in 1998 to 6 per cent in 2004; ethnic minorities – 27 per cent in 1998 to 21 per cent in 2004). The WERS researchers commented on the latter figures: “[T]his result was surprising given that the RRA 1976 was amended in 2001 placing a statutory duty on public authorities to promote race equality.”Footnote 73
The 2008 Equality Bill White Paper also reported respondents' ambiguity about positive action. There seemed to be extensive disagreement about the incidence of positive action, uncertainty about the concept and significant confusion about the law.Footnote 74 The White Paper concluded that “there is widespread uncertainty about whether, and to what extent, such measures are lawful; and an almost unanimous wish for further and accurate guidance”.Footnote 75 This reinforces the argument that law needs directly to engage with resolving organizational dilemmas about when difference should be ignored and when it should be accommodated.
C. Taking EU Opportunities (and Avoiding Threats)
The third argument is that the design of reform should address the fact that relevant EU law is in flux. This flows from ECJ decisions on positive action to date having only interpreted the old wording in the Equal Treatment Directive of 1976, with some discussion of Article 141(4). The Court has not yet considered the recent EU legislation on equality in which different positive action provisions were enacted.Footnote 76 There will be a degree of uncertainty until it does.
It seems likely that the new measures will be interpreted so that the possibilities already approved by the ECJ in the gender field are covered. So it is to be expected that the Court will permit partial tie-breaks in the allocation of jobs or promotions in favour of members of protected groups that are systematically under-represented to a sufficient degree. EU law is clear, however, that absolute and unconditional priority must not be given to anyone. Rather there must be objective assessment of all candidates that takes their specific personal characteristics into account.Footnote 77 Secondly, a more liberal approach to the allocation of training posts to members of persistently under-represented protected groups may sometimes be allowable, although the case law is less clear on this.Footnote 78 Finally, targeted benefits of various kinds are likely to be accepted so long as they can survive proportionality analysis.Footnote 79
Beyond this, however, there are not definite signs of where the Court is headed. A restrictive interpretation of modern EU positive action provisions remains a possibility.Footnote 80 Equally, there is potential for significant liberalisation, evidenced by the opinion of Advocate General Poiares Maduro in Briheche v. Ministre de l'Interieur.Footnote 81 The actual ECJ limited itself to explicit recognition that Article 141(4) may be more expansive than Article 2(4) of the Equal Treatment Directive 1976. But radical shifts of position were carefully dissected by the Advocate General.Footnote 82 A further significant question is how the ECJ will translate its findings into the non-employment sphere.
This creates both an opportunity and a threat. Past shifts in the Court's position on positive action tell us that the ECJ is able to switch direction,Footnote 83 perhaps drawing on arguments and learning developed through wider policy and scholarly debate. I would extrapolate that the ECJ may be receptive to such argumentation, for example that draws on Members States' experience and analyses of when positive action is worthwhile and defensible. Consideration of non-employment cases may additionally open up the debate.
There are also good conditions currently for the ECJ to take account of the case for an expansive approach to positive action. EU policy discourse on equality has recently evidenced typical features of the management trend related to diversity.Footnote 84 Moreover there are signs of especial interest in positive action within the EU policy community.Footnote 85
This indicates that the process by which the UK arrives at and implements any reform to the law on positive action should be designed to strengthen its position in an eventual challenge before the ECJ. The deeper the analyses, empirically and theoretically, that inspire the UK reforms and their implementation, the more it will be possible legitimately to claim that evolving EU law should accommodate what has been done.
The threat arises if “EU proofing” of this sort is not built into the reform. This would amplify the danger of an ill-conceived organisational initiative being the catalyst for ECJ review. Aside from placing an immense strain on a probably well intentioned organisation, this could destabilise the entire positive action reform project, in the UK and beyond. The UK Government and others would be entitled to intervene to defend their position. Yet an ECJ faced with ill-judged conduct that UK reform had given rise to may well feel compelled to return to default, equal treatment inspired restrictions and limitations.
D. Sketching the Experimentation Model for Law Reform
These considerations support a model for the new law that addresses the needs, first, to build knowledge in this area and, secondly, to give real help to organisations to work out what they should do. It is these twin concerns that underpin the notion that law should enable and inform experimentation by organisations with positive action initiatives. The law needs to perform three functions. First, the space needs to be delineated within which organisations are free, subject to guidance, to try things out. Secondly, organisations need to be steered, procedurally and substantively, in coming up with positive action initiatives. This is partly to avoid legal pitfalls but also, more positively, to enhance the quality of organisational strategising. Thirdly, there is a need to capture what is learned in different organisational settings and for mechanisms to re-introduce discoveries into the experimental loop.
A background constraint derives from the disagreement already adverted to about the ideal of equality. Yet for organisations non-accidentally to subject members of certain social groups to systematic disadvantage and exclusion contravenes any tenable version of the ideal of equality. In spite of ongoing contestation about the ultimate goal or aim of equality law, therefore, it is uncontroversial to promote informed scrutiny by organisations of their activities and to steer them towards non-harmful, purposeful steps to reverse instances of systemic bias.
The first task, therefore, is to locate the outer limits of the experimental zone. The burning question is how to avoid positive action being unfair to those not within the targeted identity group. How to keep at bay the spectre of injustice that haunts positive action initiatives? The risk is of positive action being at the cost of a victim of equivalent group based disadvantage of a different kind and, even worse, to the benefit a person who had not themselves been damaged by systemic unfairness the positive action was designed to reverse.
An obvious example is of a working class white male overcoming systematic disadvantage to ready himself for a particular competition, only to be dis-preferred on identity grounds (say because of a tie-break scheme in favour of women) and conceivably to the benefit of someone who hadn't themselves suffered from the group effect the positive action was designed to redress (say if the measure was to reverse effects of women as a group being more absent from paid labour and the individual female beneficiary had an unbroken service record). Of course it is frustrating that only positive action seems to provoke outcry about such eventualities, when existing systems often have unjust effects to the undeserved advantage of members of traditionally dominant groups. But this kind of injustice is grave whatever the circumstances and law should seek to prevent it.
As we have seen, however, EU law is prescriptive in this regard. The outer boundary of what it permits responds precisely to this concern. Subject to some uncertainty about training posts, the most that has been allowed regarding substantial benefits like jobs or promotions is a limited tie-break. And the cases are adamant that any such scheme must provide for individualised, objective consideration of all candidates. EU law therefore requires that organisational experimentation with positive action be delimited in a way that fully caters for the anxiety expressed here. Award of a substantial benefit cannot lawfully be influenced by identity unless consideration is given to the claims of all candidates to the good in question.
Delineating the safe zone for experimentation therefore requires law to spell out this limit, including taking a view on how it restrains positive action in the allocation of substantial benefits outside the employment sphere. Lack of clarity in the case law suggests that the safe course legally would be to apply the same limit to the allocation of training posts. In any event, the risk of unfairness identified above gives principled support for law to require organisations planning to award substantial benefits in part on the basis of identity, to provide for candidates to make a case that the scheme should not operate to their detriment. As EU law only allows such intrusive positive action in tie-break situations, by definition all candidates would be equally qualified. Hence individual arguments for exemption from the scheme should not, logically, relate to non-identity criteria for award of the benefit. Rather reasons specific to the individual for exemption from the positive action scheme would need to be advanced, for example that they had experienced equivalent hardship to that the scheme was meant to redress.Footnote 86
The next issue is how law should structure decision-making within the experimental space so defined, both to protect organisational initiatives from legal liability and to ensure defensible experimentation in the ultimate aim of creating a virtuous circle of action and learning. Existing EU law is again prescriptive about the reasoning processes that must be satisfied for positive action measures to be justifiable. The emphasis in relation to tie-break schemes has been on showing sufficient under-representation (in addition to respecting the limit above).Footnote 87 Elsewhere the cases have required organisations to establish proportionality, meaning that the aim pursued is consistent with EU provisions on positive action, that the means are “appropriate and necessary” to achieving that aim and that what is done is reconciled, so far as possible, with the principle of equal treatment.Footnote 88 Logically it is right, however, to require proportionality analysis across the board, so that the same justificatory steps are applied to decision-making about the more as well as the less drastic forms of positive action. The implication would be that a tie-break scheme could only ever be proportionate if there were the requisite under-representation. Equally, it is intrinsic to proportionality analysis that stringency of scrutiny varies with the intensity of the positive action measure in question. The less potential effect on members of other identity groups is involved, the less scrutiny is needed.Footnote 89 As a start, therefore, UK law should specifically require organisations proposing positive action to reason through the stages of an EU proportionality analysis in line with the ambition of the particular project.
But the law needs further to respond to the need for learning and to enhance the quality of decision-making in this difficult area. It should therefore direct organisations towards gathering and analyzing data to surface group-based disadvantage over which they have control or influence. The process should mandate diagnosis of underlying causes and, if identity-based explanations emerge, informed evaluation of different options for addressing these, requiring consideration both of effectiveness and of avoiding ancillary unfairness. Equally, the emphasis on informed experimentation militates towards involving those affected. It is also inherent to the notion of experimentation that the effect of what is tried is measured and the usefulness of carrying on reviewed in a timely way.
Beyond this external help should be made available through law facilitating organisational access to knowledge and advice about positive action, especially about what is worthwhile to try.Footnote 90 Ongoing institutional involvement of this kind would be the vehicle for capturing learning and feeding this back into organisational deliberation. These tasks would naturally fall to the EHRC, possibly with provision for direct funding by organisations that opt for positive action. In addition to positioning the EHRC to give tailored support in context, this would enable it to gather and spread learning from experience. It therefore seems to me that the Equality Review's proposal for the EHRC to vet proposed positive action measuresFootnote 91 had potential as one means of embedding an automatic route to support.
Adherence to these requirements would ensure organisations acted justifiably. As such it would be defensible, in principle and with respect to EU law, for UK equality law to protect organisations from liability for unlawful discrimination in respect of positive action schemes that observed such strictures. This would go a considerable way to reconciling the contradictory messages to which equality law gives rise, while protecting and stimulating sound organizational innovation. It is true that the EU legal uncertainties, both about the extent of permissible positive action and about its relationship to indirect discrimination, would remain. But reform of this type would supply powerful arguments, in theory and in practice, for these to be resolved to permit what the UK had done.
And aside from ensuring that organisations could justify what they had done in courts and elsewhere, this format would create considerable space for valuable initiatives. It should promote action to alter systemic bias “at source”. Any individual adverse effects could not consequently be regarded as unjust. If an organisation changed a practice because it was a root cause of group bias this would augment fairness for everyone. While there would be individual “losers”, because fairer processes in such circumstances should by definition produce less biased decisions, this would enhance, not reduce, individual fairness. The net effect would be to neutralise unfair advantage that had previously been accorded to those who would have received the benefit (or avoided the detriment) under the old, systematically biased, approach.
In any event, a great deal could be tried that would not even come close to awarding substantial benefits on the basis of identity. Obvious examples are finding ways to accommodate particular needs related to identity or to search out valuable skills and experience gained in ways that differ from the “norm”. And where organisations concluded that more draconian measures were appropriate, the decision-making process would of its nature force them to articulate a compelling case.
The reasoning and information processes built into this model should also assist in finding the appropriate site or level for an initiative. It might even encourage collaborative efforts across sectors, or within particular industries, enabling more to be attempted and discovered. An experimental approach would additionally encourage organisations to identify pockets of disadvantage that crossed identity boundaries. Experimentation might then relate to social sub groups whose members were at risk of distinctive disadvantage because of inter-secting identities.
The ideas explored here are evocative of the approach taken in policy analyses about tackling inequality other than through anti-discrimination rights. The recent consultation paper, New opportunities, Fair chances for the future, is shot through with evidence based assessments of targeted interventions to equalize life chances. Looking back it was observed that: “Over the past 10 years, the Government has worked hard to identify the factors that have the most impact on social mobility and has taken steps to remove the barriers preventing it for younger generations.”Footnote 92 Equally, positive equality duties were situated within this narrative, characterised as measures to enable public bodies to acquit their “crucial role in helping people to fulfil their potential and in removing the barriers that hold people back”.Footnote 93 It was extrapolated that something analogous was needed to address the fact that: “Co-existing and interwoven with these specific inequalities lies the persistent inequality of social class.”Footnote 94 In developing policy, again, evidence is to be fed in from the National Equality Panel.Footnote 95
The Civil Service diversity strategy discussed above also reflects aspects of the proposed positive action model, especially in the emphasis on learning across the Civil Service and externally.Footnote 96 Arguably, therefore, this essay draws out, in respect specifically of positive action, how UK equality law might integrate its prescriptive, hard-edged individual rights, with facilitating constructive organisational policy-making in the image of initiatives by central Government in its various guises.Footnote 97
The proposed model also resonates with work by Sturm documenting and assessing the emergence in some US equality law, and in some US organisations, of a “problem-solving” approach to employment discrimination.Footnote 98 She located this as responding to the incapacity of a “rule enforcement” approach to address second generation discrimination, in the sense of “subtle and complex forms of bias” that frequently involve “patterns of interaction among groups within the workplace that, over time, exclude non-dominant groups”.Footnote 99 The argument was made that this reality cannot be reduced to a stable set of precise conduct rules. Rather a problem-solving process is needed, as follows:
That process identifies the legal and organisational dimensions of the problem, encourages organisations to gather and share relevant information, builds individual and institutional capacity to respond, and helps design and evaluate solutions that involve employees who participate in the day-to-day patterns that produce bias and exclusion. An effective system of external accountability, including judicial involvement as a catalyst, would encourage organisations to identify and correct these problems without creating increased exposure to liability, and to learn from organisations that have engaged in similar efforts.Footnote 100
Although my argument derives from different imperatives and rationales, there are consonances in the emphasis on structuring organisational decision-making and on enhancing knowledge about combating group based inequality,. At the same time, Sturm's work is of both wider and narrower scope. It looks at the entire range of potential activities by organisations to combat inequality, but concentrates on the conduct of employment relations. My suggestions concern only positive action, but encompass this in the provision or goods and services or, indeed, in the performance of any non-employment related organisational function.
More fundamentally, there is dissonance in that Sturm's analysis of what public legal actors should do has markedly concentrated on courts and judges, side-lining the potential impact of legislators.Footnote 101 This might in part account for two controversial aspects of her theorising. The first is that law should be seen as malleable and fluid to the extent that organisations' optimal role includes elaborating, or assisting (courts especially) in the enunciation of, legal norms.Footnote 102 The second and connected point is that intermediate actors, from employees and lawyers to ngos and government agencies, have a critical role to play in ensuring the efficacy, legitimacy and accountability of organisational activity regarding equality.Footnote 103 More recent work has responded to scepticism about courts' readiness to play the distinctive supervisory role assigned to them, by concentrating on the possible contribution of institutional intermediaries, in the nature of non-court public bodies, or conceivably publicly minded private entities, in making the proposed system work.Footnote 104
I would tentatively hypothesise that the tendency to side-step the possibility of elected law-makers enacting equality legislation modelled on this vision might be explained by the constitutional, legal and political culture in the US, in which thoroughgoing legislative overhaul of equality law appears practically unimaginable. It may follow that anyone arguing for significant change must accomplish the considerable feat of doing so in ways that leave the underlying framework of law undisturbed, while also giving a plausible account of how existing judicial pronouncements either are consistent with or even require the re-ordered system.Footnote 105 Incidentally, this suggests that there may be lurking challenges in the US, distinct to those in the EU and UK, in dovetailing modern, more systemic approaches to the organisational pursuit of equality with existing legal interventions.Footnote 106 Be that as it may, it is evident that an important difference in the UK and EU is that legislative innovation has for some time been a feature of the regulatory landscape.
Even so, I am not denying the immense complexity of defining in practice, or theorising in principle, the relationship between legal and non-legal norms. The arguments I have made for the design of law reform in fact proceed from the inherent difficulty of determining when organisations should prioritise equal treatment and when targeted action in pursuit of equality is worthwhile and defensible. But there is no reason for legislation not to grapple directly with the problem and strong positive arguments for it to do so, not only to steer organisations, but to guide judicial responses. The case for law explicitly to frame organisational freedom, and to direct what is done procedurally and substantively within that sphere, rests partly on its potential influence on judges domestically and in the EU. In this way an aim is for organisational experience over time to influence the development of the law on positive action. The proposed model thereby confronts the realities of multi-layered judicial control, while channelling organisations towards higher quality deliberation about, and implementation of, positive action measures. Central is that law's processes should evoke humility in everyone concerned, whether organisational actors, EHRC advisers or the judiciary, in confronting the dilemmas thrown up by positive action, embedding the pursuit of learning by defensible experimentation into attempts to reverse systemic group based inequality, exclusion and disadvantage.
A feature of the suggested regulatory design that should be assessed over time is its voluntariness since experience begs questions about whether organisations will use new freedoms that are given to them. Still, a permissive system seems inevitable for now because there is insufficient appetite for co-ercion.Footnote 107 It is also arguable that the notion of experimentation fits uneasily with legal proscription. Finally, there may be a side benefit in an initial voluntary approach, from learning by comparing what is done by organisations subject to positive duties and those that aren't. Any contrasts (or lack of them) might be instructive as to the impact of different legal interventions.
A further troubling issue concerns the implications for individual litigation of organisations uncovering evidence of non-accidental group bias. This may well establish that unlawful discrimination has occurred, ensuring liability if individual claims are pursued in respect of this. Indeed the discovery of such evidence may provoke claims. Arguably this is as it should be and organisations should factor these effects into their decision to self-appraise. One worry, though, is that this possibility will prevent organisations self-scrutinising, or at least stop them doing so in a meaningful way. Another is that those who take this path will suffer heavier legal penalties than those who don't, meaning that seriousness about tackling inequality will worsen an organisation's position (and liability record) relative to those that are complacent.
This presents a knotty dilemma along a fault-line which underlies much of UK and EU equality law. First, any attempt to craft remedial mechanisms that go further than the model proposed here in differently balancing interests in, on the one hand, organisations tackling systematic inequalities with, on the other hand, individual victims receiving redress, raises even more intractable questions of principle and of EU (and possibly ECHR) law on the sufficiency of national remedies for unlawful discrimination. Secondly, the same conundrum is raised wherever equality law pushes organisations towards serious evidence gathering, as has at times happened in response to equal pay laws and may now occur in the performance of positive duties. But it is certainly sobering that a rational protective response to UK equality law remains, subject to a few exceptions, passively to ignore relevant identities while leaving to others the (quasi impossible) task of identifying and challenging systemic causes of inequalities. It is true that the positive duties make this stance harder for some organisations, but even they do not make it impossible given the scope for the duties to be bureaucratised to minimal effect.
Turning to the implications of this analysis for the general positive action provisions in the Equality Bill,Footnote 108 their oddest feature is perhaps that they do not track the limits of EU law.Footnote 109 This is peculiar given the intention having been for UK law to allow positive action to the extent permitted by EU law.Footnote 110 At the same time, the proposed design, by which these provisions are to be backed up with “comprehensive and authoritative guidance from the Equality and Human Rights Commission,”Footnote 111 is inadequate. Not only does it attribute quasi-magical powers to the EHRC in one fell swoop to resolve the many and varied challenges inherent in taking positive action, it does not envisage nearly enough tailored, ongoing help to organisations. While the aspiration to liberalise the law on positive action is an important start, therefore, the execution has not to date apprehended the magnitude of the challenge this entails. In this way the proposals have so far missed the unique opportunity provided by enactment of a single equality act effectively to deploy law to assist organisations, public and private, in optimising their use of positive action to shift deep-seated patterns of inequality, disadvantage and exclusion.
