Equality: towards an outcome-focused duty
Sandra Fredman* and Sarah Spencer† argue that a new duty on public bodies to promote equality should be action-based and goal-oriented but should allow those bodies greater autonomy in how they deliver equality.
The perceived limitations of statutory anti-discrimination law, with its basis in an individual complaints-led model, have led to the construction of an alternative approach to equality, based on a proactive model. Pioneered in Northern Ireland1 , the proactive model has gathered increasing momentum in Britain, where the race equality duty (Race Relations (Amendment) Act: An EOR Guide) has been followed by a similar duty in respect of disability, which comes into force in December 2006 (Disability Discrimination Act 2005: an EOR guide), and gender, which comes into force in April 2007 (Equality Act 2006: a guide).
The new developments hold much promise in overcoming the limitations of the individual complaints approach and some positive outcomes are already visible. Racial inequality is, to a limited extent, more likely to be taken into account in developing policy and services, setting targets and performance measures, and by the audit and inspection bodies that monitor delivery. However, the new approach has not led most public bodies to review their practices and implement reforms in the way parliament intended. Implementation of the race equality duty has demonstrated the ease with which the positive duty can become an exercise in procedure and paperwork, rather than in institutional change (Race equality: an ongoing obligation). According to the consultation paper on the proposed gender duty, the general view of the race duty is that it is "overly bureaucratic, process-driven and resource intensive"2. The disability duty and the proposed gender duty make some progress towards addressing these difficulties, but do not go far enough.
In addition, as work progresses towards a single equality Bill in the UK, the further challenge arises of formulating a duty that can apply to all the strands. The need for parity of protection from discrimination on grounds of age, sexual orientation and religion or belief suggests the need for a generic positive duty in Britain, and the Discrimination Law Review has indicated that this is under consideration for a single equality Act. There will be a concern to ensure that the framing of the new duty proves both more effective and less process-driven than the separate duties it replaces.
In this article, we clarify the limitations of the existing duties and suggest a possible alternative approach, which requires action while leaving public bodies with a greater level of autonomy to decide how those outcomes are delivered.
Moving on from the current framework
The detailed requirements of the current positive duties differ, but the core duty is that a public body must pay due regard to the need to eliminate unlawful discrimination and promote equality of opportunity. The goal, therefore, is equality of opportunity and the duty is to pay due regard. We argue that both the goal and the duty are problematic. "Equal opportunity" is too vague and too limited to function as a workable target. The duty to pay "due regard" merely requires a body to consider the need to promote equality, not to take any action.
We propose that the new duty should specify the equality goals, moving beyond equality of opportunity, and specify a clearer duty, moving beyond due regard. We propose a goal-oriented, action-based and progressive duty.
The goal: "equality"
Equality of opportunity
Equality of opportunity is a broader concept than the formal version of equality, which requires only that similarly situated people be treated equally. Recognising that the same treatment might perpetuate disadvantage by failing to address existing discrimination and disadvantage, equality of opportunity aims to equalise the starting point. However, equality of opportunity can have a range of applications. At its narrowest, it requires the removal of barriers at the demand side. This may open the doors to excluded groups, but does not mean that they have the resources to progress through the doors. For example, requirements for educational qualifications that are not strictly necessary for the job may be relaxed, but disadvantaged groups might still lack the training that is necessary for a wide range of jobs. A broader understanding of equal opportunities would require that resources be provided to make sure that members of disadvantaged groups can make use of new opportunities. In the above example, it would require that training be provided; in other contexts, childcare or transport might be needed. The legislation does not make it clear which of these approaches to equality of opportunity should be taken, or indeed how equality of opportunity is to be defined. Instead, there is an emphasis on monitoring, which assesses outcomes rather than opportunities. However, while "equality of outcomes" can be a meaningful goal in some contexts for groups (eg average attainment of ethnic minority pupils at GCSE level) it can mask inequality within groups. It can also focus on quantitative measures to the exclusion of qualitative approaches and is not applicable as a measurement for individuals, who have differing priorities.
Capabilities
The Equalities Review interim report3 proposes that the guiding principle be equality of capability. Capability - a concept that has been developed by Nobel laureate Amartya Sen - is widely used to assess development, and refers to an individual's capacity to achieve the "functionings" and goals they value. But since social, economic and physical constraints do not operate evenly through the community, some individuals might require more or different resources to others in order to achieve the goals and "functionings" that they value4 . The focus on capabilities aims to ensure that policymakers frame policies in such a way that individuals are able to make real choices and act on them.
A four-dimensional concept
We agree that the concept of capabilities is a useful one, and draws on international experience to underpin policy. However, to those who are not immediately acquainted with the way in which it has been formulated, the term may be misleading. In particular, on its own it may suggest an emphasis on individual capacity rather than on removing external barriers to equality, and it may seem to ignore the qualitative dimension of equality - equality of dignity and respect - which is particularly important in relation to services. Thus as a legislative concept, it might be vague and difficult to apply.
We suggest that the capabilities notion be given more explicit content, to ensure that it can provide meaningful guidance. We suggest that four central aims of equality be spelled out, drawing out the implications of the capabilities notion. Each can be more or less significant or appropriate in differing circumstances5 . They are:
Equal life chances: Equality duties should aim to break the cycle of disadvantage associated with discrimination, aiming at fair representation, such as in employment or in parliament, and pursuit of equality of outcomes for groups, such as parity in exam results.
Equal dignity and worth: Equality duties should address stigma, harassment, humiliation, degrading treatment and violence.
Accommodation and affirmation: Equality duties should go beyond identical treatment in meeting needs, and should accommodate and affirm different identities, aspirations and needs.
Equal participation in society: an equality goal in its own right, as well as a pre-requisite of good relations.
The diagram below (see table below) illustrates the complementary nature of these four dimensions of equality and examples of the differing issues that might be addressed to meet those objectives.
We suggest that, as in Canadian equality legislation, a single equality Act should begin by specifying the four aims, to guide those implementing the positive duty. The proposed terms of the duty are discussed below.
Functions of a positive equality duty
The complaints-led model has six limitations, each of which the positive duty needs to address:
1. Reliance on individual litigation means that much unlawful discrimination is not addressed if there is no complaint. It puts the onus on the individual victim, not the organisation, and it is not necessarily the worst offenders that are penalised. The positive duty shifts the responsibility to the organisation to identify and address unlawful discrimination, regardless of receipt of a complaint.
2. Statutory concepts of discrimination and equal pay are individualised and retrospective. They require proof that an individual perpetrator has caused the inequality and that an individual complainant has suffered detriment, providing individual remedies in the form of compensation. This makes it difficult to address institutional or group-based discrimination. Under the positive duty, evidence of group-based inequality (eg under-representation) triggers a response. There is no need to prove that the body has caused the inequality; instead, the responsibility falls on the body because it is in a position to remedy it.
3. Complaints are retrospective, so they do not automatically impact on future policy. The proactive duty should prompt organisations to pre-empt any detrimental impact of future policy (through impact assessments) and to introduce policies that promote equality.
4. Complaints are adversarial and conflictual and are insufficient to motivate an organisation to review its policies or practices. This is a poor change-management strategy. The law needs to be a catalyst for a broader range of incentives and sanctions that reflect what motivates decision-makers and drives organisational priorities (such as exposure through transparency, audit and inspection, and linking performance to eligibility for public contracts).
5. The complaints-led model addresses only the inequality that arises from discrimination, rather than broader barriers to equality, external or internal to the organisation. A positive duty to promote equality should identify and address broader causes: for example, the lack of transport or childcare preventing take-up of employment.
6. The complaints-led approach does not address the qualitative dimension of equality, respect for individual dignity and the worth of each individual.
We suggest that the framing of a positive duty to promote equality must be designed to fulfil a broader range of objectives than the complaints-led model. These are to: address discrimination even where no complaint has been made; address institutional patterns of discrimination where no single individual is responsible; ensure policymakers anticipate a potential detrimental impact of new policies and develop measures designed to promote equality; and activate a range of levers that motivate decision-makers to review, diagnose and implement reforms, including those that foster an inclusive culture of mutual respect.
Duty on whom?
Before we consider the terms of the positive duty it is necessary to consider to which public bodies it should apply. One of the weaknesses of the current race duty is that there are different specific duties for different bodies (those for parish councils are different from those for schools and different again from those for larger public bodies). The intention was that the obligation should be proportionate, but the result is some confusion about who is required to do what. In the new model of duty that we propose, the obligation on each organisation to act only to a proportionate extent will be clear, so that each authority can be subject to the same duty. A statutory code of practice could provide guidance on the steps that could be appropriate for different types of authorities.
We will not examine here the key question of whether, or to what extent, the positive duty should extend to the private sector. Rather, we will focus on defining a duty for the public sector that will deliver. We recognise the political constraints on any new "regulatory burden" on the private sector, and that private enterprise should not be assumed to bear the same level of responsibility as the public sector for delivering equality.
However, we suggest that when the role of the private sector is considered these constraints should not be the starting point for that discussion. Rather, the starting point is the overwhelming importance of the private sector as the employer of more than 75% of the labour force and as an increasingly significant service provider, not least for some of the most vulnerable in society. Its potential strategic importance in addressing inequality is clear. The challenge, for a later date, is to identify the most appropriate means to realise that potential.
What kind of duty?
The existing duties require public bodies to have "due regard" to the need to eliminate unlawful discrimination and to promote equality. Due regard leaves the authority, having taken that need into account along with its other priorities, to decide whether or not to act. We suggest this formulation is too weak, for two reasons:
It does not make sense to require the authority to do no more than pay due regard to the need to eliminate unlawful discrimination because, by definition, unlawful discrimination is unlawful. This is true too for equal pay for work of equal value, which is a statutory requirement. The authority should be required to act, not only to have due regard to the need to do so.
In relation to the promotion of equality, the authority has discharged its duty if it pays due regard, but decides to do nothing. The law specifies no end result. The courts would be unlikely to find a breach of the duty unless the authority had acted wholly unreasonably in the way it responded to considerations, or possibly if its response was disproportionate.
Proposal for action-based, goal-oriented approach
We need a formulation that requires action that is proportionate to the inequality identified, and to other competing considerations that the authority should take into account. There is experience abroad from which we can learn. We draw on the terms of the International Covenant for Economic, Social and Cultural Rights (ICESCR) and the South African Constitution in suggesting that the terms of the duty could be that: "A public authority shall, in carrying out its functions, take such steps as are necessary and proportionate to eliminate discrimination and to achieve the progressive realisation of equality (as defined)."
This comprises the following elements:
It is action-based. It requires the public body to take steps to eliminate discrimination and achieve equality, rather than just paying due regard to the need to do so.
It is goal-oriented. Instead of a vague notion of equality of opportunity, it specifies the goals to which the body is directing its actions: to eliminate unlawful discrimination and to achieve equality (defined as equal life chances, equal respect, accommodation and affirmation, and equal participation).
It is progressive. The public body does not need to achieve the goals immediately, but it must take immediate action to make progress towards the goals. It is dynamic, requiring ongoing action.
It requires action that is necessary and proportionate. This means that the authority can set its own priorities, within available resources and in the context of competing aims. Its decision on priorities must be measured against the standard of proportionality and necessity - a clear and strict standard. The body need not do everything at once, but it must base its priorities on evidence and consultation. Where steps to achieve equality are not taken, or are taken too slowly, the body must be able to show that this was because it was pursuing other legitimate aims, and that those other aims could not be achieved in an alternative way that was compatible with furthering the equality agenda. This is a well-known standard that is found in anti-discrimination law, human rights law, and the law of judicial review, and provides an appropriate balance between autonomy and obligation.
Framework for action: the specific duties
The current duties are supported by specific duties. The question is whether or not a generic statutory duty as defined above should also be backed by specific duties and, if so, what they should be. We are aware of the concern that the current list of specific duties relating to race, gender and disability could grow when the three additional equality strands are also covered. In that context, the much stronger formulation of the general duty we have proposed would reduce (but not eliminate) the need to specify the steps that an authority should take to deliver on the duty. This would leave greater autonomy for the authority to decide what action was necessary and proportionate in relation to each equality strand and would reduce the danger that it adopts a tick-box approach in fulfilling those requirements.
To fulfil the general duty, each authority would, in practice, need to take steps: to get baseline evidence on discrimination and equality across its functions; to diagnose the causes of inequality identified; to consult; to have an action plan setting out the necessary and proportionate steps it proposed to take; and to monitor progress.
These basic steps should be set out as specific duties to ensure a level of transparency that is essential to empower local organisations and individuals to engage with the authority on its record, and for the Commission for Equality and Human Rights (CEHR) and inspectorates to monitor delivery.
The wording of the current race equality duty suggests that this "review-diagnosis-consult-action-monitor" approach was parliament's intention. But the specific duties under the race legislation do not in practice require it. An authority is required to have a race equality scheme that must set out its arrangements for consultation, impact assessment, monitoring and training, rather than a requirement to fulfil those arrangements.
The Commission for Racial Equality's (CRE) statutory code of practice and its guidance, for instance on race equality impact assessments, indicate that action is required but in the absence of case law it is not clear how narrowly or broadly the courts will interpret the duties. Larger authorities must monitor their workforce, but there is no requirement to act if that monitoring reveals significant under-representation. To the extent that the law does require any diagnosis, action and monitoring, it fails to differentiate these separate functions.
Five specific duties are required. None is an end in itself but, in the wording of the current race duty, they are "for the purpose of ensuring the better performance" of the general duty:
1. A baseline assessment of the current situation and diagnosis of causes of any discrimination or inequality identified in any of the equality strands. Note that the lack of quantitative evidence of a problem cannot be an excuse for inaction. Qualitative evidence, including from consultation, may be the basis for action in the absence of an authoritative evidence base for decision-making, or until one is available.
2. Consult and involve affected parties. Participation is important not only as a means to better decision-making but also as an equality goal in its own right.
3. Set out, in a current action plan, the proposed "necessary and proportionate steps", and proposed timescale for implementation. Crucially, this needs to be reflected in the organisation's operational or business plan.
4. Assess the potential impact of new policies or services on equality, where initial screening suggests impact could be significant.
5. Monitor and make public the baseline assessment and annual progress in the authority's "progressive realisation of equality" across the equality strands.
The specific duties (or a subsequent statutory code of practice) could specify a common format for baseline monitoring, for ease of comparison. This transparency requirement would make it possible for the CEHR to fulfil its responsibility (ss.11 and 12 of the Equality Act 2006) to monitor the effectiveness of the equality enactments and outcomes respectively, and to use its s.32 power to assess the extent to which a particular authority has complied with its statutory duty.
We considered whether a "transparency" requirement could be sufficient, in practice requiring the action and monitoring on which it would report. It would not. Data can mislead; and enforcement action, in the event of persistent inequality, needs to focus on a specific requirement to act.
We also considered whether it would be sufficient to indicate in a statutory code of practice that some or all of these steps were to be considered among the "necessary and proportionate steps", rather than setting them out as statutory requirements. The disadvantage of that is the reduced likelihood that the authority would take these necessary steps, and the lack of consistency of approach between authorities. If these minimal steps are necessary, that requirement should be clearly set out in law.
Balance between structure and autonomy
Our framework aims to achieve an appropriate balance between structure and autonomy. The legislation would set out, in the form of the general duty, the aims that the body is required progressively to achieve (that is, equal representation; equal dignity and worth; to accommodate and affirm diverse identities and achieve equal participation) and the requirement to take necessary and proportionate steps to do so.
The specific duties would provide a structure for delivery of the duty through the stages outlined above, but the body would retain the autonomy to decide what the necessary and proportionate steps might be, from the full range of measures an authority might deploy to achieve equality.
We are thus not suggesting that the steps necessary to deliver equality across the six strands of discrimination covered by legislation should be set out in regulations: for example, requirements in relation to training, promoting positive attitudes or procurement. That approach would have the disadvantage of proscribing the "how" rather than forcing the authority to take ownership of the agenda. It would also require different specific duties for each strand, which would be complex, bureaucratic and could effectively paralyse the organisation. We propose that the body be able to respond flexibly and appropriately, revising its strategy where necessary, within the framework outlined above. A statutory code of practice could provide guidance on action, recognising that the steps will change over time. The code of practice should also provide guidance on the differing functions of consultation, on who should be encouraged to participate and how effective engagement might be achieved (see box below).
Avoiding responsibility gaps
Clearly, a public body cannot be responsible for aspects of unlawful discrimination or inequality over which it has no control. However, it is important that action on barriers to inequality should not fall into the gaps between different bodies. There are two ways to address a situation in which a body is dealing with an inequality whose source is beyond its control. The first is to draw this to the attention of the body under whose control it does fall. For example, an education authority might need to draw the attention of housing authorities or social services to the need to address sources of inequality relating to these services that are affecting children at school. Similarly, a public employer might need to draw the attention of transport authorities to the need to address transport difficulties that affect the ability of potential employees to get to work.
Second, a secretary of state's overall responsibility for a sector includes a coordinating role. Bodies that are dealing with inequalities beyond their control should draw this to the attention of the secretary of state, who could identify the means to address it. This would seem to fall within the relevant state department's general duty, but a specific duty could be considered to make this explicit. In relation to legislative proposals from the department itself, a further option would be to include a requirement, similar to that under s.19 of the Human Rights Act 1998, that the secretary of state reports to parliament on the equality impact of any new legislation, clarifying beyond doubt that the potential impact of new legislation on equality should be taken into account in the policymaking process.
Monitoring performance: the audit and inspection framework
The mainstream audit and inspection bodies have recently been encouraged by the CRE to build race equality into their regular and thematic inspections. The CRE has memoranda of understanding with many of those bodies, including the Audit Commission and Healthcare Commission, clarifying their respective roles and it produced a guide for inspectorates showing how inspection can take account of the general and specific race duties. Some inspectorates now see monitoring for compliance and outcomes on race equality as part of their core business.
However, at a time when this equality focus needs to extend across the equality strands, the main inspectorates are being reduced in number from 11 to four and the nature of the inspection process is changing. The importance of the inspectorates in monitoring performance on equality should be consolidated by inclusion of this specific responsibility in the parent statute of each inspectorate or accompanying regulations. They would act with the support of the CEHR, as the expert body (but one without the resources to carry out regular inspections). This reform would mean that the CEHR would not have to persuade a reluctant inspectorate to fulfil this role, nor duplicate that role using its own investigation powers.
Compliance
The CEHR has been given additional enforcement tools for the positive duties, recognising that judicial review and formal investigations alone are expensive, while compliance notices are an inflexible means to enforce the specific duties. The greatest obstacle, however, has been the difficulty of enforcing a "due regard" general duty and specific duties that focus on making arrangements rather than requiring action. Our proposals to redress those weaknesses would considerably enhance the capacity of the CEHR and the courts to act if an authority fails to take the "necessary and proportionate steps" that its baseline assessment had indicated were required.
Using the powers in the Equality Act 2006, the CEHR could assess compliance (s.32), requiring information from the authority for that purpose, perhaps prompted by concerns identified by an inspectorate or by data made public under the authority's specific duties. It could also enter into a binding agreement (s.23) under which the authority agreed to take specific action (necessary and proportionate steps), thereby forestalling enforcement action by the CEHR.
Conclusion
The positive duties on public bodies were intended to ensure that proactive steps are taken to eliminate discrimination and promote equality. The framing of the duties has not proved optimal and the new gender and disability duties make only limited progress in addressing the weakness in the approach. It is necessary to frame a duty that extends across the equality strands and that unequivocally requires public bodies to act, while reducing the focus on procedure.
To achieve this, we propose a general duty that an authority be required to take such steps as are necessary and proportionate for the progressive realisation of equality, where equality is defined broadly as capability and more specifically as: equal life chances; equal dignity and worth; the accommodation and affirmation of diverse identities; and equal participation across the equality strands. This should be accompanied by a specific duty structured in five distinct phases: a requirement for public bodies to publish baseline evidence on discrimination and equality across its functions; to diagnose the causes of inequality identified; to consult; to have an action plan setting out the necessary and proportionate steps, reflected in the organisation's business plan; and to monitor and make progress public. This would obviate the need for extensive and differing procedural duties applied to each strand and would empower local stakeholders, audit and inspection bodies and the CEHR to hold the public body to account should it fail to deliver on this agenda.
Our proposal, as with the current positive duties, is intended to provide a strategic framework for addressing inequality, not new substantive rights or services to individuals. That may be a necessary part of an action plan implemented to achieve the progressive realisation of the authority's equality goals, but is not part of the framework itself.
1 McCrudden, C (2004) "A review of issues concerning the operation of the equality duty" in McLaughlin, E and Faris, N (eds) The section 75 equality duty - an operational review Volume 2, Northern Ireland Office Belfast; McCrudden, C (2001) "Mainstreaming equality in the governance of Northern Ireland" in Harvey, C (ed.) Human rights, equality and democratic renewal in Northern Ireland, Hart: Oxford.
2 Advancing equality for men and women: government proposals to introduce a public sector duty to promote gender equality, Women and Equality Unit, available at: www.womenandequalityunit.gov.uk/legislation.
3 The equalities review: interim report for consultation, Cabinet Office, March 2006.
4 Sen, A (1999) Development as freedom, Oxford University Press, p.75.
5 A dimension might indeed be inappropriate in some circumstances: eg equal representation in the workplace is not sought for each age group or members of a particular religion, although the latter has been a goal in Northern Ireland.
* Sandra Fredman is professor of law at the University of Oxford and Fellow of Exeter College, Oxford, sandra.fredman@law.ox.ac.uk.
†Sarah Spencer is associate director, Centre on Migration, Policy and Society, University of Oxford, and a former deputy chair of the Commission for Racial Equality, sarah.spencer@compas.ox.ac.uk.
This article is based on a paper that was presented at an Equality and Diversity Forum seminar on 11 April 2006 and formed the basis of a submission to the Cabinet Office Equalities Review (see Equality Bill establishes CEHR from October 2007) and the Discrimination Law Review.
Equal life chances |
Equal dignity and worth |
Representation of women in boardroom Parity GCSE results Equal pay Housing and school segregation |
Harassment Homophobic bullying Degrading treatment of the elderly Domestic violence Hate crimes |
Affirming and accommodating difference |
Equal participation |
Site provision for Gypsies and Travellers Reasonable adjustment for disability Freedom of religious dress Civil partnership |
Transport for older people Meaningful consultation Childcare; flexible retirement School exclusions Children's right to participate in health and education decisions |
Functions of consultation Gaining information on discrimination and inequality from those affected and insight into causes. Giving information to those affected: transparency and accountability. Influencing decision-making: consultation or co-decision? Monitoring compliance. Who should participate Stakeholders: eg trade unions, non-governmental organisations. Individuals affected. Those delivering the duty: employers, service providers, schools etc. Other public bodies: coordinating action. How to achieve participation Capacity building: ensure all have voice, prevent consultation fatigue. Accessibility: accommodate needs of different
groups, such as Gypsies and Travellers, faith and linguistic minorities. |