International Perspectives on Positive Action Measures a COMPARATIVE ANALYSIS in the EUROPEAN UNION, CANADA, the UNITED STATES and SOUTH AFRICA

International perspectives on positive action measures A COMPARATIVE ANALYSIS IN THE EUROPEAN UNION, CANADA, THE UNITED STATES AND SOUTH AFRICA

European Commission This publication is supported under the European Community Programme for Employment and Social Solidarity - PROGRESS (2007-2013).

This programme is managed by the Directorate-General for Employment, Social Affairs and Equal Opportunities of the European Commission. It was established to financially support the implementation of the objectives of the European Union in the employment and social affairs area, as set out in the Social Agenda, and thereby contribute to the achievement of the Lisbon Strategy goals in these fields.

The seven-year Programme targets all stakeholders who can help shape the development of appropriate and effective employment and social legislation and policies, across the EU-27, EFTA-EEA and EU candidate and potential candidate countries. PROGRESS mission is to strengthen the EU contribution in support of Member States’ commitment. PROGRESS will be instrumental in:

· providing analysis and policy advice on PROGRESS policy areas; · monitoring and reporting on the implementation of EU legislation and policies in PROGRESS policy areas; · promoting policy transfer, learning and support among Member States on EU objectives and priorities; and · relaying the views of the stakeholders and society at large

For more information see: http://ec.europa.eu/employment_social/progress/index_en.html International perspectives on positive action measures A comparative analysis in the European Union, Canada, the United States and South Africa

European Commission Directorate-General for Employment, Social Affairs and Equal Opportunities Unit G.4 Manuscript completed in January 2009 The information contained in this publication does not necessarily reflect the position or opinion of the European Commission. The consortium producing this report comprised the Centre for Inclusion and Diversity, University of Bradford, European Roma Rights Centre and Ludwig Boltzmann Institute of Human Rights.

The Research Consortium University of Bradford Project Director and Principal Investigator Professor Uduak Archibong Research Officer Dr Jite Eferakorho Project Team and Experts Dr Karl Atkin (University of York), Professor Carol Baxter (NHS Employers), Dr Aliya Darr (NHS Employers/University of Bradford) and Professor Mark Johnson (De Montfort University) Legal team Professor Mark Bell (University of Leicester) and Professor Lisa Waddington (University of Maastricht) Statistician Andy Scally (University of Bradford) European Roma Rights Centre Tara Bedard, Tatjana Peric and Savelina Velislavova Russinova Ludwig Boltzmann Institute of Human Rights Katrin Wladasch Non-European Country Experts Canada Professor Pat Bradshaw (York University) South Africa Professor Oluyinka Adejumo (Kwa-Zulu Natal University) USA Professor Phyllis Sharps (Johns Hopkins University) Acknowledgements The consortium members would like to thank all the participating organisations, government offices, companies and individuals who took part in the survey, consensus workshops, expert panels, interviews and conference calls. We are particularly indebted to Priority Research Limited for co-ordinating the survey. We would like to thank Kellie Barnes for her administrative support throughout the study and her co- ordination of the conference. We are also grateful to Jan Wilcock for her help in producing the graphics for the report and to Fahmida Ashraf for her ongoing support and assistance. The consortium would also like to acknowledge the commitment and considerable assistance made by members of the steering committee, management board, and particularly the members of the Directorate-General for Employment, Social Affairs and Equal Opportunities. Despite all the above contributions, the consortium members remain responsible for any errors or misunderstandings reflected in this report.

Europe Direct is a service to help you find answers to your questions about the European Union

Freephone number (*): 00 800 6 7 8 9 10 11

(*) Certain mobile telephone operators do not allow access to 00 800 numbers or these calls may be billed.

A great deal of additional information on the European Union is available on the Internet. It can be accessed through the Europa server (http://europa.eu). © European Communities, 2009 Reproduction is authorised provided the source is acknowledged. Cataloguing data as well as an abstract can be found at the end of this publication. Luxembourg: Office for Official Publications of the European Communities, 2009 ISBN 978-92-79-11156-3 DOI 10.2767/13963 Printed in Luxembourg

Pr i n t e d o n w h i t e c h l o r i n e -f r e e p a p e r Contents Contents Continuing the Diversity Journey

Executive summary ...... 5

1. Introduction and methodological approach...... 9

2. Contextualising the study...... 15

3. Exploring positive action from a legal perspective...... 21

4. Perceptions of positive action in the European Union...... 33

5. Perceptions of positive action in Canada, the United States of America and South Africa...... 45

6. Positive action measures across different discrimination grounds, organisations and sectors...... 51

7. Conclusions and recommendations ...... 61

Bibliography...... 69

4 Executive summary International perspectives on positive action measures

There is widespread recognition that Definition and a non-binding legislative instrument, the problem of discrimination in em- understanding of positive such as a Commission recommenda- ployment and service provision will action tion or a Council resolution. Alterna- not disappear on its own. The role of tive mechanisms could include a joint governments in promoting the use of The study highlights confusion and in- declaration by the social partners. positive action measures to address consistency in the terminologies used discriminatory practices is vital. At the to describe positive measures across Context of equality and EU level, there has been a significant the study countries. Whilst European diversity expansion of legislation which pro- countries were more likely to talk about vides the basis for activities to prevent ‘positive action’, the term ‘affirmative Positive action measures were gen- and tackle discrimination. In spite of action’ was more commonly used in the erally undertaken by organisations the wealth of EC legislation support- non-European countries. In compari- within the framework of written ing the use of positive action however, son with target setting, there was very equality policies, mission state- limited progress has been made in de- limited use of quotas and their strong ments and annual reports. Whilst fining the parameters for positive ac- association with preferential treatment a good number of organisations tion and its application. attracted a negative response amongst had a written policy on equality study participants. On the whole, it and diversity, larger organisations This study was undertaken to help the was felt that greater clarification of the were not as effective in setting ap- EU develop a framework for better nature and purpose of positive action propriate targets for employment understanding the role that positive measures was needed which would in and service delivery. Equality moni- action measures can play in practice turn generate a more positive attitude toring was considered to be key to in preventing or remedying discrimi- towards the utility and application of the development of positive action nation, and to gain an insight into positive action. with gender being the most widely the kind of practical positive action monitored and sexual orientation measures already being taken in the To this end, the legal definition pro- the least monitored ground. In the EU (and in the EFTA-EEA coun- duced by this study described positive light of the centrality of monitor- tries Iceland and Norway), as well as action as ‘consisting of proportionate ing in promoting sustained posi- the possible costs and benefits of the measures undertaken with the pur- tive action measures, organisations positive action measures. The study pose of achieving full and effective need to gather disaggregated data in also sought to examine how legal equality in practice for members of key sectoral fields on all grounds of frameworks, policies and practices of groups that are socially or economi- discrimination. The lack of disag- positive action in the European Union cally disadvantaged, or otherwise face gregated data in key sectoral fields compare with Canada, the United the consequences of past or present at present, however, means effec- States and South Africa. discrimination or disadvantage’. tive positive programmes cannot be comprehensively developed and The research adopted a mixed meth- Further to this definition, there is a implemented. This problem could odological approach with three distinct strong need for the EC to work with be tackled by including positive ac- phases of data collection. The first phase pan-European special interest bodies tion as an integral part of a wider consisted of a literature review and the including civil society organisations, organisational corporate mission, development of a working definition of issue-oriented lobby groups and trade workforce planning and service de- positive action. The second phase in- unions to develop and communicate velopment, working closely with the volved running an online survey about a shared understanding on which to relevant governmental bodies. At the the use of positive action measures in the base a common EU approach to posi- same time, positive action needs to 27 EU Member States, two EFTA-EEA tive action. In order to foster a shared be mainstreamed as part of a broader countries as well as Canada, the United understanding of what is meant by normative change and supported by States and South Africa. In the final positive action within the European institutions with proper mentoring phase, a comparative in-depth study of Union, the EC should promote dia- and training. This might include a positive action practice in 11 countries logue with civil society organisations programme of education and train- was undertaken drawing upon data col- and the social partners. Based on such ing including seminars and events lected from consensus workshops, in- dialogue, EU-level guidance on the to increase knowledge and practice terviews, legal frameworks and policy meaning of positive action should be of positive action, its benefits and its documents in each of these countries. developed. This could be in the form of role within diversity strategies.

6 1. Introduction: New BusinessExecutive Horizons insummary Europe

Drivers for positive action identified as important factors, it was Outcomes and impacts evident that individuals received rela- The study identified legislation as the tively less support from line managers. A major problem identified by the main driver for positive action. Other These factors coupled with a number study in relation to the effectiveness key drivers include altruism, moral/ of barriers (discussed below) serve to of positive action relates to the lack ethical considerations, business rea- militate against successful implemen- of systematic monitoring in terms sons, demographic changes, corporate tation of positive action. In light of of outputs and outcomes. Organisa- social responsibility, organisational these difficulties, the European Com- tions were struggling to develop ro- policy and grassroots efforts. Negative mission needs best-practice networks bust evidence, and tended to rely on factors were seen to drive positive ac- to support Member States in dealing ‘soft’ measures to assess the impact of tion in some instances where organisa- with uncertainties and to ensure par- positive action. Whilst on the whole, tions set up programmes for political allel translation and application of the positive action measures are seen to be and financial gains with little genu- EU approach to positive action. These effective in raising awareness of equal- ine interest in the essence of positive networks should operate at national ity issues in organisations, having real action. Given that an organisation’s and cross-sector levels to enable the impact on minority groups, and im- equality and diversity objectives can sharing of ideas, approaches and ac- proving an organisation’s image and be hampered by a lack of understand- tivities, and encourage organisations reputation, their potential contribu- ing of the rationale for positive action to move from intention to action. The tion to business success is less well rec- (and the need to continue to make the EU institutions and Member States ognised and is not always deemed to case for and measure the impact of are recommended to introduce legal translate into better financial results. positive action), undertaking research duties to implement positive action Positive action initiatives were largely around the economic advantages of measures where necessary to achieve time-limited and not seen as long- positive action may prove beneficial. full equality in practice. term measures. Generally, the groups The research should include positive that were reported to benefit the most action’s role within talent manage- Barriers to positive action from positive action initiatives were ment, and the role that senior manag- minority ethnic groups and women, ers and line managers can play in im- Limited human and financial resourc- and the least likely to be beneficiaries plementing positive action. Additional es in addition to time were the most were considered to be lesbian, gay, bi- attention might be paid to the image frequently cited barriers to positive sexual and transgender individuals. of organisations among potential cus- action, particularly in the European tomers or users — and, in the case of countries. Lack of awareness of the Given the lack of evaluation tools used charities and public sector bodies, the benefits of positive action measures by organisations in relation to positive image among those who fund or might within the workforce and in the wider action initiatives, it is imperative to donate to them. society, and the role of the media in develop tools to assist organisations problematising these measures and to establish baseline data to facilitate Support for positive action rendering outcomes as tokenistic, positive action implementation and were also seen by study participants as design robust strategies to support Support of the wider society is es- major barriers to the acceptability of the evaluation of the effectiveness of sential for the success of position ac- positive action. To address this prob- measures taken. In order to evalu- tion programmes. Findings from the lem governments should undertake to ate the need for, and effectiveness present study yielded divergent results educate the general public through ‘so- of, positive action, data collection is regarding support for positive action cial marketing’ about positive action, required. Whilst respecting data pro- and identified successful initiatives as in order to address widespread misun- tection legislation, Member States those based upon meaningful involve- derstandings that appear to exist, and should ensure that organisations may ment of target groups in relation to to facilitate the linking up of various engage in data collection where this their design, planning, implementa- stakeholders already engaged in such is designed to facilitate and analyse tion and evaluation. There was also measures. Widespread awareness- positive action measures. Our find- evidence of commitment of some raising campaigns of both the need for ings show that the future for positive public sector organisations to push the positive action measures for disadvan- action is hopeful and many organisa- boundaries of policy, to develop a much taged groups and the benefits of such tions have plans to introduce positive broader application of positive action. measures for wider society will also action initiatives or programmes in Whilst strong individual commitment promote a wider acceptance and posi- the future. Research therefore should and senior management support were tive attitudes towards positive action. be undertaken to map the current

7 International perspectives on positive action measures

situation of ‘disadvantage’ with re- of others, which might be a reflection of tional issues for all stakeholders in- gard to the different fields in which the particular context or ‘politics’ of that cluding leaders and human resource positive action can be applied, e.g. country. Member States should revise managers, aiming to bring greater employment, education, housing, national legislation where this prohibits equality and diversity to organisa- healthcare, etc. relating to these dif- or restricts the opportunity for public, tions. Positive action measures are ferent grounds. Furthermore, there is private or voluntary organisations to widely perceived as politically con- so little evaluation of ‘good practice’ take positive action. In practice, posi- tentious and require sensitive han- in positive action that development tive action measures tend to focus more dling and careful introduction or of a model for evaluation might also on training and improving employment renewal within organisations. The be advantageous. opportunities rather than service deliv- context for positive action is com- ery. The introduction of positive action plex — its development will increas- Positive action in practice within organisations can produce some ingly require a firm evidence base of negative consequences or backlash such best practice to demonstrate that it Examples of positive practice in organi- as negative stereotypes, stigmatisation, remains an effective avenue towards sations confirm confusion about the lack of proper oversight, dishonest be- progress within pluralist cross-cul- scope of positive action measures and haviour and malpractice. It is notewor- tures, and that its benefits outweigh the overlap with other complementary thy that the qualitative data from con- the dilemmas it can sometimes raise. measures such as equality and diver- sensus workshops and interviews did In the introduction and revision of sity monitoring and impact assessment. not provide any examples of measures EC anti-discrimination legislation, There is an incredibly diverse range of related to religion or belief. it should be ensured that public, pri- activities that fall under the heading of vate or voluntary organisations who positive action. Many countries focus on This study revealed ethical, political, wish to engage in positive action are specific groups, perhaps at the expense social, structural, strategic and opera- entitled to do so.

8 Introduction and

1 methodological approach

9 International perspectives on positive action measures

1. Introduction and methodological approach In this chapter we set the context and objectives of the study, discuss the design of the study and the development and application of the methods of data collection to the different phases of the study. We present an overview of the approaches to the literature review and development of the working definition of positive action. This chapter explains the procedure for developing the online survey and provides an overview of different methods utilised in the in-depth study and dissemination of the study.

1.1. Overview of the study In spite of this extension of EC law, and the existing legal framework set out the significant expansion in the- per in other studies (De Vos, 2007). It In recent years there has been a signifi- sonal and material scope which has oc- will help the Commission gain a bet- cant expansion of legislation by the Eu- curred, enforcement of the legislation at ter insight into what kind of practical ropean Commission in the area of equal national level is still to occur primarily positive action measures are already treatment. The year 2000 saw the adop- through the action of individual victims being taken in the European Union tion of two Directives prohibiting dis- who decide to challenge (allegedly) dis- (and in the EFTA-EEA countries), as crimination on the grounds of race and criminatory behaviour before courts. well as the possible costs and benefits ethnic origin (1), and sexual orientation, This enforcement mechanism was also of the positive action measures. The religion or belief, disability and age (2). provided for in the earliest Gender Non- study also seeks to examine how legal These were followed by amendments discrimination Directives which date frameworks, policies and practices of to the long-standing Directive (76/207/ back to the mid-1970s, and it has been positive action in the European Un- EEC) on equal treatment for men and the subject of criticism over the years ion compare with Canada, the United women, the adoption of a ‘goods and (Blom et al., 1995). This is because vic- States and South Africa. services’ Gender Non-discrimination tims face a range of obstacles in bringing Directive in 2004 (3) and, more recent- discrimination cases, amongst which fi- The Centre for Inclusion and Diver- ly, the adoption of the ‘recast’ Gender nancial and emotional costs are some of sity at the University of Bradford was Directive (4). As a consequence of this the most significant (Bell, 2005). commissioned to lead this 15-month wealth of legislation, the EC is now the project in collaboration with the Eu- dominant force setting the tone and Positive action, including the notion ropean Roma Rights Centre and the content of national non-discrimination of positive duties to promote equality Ludwig Boltzmann Institute of Hu- and equality laws across the 27 Member through, for example, contract com- man Rights. The research team sought States. In addition, specific work has pliance programmes, is one means of to undertake four tasks during the been undertaken regarding positive ac- addressing the limitations and restric- course of the study: tion for Roma in the field of education tions inherent in an individual en- by the European Network of Independ- forcement model based on litigation. 1. develop a working definition of ent Experts on Fundamental Rights (5). Nevertheless, while the use of positive positive action; action measures is recommended as a method of realising equality of oppor- 2. conduct an online survey in the 1 Council Directive 2000/43/EC. tunity, there is a dearth of empirical 27 EU Member States and the 2 Council Directive 2000/78/EC. literature on the use of positive action EFTA-EEA countries; across Europe. 3 Council Directive 2004/113/EC. 3. produce in-depth case studies of 4 Council Directive 2006/54/EC. This report is the result of a study to eight EU countries, plus Canada, 5 European Network of Independent Experts help the European Commission to the United States and South Africa; on Fundamental Rights (2005) Thematic better understand the role that positive Comment No 3, The protection of minorities in the European Union, at: http://ec.europa. action measures can play in practice 4. disseminate the findings of the eu/justice_home/cfr_cdf/doc/thematic_com- in preventing or remedying discrimi- research through a seminar and ments_2005_en.pdf nation, building on the knowledge of publications.

10 1. Introduction and methodological approach

In addition to this report, the study In the context of this project, the phrase implementing positive action (fo- has produced important material that ‘positive action measure’ should be re- cusing on best practice) and les- provides valuable additional informa- garded as an umbrella term covering sons learned; tion and insight into all aspects of the all kinds of activities, initiatives, strat- study. These are as follows. egies and interventions, which seek to • exploring the perceived effective- achieve one or more of the goals iden- ness of the actions undertaken and • A synthesis report which provides tified for positive action in both em- how this could be improved; a comparative analysis of all as- ployment and service provision. pects of the study is presented in • identifying key messages and PAMECUS Vol. 1. 1.3. Aims and objectives of good practice to inform policy the study and practice. • A survey report which collates and summarises the responses from the The study involves individuals who are 1.4. Structure of this report online survey with comparative responsible for designing and imple- analyses by subgroup is presented menting positive action measures e.g. This chapter sets out the background in PAMECUS Vol. 2. human resources personnel, equal- to the study, its formal aims and ob- ity and diversity leads, cohesion and jectives and addresses the methodol- • Eleven country reports on the service development managers, chief ogy and design of the study. Chapter situation of positive action meas- executives and other senior managers 2 presents a review of the literature. ures, including examples in Aus- with responsibility for equality. The Chapter 3 explores positive action tria, Canada, France, Hungary, following are specific objectives: from a legal perspective and provides Ireland, the Netherlands, Slova- the working definition of positive kia, South Africa, Sweden, the • exploring the historical, social and action applied in the current study. United Kingdom and the United political context within which le- Chapter 4 discusses the perceptions of States are presented in PAMEC- gal frameworks on positive action positive action in the European Union US Vol. 3. have been developed across both and EFTA-EEA Member States. Chap- employment and service provision; ter 5 examines how positive action is 1.2. Defining positive action perceived in Canada, the United States • assessing perceptions, understand- and South Africa. Chapter 6 analyses Evidence from previous studies (e.g. ing and the rationale for develop- the perceptions of positive action held Archibong et al., 2006b) show con- ing and implementing strategies by respondents from all the countries fusion and mixed interpretations of for positive action, covering the participating in the survey, focusing positive action. Whilst organisations discrimination grounds of age, specifically upon differences across are permitted by law under certain disability, race, religion and belief discrimination grounds, sectors and circumstances to take positive action and sexual orientation; aspects of organisation types. Chapter 7 presents measures, no legal definition exists gender which intersect with other conclusions and recommendations of the concept. Therefore a clearer grounds will also be considered; including a comparison of positive understanding of positive action is action measures in the European and required, as the current confusion is • identifying types, range of positive non-European countries. unhelpful in the design of measures. action measures and how wide- To this end, the legal definition pro- spread they are across the private, 1.5. Methodological duced by this study described positive public and third sectors, and the approach action as ‘consisting of proportionate lengths of time these have been in measures undertaken with the pur- operation; 1.5.1. Design pose of achieving full and effective equality in practice for members of • examining the outcomes and im- A mixed methodology using both groups that are socially or economi- pact of positive action measures in qualitative and quantitative meth- cally disadvantaged, or otherwise face participating organisations; ods was employed, in order to meet the consequences of past or present the aims of the study, with a ration- discrimination or disadvantage’. A • obtaining the views of organisa- ale that seeks elaboration, enhance- full legal definition of positive action tions about the actual or perceived ment, illustration, (and) clarification measures as used in the study is pro- benefits, including relative success, of the results from one method with vided in Chapter 3. and main obstacles and barriers of the results from another’ (Adamson

11 International perspectives on positive action measures

et al., 2004). Combining both meth- measures as currently implemented in the survey. The questionnaire is di- odologies allowed us to triangulate non-European and EU countries. Uti- vided into 10 sections and consists the study, thereby providing us with a lising extensive electronic and library of 35 questions covering information more complete picture of the situation resources as well as other facilities, we about implementation of equality in different countries and ensuring conducted a comprehensive review of and diversity policies, understand- greater validity of the study findings the wider literature on the research, ing and perceptions of positive ac- (Tashakkori and Teddlie, 1998). practice and theoretical debates on tion and use of positive action meas- positive action. Information gathered ures, including outcome measures. The research was carried out in three from the literature study informed The survey also asks organisations to distinct phases: a literature review and the task of providing a definition of identify possible barriers to positive development of working definition of the concept of positive action. The in- action and about their future plans to positive action; an online survey in the formation also identified themes and conduct positive action. 27 EU Member States; two EFTA-EEA dimensions to be built into the other countries and three non-EU countries phases of the study. A total of 632 respondents completed participating in the study and compar- the survey. The majority of them were ative in-depth study. Before beginning This phase involved the research from Italy (12.2 %), the United King- the study, we held two meta-planning team conducting a comprehensive dom (9.2 %), Austria (7.3 %), Belgium meetings with some members of the review of the literature and building (6.5 %), Canada (5.5 %) and Germany steering group and the EC official, to on previous studies in order to have a (5.2 %). Other countries provided less discuss the context and focus of the clearer understanding of positive ac- than 5% of the responses (7). In gener- study, sources of data and how these tion. The working definition was put al, 40 % of the respondents were from would be collected and strategies for through a rapid consultation process the eight European countries and engaging with different stakeholders. for scrutiny by consortium members, 11 % from non-EU countries involved A detailed discussion of the meth- steering committee, EC official and in the in-depth study. Equality and odological approach employed within an expert panel comprising legal, hu- diversity managers represented the the study is presented in PAMECUS man resources and equality and diver- highest pool of respondents (23 %) Vol. 1 — Synthesis report. sity personnel. The working definition while managing directors were next and responses to the definition are (20 %). Nearly half of the respondents Prior to commencing the fieldwork, presented in Chapter 3. had been in post for five years; over the Bradford research team was grant- 30 % in post between one and five ed ethics approval to ensure that the 1.5.3. Online survey years and 12 % under one year. study was undertaken in line with the University of Bradford Code of Re- The online survey was conducted In terms of sector distribution in the search Ethics and in compliance with between 15 May and 24 October total sample, the vast majority were the Data Protection Act. Our partners 2008. The survey involved the 27 EU from the public sector (39 %), then at York University in Canada, Johns Member States plus two EFTA coun- the voluntary sector (37 %) and the Hopkins University in the United tries (Iceland and Norway) and three private sector (19 %). However, the States and University of KwaZulu Na- non-European countries (Canada, private sector comprised a larger pro- tal in South Africa were also granted the United States and South Africa). portion in North America and South ethical approval. We utilised different strategies to col- Africa, at 43 %. Whilst the majority lect data from multiple constituen- of the public sector respondents were 1.5.2. Literature review and cies within public, private and third from colleges/universities (24 %), development of the working sector organisations. An online ques- most of those from the voluntary definition of positive action tionnaire (6) was launched in May sector came from education (20 %), 2008 with an option for respondents and private sector respondents were There is already a good deal of infor- to complete the questionnaire in predominantly from the professional mation on positive action measures in English, French or German, and was and business services sector (29 %). reports and studies such as the research later translated into five additional Respondents came from organisa- by Dhami et al. (2006), Archibong et al. languages (Bulgarian, Hungarian, tions of varying size with the majority (2006b) and Baxter et al. (2008). Before Italian, Polish and Slovak) for use in (40 %) from public/private organisa- commencing fieldwork, an in-depth literature review was conducted to 6 A copy of the questionnaire is available in 7 A full breakdown of the respondents is avail- explore the context of positive action Pamecus Vol. 1 — Synthesis report. able in Pamecus Vol. 2 — Survey report.

12 1. Introduction and methodological approach

tions with over 1 000 employees. Over al., 2006). Changes were made to the consensus workshops in nine of the 50 % of those from the voluntary sec- original EU countries in negotiation 11 case-study countries. tor/NGOs had between one and 10 with the European Commission in paid and unpaid staff. A quarter of the order to capture as many examples of Excepting France and Sweden, half- respondents were from organisations positive action (on all grounds) and day consensus workshops were held in with a branch/subsidiary or significant lessons for the effectiveness and repli- each country, as the main focal point customer base in European countries cability of these measures. In addition, for data collection activities during other than their own; whilst one in we also considered the need to work this phase. During each workshop, five organisations has a branch/sub- with countries represented by mem- two discussion groups were held with sidiary or significant customer base in bers of the project team to ensure ease representatives from all stakeholders other countries outside Europe. of access to the required participating including employers covering private, organisations. public and third sector organisations, 1.5.4. Comparative study on campaigning bodies representing dis- positive action This comparative study allowed for advantaged groups, employer associa- further elucidation of the need for, tions and trade unions. Each workshop A comparative case-study approach use of and impact of positive action began with a plenary session present- was adopted in exploring legal frame- measures. This aspect of the study ing a keynote address and an overview works and practical applications of specifically aimed to identify the his- of the research study. This was followed positive action measures in selected torical, social and political context by group discussions, facilitated by non-European and European Un- of positive action, to explore its im- the researchers, culminating in a final ion countries (including EFTA-EEA plementation, identify methods of plenary session. A total of 272 people countries). The case-study approach measuring its impact, seek the views took part in 18 heterogeneous small facilitated a sufficiently thorough in- of organisations regarding its bene- group discussions. The groups were vestigation of each country within fits, success and challenges and invite presented with a series of questions to its real-life context, using multiple suggestions for overcoming challeng- structure the initial discussions, but sources of evidence. The findings of es to the successful implementation they were not precluded from discuss- the comparative study are presented of positive action. In addition to ing other pertinent issues. in Chapters 3, 4, 5 and 6. an in-depth literature review of the context and application of positive Themes elicited from workshops The non-European countries cho- action in the selected EU and non- were further validated by conduct- sen for an in-depth comparative case EU countries, the comparative case- ing targeted follow-on face-to-face or study were Canada, the United States study data were collected by means telephone interviews with individu- and South Africa. These countries of participatory methods, including a als identified from consensus work- were selected because of their his- consensus workshop, interviews and shops who were willing to discuss tory and credibility regarding anti- the analysis of the legal framework their views in more detail and those discrimination laws and affirmative and policy documents of participat- people who were unable to attend but action measures they practised. For ing organisations. The in-depth study wished to contribute. More specific Canada and the United States, the fieldwork was conducted from 10 areas, mostly context-specific ques- measures are well established, but in July to 24 September 2008. tions, were covered. A total of 141 South Africa the measures and legal detailed interviews were conducted framework have been more recent. Consensus workshops and between July and September 2008. interviews With the consent of the interviewees, The European countries involved in most interviews were taped and lasted the case study are Austria, France, A consensus workshop method was approximately 30 to 60 minutes. The Hungary, Ireland, the Netherlands, adopted in bringing together the wis- interviews took a semi-structured, Slovakia, Sweden and the UK. In gen- dom, interpretations and experiences in-depth form with key actors, using eral, these countries were selected on of all stakeholders for the best possible an interview pro-forma to facilitate the basis of geographical importance, results and decisions in the context of ‘guided conversation[s]’ (Fielding, covering different regions in Europe, positive action activities in each coun- 1993, p. 144). size and experience of positive action try (Spencer 1989; Stanfield 2002). measures on the different grounds In order to gain a deeper insight into Questions for the workshops and of equality, and evidence from a re- peoples’ perceptions and experienc- interviews were designed to explore view of the literature (e.g., Dhami et es of positive action, we conducted understanding and commitment to

13 International perspectives on positive action measures

positive action, and identify exam- provided an overview of how legal 1.6. Data analysis ples of good practice to determine the frameworks, policies and procedures level of ownership within the proc- supported different positive action The survey data were analysed using ess of designing and implementing measures in the countries involved. the statistical package for the social sci- such initiatives, the extent to which The legal analysis was undertaken us- ences (SPSS) employing a combination positive action measures addressed ing a template and was based around of descriptive and inferential statistics. the whole employment cycle and the four key questions, which are pre- Because of the difficulties in establish- goals of positive action — why the ac- sented in Chapter 3. ing a suitable sampling frame, and the tions were set up, who was responsi- subsequent use of snowballing to im- ble, how they were designed and what Expert panel prove the response rate, the sample is thoughts were put into the design and unlikely to be entirely representative of who was consulted about the initia- An expert panel approach was used the population of all organisations in tive. Workshops and interviews were at two stages in the project. It was the participating countries and cannot conducted to check if the experiences first utilised during the development be treated as if it were a random sam- of positive action matched the aspira- of the working definition of positive ple. Nevertheless, it was considered to tions of the initiatives. action and then through a telecon- be useful to perform cross-tabulations ference conducted for confirmabil- of responses with organisational char- Documentary analysis — ity (8) involving three members of acteristics and use a p-value threshold policy and legal framework the research team and three panel of 0.05 to provide an indication of char- members from education, health and acteristics that appeared to be related In addition to the consensus work- human resources backgrounds. The to a response considerably above or be- shops, this phase involved a detailed confirmability was necessary to assess low the mean response of the sample in analysis of organisational policies / trustworthiness of the US consensus each section of the questionnaire. The legal frameworks underpinning posi- workshop data, as participants were chi-squared or Fisher’s exact tests were tive action measures in each of the 11 drawn from a less divergent pool. used, as appropriate. countries. Documentary analysis has All panel members were selected for a long history in social science and their substantial academic and prac- To introduce transparency and a sys- can be a valuable source of data (Prior, tical experience of the issues involved tematic approach, we used a framework 2003). Acquiring documents can be a in this study. Panel members were in- approach to qualitative data analysis relatively economical and expedient itially presented with an overview of (Silverman, 2001) on the consensus method of data collection and can be the main issues, particularly the key workshops, interviews, policy analysis undertaken either quantitatively or statements identified from the con- and expert data. A thematic framework qualitatively. The researchers com- sensus workshop in the United States. was cross-referenced with project aims piled an initial list of all the required They were then asked to present ad- and objectives, as well as the main dis- documents based on the project aims ditional views on their understand- cussion points highlighted in the litera- and objectives. Interviewees were ing, drivers, effectiveness and impact ture review. Documentary analysis was requested to provide relevant docu- of positive (affirmative) action in the undertaken using a context-based con- ments. The documentary analysis United States. tent analysis approach (Prior, 2003).

8 Confirmability refers to the degree to which the results could be confirmed or corroborat- ed by others.

14 2 Contextualising the study International perspectives on positive action measures

2. Contextualising the study In this chapter we present an overview of the literature and current research and debates around positive action. A review of EC legislation and policy, including an analysis of the legal framework of study countries is provided in Chapter 3. However, research that looks at the effects of the legislation on positive action will be presented in this chapter. We examine the various meanings and range of positive action, as well as related theoretical debates. The chapter also highlights important policy developments and implications for the implementation of positive action measures in some countries involved in this study. The literature review helped provide sufficient insights into exploring the empirical data collected from the survey and the in-depth country study. With respect to the country-specific literature on positive action policy and practices (and legal analyses), consortium partners have assisted with some of the review of pertinent publications in order to overcome language barriers.

2.1. Conceptualisations of international team working, whereas ply this concept through workforce positive action Miller and Rowney (1999) prefer to diversity measures, but above all use ‘managing diversity’. Diversity positive action is embedded within a The understanding of the term ‘posi- management is an emerging para- larger political context. These factors tive action’ and related terminology digm utilised by human resource pro- can impact on the nature of positive vary significantly across countries, fessionals to maximise the potential of action and initiatives and can lead to sectors and discrimination grounds. diverse staff in fulfilling the mission of misinterpretation of positive action, This situation is further compounded the organisation (Dhami, et al., 2006; which can in turn result in a lack of, by the fact that terms such as ‘posi- Thomas and Ely, 2002). Dhami et al. or limited engagement with, positive tive action’, ‘reverse discrimination’, (2006) define diversity management action (Chater & Chater, 1992; Johns, ‘positive discrimination’, ‘affirma- as ‘a process intended to create and 2005; Archibong et al., 2007). tive action’ and ‘corrective action’ are maintain a positive work environment used synonymously in myriad con- where the differences of individuals McCrudden (1986) provides ‘five texts (Adam 1997; Archibong et al., are valued so that all can reach their types (9) of positive action … not in 2006a). Alternative labels for positive potential and maximise their contri- the sense of what is legally permis- action include ‘constructive action’ butions to the organisation’s strategic sible, but in how the term appears to and ‘structural initiatives’, ‘diversifica- goals’ (p. 22). Whilst positive action be used in common parlance’ (p. 223). tion strategies’ (Groschi and Doherty, has a legal basis, diversity manage- Burrows and Robinson’s (2007) work 1999) and ‘mainstreaming projects’ ment is fundamentally a management ‘explores some of the ways to con- and ‘balancing measures’ (McCrud- technique, which complements posi- ceptualise positive action measures den 2007). Whilst these terms may be tive action. Complementary measures in relation to sex discrimination in considered to be related and border- can promote success and sustainabil- the context of the law of the EC and line cases, Archibong et al., (2006a) ity, and in turn maximise the impact of Great Britain’ (p. 24). The authors posit that the terms should, however, of positive action (Welsh et al., 1994). argue that the EC law is more progres- remain distinct from positive action sive and consequently provides in- itself. A more detailed description of Archibong et al. (2007) describe posi- sights into possible legislative reform some of these concepts is provided in tive action as having three significant in United Kingdom. The authors also Chapter 3 under the legal annotation conceptual dimensions: the legislative, express frustration with the slow pace of the definition. the executive or practice, and the political, which includes its communica- 9 McCrudden’s typology of positive action: (1) More broadly, Iles and Hayers (1997) tion or surrounding debate. They posit eradicating discrimination; (2) facially neutral but purposefully inclusionary policies; (3) out- use the overarching ‘diversity compe- that while the statutory bodies explain reach programmes; (4) preferential treatment tence approach’ to describe effective the legislative concept, managers ap- in employment and (5) redefining ‘merit’.

16 2. Contextualising the study

of legal reform and call for United Another study examined positive ac- cited statistics showing women were Kingdom to align its legislation with tion measures in the health and edu- significantly under-represented in top that of the EC in relation to conditions cational institutions (Archibong et al., corporate leadership positions, with of women in employment. Their arti- 2006b) and suggests that there was only 3 % as chairs of boards and 1.9 % cle provides a more detailed analysis mixed interpretation or confusion be- as chief executive officers. of positive action and draws from the tween equal opportunities practices literature to illustrate core underlying and positive action policies. Gener- Parker et al.’s (1998) qualitative re- principles (for example, equality) that ally, most of the participants in the search found that banks’ implemen- might potentially present an obstacle study felt that there were more posi- tation of equal opportunities poli- to future legislative amendments. Cit- tive action activities within the NHS cies made a career in management a ing McCrudden’s analysis, the authors and educational institutions that were practical possibility for many women, present three types of positive action geared towards women and black and whilst British Telecommunications ‘that might be utilised in the context minority ethnic groups than for those (BT) were commended by Liff (1999) of a revision of employment legisla- with disabilities. for their progressive positive action tion’ (Burrows and Robinson, 2007, approaches including targeted recruit- p. 26) namely purposefully inclusion- Murphy’s (1993) review of a three- ment activities, training directed at ary policies, outreach measures and year positive action project in North- minority employees, encouragement preferential treatment. ern Ireland found programme expan- to job sharers and childcare support. sion, a high success rate of women Examples of positive action initiatives 2.2. Positive action-related finding jobs and benefits across the based in the voluntary sector include research in different community. The project had become a government-funded ‘positive action sectors a catalyst for links regionally, nation- training in housing’ (PATH) scheme ally and with partners across Europe. aimed specifically at recruiting people This section summarises positive There are many other organisations from under-represented ethnic mi- action-related findings within the lit- involved in offering positive action nority backgrounds and offering them erature emerging from the different schemes, such as the Museums Asso- management training opportunities sectors. The literature reveals little ciation. These schemes are monitored (Echiejile, 1994; Bowes and Sim, 2008; research on positive action in the dif- and evaluated regularly and are per- PATH, 2008). ferent sectors (Shields and Price, 2001; ceived to be successful (Museums As- Carter, 2000; Sheffield et al., 1999; Bag- sociation, 2008). In addition, there are 2.3. Measurement of impact ilhole, 1999; Iganski et al., 2001), most some positive action related projects / success of positive of which were from the private and that are not named as positive action. action public sectors. In general, there was a These include initiatives such as the dearth of research relating to positive ‘cultural understanding in leadership The review did not find evidence of a action in the voluntary sector. and management project’ (CULM) discrete list of outcome criteria char- (Archibong and Burford, 2007) which acterising ‘successful’ interventions. There is a blurring of categories be- was acknowledged to have created However, success is largely dependent tween the public and privatised sec- a confidential and safe environment on what the initiative sought to achieve tors, and this has presented some for answering awkward questions and through its aims and objectives. As a challenges with respect to deciphering remaining open and unconstrained result, there is a shortage of evaluative public from private sector organisa- about views. literature for positive action initiatives. tions. Most of the studies dealt with Measures of success and evaluation of positive action measures in the public Mathur-Helm (2005) found that in interventions were generally either sector in health and education. One spite of the various affirmative ac- omitted altogether from the literature, of them examined several positive tion and equal opportunity initia- or were incomplete (Band and Parker, action initiatives within the UK Na- tives, women in South Africa were 2002), and ‘success’ was instead dis- tional Health Service (NHS) trusts, constantly faced with career advance- cussed elusively. Nevertheless, there ranging from work experience place- ment barriers owing to the prevalence is some indication of encouraging ments and mentoring schemes, to ap- of patriarchy in organisations, which outcomes following positive action prentice schemes designed to promote made it ‘difficult to rise to senior and initiatives in the health and education equality of opportunities and diversity executive management levels, and not sectors (Payne and Huffman, 2005). It within the workforce (Baxter et al., benefiting from government poli- is imperative that those either directly 2008) but not carefully coordinated cies and legislation to advance their or indirectly affected by it have a clear nationally (NHS Employers, 2005). careers’ (p. 58). Mathur-Helm’s study understanding of its purpose, and that

17 International perspectivesInternational on positive perspectives action measures on positive action measures

initiatives that are costly to implement versity contracts’, establishing a Cen- representation was important and this can demonstrate clear utility and val- tre for Management of Diversity and study found that women in the indus- ue for money. extending voluntary ‘covenants’. In try tended to have been targeted by the light of these developments in the recruitment campaigns or have read Dhami et al. (2006) echo our observa- Netherlands, workshop participants literature specifically aimed at attract- tion, as they describe measurement of confirmed that the term enjoying cur- ing them. However, meeting targets or the effectiveness of affirmative action rency is ‘diversity management’, while increasing numbers does not neces- policies in the United States as a diffi- the literature uses ‘affirmative action’ sarily confirm ‘success’. Research into cult endeavour. They assert that most (e.g. Dhami et al., 2006; Vries and Pet- positive action in the aviation indus- studies on this subject focus on the tigrew, 1994) and ‘positive action’ (e.g. try found that whilst positive action economic attainment of ethnic minor- Bacchi, 2004) to describe similar ac- led to increased female representa- ity groups, but whilst some measure tivities in the Netherlands. tion, women were also far more likely gross outcomes, others focus on labour to leave the industry. Whilst women force participation, and yet others on Recent studies have engaged in an were successfully recruited through earnings. These factors all affect the empirical assessment of the Fair Em- positive action, this could not neces- evaluations made. Whilst Stephanop- ployment Act in Northern Ireland and sarily be considered to be synonymous olous and Edley’s (1995) review of the analysed the patterns of affirmative with success, as many experienced effectiveness of affirmative action in action agreements between the Fair inherent difficulties in working in a the United States found that, overall, Employment Commission and em- male-dominated organisation (Davey the extent to which affirmative action ployers in Northern Ireland between and Davidson, 2000). had expanded minority employment 1990 and 2000 (Heaton and Teague, in skilled positions was unclear, the 1997; Osborne and Shuttleworth, 2004; Agocs’ (2002) study in Canada ar- programmes considered were effec- McCrudden et al., 2004). Heaton and gued that formalised employment tive, but could possibly be implement- Teague argued that the tension between equity (an alternative expression of ed in a fairer manner. In other studies, a positive institutional context for af- positive action) programmes, with Holzer and Neumark (2000) note clear firmative action and negative ground- mandatory goal-setting and vigorous evidence of better medical care to mi- level religious circumstances could be enforcement by government authori- norities and low-income people from better managed in a climate of peace. ties, has a significant impact on re- affirmative action in medical schools. More recently, Osborne and Shuttle- sults. The study admonished organi- Holzer and Ihlanfeldt (1998) suggest worth (2004) considered the effects sations to adopt ‘mandatory equality that customers often like being served of the legislation ‘a generation on’ and policy rather than voluntary for em- by ‘co-ethnics’, implying that minor- highlighted the success of affirmative ployers’ (Agocs, 2002, p. 22). Whilst ity customers might be happier (and action measures in securing change, the author noted that, relative to oth- white customers less happy) as a result particularly in producing a substantial er countries, Canadian employment of affirmative action. improvement in the employment pro- equity legislation was advanced and file of Catholics, who are now well rep- broad in scope and coverage, the gap Evaluation of the effectiveness of af- resented in senior-level jobs. between the promise of the policy firmative action statutory laws and and the limited results was attributa- policy instruments in the Nether- Reports frequently state that initiatives ble to a lack of support for implemen- lands has been seen largely as a ‘bu- have been ‘successful’ because they tation and enforcement by political reaucratic monstrosity’ (Glastra et have led to increased minority group leaders and employers and a lack of al., 1998) due to the added burden it recruitment. In fact, most of the posi- commitment and resources. Thomas places on employers. Despite the le- tive action literature focuses on the and Jain (2004) attempted to look at gal consequences (recourse to crimi- recruitment stage of the employment the potential lessons for South Africa nal law proceedings) most firms did cycle (Secker, 2001; Refugee Council, based upon the Canadian experience not comply fully, primarily because 2006; Ward, 2006). Certainly there is and concluded that ‘employment employers saw the reality of minor- some justification for this focus on equity must be viewed from both ity (un)employment ‘as a supply-side reaching targets. For example Dainty macro — and micro — perspectives rather than a demand-side problem’ et al. (1999) found that, unlike men, … the real challenge existed to go be- (Dhami et al., 2006, p. 44). As an al- women were unlikely to be advised yond compliance in the ensuring of ternative to any form of legal compli- by friends and family or same-sex top management commitment to the ance, the government proposed a set role models to join the construction sentiments that underlie the legisla- of voluntary measures, including ‘di- industry. Thus, a physical increase in tion and the holistic development of

18 2. Contextualising the study

people and organisational cultures and this has been interpreted as suc- as part of a wider portfolio of meas- that are free of historical discrimina- cess. Nevertheless, evidence suggests ures designed to induce change at an tion’ (p. 51). that self ratings are not always reli- organisational level. This literature able and valid measures of success review, however, did not find any Other studies have reported im- (David and Sutton, 2004). study that had evaluated the effect proved beliefs and attitudes amongst of positive action on culture that was those who have taken part in positive All of the success factors discussed based on evidence and measurement. action schemes (Brew and Garavan, have been associated with changes Furthermore, we found little evidence 1995; Band and Parker, 2002). In in the individuals themselves. How- of longitudinal research into the suc- Band and Parker’s (2002) evaluation ever, it has been noted that individu- cess of positive action interventions. of ethnic minority undergraduate als do not operate within a vacuum, Typically, where an intervention was mentoring schemes, successful out- but within the broader context of an reviewed, it was evaluated in the im- comes included satisfaction among organisation in which factors such mediate to mid-term. Nevertheless, those being mentored that the scheme as perceptions of fairness, threat and the review did find encouragement in had met their expectations, especially utility can individually or collectively a study by Payne and Huffman (2005) in terms of career development, self- impact on the success of positive ac- which found that mentoring of US confidence and study skills. Suc- tion programmes (Kottke and Agars, army officers was positively related to cess was also measured in terms of 2005). Anderson (2004) referring to affective commitment and negatively the enthusiasm and commitment of a case study, has claimed that positive related to turnover behaviour. It also mentors and their satisfaction with action training can help raise aware- provided longitudinal evidence, as the level of benefit provided to their ness and understanding of organisa- affective commitment partially me- students. Several initiatives have re- tional attitudes but strongly implies diated the relationship between men- ported that participants enjoyed the such initiatives will have a limited toring and actual turnover behaviour initiative (Brew and Garavan, 1995), impact unless they are implemented 10 years later.

2.4. Summary

Whilst policy initiatives around positive action typically apply to all sectors, there is a general tendency for the public sector, which is dependent to a larger extent on public funds, to implement the letter of the law with regard to positive action measures. Furthermore, mechanisms built into positive action, such as impact assessments, guarantee that public sector organisations have an incentive to implement positive action measures. On the other hand, private sector organisations, while under less scrutiny to implement positive action measures, are driven by business imperatives. Documentary policy evidence on third sector organisations suggests a higher level of implementation than public sector organisations. More practitioners within the third sector have attributed their commitment to positive action measures to the vision and mission of their organisations, as opposed to monetary (private sector) or assessment (public sector) incentives. However, we acknowledge the general lack of research evidence on the impact of positive action in the third sector as a limitation to this literature review.

This rapid appraisal of positive action literature reveals ethical, political, social, structural, strategic and operational issues for leaders and human resource managers, aiming to bring greater equality and diversity to organisations. Positive action measures are widely perceived as politically contentious and require sensitive handling and careful introduction or renewal within organisations. The context for positive action is complex — its development will increasingly require a firm evidence base of best practice to demonstrate that it remains an effective avenue towards progress within pluralist cross-cultures, and that its benefits outweigh the dilemmas it can sometimes raise.

Exploring positive action 3 from a legal perspective International perspectives on positive action measures

3. Exploring positive action from a legal perspective Given the importance of the law, in this chapter we discuss positive action from a legal perspective. The first part of the chapter explores positive action from the perspective of EC law, and discusses both the relevant legal provisions and the case-law of the European Court of Justice. An extensive study of EC law in this area was also the basis for the working (legal) definition of positive action that was developed for use within this study. The working definition and annotation are given in the second section of this chapter. Moreover, the reaction of respondents to the working definition, as given in the survey, is noted. Lastly, this chapter provides a comparative overview of the laws regulating positive action in the 11 countries that have been the focus of case studies.

3.1. European Community For many years, European Com- regard to the other grounds covered law (10) munity law only addressed non-dis- by Article 13 EC. crimination, and therefore positive Given that all positive action must be action, with regard to gender. This The original Gender Equal Treatment adopted and implemented within a le- situation changed with the incorpo- Directive, which addressed employment gal framework, it is important to have ration of Article 13 in the EC Treaty discrimination, provided an exception an understanding of the scope for, and of 1999, which led to the Community from the prohibition of discrimina- limitations imposed upon, positive ac- obtaining the competence to take action for positive action measures in tion as established by the law. Within tion to combat discrimination on the Article 2(4): ‘this Directive shall be with- the European Union, the basic legal grounds of sex (11), racial and ethnic out prejudice to measures to promote framework with regard to positive origin, religion or belief, sexual ori- equal opportunity for men and women, action is established by a number of entation, age and disability, and with in particular by removing existing in- Council directives and the EC Treaty the subsequent adoption of the Racial equalities which affect women’s oppor- itself, as well as related case-law from Equality (12) and Employment Equality tunities …’. This early provision already the European Court of Justice. In es- Directives (13). Nevertheless, in order reveals an essential element of the Com- sence, these instruments allow Mem- to gain an understanding of the scope munity’s approach to positive action: ber States the possibility to permit for adopting positive action measures there is no obligation on Member States positive action, but only within cer- under Community law, it is important or other parties, to permit or adopt posi- tain limits. Within each Member State, to begin by considering the relevant tive action measures. Instead, as De Vos and also within the non-European provisions within the Gender Equality (2007) has noted: ‘the Community’s pos- countries covered in the PAMECUS Directives of the 1970s and 1980s (14), itive action provisions provide Member study, the limitations to positive ac- and related case-law. To date, no ECJ States with a policy option which, within tion are (further) established in vari- case-law has arisen on this issue with the general limits [established by EC law, ous legal instruments, which include MB/LW], is essentially used at their dis- national constitutions and case-law. 11 As noted, earlier versions of the EC Treaty cretion’ (De Vos, 2007, p. 38). EU Member States cannot allow posi- also allowed some scope for the adoption of tive action which exceeds the limits directives addressing gender discrimination. The European Court of Justice (hence- established by EC law; but neither is 12 Directive 2000/43/EC implementing the forth: the Court) had various oppor- there any obligation on them to al- principle of equal treatment between persons tunities to consider the meaning of low the maximum scope for positive irrespective of racial or ethnic origin, OJ L Article 2(4) of the Directive (15). In its action as provided for in EC law, and 180, 19.7.2000, p. 22. 13 Directive 2000/78/EC establishing a general they may impose further restrictions 15 See C-450/93, Kalanke v Freie Hanses- that EC law does not require. framework for equal treatment in employment and occupation, OJ L 303, 2.12.2000, p. 16. tadt Bremen [1995] ECR I-3069; C-409/97, Marschall v Land Nordrhein-Westfalen, 14 In particular Directive 76/207/EEC on equal [1997] ECR I-6363; C-158/97, Badeck v Hes- 10 Parts of this section are based on an ex- treatment for men and women as regards ac- sischer Ministerpräsident, [2000] ECR I-1875; tract from Waddington, L. and Bell, M. (2001), cess to employment, vocational training and C-407/98, Abrahamsson and Anderson v Fo- ‘More equal than others: distinguishing Eu- promotion, and working conditions, OJ L 39, gelqvist, [2000] ECR I-5539; Case C-476/99, ropean Union equality directives’, Common 14.2.1976, p. 40 (henceforth: original Gender Lommers v Ministerie van Landbouw, Natuur- Market Law Review, 38, pp. 587–611. Equal Treatment Directive). beheer en Visserij, [2002] ECR I-2891.

22 International perspectives on positive action measures3. 1.Exploring Introduction: positive New action Business from Horizons a legal perspective in Europe

most controversial decision, Kalanke, flexible in nature and guarantee an the positive element of that article, the Court stressed that ‘as a derogation objective and individual assessment notably the possibility of conferring from an individual right laid down in of all candidates (19). ‘specific advantages in order to make the Directive, Article 2(4) must be in- it easier for the under-represented terpreted strictly’ (16). The Court has also addressed positive sex to pursue a vocational activity’. action in the context of working con- At first sight, Article 5 of the Racial As of 1999, the Treaty of Amsterdam ditions. In Lommers, which concerned Equality Directive seems to be more resulted in the insertion of a new pro- a provision which offered female staff restrictive than Article 141(4) EC. vision in the EC Treaty which con- access to childcare facilities, but only Similarly, Article 7(1) of the Employ- cerned positive action, Article 141(4) allowed male staff such access in situ- ment Equality Directive only lifts the reads: ‘with a view to ensuring full ations of ‘emergency’, the Court held ‘positive action as compensation for equality in practice between men and that ‘it is not places of employment past wrongs’ dimension from Article women in working life, the principle which are reserved for women but 141(4) EC (24). Nonetheless, one can- of equal treatment shall not prevent enjoyment of certain working con- not conclude that the scope for posi- any Member State from maintaining ditions designed to facilitate their tive action under the two directives or adopting measures providing for pursuit of, and progression in, their of 2000 is more limited than that al- specific advantages in order to make careers …’ (20). The Court regarded lowed for with regard to gender under it easier for the under-represented this measure as forming ‘part of the Article 141(4) EC. Firstly, in the light sex to pursue a vocational activity restricted concept of equality of op- of the decision in Abrahamsson, it is or to prevent or compensate for dis- portunity’ (21), which was allowed un- not evident that these textual differ- advantages in professional careers’. der Article 2(4). ences will result in a broader scope for Although this provides a more posi- positive action in favour of women, tive formulation than that found in In the meantime, the original Equal in comparison with the other Article 2(4) of the aforementioned Treatment Directive has been replaced Article 13 EC grounds. In that case, Directive, the Court’s interpretation by the ‘recast’ Directive (22), which cod- the Court declined an opportunity to of Article 141(4) EC suggests it does ifies all of the older Gender Equality make a fresh start on gender-based not significantly increase the scope Employment Directives. Article 2(4) positive action in Community law, for positive action (17). has been deleted, and instead, all gen- but rather it wove Article 141(4) EC der-based employment positive action into the principles already estab- Since Kalanke, the Court has consist- schemes find their legal foundation in lished through its existing case-law. ently maintained that it will not ac- Article 141(4) EC (23). Moreover, the position, past and cept positive action schemes based on present history of disadvantage and gender which produce ‘equal results’ Turning to the newer instruments, discrimination, and the nature of through automatic mechanisms at Article 5 of the Racial Equality Direc- the barriers experienced, of the eight the selection stage. At the same time, tive provides: ‘with a view to ensuring grounds, or groups of people, covered it must also be acknowledged that the full equality in practice, the principle by Article 13 EC are not the same, Court is willing to permit a wide range of equal treatment shall not prevent and this may influence the finding of positive action measures, including any Member State from maintain- as to what kinds of positive action strict quotas, prior to the point of em- ing or adopting specific measures to measures are compatible with EC ployment selection. For example, in prevent or compensate for disadvan- law. Following the wording of the di- Badeck, the Court was prepared to ac- tages linked to racial or ethnic origin’. rectives, one could argue that where cept measures which imposed a strict Whilst this text closely follows that a group experiences a particularly quota reserving at least 50 % of train- found in Article 141(4) EC, it omits severe form of disadvantage, more ing places for women, and requiring at radical and far-reaching forms of pos- least 50 % of all candidates invited to itive action should be allowed, than 19 Marschall, par. 35. interview to be women (18). Moreover, where lesser degrees of disadvantage the Court has not rejected all forms of 20 Lommers, par. 38. exists. This would suggest that a ‘one positive action at the point of selec- 21 Ibid. tion, but does require that these are 22 Directive 2006/54/EC on the implementa- 24 ‘With a view to ensuring full equality in tion of the principle of equal opportunities and practice, the principle of equal treatment shall not prevent any Member State from maintain- 16 Kalanke, ibid, p. 3078. equal treatment of men and women in matters of employment and occupation (recast), OJ L ing or adopting specific measures to prevent 17 Abrahamsson and Anderson v Fogelqvist. 204, 26.7.2006, p. 23. or compensate for disadvantages linked to any of the discriminatory grounds referred to in 18 Badeck, pars. 55 and 63. 23 See Article 3 of the ‘recast’ Directive. Article 1.’

23 International perspectives on positive action measures

size fits all’ approach (25), which the persons, the principle of equal treat- 3.2. Working definition of Court adopted in its gender case-law in ment is without prejudice to the right positive action the 1990s, would not be appropriate in of Member States to maintain or adopt the context of the Racial Equality and provisions on the protection of health Drawn from EC law, a working defini- Employment Equality Directives. and safety at work or to measures tion of positive action was developed aimed at creating or maintaining pro- for use throughout the current study. However, given the lack of case-law visions or facilities for safeguarding at the European level on the new or promoting their integration into For the purpose of this study, positive grounds, there remains a variety of the working environment’. The lat- action consists of proportionate meas- positive action schemes which have ter element of this provision appears ures undertaken with the purpose of yet to be tested for their compatibility to stem from proposals of the Dutch achieving full and effective equality in with European Community law. For delegation (28), reflecting a desire to practice for members of groups that example, whilst in Badeck the Court protect existing elements of Dutch law are socially or economically disad- was willing to permit training schemes which provide for preferential treat- vantaged, or otherwise face the con- which reserved 50 % of the places for ment of disabled persons in order to sequences of past or present discrimi- women, how would the Court regard support their reintegration into the nation or disadvantage. In order to training schemes which are exclusive- workforce (29). It is less clear how the achieve this, positive action measures ly provided for persons of a particular reference to health and safety law in are designed to achieve one or more of ethnic origin (26)? this context relates to positive action the following goals: for disabled persons. The most logical A further potentially challenging issue explanation is that this provides Mem- • preventing or compensating for is the compatibility of employment ber States with the possibility to adapt disadvantages and discrimination, quotas for people with disabilities with their health and safety regimes to take whether these arose in the past or the Employment Equality Directive. A account of the particular situation of are still ongoing; significant number of Member States disabled workers. This reinforces cer- of the European Union provide for tain existing obligations on employers • promoting substantive equality by some form of (obligatory) quotas (27), imposed under Health and Safety Di- taking into account the specific and, in countries such as France and rectives (30). Nonetheless, there is also situation of members of disadvan- Germany, quotas are regarded as an a risk that excessively protectionist taged groups and breaking the cy- intrinsic element of disability employ- measures ostensibly designed to guar- cle of disadvantage associated with ment policy. Such schemes would nat- antee the health and safety of workers membership of a particular group; urally fall foul of the test established in with a disability, could in fact result in Kalanke, though, as noted above, the the exclusion and denial of equal treat- • redressing under-representation Court may regard the different social ment to people with disabilities (31). and promoting diversity in partici- context for each of the grounds of dis- pation of all groups in social, eco- crimination as justifying a change in nomic, cultural and political life. 28 Council of the EU (2000) ‘Outcome of pro- the scope for positive action. Moreo- ceedings of the working party on social ques- ver, Article 7(2) of the Employment tions of 14 and 28 March 2000’, 6941/00, Brus- • Positive action measures achieve Equality Directive provides addi- sels, 31 March 2000, pp. 5–6. these goals by influencing the way tional protection for positive action 29 Specifically Article 7 of the Wet op de (re) in which social goods, such as em- in respect of people with disabilities. integratie arbeidsgehandicaten. See Wad- ployment, education, housing or It states that ‘with regard to disabled dington, L. (2000), ‘Tweede-generatie richtlij- healthcare, are allocated. nen Gelijke Behandeling: de nieuwe Richtlijn inzake gelijke behandeling ongeacht ras of 25 By this is meant judgments which purport etnische afstamming en de Kaderrichtlijn ge- Positive action covers a wide range of to set out what kinds of positive action are al- lijke behandeling in arbeid en beroep’, Sociaal measures, but, in the light of EC law, it ways prohibited, for all Member States and in Recht, 12, pp. 357–362. is not viewed as including automatic all circumstances. 30 See, for example, Directive 89/391/EEC on and unconditional preferential treat- 26 For example, this is permitted under Sec- the introduction of measures to encourage im- ment for women (or men) in selection tion 37 and Section 38 of the British Race Rela- provements in the safety and health of workers for employment (e.g. quotas). Taking tions Act 1976. at work, OJ L 183, 29.6.1989, p. 1, Article 15. into account EC legislation, positive 27 For further information see Waddington, 31 For a consideration of this problem from a action also has a broader meaning in L. (1996), ‘Reassessing the employment of peo- British perspective see Davies, J. and Davies, relation to disability, where it includes ple with disabilities in Europe: from quotas to W. (2000), ‘Reconciling risk and the employ- anti-discrimination laws’, Comparative Labor ment of disabled persons in a reformed welfare measures aimed at creating or main- Law Review, 18, pp. 62–101. State’, Industrial Law Journal, 29, pp. 347–377. taining provisions or facilities for safe-

24 3. Exploring1. Introduction: positive New action Business from aHorizons legal perspective in Europe

guarding or promoting the integration such as quotas for ethnic minorities. In relation to disability, the Employ- of disabled persons into the working The European context is distinct in ment Equality Directive uses the environment. This includes preferen- two respects. First, there has generally heading positive action to permit tial treatment for disabled persons, been less recourse to strong forms of measures that would otherwise fall such as quotas. preferential treatment, with the excep- within the category of positive dis- tion of disability and, to a lesser ex- crimination (34). 3.3. Annotation to the tent, gender. Secondly, in its case-law definition of positive on gender equality, the Court of Jus- 3.3.2. Beneficiaries of positive action tice has held that measures which give action absolute and unconditional priority to As is clear from the definition given the under-represented sex (typically As is clear from the definition given above, positive action measures can women) at the point of employment above, positive action measures are be targeted at many different groups, selection, constitute unlawful dis- designed to improve the position of serve related but different goals, and crimination against members of the socially or economically disadvan- impact on many different areas of life. other sex (32). As a result, it is helpful taged groups and groups which face Furthermore, positive action measures to distinguish in the terminology used the consequences of past or present can apply at the national level, and be between different types of measures discrimination or disadvantage. established by legislation or govern- targeted at disadvantaged groups. However, since groups, by their very ment policy, or at the workplace level nature, consist of numerous indi- and only be based on the efforts of ‘Positive action’ is generally taken to viduals, the beneficiaries of positive one individual, or may fall in between include all those measures which seek action measures are also individuals these two extremes. As a consequence, to benefit socially or economically who are themselves members of the positive action measures can be very disadvantaged groups, but which do target group. By targeting and benefit- diverse indeed. not involve unconditional preferential ing (sufficient numbers of) members treatment in the allocation of social of the group in question, positive ac- In order to provide further insight goods based purely on an individual’s tion aims to ensure that the group as a into the kinds of measures that can be characteristics. In contrast, ‘positive whole will cease to be disadvantaged. regarded as positive action, it is help- discrimination’ connotes those meas- ful for this annotation to discuss and ures which go further, such as ad- Not all groups are the (potential) target elaborate on various aspects of the mitting ethnic minority students to of positive action measures — instead, definition of positive action, as devel- university with lower entrance quali- such measures are directed towards oped for the current study. fications than those of other students. socially or economically disadvan- In the context of gender equality in taged groups and those which face the 3.3.1. Terminology - positive action, the labour market, this distinction consequences of past or present dis- positive discrimination, between positive action and positive crimination or disadvantage. Since the affirmative action discrimination reflects what EC law level of social and economic disadvan- permits; positive action is lawful and tage, as well as the history of discrimi- Although the term ‘positive action’ positive discrimination is unlawful. nation, is sometimes influenced by the has been selected for this report, Whilst it is frequently assumed that state of economic development within there is little consensus within the the Court of Justice will apply simi- a country or sector, and by cultural academic literature on precisely what lar principles to other forms of dis- perceptions, it is important to estab- this entails. This confusion is exac- crimination (33), this has not yet been lish and identify real disadvantage be- erbated by the use of other related tested through litigation. Moreover, fore proceeding to develop a positive terms, such as ‘affirmative action’, it remains difficult to maintain a wa- action scheme. ‘positive discrimination’ or even ‘re- tertight distinction between positive verse discrimination’. action and positive discrimination. The examples given in the report also reveal that positive action meas- ‘Affirmative action’ is a term which ures can be targeted at many differ- is used less frequently in Europe, but 32 See e.g. para 27, Case C-319/03 Briheche ent groups. In contrast, European [2004] ECR I-8807. which is the dominant term in the Community law only addresses non- USA. In that context, it has been asso- 33 European Commission, ‘The application of discrimination, and therefore posi- ciated with a wide range of measures Directive 2000/43/EC of 29 June 2000 imple- tive action, with regard to six named menting the principle of equal treatment be- including strong forms of preferential tween persons irrespective of racial or ethnic treatment for disadvantaged groups, origin’ COM(2006) 643, p. 7. 34 Article 7(2), Directive 2000/78/EC.

25 International perspectives on positive action measures

categories, namely those groups establish positive action measures they are pursuing. For example, where identified on the basis of sex, race or in its capacity as an employer and a a group experiences particularly severe ethnic origin, sexual orientation, reli- provider of services and grants. Such disadvantage, exclusion and discrimi- gion or belief, disability and age. This initiatives can be taken by central gov- nation, the proportionality principle implies that any positive action scheme ernment, as well as by federal govern- implies that more radical positive ac- directed at such groups, whether at the ment and local authorities within the tion measures will be justified — than national or the workplace level, must scope of their powers. Private employ- in the case of a group which experienc- be in conformity with Community ers and service providers, as well as es less disadvantage. Furthermore, the non-discrimination law. The limits of employers and service providers in the principle implies that once a positive Community law with regard to positive public sector, such as universities and action measure has achieved its aim, action based on sex and disability have hospitals, can also implement positive and the situation of social or economic been noted briefly in the definition and action measures, as can voluntary sec- disadvantage or the consequences of above, and are based on EC legislation tor organisations. Such measures can past or present discrimination or dis- and case-law (35). Community law im- be adopted by both large and small advantage have been eliminated, the poses no direct limits on positive action organisations, and cover the whole of measure should be discontinued. How- targeted at groups which do not fall into their operations, or just a specific part ever, one should not conclude from this these categories, such as people with or region in which they operate. that all positive action measures should a criminal record, or asylum-seekers. automatically be temporary in nature. However, national law may also limit the 3.3.4. Positive action must be Some forms of exclusion and disadvan- possibility of directing positive action at proportionate tage are so embedded that only long- certain groups, and may indeed be more term positive action schemes can help restrictive than European Community Positive action measures provide for to correct them and where a group-re- law in this respect. benefits for members of disadvantaged lated characteristic is likely to lead to a groups, in terms of enhanced access permanent reduction in (employment 3.3.3. Bodies responsible for to certain social goods, and serve spe- and educational) possibilities, such as implementing positive action cific goals. However, positive action is the case with certain kinds of (intel- measures can have negative effects on lectual) disability, permanent positive It is up to the State to decide whether, individuals who are not members of action schemes can be proportionate and how far, to permit positive action the target group, in that they may find and justified. measures (within the boundaries al- it more difficult to access social goods lowed by EC law). However, once the than would otherwise be the case. This 3.3.5. Positive action and data State has provided that positive action is because social goods, such as em- collection measures are permissible, a wide vari- ployment opportunities or housing, are ety of actors can then implement spe- finite, and a distribution of such goods Positive action is often closely associ- cific measures and programmes. The which favours one group, must inevi- ated with data collection, although State itself can both legislate to estab- tably lead to reduced opportunities for the two are distinct. Data collection lish positive action schemes, such as other groups. Nevertheless, positive refers to mechanisms for gathering through the establishment of a (bind- action measures are necessary and ap- information on the situation of dis- ing) employment quota designed to propriate when, inter alia, the previous advantaged groups. These take a wide benefit people with disabilities, or distribution method disproportionate- variety of forms (36). On the one hand, ly disadvantaged members of certain quantitative data collection might en- groups, and the positive action meas- tail the disaggregation of statistics by 35 Specifically Article 2(4) of Directive 76/207/ ure serves to correct this disadvantage. reference to gender or age, or alterna- EEC on the implementation of the princi- tively monitoring the proportion of job ple of equal treatment for men and women as regards access to employment, vocational However, positive action measures can training and promotion and working condi- only be justified as long as the target tions, OJ L 39, 14.2.1976, p. 40 and the related group remains socially or economically 36 For a range of examples, see Makkonen, T. case-law (for a good overview of the case-law disadvantaged, or the consequences of (2006), European handbook on equality data, see Gijzen, M. H. S. (2006), ‘Selected issues in Luxembourg: Office for Official Publications equal treatment law: a multi-layered compari- past or present discrimination or disad- of the European Communities; European son of European, English and Dutch law’, The- vantage remain ongoing, and the posi- Commission (2004), Comparative study on the sis, Intersentia, pp. 227–234 (Section entitled: tive action measure serves to eliminate collection of data to measure the extent of dis- ‘From Kalanke to Briheche’); and Article 7 of or reduce this disadvantage. In essence crimination within the United States, Canada, Directive 2000/78/EC establishing a general Australia, the United Kingdom and the Nether- framework for equal treatment in employment this means that positive action meas- lands, Luxembourg: Office for Official Publi- and occupation, OJ L 303, 2.12.2000, p. 16. ures must be proportionate to the aim cations of the European Communities.

26 3. Exploring1. Introduction: positive New action Business from aHorizons legal perspective in Europe

applicants from ethnic minority com- infrastructure in urban areas with a large The reasonable accommodation re- munities. On the other hand, qualitative ethnic minority population is as good as quirement obliges employers and other data collection methods could include a that available in other parts of the city. providers of social goods not to ignore survey of lesbian and gay persons in or- disability, as is the case with regard to der to discover their experiences in the Whilst it might be argued that main- most elements of non-discrimination healthcare system. streaming is akin to positive action, it is law (40), but specifically to take disabil- arguably clearer to see the two as distinct. ity into account. A reasonable accom- Collecting data does not, by itself, meet Whereas positive action involves target- modation requirement prohibits, for any of the goals of positive action which ed measures that, inter alia, attempt to example, an employer from denying an are identified in the definition adopted compensate for specific disadvantages, individual with a disability an employ- by this project. Compiling statistics does mainstreaming has a broader agenda. ment opportunity, by failing to take ac- not compensate for disadvantages or Its methods are more procedural in count of the protected characteristic, prior discrimination, nor does it redress nature: incorporating the promotion if taking account of it — in terms of under-representation. of equality into decision-making and changing the job or physical environ- service delivery. Mainstreaming seeks ment of the workplace — would enable The main reason for data collection is to change the mindset of policymakers the individual to do the work. to identify where inequalities are cur- in order that equality becomes a central rently being experienced. This then concern. If this approach works effec- Perceived in this way, the notion of forms the context against which positively, it is likely to prompt policymakers reasonable accommodation can be re- tive action initiatives can be taken. to consider taking positive action. For garded as based on a ‘difference’ model For example, if data collection reveals example, mainstreaming might mean of discrimination. This model recog- a low level of participation by young that when a national museum seeks to nises that individuals who possess the black men in law courses at university, raise its number of visitors, it takes into relevant characteristic are different in a then it would be appropriate to design account whether certain groups are cur- relevant respect from individuals who positive action schemes in response, rently under-represented, such as Mus- do not, and that treating them similarly such as recruitment programmes tar- lim women. Once this issue is placed can lead to discrimination. It requires geted at this group. on the agenda, then the museum might that, for example, employers treat some take the next step of designing positive individuals — persons with disabili- 3.3.6. Positive action and action measures in response, such as a ties who would be qualified if the em- mainstreaming targeted outreach programme. ployer modified the job to enable them to perform it — differently from other Mainstreaming is an approach which 3.3.7. Distinguishing positive individuals. This is an asymmetric no- seeks to promote equality by mobilising action from reasonable tion and requires that some definition all areas of law and policy (37). Rather accommodation or classification of the covered group be than relying only on specific measures, included in the legislation. such as anti-discrimination legislation, At first glance it may seem that the ob- mainstreaming implies that equality ligation to provide for a ‘reasonable ac- Directive 2000/43/EC for a legal definition of the needs to be pursued in all activities. In commodation’ (38) is a particular form concepts of direct and indirect discrimination. practice, this means that equality needs of positive action, as it provides for ‘ad- 40 Conventional employment non-discrimina- to be taken into account during policy vantages’ to individuals who fall within tion legislation, including the long-standing EC gender-based directives and the Racial Equality formulation, implementation and evalu- the group of persons with a disability. Directive (2000/43/EC), is based on the premise ation. For example, it might be assumed However, this impression is mislead- that employers should not take into account that equality issues are not especially ing and the obligation to provide for a certain characteristics such as gender or race. relevant to transport policy. In contrast, reasonable accommodation can better These characteristics are generally classified as irrelevant, and not pertinent to the employ- mainstreaming demands that trans- be characterised as a particular kind of ment decision. For this reason, non-discrimi- port policy is used to promote equality. non-discrimination legislative provi- nation legislation usually adopts a symmetric This could entail ensuring that public sion, related to, but not synonymous approach, meaning that both the dominant transport is accessible for persons with, the established forms of direct and group (e.g. men, ethnic majority) and the dis- 39 advantaged group/minority group (e.g. women, with disabilities or that the transport indirect discrimination ( ). ethnic minority) are protected by the discrimination prohibition. This can be regarded as a 37 Centre for Strategy and Evaluation Services 38 As provided for in Article 5 of Directive ‘sameness’ model of discrimination. Under this (2007), ‘Non-discrimination mainstreaming 2000/78/EC. model, discrimination occurs when individu- — instruments, case studies and ways for- als who are fundamentally the same are treated ward’, Brussels: European Commission. 39 See Article 2 of both Directive 2000/78/EC and differently for illegitimate reasons.

27 International perspectives on positive action measures

The obligation to provide for a rea- take, for example, the Leeds hospital those measures which are closer to sonable accommodation can be dis- internship programme described in positive action share some or all of the tinguished from positive action from section 6.7.1, which was limited to following characteristics. Firstly, they a procedural perspective as well. Un- people with black and minority ethnic are targeted at a well-defined social like most forms of positive action backgrounds. At the other end of the group. Secondly, they seek to redress which are aimed at members of so- spectrum, there are measures which disadvantages in a specific setting, cially or economically disadvantaged loosely benefit disadvantaged groups such as access to education or employ- groups, reasonable accommodations but which are traditional elements of ment. Thirdly, the necessity for posi- generally possess an individualised national social welfare policy, rather tive action will be subject to periodic character (41). For this reason, sta- than positive action. Providing State review. As discussed earlier, this does tistical data revealing a numerical pensions for older persons is a good not mean that positive action must be imbalance of a particular group of example. Pensions seek to address the time-limited, but it is not automati- workers, such as women or ethnic social reality that older people gradu- cally assumed to be indefinite. minorities, in a particular employer’s ally leave the labour market and need workforce are largely irrelevant for an alternative means of income. Whilst 3.3.9. Response to the working decisions concerning reasonable ac- this could be viewed as compensatory definition of positive commodation. Reasonable accom- in nature, pensions are more typically modation can also involve regular regarded as a basic social entitlement The legal definition developed for the and ongoing expenditure, such as which flows from the contribution to current study was positively received a provision of personal assistance, society (financial or otherwise) which by a wide range of professionals with rather than a one-off decision to individuals make during the earlier pe- experience in the field of diversity in- award a woman or a member of an riod of their lives. Moreover, pensions cluding an audience at an internation- ethnic minority a job or training. On are not a temporary means of tackling al conference on ‘Gender and positive this basis, the reasonable accommo- disadvantage, but an indefinite feature action’ at the Izmir University of Eco- dation obligation is not susceptible to of national welfare regimes. Finally, nomics in Turkey in June 2008. This is the problems of under-inclusiveness State pensions are often not targeted confirmed by the survey result which or over-inclusiveness which can dog at overcoming disadvantage; indeed, shows that 81 % of respondents felt classical positive action measures. those who have made more contribu- that the definition was broad enough Lastly, the requirement to make rea- tions through the taxation system may to cover their organisation’s activities. sonable accommodation is generally receive a higher pension than those The approval was strongest amongst mandated by law, as is the case in who are more economically vulner- the following groups: chief executive/ the Employment Equality Directive, able. An alternative example would managing director (CE/MD) (87 %) whilst employers and providers of be the provision of free education for and voluntary sector/NGOs (86 %). other social goods are generally left young people, which is preferential No other group by country or any with a choice of whether or not to treatment in the sense that the same other descriptor was significantly provide for positive action. opportunities are not available to below the overall response level. older persons, but it would be odd to 3.3.8. Distinguishing positive action construe the education system as an When asked if the definition would be from general policies to age-related form of positive action. easy to apply in their organisation, 66 % promote social inclusion of the respondents said that it would. Between these two extremes, there are The most positive sector was the volun- A final issue to consider is the dif- measures which fall on the bound- tary sector/NGOs at 77 %, in particular ficult boundary between positive ac- ary between social policy and positive those organisations in education and tion and general policies to promote action. For example, is State provision training, with a response of 86 %. Con- social inclusion. Some initiatives are of financial assistance to persons with versely, public sector respondents had easily categorised as positive action; disabilities or single parents to en- a significantly lower rate of agreement courage them to take up a low-paid that the definition would be easy to ap- 41 An exception to this is the notion of an- employment a form of positive action ply within their organisation at 48 %, ticipatory reasonable accommodation, as is (designed to compensate for disad- with college/university education sub- applied in the United Kingdom in the context vantages) or an element of the State category the lowest at 45 %. This was of access to goods and services, which is targeted at the group of people with disabilities social assistance regime? Given the rather surprising considering that pub- as a whole. This is not discussed further in this variety of such measures, a complete lic sector organisations are more likely annotation. inventory is impossible. Nevertheless, to be driven by legislative requirements

28 3. Exploring positive action from a legal perspective

and would therefore be expected to feel employers of more than 25 persons must of religion, sexual orientation and dis- more comfortable with the legal defini- ensure that they employ one disabled ability (49). Universities must prepare tion of positive action used in the study. person for every 25 employees (46). With annual plans reviewing the measures both of these laws, however, the possi- needed to promote equal rights for 3.4. The legal framework bility exists for the employer to make a students. An equivalent duty to pre- in European and non- payment as an alternative to complying pare annual plans exists in relation European countries with the quota. In Ireland, the Disability to educational authorities delivering Act 2005 requires public bodies to en- schooling, preschooling and school- This section of the report provides a sure that 3 % of their workforce is com- age childcare (50). This approach, which summary of the legal situation regard- posed of disabled persons ‘unless there focuses on organisational plans, is also ing positive action in the countries se- is a good reason to the contrary for not evident in the UK. Here public authori- lected for case study within the EU (42) doing so’ (47). ties are under a legal duty to promote and outside the EU (43). The legal equality on grounds of race, disability analysis in Hungary and Slovakia was Aside from disability, the other main and gender (51). This means that, inter limited to provisions for positive ac- example of a positive action scheme alia, many public authorities have to tion targeting Roma, in line with the whose terms are directly found within prepare ‘equality schemes’ which set objectives of the overall study. The in- statute was in Northern Ireland. In or- out their arrangements for promoting formation provided below is based on der to overcome the historical under- equality. In Hungary, there is less detail the answers given by national experts representation of Catholics in the po- within the legislation, but public bod- to a legal questionnaire. For more de- lice, the Police (Northern Ireland) Act ies and organisations with more than tailed information on the relevant na- 2000 created a quota scheme based on 50 employees must draw up an equal tional legal provisions, please consult the principle that one Catholic per- opportunities plan (52). the country case-study reports (44). son should be recruited for each non- Catholic appointed. The examples discussed above set Does legislation establish broad objectives for the organisations positive action measures? Does legislation establish concerned, but there are also instanc- obligations on public or es where more discrete measures are This question sought information on private sector organisations taken. In Northern Ireland, employers situations where legislation directly to take positive action? with more than 10 employees are un- established a positive action scheme, der an obligation to monitor the reli- as opposed to situations where legisla- Within the EU Member States under gious composition of their workforce. tion encouraged or required organisa- examination, there was a variety of If there is not ‘fair participation’ from tions to create their own positive action practice regarding obligations to take both the Catholic and Protestant com- schemes. Unsurprisingly, there were positive action. In general, no statu- munities, then there is a legal duty to relatively few examples of positive ac- tory obligations were identified in re- take ‘affirmative action’ (53). In France, tion schemes designed by statute. Those lation to Ireland, the Netherlands and employers must take into account age which most clearly fell into this category Slovakia. In contrast, several States had and disability when selecting employ- related to disability and the creation of enacted legislation which required or- ees for redundancy (54). Where the quotas for the employment of disabled ganisations to take steps to promote firm employs more than 50 persons, persons. Both France and Austria have equality. In Sweden, various statutes re- such quotas. In France, public authori- quired employers to take ‘goal-orient- 49 Equal Treatment of Students at Universities ties and private employers with more ed work’ in order to promote gender Act 2002. than 20 full-time workers must ensure equality and ethnic diversity (48). Similar 50 Act Prohibiting Discriminatory and Other that 6 % of the workforce consists of duties applied in relation to universi- Degrading Treatment of Children and Pupils disabled persons (45). In Austria, all ties, but these also covered the grounds 2006. 51 In Northern Ireland, the duty applies to a longer list of grounds, including religion, age, 42 Austria, France, Hungary, Ireland, Nether- 46 Act on the Employment of People with Dis- sexual orientation and persons with depend- lands, Slovakia, Sweden, UK. abilities (Behinderteneinstellungsgesetz) BGBl. ents (s. 75 Northern Ireland Act 1998). Nr. 22/1970, last amended by Federal Law Ga- 43 Canada, USA, South Africa. zette I Nr. 82/2005. 52 Article 63(4), Equal Treatment Act. 44 These will be available from the Pamecus 47 s. 47(4). 53 Article 55(2), Fair Employment and Treat- website: http://www.brad.ac.uk/acad/health/ ment (Northern Ireland) Order 1998, SI 3162 research/cid/pamecus.php 48 Equal Opportunities Act 1991; Law on (NI 21). Measures against Ethnic Discrimination in 45 Article L5212-2, Labour Code. Working Life 1999. 54 Article L1233-5, Labour Code.

29 International perspectives on positive action measures

extra conditions must be met before have more than 50 employees and legislation which were very similar to employees over the age of 50 can be are bidding for contracts worth more the approach found in the EU direc- made redundant. Disabled workers are than USD 50 000 (approximately EUR tives. In the case of Ireland, there were also exempted from the normal com- 38 750) (Hepple et al., 2000). Section additional provisions on positive ac- petition process for recruitment to the 202(1) states: ‘the contractor will take tion applying to areas outside employ- civil service, whilst special protections affirmative action to ensure that appli- ment. Section 14 of the Equal Status are in place regarding disabled work- cants are employed, and that employ- Acts 2000–04 permits ‘preferential ers who face dismissal. ees are treated during employment, treatment or the taking of positive without regard to their race, color, re- measures which are bona fide intend- In relation to Canada, the United ligion, sex or national origin’ (56). ed to: (i) promote equality of oppor- States and South Africa, there seems tunity for persons who are, in relation to be a more extensive use of legisla- In South Africa, the Employment to other persons, disadvantaged …’. tion to oblige organisations to take Equity Act 1998 applies to public au- A comparable exception permitting positive action (55). In Canada, the thorities and employers with more positive action in fields outside em- Federal Employment Equity Act 1995 than 50 employees. This creates a le- ployment can be found in the British aims to remedy past discrimination gal duty on employers to ensure that Race Relations Act 1976, where Sec- against women, persons with disabili- their workforce is representative of tion 35 authorises measures aimed at ties, Aboriginal peoples and members the South African population, with particular racial groups in order to of visible minorities. It applies to fed- specific reference to ethnic origin, meet ‘the special needs of persons of eral public authorities and federally gender and disability. Designated that group in regard to their educa- regulated employers with more than organisations are obliged to report tion, training or welfare …’. 100 employees. These organisations annually or biannually on the com- must draw up employment equity position of their workforce to the There was some divergence between plans. In addition, those who receive a Commission for Employment Equity. the Member States regarding the contract from the federal government They must also consult their work- grounds of discrimination in respect for more than CAD 200 000 (around force on the measures being taken to of which positive action was permit- EUR 132 000), and who employ more achieve employment equity. Section ted. In relation to employment, Aus- than 100 employees, must sign a com- 15(1) refers to the goal of equitable tria, Ireland and the UK had provi- mitment to implement employment representation in all occupational sions permitting positive action on equity in line with the act. In Canada, categories and levels of the organi- all grounds. In the case of disability there are also significant examples sation. The national expert reported discrimination law in the UK, this of obligations to take positive action that this is frequently interpreted as was implicit; the Disability Dis- within provincial legislation. These implying quotas, although this is not crimination Act 1995 only prohibits are not mentioned here for reasons of expressly provided for within the leg- discrimination against disabled per- space, but more detail can be found in islation. In addition, the Broad Based sons. Consequently, positive action the Canadian case-study report. Black Economic Empowerment Act for disabled persons could not be 2003 allows for quotas to be adopted challenged as discrimination against A similar pattern of imposing obliga- in specific sectors via transformation non-disabled persons. The same le- tions on federal contractors can be charters and codes of practice. gal approach regarding disability was found in the United States. In 1961, also reported in Sweden. Executive Order 10925 introduced a What forms of positive duty on federal contractors to ‘take af- action are permitted, but not In several countries, there was a ten- firmative action to ensure that appli- required, by legislation? dency to permit positive action in cants are employed, and that employ- relation to socio-economic disadvan- ees are treated during employment, There was considerable variation tage, with less emphasis on discrimi- without regard to their race, creed, between the EU Member States in nation grounds. In the Netherlands, color or national origin’. Executive Or- their approach to permitting posi- legislation only expressly permitted der 11246, introduced in 1964, obliges tive action. Austria and Ireland had positive action in relation to gender, federal contractors to have a written incorporated texts into their national race and disability (57). In France, affirmative action programme if they 56 The full text of Executive Order 11246 57 Article 2, Section 3, General Equal Treat- 55 ‘Positive action’ is not the terminology used is available at: http://www.dol.gov/esa/ ment Act (Algemene Wet Gelijke Behandeling); in these States, but it will be adopted here in ofccp/regs/statutes/eo11246.htm, accessed Article 5(1), Equal Treatment Act Women/ order for consistency throughout the report. 28.11.2008. Men and Article 7:646(4), Civil Code.

30 3. Exploring positive action from a legal perspective

there is no general statutory provision Are any forms of positive litigation. In 2006, the Supreme Court on positive action by reference to dis- action prohibited by held that a programme by Uppsala Law crimination grounds, although a wide legislation? School, which awarded 10 % of places range of measures tackling socio-eco- to students with a foreign background, nomic disadvantage were reported. In some EU Member States, the limits constituted unlawful discrimination on By focusing measures on disadvan- to positive action are expressly found grounds of ethnic origin (64). As is well taged neighbourhoods, such positive within legislation. In Hungary, the known, there has been extensive litigation action schemes indirectly assist per- Equal Treatment Act provides protec- in the United States on the legality of pos- sons of migrant origin. In Slovakia, tion for measures which it calls ‘pref- itive action and it is impossible to present legislation passed in 2008 permits erential treatment’, but only if these are in detail this rich body of case-law within public authorities to take ‘affirmative based on an act, government decree or the confines of this report. In brief, the measures’ where these are focused a collective agreement (60). Moreover, US Supreme Court has held that any ra- on socio-economic disadvantage and such measures must be time-limited or cial classification must be subject to ‘strict disadvantages linked to age and dis- limited by reference to when a particu- scrutiny’, in other words, the programme ability (58). In Hungary, it was also re- lar condition is achieved. Similarly, in must pursue a compelling interest and it ported that positive action was mostly Slovakia, the 2008 legislation provides must be narrowly tailored (65). Consider- by reference to socially disadvantaged a list of statutory restrictions determin- ing some examples from recent case-law, groups, rather than explicitly identifying when positive action can lawfully in Grutter v Bollinger the Supreme Court ing Roma as the target group. be taken, which includes the statement accepted that race could be taken into ac- that positive action can only be taken count in law-school admissions in an ef- In relation to the non-EU countries by public authorities and only if ‘there is fort to combat the under-representation under consideration, the main fo- existing provable inequality’ (61). In the of ethnic minorities (66). The Supreme cus of their efforts was the statutory Netherlands, there is no statutory pro- Court was, though, unwilling to extend schemes for positive action discussed tection for positive action on grounds this to a scheme where race was used in in the previous section. In Canada, it of sexual orientation, religion or age (62). the allocation of children to particular was noted that most provincial human It should be noted though that policies high schools (67). rights statutes permitted special or af- to promote employment within certain firmative programmes to improve the age categories will not be considered as In Canada, it was also reported that situation of disadvantaged persons. unlawful discrimination if such policies case-law requires a rational connec- In South Africa, organisations not are laid down by or pursuant to the Act tion between the positive action pro- formally bound by the duties within on Equal Treatment on the Ground of gramme and its ameliorative purpose. the Employment Equity Act could, Age in Employment (63). As mentioned In South Africa, a recent decision nevertheless, volunteer to accept its above, there is no express provision for from the Supreme Court has clarified obligations and pursue positive action positive action in French anti-discrim- that positive action cannot be used as under this framework. In the United ination legislation and it is clear that a justification for otherwise arbitrary States, Section 706(g)(1) of Title VII positive action which is overtly based on recruitment decisions (68). of the Civil Rights Act 1964 allowed ethnic origin would be unlawful. courts to impose positive action as a 64 The State v Lönn and Midander, Case remedy where unlawful discrimina- In Sweden, the limits to positive ac- T_400/06, 21 December 2006. See further: tion had occurred: ‘if the court finds tion have recently been the subject of Numhauser-Henning, A. (2007), ‘Report on that the respondent has intentionally measures to combat discrimination: Directives 2000/43/EC and 2000/78/EC, Country report: engaged in or is intentionally engaging 60 Article 11(1). Sweden’, p. 16, available at: http://ec.europa.eu/ in an unlawful employment practice 61 Article 8(a), Law Amendment 85/2008 to employment_social/fundamental_rights/pdf/ charged in the complaint, the court Law 365/2004 Act on Equal Treatment. legnet/svrep07_en.pdf, accessed 28.11.2008. may enjoin the respondent from en- 62 The absence of statutory protection means 65 Fredman, S. (2002), Discrimination law, gaging in such unlawful employment that measures providing benefits based on Oxford: Oxford University Press, p. 147. practice, and order such affirmative sexual orientation, religion or age are likely 66 Grutter v Bollinger 539 USA; 306, 123 S Ct action as may be appropriate’ (59). to be regarded as discrimination (against the 2325 (2003). groups that do not benefit) in the areas falling within the scope of non-discrimination laws. 67 Parents involved in community schools v 58 Article 8(a), Law Amendment 85/2008 to Conversely, where there is no applicable prohi- Seattle School District No 1 et al., 127 S Ct 2738 Law 365/2004 Act on Equal Treatment. bition of discrimination, such as age discrimi- (2007). nation in access to goods and services, positive 59 See further, Connolly, M. (2006), Discrimi- action is, by default, permitted. 68 Supreme Court of Appeal (17 September nation law, London: Thomson Sweet & Max- 2008) Martin Gordon v Department of Health well, p. 365. 63 Article 7, Section 1a. KwaZulu-Natal (337/2007) [2008] ZASCA 99.

31 International perspectives on positive action measures

3.5. Summary

The predominant feature of positive action within EC legislation is its permissive nature. There do not appear to be any obligations to take positive action and consequently it is unsurprising to find that national practice varies greatly in this area. This diversity relates both to the grounds where positive action is permitted or required, as well as the extent to which national legislation imposes duties on organisations to engage in positive action.

The main restraint imposed by EC law has been in relation to gender equality, where the Court of Justice has not permitted measures which confer automatic and unconditional priority on the under-represented sex. In the absence of any other case-law, it is difficult to foresee whether the Court will adopt an identical line of reasoning in relation to positive action for other discrimination grounds, or positive action in areas outside the labour market (such as education). Directive 2000/78/EC indicates that more far-reaching measures are permitted in relation to disability. This corresponds with national practice; for the most part, disability is the only ground where Member States have taken measures such as quotas. The three non-EU case studies illustrated a greater willingness to oblige public and private sector organisations to promote equality and to take positive action. Significantly, there is evidence of comparable instruments emerging within the EU, most notably in Sweden and the UK.

32 Perceptions of positive action

4 in the European Union

33 International perspectives on positive action measures

4. Perceptions of positive action in the European Union This chapter presents a detailed analysis of the perceptions of positive action held in the Euro- pean Union, in particular by study participants from the eight European countries that participated in the in-depth study. The analysis draws heavily upon data collected from the consensus workshop and interviews held in each country, in addition to the findings from the survey and analysis of the legal framework and relevant policy documentation relating to positive action in each country. However in France, the findings were drawn from secondary analysis of a French country report on measures to combat discrimination (69). The chapter begins by comparing understanding of positive action as practised in organisations within the European Union, particularly in the eight case-study countries, including the type of measures that individuals considered constituting positive action. Drivers for positive action are then discussed as well as factors which are considered to contribute to the effectiveness of positive action measures. As part of this discussion, we assess how organisations measure the impact of positive action projects and identify groups that appear to have benefited the most and least from targeted initiatives. Finally the chapter identifies the main barriers that have hindered progress in implementing positive action initiatives and outlines plans regarding positive action measures for the future.

4.1. Understanding for disadvantage. Participants in these specific groups, with the aim of creat- of positive action(69) countries made associations between ing a more equitable society. For the positive action and the removal of Dutch participants, this included leg- There was no consistent understand- barriers, social justice, global justice, islation, training, organisational de- ing of ‘positive action’ amongst mem- advocacy and empowerment. Nev- velopment and coaching. There was bers of the European countries taking ertheless, despite providing us with a common agreement that positive ac- part in the study. Participants repre- clear definition for this term, not all tion measures had to be implement- senting countries in Europe displayed participants considered positive action ed as part of a multi-angled strategy differing levels of familiarity with the to adequately capture or reflect their in order to be successful and effective term and varying levels of reluctance understanding. In the UK, positive in the long run. If not supported by to use it to describe their activities. For action was described as a ‘conceptual the whole society, any measurements countries such as Hungary, Ireland mess’ and suggestions were forwarded aiming at more equality were consid- and the United Kingdom, positive ac- to replace it with the term ‘balancing ered useless. According to the partic- tion was understood as constituting measures’ as a definition that would ipants, in order to make a difference, specific measures to redress past dis- make it more easily understood and the application of positive action crimination directed towards a partic- acceptable to the general public. measures needs a critical mass. They ular group and with the aim of equal- considered positive action measures ising the position of that group with In other European countries, the had to be part of a coherent perma- that of the majority society. There was term positive action was seen in a nent procedure following a target- a general consensus in these countries different light. In the Netherlands, oriented strategy open to change. that positive action measures should participants considered positive ac- This was in line with their argument equalise social inequalities, eliminate tion measures to be an old-fashioned that positive action measures should disadvantage and even compensate strategy and struggled with the con- never be implemented within one de- cept. They preferred to see positive partment or one hierarchy level only, 69 See Latraverse, S. (2007), Report on mea- action as an important tool within a but that they should be implemented sures to combat discrimination: Directives wider diversity management strategy in all business areas. 2000/43/EC and 2000/78/EC. Country report which included all methods designed France, available at: http://europa.eu.int/comm/ employment_social/fundamental_rights/policy/ to counteract the effects of exclusion, Similarly in Sweden, positive action aneval/mon_en.htm discrimination and stereotyping of was not a commonly used term and

34 4. Perceptions of positive action in the Europe Union

nor was it found in the relevant leg- equality for all within French society sectors. Finally, in the face of de facto islation, which speaks about ‘active and the understanding that, in law, no inaccessibility of higher specialised measures’. Confusion arose in relation section of the French population can elite schools to children of disadvan- to the concept as Swedish legislation claim rights as a minority seems to taged neighbourhoods, one of these and policy stress the need to take ac- have worked against policies targeting schools — the Institut d’Etudes Poli- tive measures, yet on the other hand specific ethnic groups designated in tiques —has entered into partnership impose a number of bans and restric- term of their origin (see legal analy- with schools in these neighbourhoods tions, including a restrictive case-law. sis of France). Relevant categories for in order to select their best students Most of the people interviewed in positive action are accepted only to the and create a parallel recruiting sys- Sweden were unable to provide a defi- extent that they rely on neutral criteria tem. Since its inception, this scheme nition of positive action because of its devoid of identity content, such as so- has been replicated elsewhere. perceived complexity. cio-economic considerations. Conse- quently there are only policies target- Amongst the European countries, Slo- In Austria, workshop and interview ing the Roma population — who are vakia showed the least understand- participants placed greater emphasis not designated by their ethnic group ing of the concept of positive action upon the benefits of positive action — aimed at accommodating their way and issues around equality and equal measures. Participants stressed that of life in terms of housing and educa- opportunities were also discussed in positive action measures were a gain tion. Positive action for disabled peo- very abstract terms. Most workshop for everybody and should therefore be ple is permitted, but at present similar participants and interviewees did not taken into account in all sectors of soci- action cannot be undertaken for les- possess an in-depth understanding of ety and for all groups needing it. There bian, gay, bisexual and transgender positive action, in that almost any ini- was opposition expressed to the pros- (LGBT) individuals. tiative could be understood or labelled pect of positive action simply provid- as such, regardless of its goals. Some ing opportunities for preferential treat- In France, a number of initiatives participants understood positive ac- ment for one or two groups. Rather the could be seen as constituting position as any initiative helping Roma, approach preferred was to adopt an at- tive action, although they may not and identified initiatives such as - tar titude of openness to who needs which have been packaged using this termi- geted training, recruitment, scholar- kind of empowerment and at what nology. For example, in 2005, there ships, housing projects and language time. For this process to be effective, were plans to introduce large-scale support. The participants admitted creating awareness about difference vocational training and support to that such measures could address all and discrimination in its structural di- integrate unemployed people into the Roma as an ethnic group, but when mension was considered essential. workforce, creating regional centres talking about concrete examples, the of employment and focusing many discussion tended to focus on meas- In some respects this resonated with programmes and services to the ben- ures which in other contexts would attitudes held towards positive action efit of the long-term unemployed. In be considered tools for overcoming in France. As was highlighted in Chap- addition, under the umbrella of social social exclusion. On the other hand, ter 2, positive action was not widely cohesion, attempts were made to tar- the participants realised that not only employed or commonly understood get unemployed, disadvantaged youth poor or socially excluded Roma face in France as a measure for tackling and long-term elderly unemployed discrimination and that greater atten- inequality. The lack of familiarity with through apprenticeship and vocational tion should be paid to other groups of the concept, we believe, explained why training in the private and public sec- Roma as well. Slovak organisations felt we were unable to attract sufficient in- tor. Plans were also made for massive that the lack of legal provisions on pos- terest from individuals working in the investment in public housing in order itive action in the past had contributed French public, private and third sector to benefit disadvantaged neighbour- to this lack of common understanding to attend a consensus workshop. Con- hoods and provide emergency hous- of the term. They also indicated a need sequently, in the absence of data col- ing. Similarly, in terms of education, for the government to clarify some lim- lated from individuals about positive there were plans to concentrate on itations in the current legislation and action in France, we compared data children in serious difficulty in order raise awareness of positive action. The collated from consensus workshops to improve equality of opportunity by fact that it was only introduced in law and interviews held in other European offering specific support to identified in 2008 may explain why Slovakia was countries with our analysis of positive children in disadvantaged sectors and one of the countries with the lowest action in France as drawn from exist- the allotment of specific budgets for number of responses in the survey. It is ing literature. As such, the principle of schools in designated geographical also important that, for both Hungary

35 International perspectives on positive action measures

and Slovakia, most programmes which tion. But this level of confusion was In Austria, equality legislation was may be considered positive action for particularly high in Austria (48 %), classed as a positive measure per se, Roma are linked to social inclusion Greece (42 %), Hungary (33 %), Bul- owing to its underlying aim of achiev- policy and general strategies to address garia (32 %) and the combined eight ing equality of opportunities, remedy- this group’s situation. EU case-study countries (23 %). ing deficits and recognising everyone’s rights. In contrast to these countries, In some of these countries, a negative legislation was not considered to have consequence of introducing positive 4.2. Drivers for positive played an important role in promot- action within organisations was back- action ing positive action in countries such lash from individuals who did not as the Netherlands, where organisa- fully understand the rationale behind For European countries, a number of tions did not agree with moves to- positive action and saw it as ‘favourit- factors served as an impetus to imple- wards introducing positive measures ism’ towards particular groups. In the ment positive action, some of which and openly disobeyed laws to collect Netherlands, it was felt that positive were common to several countries. data on the ethnic background of their action was very much connected with The survey showed that legislation employees with no sanctions imposed quotas and other forms of preferential (47 %) featured as the most impor- by the government (see legal analysis treatment. As such, preferential treat- tant driver. In countries such as the for the Netherlands). ment aiming to increase the represen- UK and Ireland, the legislative pro- tation of specific groups attracted a vision for positive action had been a To a certain extent, participants in mixed response. The importance of the powerful incentive to make changes, the Netherlands believed that, rather quota system in making women more particularly in relation to disability in than legal pressures, the desire to help visible in public services was consid- Ireland, where the law guarantees the others was a more compelling ration- ered a positive outcome of preferen- use of quotas. The legislative frame- ale for implementing positive action tial treatment but, at the same time, it work in these countries placed spe- measures. This was mentioned in line was recognised that there were preju- cific duties on employers and service with the ambition to attain a good im- dices towards people who were sus- providers to be proactive in ensuring age and avoid social shame. This was pected of having been employed ac- equal treatment to everyone, regard- considered to be relevant for public cording to a quota system leading to less of difference. In Ireland, nine bodies and government departments a certain reluctance to come forward grounds (gender, marital status, fam- to a higher extent than for private amongst people who would be em- ily status, sexual orientation, religion, companies, as they need to represent ployed preferentially within this sys- age, disability, race and membership the diversity in society in the organi- tem. Some Swedish participants also of the Traveller community) are cov- sation to be legitimised, considering it linked their understanding of posi- ered by this legislation, whilst six part of their public duty to act as role tive action to quotas or preferential are covered in the UK (age, disabil- models. Recent initiatives in the ju- treatment as a way of reducing dis- ity gender, race, religion and sexual diciary to raise the number of judges advantage. Many of the interviewees orientation). The survey also showed with a migrant background was moti- thought that positive action was of- that, amongst the EU countries, the vated by the awareness that allochtoon ten seen as working against the prin- UK (84 %) and Ireland (83 %) had the (minorities) were not represented in ciple of equality and the need to ‘treat significantly highest proportion of the judiciary and that this factor could everyone in the same way’ was raised respondents with a written equality lead to failures in perception, lack of frequently by its opponents. It was and diversity policy. Analysis of the knowledge about structural discrimi- pointed out that, in Slovakia, there documents provided by organisations nation and to prejudice influencing was a failure to distinguish between based in both of these countries, such attitudes towards clients and even positive action and positive discrimi- as mission statements and annual re- court rulings. nation, with some participants ad- ports, also reflected a commitment mitting to the legitimacy of quotas, to equality beyond rhetoric. At the As in the Netherlands, the moral case even though serious doubts were lowest end, we had Slovakia (29 %), for positive action was also a major im- expressed about their effectiveness. Greece (25 %), Bulgaria (24 %) and petus for the implementation of posi- This accords with findings from the Poland (13 %) with equality and di- tive action in Hungary, particularly in survey which show significant confu- versity policies. Overall, 59 % of the improving the disadvantaged situation sion about positive action. Overall, eight EU countries involved in the of the Roma community. Similarly, in 19 % of respondents confused posi- comparative study had a written pol- Slovakia, it was apparent that projects tive action with positive discrimina- icy on equality and diversity. targeting Roma were not designed

36 4. Perceptions of positive action in the Europe Union

under any legal provision or justified dressed Roma in general, while some ment their equality policies in order from a human rights perspective, but addressed specific concerns of Roma to achieve real practical outcomes. were rather based on an understanding women and many educational and Considerable stress was placed upon of the poor social conditions which ex- employment training programmes the need to implement positive action ist within Roma communities. targeted young Roma. Demand for in order to ‘practice what we preach’ an increasing labour force was com- and ‘ensure public confidence in the In addition to legislative require- pelling enterprises to be much more service’. Further, organisations had es- ments, the prospect of improving open to employing Roma, as was the tablished diversity policies and a strong organisational business performance availability of funding for Roma pro- corporate image promoting fairness was seen as a major driving force for grammes that motivated some organi- and equality, which the use of positive the implementation of positive action sations to start working with them. action measures helped to reinforce in measures. In response to changes in It is notable from the survey that different ways. Within the UK health the population, organisations were many of the organisations based in sector, positive action strategies were increasingly using such measures to Bulgaria (84 %) and Hungary (75 %) seen as a useful way to build up greater create workplaces that represented had set themselves specific targets trust in local communities through local communities through targeted for service use/customer profile and targeting recruitment strategies and recruitment and retention practices. service delivery in relation to disad- employment accreditation around is- In countries such as Austria, Ireland, vantaged groups. However, the sig- sues relating to disability (‘two ticks’70), the Netherlands and Sweden, partici- nificantly lowest responses for setting lesbian, gay, bisexual and transgender pants talked about positive action in specific targets were recorded in the people (‘Stonewall Diversity Champi- the context of good business practice UK (45 %) and Ireland (33 %). This ons’71), age (‘Age Positive’72) and people and saw promoting equality through is interesting considering these coun- with mental health problems (‘Mind- positive action as giving businesses a tries’ long histories of equality legis- ful Employer’). In Sweden, within the competitive advantage. The relevance lation, particularly in the UK with its higher education sector, increased di- of public image and the public rela- positive duty requirements. In terms versity was seen to ‘improve the quality tions aspect of positive action meas- of equality and diversity targets in of education’ and due to the require- ures were also mentioned. For Sweden service delivery, the highest signifi- ments for policy documents on action and the Netherlands, there was greater cant proportions in the EC countries measures set by Swedish law, most in- emphasis upon the business case in were found in organisations based in terviewees spoke about the wealth of relation to the private sector organisa- Bulgaria (76 %), Hungary (75 %) and policy evidence in support of diversity tions, where it was felt to be no longer Italy (66 %) and with the lowest being initiatives. Within Ireland, for some acceptable to present a ‘white male Germany at 24 %. private sector businesses, commitment team’ or to perpetuate a workforce to their equal opportunities policy was made up of ‘blue eyed, blond haired Participants in Slovakia and Hungary part of their larger institutional stra- Swedes’. Employing a multi-ethnic talked about the need to combat un- tegic goal. In Austria, where acknowl- workforce was also seen as a way of employment, low education attain- edgement of the rising relevance of so- capturing new markets by attracting a ment levels and the overall substand- cial responsibility was considered to be diverse clientele. In contrast, countries ard housing conditions prevailing high, positive action was used as a tool such as the UK discussed the need for in many Roma communities. Again, by management for raising employee public sector organisations to become positive action in relation to Roma satisfaction and efficiency, leading to more representative in order to better involved a similarly strange combi- benefits and opportunities for both meet the needs of existing service us- nation of motives (including per- management and employees. ers, particularly in the health sector. ceived budgetary and other threats In addition, businesses were driven due to demographic development) Organisations in most countries to comply by the fear of litigation and meaning that due to higher birth talked about the close alignment of compensation claims. rates amongst Roma families, social positive action with their organisa- inclusion costs will continue to grow tional mission and how it enabled In Hungary and Slovakia, positive ac- unless real change happens (the same tion measures for Roma were less clear high cost of failure to take action). 70 http://www.direct.gov.uk/en/DisabledPeo- in terms of their incentives and tended ple/Employmentsupport/LookingForWork/ DG_4000314 to be characterised by mixed motiva- Participants from the different EU tions. Most positive action measures countries also explained how posi- 71 http://www.stonewall.org.uk/workplace/ by private and civil organisations ad- tive action enabled them to imple- 72 www.agepositive.gov.uk

37 International perspectives on positive action measures

them to fulfil their objectives more For participants from most coun- critical reflection and realistic out- effectively. For example, in Ireland, tries, the availability of resources, come measures. Sweden in particu- a number of projects had been set up including financial resources, were lar emphasised the importance of to meet the specific needs of Travel- considered to be very important in producing practical plans, the need ler communities. In Sweden, compa- putting positive action measures into for setting certain goals, as well as nies spoke about equality and diver- operation. Participants from Aus- continual education and training in sity as part of their mission and saw tria and the Netherlands were most this field. clear connections between positive vocal in identifying a range of suc- action and their specific work du- cess factors and the circumstances Both in Slovakia and Ireland, it was ties, for example the police service. in which positive action measures felt that the effectiveness of posi- In Hungary, participants also made would work most effectively. For tive action measures for Roma and links between positive action meas- them, success would require a posi- Traveller communities respectively ures, the existence of their organi- tive attitude from those leading the was dependent upon a range of fac- sation and more widely to govern- initiative, as well as the development tors which had hitherto been inad- mental policy on Roma integration of tailor-made strategies that would equately addressed. Although the in Hungary. vary depending on the target group, nature of the projects were not iden- the sector and the organisational tical, feedback from projects target- Having a sense of ownership of a culture. Reviewing changes of the ing Roma and Traveller communi- project was also a strong factor that initial situation in relation to the for- ties in both countries highlighted determined its success. In Ireland, mulated goal was essential. Empow- issues which need to be given more commitment or buy-in from sen- erment of staff members and a strong attention in planning such projects. ior management was perceived as commitment of senior managers and These can be summarised as the an added incentive for positive ac- other leaders within the organisa- participation of the communities tion to succeed. Similarly, in Sweden tion were key in driving initiatives. in the development of positive ac- senior-level management support Alongside this, there would need tion initiatives and motivation to was felt to be crucial and involve- to be awareness of the relevance of improve their position as necessary ment at the outset from the target discrimination and structural bar- for positive action measures. This group itself was equally important. riers for certain groups in society resonates with the general negative It was only in the UK, however, that incorporated within the organisa- attitude towards displayed towards we found evidence of community tion, which would have to include a members of the Roma community groups and grassroots organisations clear commitment and awareness at in Hungary. Similarly, the develop- who, fuelled by dissatisfaction with all hierarchy levels, and explicitly at ment of positive action initiatives the existing status quo, had lobbied the level of top management, an in- posed a series of moral questions in for positive action measures and volvement of all departments of an Hungary, where Roma projects were succeeded in setting up specialised organisation in the development and being developed by organisations services to meet their own needs. implementation of these measure- who were attracted by the availabil- ments. To be effective, there would ity of funding for positive action but 4.3. Support for positive need to be constant discussion about did not necessarily understand the action the benefits of diversity which must needs of Roma communities. be an ongoing practice. Otherwise Despite some confusion and consider- the risk of losing the basic require- The survey shows that initiatives able variation in people’s understand- ment of overall commitment was such as creating networks and fo- ing of positive action, the concept considered high. rums, training and leadership pro- received overwhelming support from grammes and confidence/commu- participants in all the European coun- For countries such as Sweden, the nication-building programmes were tries taking part in the study. Never- Netherlands and the UK, the impor- most likely to be monitored, par- theless, this was not unconditional tance of evaluation for measuring ticularly with regard to gender, dis- support; it was felt that positive ac- effectiveness was stressed, as well as ability and ethnicity. The European tion would only be effective in certain the need to adopt a learning attitude. country with the significantly high- conditions and there appeared to be Whatever the scale of the initiative, est response above the overall 72 % a considerable amount of consensus it was felt that positive action could responses describing the measures between countries as to the nature of not be undertaken in an ad hoc man- instituted as positive action was Aus- these success factors. ner but required careful planning, tria at 89 %.

38 4. Perceptions of positive action in the Europe Union

4.4. Outcomes and impacts gued that implementing strategies to to be the most important and most change the recruitment rates within desirable impact of positive action for Despite strong rhetoric about the im- a company would have been and is a companies and organisations in the portance of having evidence-based time-consuming procedure, so suc- Netherlands. Changes in recruitment strategies, it was quite clear that most cess could have been assessed only procedures have been crucial for re- of the positive action projects based after a longer period. ducing barriers and for creating more in European countries did not have equal access to employment. Cultur- systematic monitoring systems and In addition, the practice of positive ally biased tests, traditions to hire peo- output measures in place. Only over discrimination in the 1990s, i.e. em- ple via specific students associations, a third of the respondents surveyed ploying people because of their cultur- language barriers or simply prejudice from the eight case-study European al background with the aim of better had been influencing the engage- countries stated that they utilised ex- understanding the needs and interests ment of personnel. Analysis of these ternal assessment, and almost half of of that group, had not proved benefi- procedures, followed by the develop- them relied on anecdotal evidence for cial. The efforts of Dutch police forces ment and implementation of adapted assessing the results/impact/effective- over two decades to recruit members procedures had changed the scene ness of their positive action. Despite of ethnic minorities led to changes in completely and had led to a more het- this, participants were able to provide the ethnic composition of the work- erogeneous workforce. The creation of some sort of evaluation of the success force but also to the creation of the networks of certain groups targeted by of projects that had been implement- Moroccan, the Turkish and the An- positive action measures have become ed and the kind of improvements they tilles brigade, responsible for solving independent from their founding had witnessed. conflicts within their own communi- objectives and have developed their ties. Unfortunately, this did little to own agendas, representing company In Austria, it was felt that positive ac- change the culture of the Dutch police identity in diversity. tion had heralded a modified approach service, which had been very much to women and disabled people, with shaped by white male Dutch citizens Mirroring findings from the workshops specific reference made to a growth and served to perpetuate stereotypes and interviews, the survey (Figure 1) of equality policies for these groups. and barriers within the organisation. also showed that gender, ethnicity and More generally, societal awareness The subsequent high level of attrition disability were the most likely grounds and sensitivity to cultural differences led to the conclusion that a better way to be monitored. The survey also indi- appeared to have heightened in a posi- to promote the attractiveness of an cated that age was being monitored by tive sense. Furthermore, a change of employer for all was to promote the organisations to a lesser extent but there attitudes within organisations was ob- organisation as one that celebrated di- was little evidence of activities around served, concrete actions had led to a versity and equality. age in the workshop discussions and greater understanding of the needs of interviews. employees, interaction between em- Since then, the capacity to find quali- ployees and channels of communica- fied people from groups that previous- In the UK, a number of groups were tion within the organisations. ly were not attracted was considered considered to have benefited from

In the Netherlands, reflections on the FIGURE 1: DISCRIMINATION GROUNDS MEASURED IN DIFFERENT period of ethnic monitoring suggested MONITORING CATEGORIES that this did not seem to have been ef- Equality grounds measured In which categories fective in improving the representa- Gender Current employees tion of minority ethnic people in the workplace. However, it was admitted Age Applications that some benefits had arisen in that Disability Recruitment the obligation to collect data on the ethnic background of employees had Racial or ethnic origin Training contributed to awareness-raising on Religion or belief Progression & promotion diverse backgrounds and on the reality of unequal representation of peo- Sexual orientation Service user/customer pro le ple with different ethnic backgrounds Other Retention in companies, in the labour market 01020304050 010203040506070 and in society. Furthermore, it was ar-

39 International perspectives on positive action measures

positive action, namely disabled peo- tiatives related to employment were been raised with the Roma popula- ple, black and minority ethnic groups generally perceived to have had a tion in Sweden, especially in relation and women (particularly South Asian beneficial impact in terms of improv- to employment and education. women). The prescriptive nature of ing peoples’ chances of finding a job the statutory legislation in relation to (albeit at lower levels) and within or- In terms of improvements, it was felt disability, coupled with widespread ganisations, helping to increase their that positive action measures had acceptance within society of the dis- promotion prospects, establishing brought about higher levels of sen- advantage faced by disabled people, staff networks and creating opportu- sitivity to the relevance of difference were felt to be significant factors in nities for mentoring. Similarly, in re- and its implications for inequalities, improving their position. By the lation to service provision, changes to a change of attitudes and a redefini- same token, it was thought that so- patients’ lifestyles, well-being and action of roles within organisations. In ciety would be less likely to challenge cess to healthcare services were con- addition, there was increased repre- positive action set up for this group. sidered to provide sufficient evidence sentation of specific groups, lowering Other groups identified as benefiting for the success of patient education of communication barriers and high from positive action were women and through outreach work with black levels of internal promotion and op- black and minority ethnic groups, and minority ethnic patients (includ- portunities for training. Of all the Eu- who were thought to be more vis- ing recent immigrant communities ropean countries, Sweden appeared ible in non-traditional areas of work. from East Africa) in the Midlands. to undertake the most established Targeting of these groups was driven These projects tended to be short- monitoring system for positive action, mainly by local community needs, term and health-related, targeting resulting in detailed evaluations for government grants geared toward specific black and minority ethnic specific positive action projects. Other under-represented groups within or- groups. In addition, the possibility types of evaluation included Internet- ganisations or internal audit within of using positive action to introduce based evaluation forms that had to be organisations. For representatives alternative psychological therapeutic filled by all employees affected, and from the voluntary sector, neverthe- approaches to treat minority ethnic their results were compiled within the less, the worry remained that positive groups was also being considered as information database. Only a few ini- action schemes signalled no long- a positive spin-off. tiatives did not include an evaluation term commitment towards achieving component. equality. Few individuals felt that the The overall response to positive ac- position of lesbian, gay and bisexuals tion in Ireland was favourable but at As noted earlier, positive action pro- had improved greatly through posi- the same time cautious in light of the grammes in Hungary and Slovakia tive action programmes and trans- absence of evaluation tools and the have targeted Roma communities in gender individuals were thought to fact that many projects were in the both countries and it is possible to have benefited the least. early stages of implementation. The draw conclusions on similarities be- main beneficiaries of positive action tween them. In both Hungary and On the whole, individuals were not in Ireland were identified as women, Slovakia, there was significant focus able to provide a lot of information the Traveller community, disabled during discussions on data protec- about the monitoring systems they people, minority ethnic groups and tion issues, and their impact on the used in relation to positive action. It the long-term unemployed. ability to undertake positive action was apparent that not all organisa- on behalf of Roma. In both Hungary tions felt confident about discussing In Sweden, as in the UK, disabled and Slovakia, the collection of ethnic the impact of their initiatives, owing people, minority ethnic groups and data is only allowed according to strict to a lack of clarity about what their women were considered to be the legal provisions, which are often per- expectations had been from the outset main beneficiaries of positive action ceived as making it illegal (this is not in relation to their respective project through initiatives such as employ- the case, but the conditions are strict). outcomes. Within the public sector, ment training, targeted recruitment Problems are created because positive there was greater appreciation of the and mentoring schemes. Less work action measures specifically targeting need to provide evidence for the ef- had been conducted with lesbian, gay, Roma could be perceived as illegal, fectiveness of positive action initia- bisexual and transgender individuals and government policy reflects this, tives; without this evidence, there was although recognition of these groups with policies on Roma inclusion often a danger that organisations would lose was improving through the establish- referring to specific actions aimed at momentum to continue using this ap- ment of networks and forums in the ‘disadvantaged groups’ rather than at proach. Hence, positive action ini- public sector. Concerns have also Roma as an ethnic group. This dis-

40 4. Perceptions of positive action in the Europe Union

counts the historical discrimination the private sector appeared to be less The majority of countries did not dis- which has created the need for posi- prevalent. cuss future plans for positive action, tive action measures. In Slovakia, even which might in part be explained by positive action provisions in the Anti- In both Hungary and Slovakia, there the fact that within the workshops we Discrimination Act refer to positive was significant discussion on the neg- did not ask them explicitly about their action allowed on the basis of social ative reaction by non-Roma to these intentions to continue or develop. disadvantage. Data protection con- programmes, which are perceived Findings from the interviews, howev- cerns make it difficult to officially to benefit Roma only (as opposed er, suggested that organisations were monitor and evaluate the impact of to wider society). This reflects both keen to learn about best practices and positive action measures for Roma. widespread negative and prejudiced to receive suggestions on evaluation attitudes towards Roma in both coun- of positive action measures initiatives Positive action measures targeting tries, but also a lack of understand- already in existence. Notably, in Swe- Roma in Hungary and Slovakia can be ing of positive action in general. The den, some organisations talked about clearly and overwhelmingly linked to negativity can be linked to severe de- enlarging the scale of their activities social policy on Roma inclusion, and ficiencies in information campaigns and expanding them to other fields only very recently to law-enabling and awareness-raising on positive ac- and other disadvantaged groups. positive action. Most programmes tion in both countries (which is also a Some Swedish universities were keen are housed within the government quite new concept in both countries). to include social class and sexual ori- strategy on the Decade of Roma In- entation as part of their positive ac- clusion. The early national strategies In both Hungary and Slovakia, it can be tion work but were prevented from for improving the situation of Roma seen that while complex programmes doing so because of resource limita- and their financing are also linked to are needed to make adequate im- tions. A number of institutions also this government strategy. Thus, gov- provements in the situation of Roma, mentioned that, as of 2009, there ernment policy can be seen to drive this is not what is implemented in re- was a need to include age as an area positive action affecting Roma in ality. Isolated programmes exist which of concern, because of the new anti- both Hungary and Slovakia, while at may bring improvement in one area, discrimination law, yet admitted not the same time strangely limiting it. but which in the end bring negative to knowing how to deal with it. results in other areas (see case study The linking of positive action for Roma for housing schemes in the country In the survey, participants from the to government policy also clearly report for Hungary) because the full European country with the signifi- leads to discontinuity in programmes, consequences of the programme were cantly highest response above the which end before the overall goals not thought through adequately. overall 55 % of the study sample plan- have been achieved. This reflects the ning future positive action activities negative impact of changing govern- In terms of monitoring of these were the French at 78 %. This is sur- ment policy on real improvement re- projects, many implementers of posi- prising considering the difficulty faced sulting from positive action measures tive action for Roma in Hungary in attempting to arrange a workshop (especially those implemented by the were unclear how to measure their in France as a result of the different public and voluntary sectors). Most impact. Indeed, a number of pro- perceptions of the utility of positive measures considered to be positive grammes were not measured in any action in different constituencies. The action for Roma in Hungary and Slo- systematic way, except where specifi- European countries with responses vakia are explicitly linked to tenders cally required by the fact that many lower than the overall value in terms and projects, which are time-limited programmes implemented by local of future planning for positive action and not necessarily renewed (in fact, authorities in Hungary target socially initiatives were the UK (47 %) and the often not). Many are financed by the disadvantaged groups rather than Netherlands (35 %). government (and these sources also Roma specifically. The most successful change with policy shifts), but also positive action measure in Slovakia so To conclude, there was clear joint by private foundations, which also far was thought to be the programme opinion that only well-designed, com- have changing priorities driving the of Roma teaching assistants, which plex, adequately financed and par- types of programmes implemented. received the highest support in terms ticipative initiatives would bring some This is true to a lesser extent in the of being a successful programme, significant impact to Roma popula- private sector, where positive action mainly because it has existed for sev- tions. Unfortunately, in reality this is measures are supported by company eral years and some progress in the still far from the case, particularly in resources, although positive action in area of education is visible. Slovakia.

41 International perspectives on positive action measures

4.5. Barriers to positive cruitment and which makes little or cases. National governmental statisti- action no allowance for change. On the oth- cal agencies refuse to collect data on er hand, there was also little trust in race and ethnic origin in the national A number of barriers were identified Swedish institutions by some commu- census except where it concerns the as inhibiting the impact of positive ac- nities, such as Roma or immigrants, nationality and the origin of first-de- tion, some of which were common to whose past experiences made them gree ascendants. Therefore, racial and several countries, whilst others were mistrust the police and the judiciary, ethnic statistical indicators allowing more specific to a particular situation. or who had had negative experiences evaluation of the impact of policies, or with police in their own countries. created for monitoring purposes, were Disclosure of potentially sensitive not available. information was one such issue that Swedish legislation was also men- arose in relation to different grounds, tioned by study participants as an In the Netherlands, barriers were for example, in Ireland in relation to obstacle, as it bans data collection, identified that related more closely to disability and in the UK concerning making the assessment for positive ac- behaviours of the target group rather sexual orientation. The survey also tion more difficult, unlike some other than those involved with targeting. It suggested that there was reluctance countries. Lack of resources in terms was felt that advertisements address- on the part of organisations to ask of personnel, finances and time were ing preferential treatment were not ef- questions about these issues both to perceived as another important stum- fective, since people did not want to be potential and current employees. Dis- bling block. Positive action is some- defined as ‘quota people’. Participants ability monitoring was undertaken times observed as an additional ob- felt that actively approaching specific by only 26 % of organisations whilst ligation; while many universities and target groups discouraged people from sexual orientation was monitored by large companies have designated staff applying for jobs to avoid the stigma a mere 10 %. members in charge of equality issues, of being labelled the ‘affirmative ac- this is not the case with many smaller tion candidate’. To counteract these Negative attitudes held by mainstream companies and institutions, where responses, it was deemed necessary society as well as stereotypes and prej- staff members are often addition- to create a climate of approval within udices perpetuated by the media were ally burdened with diversity or equal- an organisation where people felt wel- thought to problematise positive ac- ity issues. In the field of recruitment, come and acknowledged regardless of tion and render any positive action some interviewees complained that their background and indeed because outcomes as tokenistic. In this respect, their companies ‘fire more than hire’, of their various backgrounds. whilst Swedish society appeared to be creating an environment in which it relatively enlightened when it came is very hard to recruit more women In relation to positive action schemes to gender equality and measures pro- or minorities. It was also emphasised targeting Roma in Slovakia, a number moting it, Africans and Middle East- that most positive action measures are of problems were identified that lim- ern Muslims, as well as persons with undertaken as projects, and that there ited their effectiveness in improving disabilities, were often relegated to the is a need for long-term commitment, training and employment opportu- bottom of this hierarchy among dis- as achievements take time and effort. nities and in some instances possi- advantaged groups. Additionally, in Money was also often a factor, as posi- bly even reinforced segregation. The Sweden, positive action was not un- tive action is also considered costly; in spontaneous and short-term nature derstood as bringing any benefits, and the case of positive action for persons of projects has done little to promote an awareness of its worth was lacking. with disabilities, for instance, enabling the progress of initiatives and advance Sometimes the interviewees had no physical access was perceived as in- the position of Roma at the national, support from colleagues, or met op- volving significant investment. regional and local level. The dearth position from the management, who of monitoring and evaluation pre- regarded their equality obligations as In France, progress in the field of posi- vents efficient public policy cycles, in mere formalities and preferred efforts tive action has been hampered by the which lessons learned influence fu- only to the level of satisfying the leg- failure to collect monitoring data on a ture policies. From the perspective of islation. Some employees are reported routine basis. It has been pointed out NGO participants, discontinuity was to believe that they only need to fill that although there is no general prin- grounded in the system of fundrais- up diversity quotas to have the ‘im- ciple forbidding the collection of data ing. It was made almost impossible migrant alibi’; there is also a problem based on ethnic origin, religion, age, to develop programmes and activi- of closed structures such as the police disability or sexual orientation, au- ties with longer-term goals because of force, where there is no external re- thorisation for this is required in most the requirement of funders for more

42 4. Perceptions of positive action in the Europe Union

innovative programmes, and at the Roma population have not helped to of the majority population in Slovakia. same time they were rather reluctant foster positive relations between them. This was notably most visible at a lo- to support existing services (for exam- Roma clients have often been blamed cal level, where the actions of a mayor ple, Roma health assistants or Roma for not wanting to collaborate with or an NGO have been opposed by the teaching assistants). service providers who have been quick local non-Roma population. However, to label Roma as unreliable. A number even at the national level, there is evi- As was seen in Hungary, attitudes of of well-intended programmes were dence of low levels of political will to the majority population towards the blocked by a very high level of rejection engage effectively in Roma issues.

4.6. Summary

The comparative analysis of positive action as understood and implemented within the European Union has highlighted a number of significant issues. To varying degrees, positive action has been welcomed as an effective means of redressing disadvantage suffered by specific groups within society. Participants from many countries were opposed to the prospect of preferential treatment and the notion had clearly caused resentment and hostility, rightly or wrongly, towards the use of positive action initiatives. For those who were supposed to benefit from positive action, the perception that they were being given an unfair advantage was also a cause for concern. In response to these kinds of ten- sions some countries, such as Austria and the Netherlands, emphasised the importance of positive action measures as part of a broader equality strategy rather than as a stand-alone measure.

Across the eight countries, positive action measures in existence tended to focus upon providing training and improving employment opportunities rather than service delivery. In terms of outcomes, disabled people, women and minority ethnic groups appear to have benefited the most from positive action, with legislation and social attitudes playing an important role in promoting initiatives to benefit these groups. Increasingly though, in countries facing demographic change, there were strong financial pressures and policy drivers forcing organisations to modify the way in which they recruited staff and delivered services. In the case of some groups, this was helped by the availability of grants but, generally, limited resources constitute a major barrier to implementation.

Despite support for positive action measures, progress in this area has been hindered by failure amongst organisations to collect information to inform further work. For example, the fact that not all organisations are monitoring their workforce by the different discrimination grounds means that the evidence base to underpin any strategy will be weak. This is not helped by the fact that, as we have seen with grounds such as disability and sexual orientation, some individuals might not be willing to disclose information in an environment where they might be subjected to discrimination and made to feel different. This inability to collect and monitor data systematically was also noticeable amongst organisations in which positive action schemes were operational. As a result of this, although organisations were clear about the outcome measures needed for the success of their project, most of them were unable to provide any quantitative evidence to reflect the level or indeed lack of progress made. Instead, organisations tended to rely upon ‘soft’ measures, which indicated a general sense of improvement, greater sensitivity towards the needs of marginalised groups and improved understanding of difference.

Perceptions of positive action 5 in Canada, the United States and South Africa

45 International perspectives on positive action measures

5. Perceptions of positive action in Canada, the United States and South Africa In this chapter we use evidence from the online survey, documentary analysis, consensus workshop and individual interviews to present an analysis of positive action measures in Canada, the United States and South Africa. We provide a detailed analysis of the perceptions of positive action as practised in organisations within these countries, including the type of measures that individuals considered constituting positive action, drivers for positive action as well as factors deemed to contribute to the effectiveness of positive action measures. We also examine how organisations measure the impact of positive action projects and identify groups that appear to have benefited the most and least from targeted initiatives, and the main barriers that have hindered progress in implementing positive action initiatives and outline plans regarding positive action measures for the future. In these countries, the term affirmative action is used to refer to programmes designed to remedy past and present injustices targeted at marginalised groups in society. We will therefore use the term affirmative action interchangeably with positive action in this chapter.

5.1. Understanding of In South Africa, owing to the legacy of Canada, the term ‘employment equity’ positive action apartheid and segregation among dif- is used in an attempt to distance its ferent racial groups, affirmative action positive measures from the hard-core While the conceptualisation of posi- is well received as a policy, though affirmative action reform (e.g. quotas) tive action varies widely across coun- there were misgivings about its im- in the United States, which in effect is tries, some broad generalisations can plementation. Support for affirmative synonymous with positive discrimi- be made. Most participants in the action is mirrored in the legal frame- nation (Bacchi, 2004). In Canada, af- country workshops generally agreed work of South Africa. Affirmative ac- firmative action is enshrined in the that affirmative action constitutes a tion was introduced after the disman- constitution, which confers equal set of specific measures to redress past tling of apartheid, by the subsequent rights on every individual as explic- or present discrimination targeted at government of South Africa, to redress itly stated in the Canadian Charter particular groups that have been mar- the injustices and racial imbalances of Rights and Freedoms that ‘Every ginalised within society. There were perceived to have been the result of individual is equal before and under differences, however, in terms of the the systematic impact of apartheid in the law and has the right to the equal perception and implementation of the country. It was believed that af- protection and equal benefit of the law positive action measures, and the leg- firmative action would ensure that the without discrimination and, in par- islative and policy basis for these meas- formerly disadvantaged population ticular, without discrimination based ures across countries. Unlike Canada (also regarded as designated groups in on race, national and ethnic origin, and the United States, which have a Section 1 of the Employment Equity colour, religion, sex, age or mental or long history of affirmative action, in Act No.55 of 1998 (Department of La- physical disability’ (73). South Africa it has a very nascent his- bour Republic of South Africa, 2008)) tory; hence it presents an interesting of South Africa enjoyed the same ben- In the United States, there is a close contrast with the North American efits and opportunities guaranteed for association of affirmative action with countries. Participants in South Af- all racial groups in the post-apartheid quotas and other forms of preferential rica described their experiences with Constitution. treatment within the consciousness intense emotion, which was reflected in the language used to explain their Similarly, there is a stronger legal back- 73 Charter of Rights and Freedoms, Part I of understanding and engagement with ing for affirmative action in Canada in the Constitution Act, 1982, being Schedule B affirmative action. comparison with the United States. In to the Canada Act 1982 (UK), 1982, c.11

46 5. Perceptions of positive action in Canada, the United States and South Africa

of the population. As such, preferen- ‘equal’ (treating people the same) Other drivers included business case, tial treatment aimed at increasing the and ‘equity’ (treating people fairly). corporate social responsibility, moral/ representation of specific groups has Whilst participants in the Canadian ethical consideration, leadership and often attracted a negative response workshop frowned at any reference policy. It needs to be emphasised that in the United States. Here, the major- to reverse discrimination, workshop these drivers differed in priority ac- ity of the participants expressed their participants in South Africa and the cording to the following dynamics: understanding of affirmative action as United States described affirmative participant pool, context, sector and a set of strategies used, mostly by em- action as measures to reverse dis- country. In Canada, the main impetus ployers and educational institutions, crimination. for affirmative action is business case. particularly institutions of higher This consensus was discussed in the education, to provide equal opportu- In response to questions around un- context of addressing the talent short- nities and fairness related to obtain- derstanding of positive action, the age that is developing in the Canadian ing employment or admission to uni- survey suggests that respondents labour market and ensuring that staff versities and colleges. In the United from all non-EU countries combined reflect the racial and ethnic complexity States in particular, affirmative action were significantly less (9 %) likely to of society. In addition, policy response is negatively portrayed in the media confuse positive action with positive to regulations emerged as another as a means of giving opportunities to discrimination. Interestingly, none of driver for implementing positive ac- undeserving and unqualified people. the 18 survey respondents from the tion. Whereas in large organisations, This perception is rife and literature United States evidenced this confu- employers are mandated to employ supports the claim that affirmative ac- sion. This finding contradicts the positive measures under the Employ- tion is exploited politically in the me- literature (Bacchi, 2004) which au- ment Equity Act, in smaller organisa- dia to undermine its value. tomatically links positive discrimi- tions, policies such as anti-racist and nation to affirmative action in the mission statements which embed the Although the understanding of af- United States and participants’ views business case are utilised. firmative action in Canada is based during the workshops. It is to be noted on experience within a work or or- that the contrast between the survey The impetus for affirmative action that ganisational context, there was a gen- and workshop findings is indicative emerged from the United States work- eral consensus that ‘positive action is rather than summative, owing to the shop range from the demographic im- ensuring equal access, full participa- vastly different context and variations perative, grassroots efforts, civil rights tion and advancement in all aspects in numbers of participants from the law aimed at combating segregation of Canadian society: social, political, non-EU countries who participated and business case. While legislation economic and cultural’. In addition, in the survey; 70 of the total respond- played a role, it was not significant there were other sentiments, involving ents, representing Canada 35 (5.5 %), compared to Canada and South Africa the need for a strategic, thoughtful ap- the United States 18 (2.8 %) and South and, to a large extent, this may be due proach to overcome historic barriers Africa 17 (2.7 %), respectively. to the flexibility in the way affirmative and address systemic change. There action programmes are implemented was some discussion about the con- 5.2. Drivers for positive across institutions and sectors. Many fusion around the terms ‘affirmative action described changing contexts, such as action’ and ‘positive action’, as there the ‘changing complexion’ of the na- are several different governing bodies, There was a general consensus during tion, in other words that the United policies and legislative documents that the workshops in most countries that States was becoming more diverse impact on requirements and expecta- legislation was the key driver for pos- and the ‘old ways’ were not working tions around these issues. Some of the itive action measures and, not sur- as well as in the past. They stated that various legal groups and documents prisingly, legislation was ranked the ‘grassroots’ efforts from the surround- outlined in the legal analysis include highest driver in the survey respons- ing communities could be driving af- the Employment Equity Act, the On- es. Legislative provision is a power- firmative action, so that organisations tario Human Rights Code, municipal ful driver for affirmative action, par- represented the surrounding commu- legislation and organisational policies. ticularly in Canada and South Africa, nities that they served. Organisations There was also some discussion about and to a lesser extent in the United that appeared more diverse or hiring the fact that in Canada there are no States. As shown in the survey data, diverse people would also attract more quotas, but there are goals set for spe- 71 % of respondents from these non- customers from the local community. cific minority groups and in relation EU countries reported having a writ- Others discussed the historical con- to the difference between the terms ten equality and diversity policy. text as a driver of affirmative action

47 International perspectives on positive action measures

or positive action. An educator stated survey, a finite list of options was given previously been discriminated against that, previously, African Americans and respondents had to align their re- or treated unfairly. However, there had been excluded from schools and sponses with the choice given. As will were problems associated with imple- higher education. become apparent in the analyses be- mentation of affirmative action. In the low, this is a recurrent dynamic, again United States, the problems ranged Leadership also emerged in Canada mainly resulting from sample size. from unintended consequences of the as a key driver, as evident in the quote affirmative measures, negative atti- from a workshop participant: ‘People 5.3. Support for positive tudes towards affirmative action, hos- will say that it’s one of the most di- action tile political climate and misinforma- verse teams they’ve ever worked on. tion to media manipulation. In South The two top leaders [interviewee and Findings from the survey responses Africa meanwhile, implementation her supervisor] both come from a and workshop yielded divergent and had been fraught with negative stere- place of advocacy and a professional sometimes overlapping results. The otypes, stigmatisation, lack of proper knowledge and experience, back- following factors were ranked in the oversight and malpractice. A poignant ground in this work, so a very differ- order of priority as providing support reminder of the problems associated ent, deliberate and conscious effort to for positive action activities amongst with affirmative action was captured enhance diversity. There is a difference survey respondents: strong individual by a participant who compared the between having mandates and goals. commitment from participants; lead- consequences of affirmative action The only thing that works is if the ership and senior management sup- with the apartheid system. In both leadership is walking the talk’. Besides port; positive feedback from services cases, casualties were left behind as a legislation as the key driver in South users/customers; broad support from consequence. Africa, a number of moral/ethical employees; and support from line considerations were identified as es- managers. For the workshop par- In Canada, some participants recom- sential drivers. These included justice, ticipants, support for positive action mended that in order for affirmative fairness, inclusiveness, emancipation included all the survey responses as action to be effective it needed to be and grassroots agitations as the impe- well as legislative support, corporate done alongside broader normative tus for affirmative action. responsibility and financial resources, change and supported by institutions; amongst others. In Canada, the Unit- affirmative action required buy-in In terms of the drivers for positive ac- ed States and South Africa, support from leaders (senior management) tion among survey respondents, not ranged from legislation, leadership, within organisations and in some surprisingly legislation ranked the demographic imperative, corporate cases compensatory incentives for it highest. This is quite similar to the re- social responsibility, statutory agency to be successful. Affirmative action sponses in the workshop where legis- and mission of organisations. Whilst also needed to be part of the organi- lation was the greatest driver predom- the findings in the survey and work- sation’s larger corporate strategy in inantly across countries. However, a shop might yield different and some- order to work. In cases where positive closer look at the survey data suggest times overlapping results, we need to action had been a success, one partici- that factors such as ‘internal consulta- emphasise that methodological nu- pant said, ‘It means commitment from tion’, ‘good exemplar’ and ‘work with ances potentially influenced the re- the top. Unless the CEO and top ex- stakeholders’ ranked equally highly. sponses (similar to the above question ecutives get it as their project it won’t Business benefit ranked quite low in on drivers for positive action). For in- happen. If you don’t have that, it’s very the survey responses. On the other stance, with this particular question, difficult to go further.’ hand, the workshop responses includ- the survey respondents were given a ed business case as a high lever for list of finite options to choose from; Findings in the survey suggest a sur- positive action programmes, in addi- by contrast, the workshop participants prisingly different dynamic in terms tion to a core of other drivers such as had to identify support without being of the outcome and impact of posi- organisational mission/vision, ethical/ limited as to the choice of responses tive action. While the measures used moral dimension and corporate social to tick. in the workshop to evaluate outcomes responsibility. Whilst the findings in are mostly qualitative, the measures the survey and workshop might have 5.4. Outcomes and impacts used in the survey are quantitative yielded different and sometimes over- and measured by the number of re- lapping results, we need to emphasise Many participants believed affirma- spondents. However, findings from that methodological nuances poten- tive action to be effective in providing the workshop can be quantified from tially influenced the responses. In the opportunities for groups who have the consensus statements arrived at

48 5. Perceptions of positive action in Canada, the United States and South Africa

by participants. The survey findings 5.5. Barriers to positive of affirmative action programmes. revealed that intangible factors such action Hence, in Canada we saw examples as ‘raising awareness of equality is- of the LGBT community taking ac- sues in the organisation’, ‘improving A number of barriers were identified tive ownership of specific schemes. the reputation of the organisation’ and by participants as hampering their ef- However, the survey results show that ‘enhancing the confidence of the par- forts to implement positive action. In organisations from non-EU countries ticipants’ ranked highly amongst the general it was felt that lack of proper were significantly less likely (36 %) outcome and impact measures; other education on the wider benefit of af- than overall (51 %) to have equality factors that ranked highly among firmative action in society and misin- and diversity targets in service deliv- workshop participants, such as ‘work- formation from the media led to nega- ery, with an even lower likelihood in force representation or demographic tive attitudes about affirmative action Canada at 26 %. imperative’, ‘better financial results’ as is the case in the United States. or ‘business case’ did not rank highly In contrast to the barriers identified among survey respondents. A lack of clarity and ambiguity in leg- in the country workshops, the survey islation was also thought to under- found a strikingly high percentage of When asked which groups benefited mine the success of affirmative action total respondents citing factors such the most from affirmative action in programmes. Exploitation of the ben- as lack of money and time as the major the three countries, the responses efits and rationale of affirmative action barriers to positive action measures. were consistent across the non-Eu- for political gains were seen as serious Legislation, support from line man- ropean countries in so far as ethnic/ challenges for affirmative action pro- agers and resistance from employees racial minorities and women were grammes and in some cases resulted also ranked quite low in terms of bar- concerned. In the United States, the in unnecessary litigation. riers to positive action. These differ- impact is mostly felt by women and ences between workshop and survey all minority ethnic groups, particu- There was consensus that affirma- findings have to be treated cautiously, larly in terms of job progression to tive action needed to be part of an owing to sample size issues discussed leadership roles within organisations, organisation’s larger corporate strat- in the methodology section. enrolment numbers of Hispanics and egy in order to work. Some argued African Americans in higher educa- that if affirmative action was not tion and further education institu- mainstreamed as part of a broader tions. In South Africa, the partici- normative change and supported by pants were keen to stress the benefit institutions with proper mentoring derived from affirmative action by and training, it could even be counter- ‘the kingmakers’ further described by productive. Lack of buy-in from sen- the participants as the dynasties, po- ior management within organisations litically affiliated members or fami- and in some cases lack of compensa- lies of managers. In South Africa and tory incentives also had the potential the United States, LGBT and disabled to affect the success of affirmative ac- people were seen to have benefited tion. As illustrated in South Africa, the least from affirmative action, as participants felt that some beneficiar- they are still poorly represented in the ies were not well equipped or qualified working population. This was rather to assume responsibilities for jobs. surprising given the robust disability This resulted in a ‘brain drain’ and al- legislation, in particular Americans legations of reverse discrimination, as with Disabilities Act (ADA) and In- former post-holders were expected to dividuals with Disabilities Education take on jobs that should have been un- Act (IDEA). However, in Canada dertaken by unqualified colleagues. affirmative action measures were deemed to have had a positive impact The role of targeted groups was also on these groups due to ‘a gay senior thought to play an important role in manager championing the cause in- terms of the success of projects. Self- ternally’ and systematic implemen- advocacy and visibility by members tation of accessibility strategies for within targeted groups was seen as disabled people. advantageous in furthering the aim

49 International perspectives on positive action measures

5.6. Summary

While the term ‘affirmative action’ had very divergent meanings depending on national contexts, there was an overwhelming consensus that such a programme was designed to remedy past and present injustices. To a large extent, the different meanings were informed by national legislative and policy underpinning. For example, in Canada affirmative action is very human-rights centred. In spite of the consensus around the meanings of the term, there were some instances in South Africa where inappropriate behaviour and malpractice seriously undermined the effectiveness of affirmative action programmes. The vast majority of participants agreed that legislation was the key driver for affirmative action programmes. Other drivers in close competition were business case (particularly in Canada), demographic imperative, policy and leadership. In spite of its benefits, implementation was fraught with problems, such as negative attitudes towards affirmative action initiatives, media manipulation, legislative ambiguity, lack of self-advocacy, lack of adequate finance, negative stereotypes/stigma and lack of proper oversight.

50 1. Introduction: New Business Horizons in Europe Positive action measures across 6 different discrimination grounds, organisations and sectors International perspectives on positive action measures

6. Positive action measures across different discrimination grounds, organisations and sectors In this chapter we present an analysis of the perceptions of positive action held by respondents from all the countries participating in the survey, focusing specifically upon differences across discrimination grounds, sectors and organisation types. This chapter will also make reference to workshop findings and policy analysis. The chapter begins by describing the context of equality and diversity in the participating organisations, understanding of positive action, factors influencing its practice, the support available to sustain it and the barriers to its successful practice. We report on how organisations measure the impact of positive action projects and provide examples of positive action being applied in participating case-study countries.

6.1. The context of equality policy. To some extent, these distri- these targets. It was rather surpris- and diversity butions were not surprising, as they ing to note that organisations with a mirrored the trend that public sector smaller employee base (11–25) were As evidenced in the earlier sections, organisations will have an obligation more likely to have specific targets participants emphasised the need to produce a written policy. on service delivery at 81 %, above an for positive action measures to be overall 51 % response rate by sur- part of an overarching equality and It was widely agreed by participants veyed organisations, as compared to diversity strategy in any organisa- in the study that equality and diver- organisations with a large number of tion. We considered by way of con- sity monitoring enabled employers staff (5 000+) at 37 %. text whether organisations had a to check the effectiveness of their written equality and diversity policy policies and procedures and was 6.2. Understanding and what variations might exist be- required as a precursor to the de- of positive action tween sector and organisation type. velopment and implementation of The survey data showed that almost targeted strategies. The centrality As illustrated in Figure 2, when half (49 %) of the organisations con- of monitoring in the successful im- asked which statement they thought tacted had a written equal oppor- plementation of positive action re- best described ‘positive action’ as tunities policy, though this varied flected a clear hierarchy with gender understood in their organisation, considerably between demographic being the most widely monitored the only statement which over half groups. Organisations with 5 000 (49 %) and sexual orientation (15 %) (56 %) of the sample agreed upon employees and over and represent- the least monitored ground. Those was ‘organisational commitment ing the public sector were signifi- who did monitor their equality and to equality and diversity’. Around cantly more likely to have a written diversity activities did so primarily 40 % agreed on ‘activities to combat equality and diversity policy (81 %) for current employees, followed by disadvantage and discrimination’ than those with fewer than 10 staff monitoring of job applicants. Al- and ‘action to combat stereotypes’ (24 %) and that represented volun- most half (48 %) of organisations and over a third chose ‘legal obliga- tary sector organisations. Whilst that actively monitored these diver- tion to address equality and diver- respondents from the public sector sity areas published some or all of sity’; ‘action to redress under-repre- reported more likelihood (59 %) of this information. sentation’ and ‘affirmative action to having a written policy, there were reverse specific discrimination’ to variations between areas of work, Whilst 55 % of organisations in the describe positive action. The varia- with health organisations (79 %) survey had specific targets on service tions between groups that confused and colleges/universities (64 %) user/customer profile with respect positive action with positive discrimi- more likely than voluntary sector to equality groups, local authority nation were mainly on a country basis social-care organisations (38 %) to (75 %) and social-care organisations and this has been discussed in Chapters have a written equality and diversity were significantly more likely to have 4 and 5.

52 6. Positive action measures across different discrimination grounds, organisations and sectors

6.3. Drivers for positive FIGURE 2: DESCRIPTION OF POSITIVE ACTION AS UNDERSTOOD IN ORGANISATIONS action

Organisational commitment to equality and diversity In the general sample, the most Activities to compensate for disadvantage and discrimination significant driver for their positive action initiative was legislation Action to combat stereotypes (47 %). There was a reasonably even Legal obligation to address equality and diversity response covering the other identi- Action to redress under representation fied drivers ranging from 17 % for A rmative action to reverse specic discrimination influence from funding bodies to Positive discrimination to treat some groups more favourably 36 % for internal consultation (see None of these Figure 4). Any action targeted to- Other wards encouraging organisations to 0102030405060 develop positive action initiatives needs to consider a wide range of Overall perception of the benefits of tion (average 76 % before and 85 % potential influences whose rel- positive action was assessed at the be- after completing the survey) was evance may differ according to or- ginning and at the end of the survey; rated by CE/MDs at 90 % before and ganisation type. the second time, the wording of the 94 % afterwards. Positive action is one question asked respondents to rate of the strategic aims and objectives of the statement ‘taking into account the organisation (average 69 % before the definition of positive action used and 73 % afterwards) and was rated in this survey.’ Figure 3 shows that by chief executives at 82 % before and all of the ratings increased the sec- 88 % afterwards. Voluntary organisa- ond time, with an overall increase of tions and NGOs also tended to give about 3 % in the proportion agreeing. slightly higher ratings. The ordering of the responses were the same. This indicates that a clari- FIGURE 3: PERCEPTIONS OF BENEFITS OF POSITIVE ACTION AT EARLY AND LATER fication of the nature and purpose STAGES OF THE SURVEY of positive action led to a better understanding of its actual nature. This a) Recognition of bene ts of Positive Action at the early stage of the study may have generated a more positive Positive Action would/does bene t the individuals involved response or attitude to positive action Positive Action would/does bene t the local community and helped respondents to a clearer perception of its utility. Despite the Positive Action would/does bene t this organisation apparent confusion over the nature Positive Action would be/is recognised as valuable by this organisation of positive action, clearly the majority of the respondents have a favourable Positive Action is part of the strategic aims and objectives of the organisation attitude towards it. If clarification in- -200 -100 0 100 200 300 400 500 600 creases a positive response amongst respondents, it is at least plausible Strongly disagree Tend to disagree Neither Tend to agree Strongly agree that awareness-raising amongst those b) Bene ts of Positive Action at the after stage of the study who currently have little understand- Positive Action can bene t the individuals involved ing of positive action would have an even greater effect. A consistent find- Positive Action can bene t this organisation ing was that respondents at the chief Positive Action can bene t the local community executive/managing director (CE/ MD) level or equivalent gave higher Positive Action can be recognised as valuable by this organisation ratings to certain of the items: positive action [would be/is] [can be] rec- Positive Action is part of the strategic aims and objectives of the organisation -200 -100 0 100 200 300 400 500 600 ognised as valuable by this organisa-

53 International perspectives on positive action measures

FIGURE 4: FACTORS INFLUENCING THE IMPLEMENTATION OF POSITIVE ACTION respectively). Reluctance of target groups to participate, lack of support from line managers and resistance among employees were cited by fewer Legislation than 20 % of respondents. A relatively Internal consultation low proportion (13 %) of respondents Good examples set by others thought that substantial resistance or Work with stakeholders cynicism among employees was a sig- Analysis of stang prole nificant barrier. Given the selective na- Analysis of service user/customer prole ture of the sample, we could speculate

Consultation with service users/customers that this potential barrier may be more significant in the wider population. Representations by unions/sta associations etc Based on previous discussion around Business benet (or consideration) the definition/understanding of posi- In uence from funding bodies tive action, this problem could be tack- Other led by providing better information Shareholder/investor policy and explanation to the general public 0 10 20 30 40 50 regarding the nature of positive action. 6.6. Outcomes and impact 6.4. Support for positive lower for colleges and universities at action 48 %. The lowest rating was for ‘sup- Figure 6 below shows how effective port from line managers’ at 69 %. If positive action is seen to be in vari- Figure 5 shows that the great major- this is a problem, those at CE or MD ous areas. It is of note that the highest ity of participants (85 %) agreed that level seemed unaware of it since their ratings are given to more intangible there was ‘strong individual commit- rating was 82 %. Public sector organi- items such as awareness of issues, the ment from participants’. This figure sations, and colleges and universities organisation’s reputation and people’s was a little higher for voluntary organ- were again lower than average at 62 % self-confidence, all of these being en- isations and NGOs, at 88 %, but lower and 35 % respectively. People work- dorsed by more than three quarters of in the public sector (78 %) and in large ing in equality and diversity also rated respondents. Positive action’s potential organisations with between 1 000 and this lower at 59 %. contribution to business success was 5 000 employees (73 %). Some 78 % less well recognised, with only a third agreed that there was ‘leadership and 6.5. Barriers to positive (32 %) agreeing that it translated into senior management support’; those at action ‘better financial results’. There were CE or MD level were more optimistic generally higher ratings by those at CE/ about this at 90 %, those in the public Lack of money and lack of time were MD level, and lower in the public sec- sector were lower at 70 % and colleges the most frequently cited barriers tor, especially colleges and universities. and universities lower still at 60 %. to positive action (46 % and 37 % Those who gave their current role as working in equality and diversity were FIGURE 5: SUPPORT RECEIVED FOR POSITIVE ACTION also less sure of support from lead- Figure 5: Support received for positive action ers, rating this at 64 %. Three quarters

(76 %) agreed they received ‘positive Strong individual commitment from participants feedback from services users/customers’. The only significant difference in Leadership and senior management support this rating was between sectors: with the voluntary sector/NGOs rating it Positive feedback form services users/customers higher at 81 % and the public sector lower at 70 %. Slightly lower num- Broad support from employees bers (72 %) thought there was ‘broad Support from line managers support from employees’. Again, this was higher for the voluntary sector/ -150 -100 -50 0 50 100 150 200 250 300 350 NGOs (80 %) and lower in the pub- Strongly disagree Tend to disagree Neither Tend to agree Strongly agree lic sector (60 %), but also markedly

54 6. Positive action measures across different discrimination grounds, organisations and sectors

FIGURE 6: EFFECTIVENESS OF POSITIVE ACTION area of religion or belief and one fifth addressing sexual orientation. Of the Raising awareness of equality issues in the organisation types of positive action identified in Improving the reputation of the organisation the questionnaire, there was a fairly Enhancing self con dence of participants even spread of responses indicating Providing opportunities for networking that multiple measures were com- Providing professional development monly used across the discrimination grounds. When asked when posi- Increasing representation of target groups in the workforce tive action measures were first intro- Improving service quality/customer satisfaction duced, almost half of the respondents Improving workforce quality (48 %) had introduced such measures Improving eciency more than five years previously, 40 % Improving sta retention between one and five years, and 8 % Better nancial results under a year before the survey. The Other overall picture mirrors the findings -350 -300 -250 -200 -150 -100 -50 0 50 100 150 200 250 300 350 from the workshops and interviews Strongly disagree Tend to disagree Neither Tend to agree Strongly agree as shown in Figure 7. However, the only discrepancy arises in relation to the creation of networks and forums which did not appear to be widely A broad range of measures were re- 6.7. Positive action in cited examples of positive action in ported to be used by respondents, and practice the case-study countries. This may no single measure showed a response have been because only 10 countries greater than 40 %. Only 16 % had an Overall, 72 % of the respondents said participated in the consensus work- external assessment and only 26 % that their organisation had some kind shops and interviews compared to 32 employ targets/performance indica- of initiative or programme which they countries that took part in the survey. tors. The other measures involved pe- thought could be described as positive That said, we were surprised to note riodic or ad hoc reviews, staff and user action. As shown in Figure 7, 50 % or that no country offered examples of consultation and anecdotal evidence, more respondents were implement- positive action based around religion which by their nature are likely to be ing measures in the areas of age, dis- or belief. attitudinal rather than focused on ability, racial or ethnic origin and measurable outcomes. gender, with a quarter addressing the

FIGURE 7: POSITIVE ACTION MEASURES USED IN RELATION TO DIFFERENT DISCRIMINATION GROUNDS

Measures used

Networks and forums

Training and leadership development

Help with communication/con dence building

Help with gaining further quali cations

Encouraging work/life balance

Targeted recruitment Areas used in

Positive Action Training Gender

Outreach work Racial or ethnic origin

Mentoring schemes Disability

Target setting Age

Championing schemes Religion or belief

Quotas Sexual orientation

Other Other 0 10 20 30 40 50 % 0 10 20 30 40 50 60 70 80 %

55 International perspectives on positive action measures

6.7.1. Examples of positive action The following examples of positive Disability and higher action are drawn from the case-study education in the UK In order to provide further insight countries: into the kinds of measures that can Birmingham University has estab- be regarded as positive action, we give Black and ethnic minorities lished a scheme which provides fi- concrete examples of positive action and vocational training in nancial awards to students with dis- measures which have been applied the UK abilities, including students who in the European Union, Canada, the have dyslexia, to help them with their United States and South Africa. Figure studies. The money does not have to 8 presents a breakdown of the total Leeds Mental Health Teaching Na- be spent on specific disability aids. number of positive action measures tional Health Service Trust (a hospi- Instead, the award recognises that identified by study participants within tal) in the United Kingdom encour- students with a disability often find it the case-study countries. These exam- ages people who are black or from difficult to supplement their income ples are drawn from the current study ethnic minority backgrounds to take through part-time work whilst they and reflect localised interpretations of up a range of internships. The aim is are studying. positive/affirmative action. Detailed for the internees to gain confidence description of these examples are and skills, and to remain employed in Ethnic minority parents and available in the country reports, where the National Health Service after the education in Austria analyses confirm the considerable mis- internships have come to an end. understanding about what constitutes The City of Vienna, in cooperation positive action and the overlap with Asylum-seekers and with various organisations represent- other complementary measures, such employment in Austria ing ethnic minorities, has a project as equality and diversity monitoring (MA 17) which is designed to promote and impact assessment. In addition, A scheme financed through the ESF greater involvement of ethnic minor- there was no example of positive ac- (EQUAL), FluEQUAL, increases op- ity parents in their children’s education to cover the diversity ground of portunities for asylum-seekers in the tion. The project involves informa- religion and belief. labour market, by providing them tion events, translation services and with access to German-language childcare services and also provides courses and vocational training and German-language classes for moth- encouraging them to enter the labour ers of ethnic minority children. These market. take place at the child’s school.

FIGURE 8: INTENSITY OF POSITIVE ACTION MEASURES ACROSS CASE-STUDY Various groups and housing COUNTRIES The City of Vienna housing department has established a scheme to ad- Quota dress intercultural conflict amongst (2) residents of the city’s public housing. The city employs a group of intercul- Mentoring schemes tural mediators who work in mixed (7) teams to provide support that is easy to access and accept for as many people as possible. The concept of ‘culture’ Championing schemes / as used in the scheme covers not only outreach work (12) people of different origins and cus- toms, but also addresses conflicts be- Positive action training / training and tween young and old people, and ac- leadership / confidence building (15) commodates sign-language users. Targeted Recruitment / Target setting (15)

Review of services/polices Awareness raising about positive action (19)

56 6. Positive action measures across different discrimination grounds, organisations and sectors

Ethnic minorities and to integrate Roma children unjustly Internship programme academia categorised as disabled back into for Aboriginal students in mainstream classes, and the creation Canada The Netherlands Organisation for of a requirement that schools adopt Scientific Research (NWO) has estab- and implement concrete equal oppor- Within the Bank of Montreal, students lished a grant scheme (Mozaic) to in- tunity policies in order to be eligible of Aboriginal origin were provided crease the number of researchers from for Structural Funds. with the opportunity to gain experi- ethnic minorities who are carrying out ence working in the bank, which also PhD research at Dutch universities. Roma and access offered them a scholarship towards The scheme is only open to applicants to employment post-secondary education. from an ethnic minority background. A special employment programme Vienna needs you — police Socially disadvantaged at USA Steel in Kosice targeting the officers with a migrant groups and university Roma community exists in Slovakia. background admission Within this programme, USA Steel cooperates closely with municipal au- In order to increase the number of In Hungary, Article 19/A of Govern- thorities in three surrounding areas police officers with a migrant back- ment Decree 268/2000 on the gen- (Vel’ka Ida, Saca and Kosice’s Lunik IX ground into the police force in Vien- eral rules of admission procedures of district) to identify Roma candidates na, a recruitment campaign was set up universities provides that a socially for employment at its factory. Through in November 2007 targeting Austrian disadvantaged applicant shall be ad- this programme, USA Steel subcon- nationals of migrant background. A mitted to a university faculty if he/ tracts Roma candidates through the ‘tandem-couple’ made up of a police she reaches 80 % of the admission municipal office for a period of one officer with migrant background and threshold determined for the given year, after which time the individu- a representative of the Viennese Mu- faculty. The category of ‘socially disad- als concerned may be considered for nicipal Department for Diversity visit vantaged’ persons includes, inter alia, regular employment. schools and migrant community asso- those whose parents only completed ciations to present the initiative as well elementary school. The number of Accessibility project in as to act as a role model. There are no students admitted on the basis of this Canada quotas implemented and there is no measure shall not exceed 3 % of the explicit preferential treatment of ap- maximum number of students deter- An under-representation of disabled plicants with a migrant background. mined for the given faculty. people within the employer’s workforce led to a review of potential bar- Mingo migrant enterprises in Integrated education for riers. As a result, the accommodation Austria Roma policy was enhanced and streamlined to help provide a better experience In May 2008, an office was set up Since 2004, the Hungarian Ministry for employees and managers to assess which offers free information and of Education and Culture has been and request adaptive technology solu- counselling in different languages to implementing programmes to reduce tions. minority ethnic businesses, which the segregation of Roma children in often face difficulties in accessing sub-standard schools/classes, as well ‘Ability Edge’ project in mainstream support due to linguistic as in schools/classes for children with Canada barriers and cultural differences. The mental disabilities, and foster the in- assistance entails provision of infor- tegration of Roma children in main- An initiative which involved several mation and help with implementing stream education. The government banks getting together and creating operational steps to innovation as well programmes included measures such internship opportunities for people as support in dealing with agencies as providing an integration payment with disabilities. In addition, students and authorities. to schools which included children with disabilities were provided with a with special educational needs, a pro- scholarship towards post-secondary gramme entitled ‘Last Bench’ aiming education.

57 International perspectives on positive action measures

Roma internship programme Roze in the Netherlands outside of South Africa, to enable them for reporters and editors in to take over from white post-holders. Hungary The problems faced by LGBT people in terms of accessing employment Improving the recruitment of The aim of this initiative was to and working in hostile environ- people of Chinese and mixed increase the positive presence of ments prompted the formation of a ethnic background into the Roma in the media. The public tel- network of LGBT people within the health service in the UK evision and public radio each facili- Roze trade union. A website and a tate 10-month internships for five web-based forum were set up to fa- A primary care trust introduced a individuals, during which the time cilitate exchange of experience and number of initiatives to improve the the interns attend professional skills has also served as a platform for the representation of under-represented classes, are provided with tools to organisation of public events and groups into the workforce, including address psychologically harmful sit- campaigns to enhance the visibility specific ethnic groups. In order to at- uations they may encounter, paired of LGBT people in employment. tract more applicants of Chinese and with a professional mentor, provid- mixed ethnic background, job vacan- ed with a scholarship and receive a Social fieldwork in Slovakia cies were emailed to 300 community certificate upon completion. organisations and also distributed The main purpose of this project is to through the organisation’s weekly bul- Equal chances against provide Roma people living in socially letin. A national website ‘Ethnic Brit- breast cancer for socially excluded environments and Roma set- ain’ was also used to advertise 80 posts. underprivileged women in tlements with quality social counsel- In addition, guidance for potential ap- Hungary ling and assistance. The programme is plicants on how to access NHS jobs was managed by the Social Development translated into different languages. In cooperation with Roma NGOs Fund and is open to municipalities and representatives, the organisers which have a Roma community. Based Mosaic initiative in the reached out to Roma women with on its size, the village or city can apply United States the aim of encouraging a wider pro- for funding for one or several social vision of breast cancer screening. fieldworkers. The workload is clearly From May 2008, as part of the universi- This included providing transport to defined and the fieldworkers serve in- ty’s strategy to increase diversity among screening sites and mobile screening dividual clients or families. ranked faculty from under-represented units in isolated settlements. In ad- ethnic groups, USD 1 million has been dition, through cooperation with Targeted recruitment made available to fund the Mosaic ini- Roma representatives and health for management and tiative. As part of this venture, depart- service providers, the scheme tried supervisory level in South ments are able to apply for funds to the to facilitate sustainability by encour- Africa dean or director of its division to cover aging open lines of communication the cost of items such as salary, research and cooperation. In 1993, as a means to increase the support and laboratory equipment. number of black people at manage- Accepting working place ment and supervisory level, an electric- ‘Open Up’ in Sweden model in Hungary ity company owned by the government, instituted targeted recruitment of black This development project created a In order to address the high attri- people within the organisation. For- working scheme for young persons tion rates within a poultry process- mal mentorship programmes were set with functional disabilities by SEKO ing plant, the company worked with up for lower-level personnel to be able tele Stockholm, with the goal of find- the local employment centre and to move into management positions, ing them jobs in the IT/telecom vocational training centre to set up a where existing white post-holders were sector. The project also helped com- project providing on-the job training encouraged to serve as mentors. Op- panies recruit from this group, as well for employers and employees, many of portunities were also made available for as generally supporting them to get in whom were unskilled with low levels sponsorship of black people to study at touch with persons with disabilities of education. appropriate universities within and who were users of their services.

58 6. Positive action measures across different discrimination grounds, organisations and sectors

Drugs project in Ireland Teaching assistant in Slovakia

A drugs awareness project was set up A programme was implemented in targeting Traveller parents as a result schools to fund teaching assistants of research highlighting the prob- in providing language and individual lem of drug use in this community. learning support for socially disad- The aim of the project was to reduce vantaged children in order to help the number of people using drugs as them overcome existing barriers in well as to try to normalise discussion the education system. Whilst it be- about drugs by engaging the commu- gan as an NGO initiative, this pro- nity about the problems of drug use. gramme is now funded by the Minis- To date, the 10-session course has run try of Education through the regional five times in less than two years. school offices.

6.8. Summary

This chapter presented limited evidence of private sector activity in relation to positive action. Organisations were more likely to introduce positive action measures to address imbalances in the workforce as opposed to service delivery. The contexts for positive action practice were more favourable in third sector organisations, where individuals were more likely to adopt a broader understanding of positive action in order to build confidence, empower individuals and ensure sustainability. Over- all understanding of positive action improved with awareness-raising and required sustained support from leaders and managers.

1. Introduction: New Business Horizons in Europe

7Conclusions and recommendations

61 International perspectives on positive action measures

7. Conclusions and recommendations This chapter presents the conclusions of the present study. A comparison of positive action practice between European and non-European countries is then presented, together with the comparison of practice across different discrimination grounds, sectors and organisations, and key messages will be drawn from these comparisons. These are then followed by recommendations and implications for research, policy and practice.

7.1. Conclusions positive action, whilst findings from The findings of this study show leg- the survey show greater uniformity in islation to be a key driver for posi- 7.1.1. Comparisons between EU and terms of respondents’ descriptions. tive action. However, many partici- non-EU countries pants felt strongly that the legislation There are differences in terms of theap - needs to be backed up by enforcement A general conclusion of the research plication of positive action measures, mechanisms and less bureaucracy to is that whilst the conceptualisation of and the legislative and policy basis for ensure that organisations fully engage positive action varies widely across these measures across countries. In with it. Other drivers that emerged countries, some broad generalisations some countries, participants consider from the study include business case, can be made. Terminology to describe positive action measures to be an ‘old- corporate social responsibility, moral/ remedial action to address past and fashioned’ strategy and struggled with ethical considerations, leadership and present injustices targeted at margin- applying these concepts to measures organisational policy. It is noteworthy alised groups differs. In the non-EU within their local contexts. For exam- that where there was no direct legis- countries, the term ‘affirmative action’ ple, the Dutch study participants view lative backing for positive action, the is widely used. In Europe, however, positive action as an important tool moral case was a major impetus. This the concept of ‘positive action’ is more within a wider diversity management was most noticeable where particu- commonly employed. The legal defi- strategy, which includes all measures lar communities (for example Roma) nition used throughout this study was designed to counteract the effects of were living in poor socio-economic presented in Chapter 4 and describes exclusion, discrimination and stereo- conditions. positive action as ‘consisting of pro- typing of specific groups, with the aim portionate measures undertaken with of creating a more equitable society. Findings from the present study yield- the purpose of achieving full and effec- In South Africa, however, due to the ed divergent results regarding support tive equality in practice for members legacy of apartheid and segregation for positive action. Whilst strong indi- of groups that are socially or economi- among different racial groups, what is vidual commitment and senior man- cally disadvantaged, or otherwise face termed ‘affirmative action’ is well re- agement support are identified as the the consequences of past or present ceived as a ‘stand-alone’ strategy. most important factors, the finding discrimination or disadvantage’. that participants received relatively These differences in perception and at- less support from line managers is Evidence from the study shows a titude towards positive action are mir- worrying and needs to be addressed. general agreement that positive ac- rored in the legal frameworks of both This, coupled with a number of barri- tion constitutes specific measures to countries. Whilst in some countries, ers relating to human and financial re- redress discrimination experienced there is the general tendency to move sources, may militate against success- by particular groups within society. away from enforcing positive action ful implementation of positive action. Whilst particular emphasis is laid through legislation (for example the upon the definition of positive action Netherlands), there is a stronger le- There are some differential outcomes in different countries, it is apparent gal backing for affirmative action in of positive action in qualitative com- that there is considerable variation in countries such as Canada and South pared to quantitative components of participants’ level of familiarity and Africa. Overall there was very limited this study. The survey findings reveal understanding of the concept. The use of quotas and their strong asso- that intangible factors such as ‘raising qualitative aspect of the study reveals ciation with preferential treatment at- awareness of equality issues in the or- a more divergent interpretation of tracted a negative response. ganisation’, ‘improving the reputation

62 7. Conclusions and recommendations

of the organisation’ and ‘enhancing particularly strong in some countries with gender, age, disability and racial the confidence of the participants’ are where the media was thought to prob- or ethnic origin being the most widely important parameters amongst the lematise positive action and render monitored grounds and sexual orien- outcome and impact measures; whilst any outcomes as tokenistic. Similarly tation and religion or belief being the other factors relating to workforce whilst participants from Sweden felt least monitored grounds. representation and the need for better their society appeared to be relatively financial results were more significant enlightened when it came to gender There were variations in the extent to for workshop participants. equality and measures promoting it, which sectors provided an enabling groups such as Africans and Middle environment for the development Whilst this study has highlighted clear Eastern Muslims, as well as persons of positive action. Whilst a good benefits of positive action, at the with disabilities, were not considered number of organisations had a writ- same time there are reports of nega- to be in need of positive action. ten policy on equality and diversity, tive impacts. Some of the adverse con- larger organisations were not as ef- sequences, including skills shortage In terms of formulating positive ac- fective in setting appropriate targets resulting from ‘brain drain’ and ‘ghet- tion strategies, there is a concern in for employment and service delivery. toisation’ within the workforce as a re- some countries about the need for Monitoring is an important aid to sult of targeted recruitment, were seen the target group to be involved in implementing positive action since in some countries. Exploitation of the the positive action initiative from its it provides a basis for identifying benefits and rationale of affirmative inception. In Canada, it was felt that evidence of under-representation. action for political gains were seen as a lack of self-advocacy and visibility Whilst some organisations recognise serious challenges for programmes to by members within targeted groups the value of monitoring, the absence be sustained. There is a danger that could undermine the success of af- of relevant data may have hampered this could invite unnecessary litiga- firmative action programmes. Conse- efforts to formulate appropriate position which may in turn discourage quently we saw examples of the LGBT tive action measures. (other) institutions from implement- community taking ownership of pro- ing positive action programmes. grammes. Similarly, in the UK, we A clarification of the nature and pur- found evidence of community groups pose of positive action led to a better It is evident that organisations are and grassroots organisations them- understanding of the actual nature of struggling to develop a robust evidence selves who, fuelled by dissatisfaction positive action. There was a reason- base for positive action. For some with the existing status quo, had lob- ably even distribution of responses in countries, like Sweden, this process is bied for positive action measures and relation to the drivers of positive ac- hindered by legislative barriers (e.g. succeeded in setting up specialised tion with legislation featuring highly. legislation barring data monitoring by services to meet their own needs. Therefore any action targeted to- ethnicity) whilst others, such as South wards encouraging organisations to Africa and UK, had not worked out Our findings show that the future for develop positions or initiatives needs fully how to obtain information on positive action is hopeful and many to consider a wide range of potential grounds such as disability and sexual organisations have plans to introduce influences whose relevance may dif- orientation. positive action initiatives or pro- fer according to the type of organisa- grammes in the future. tion concerned. Across EU and non-EU countries, similar attempts could be witnessed 7.1.2. Differences across organisa- There was widespread consensus by those with little genuine interest in tions and sectors about the need for organisational the ethos of positive action, to use the support for positive action, espe- situation to their advantage. This was It was widely agreed by participants in cially management buy-in. Howev- evident in some countries where posi- the study that equality and diversity er, there were disparities in different tive action had given rise to dishonest monitoring provides a tool to enable sectors, with third sector organisa- behaviours and malpractice. employers to check the effectiveness tions more likely to support positive of their policies and procedures and is action than public and private sec- In this study, there was evident lack of required as a precursor to the develop- tors. It was noteworthy that a rela- awareness both within the workforce ment and implementation of targeted tively low proportion of respondents and in the wider society about the strategies. The centrality of monitor- thought that substantial resistance benefits of positive action, and mis- ing in the successful implementation or cynicism among employees was a information from the media. This was of positive action revealed a pattern significant barrier.

63 International perspectives on positive action measures

A broad range of methods were uti- hoc reviews and anecdotal evidence. traditional hierarchy of application, lised to assess the outcomes of posi- Whilst the quantitative evidence from the qualitative data from consensus tive action. However, there was a the survey suggests a spread of posi- workshops and interviews did not tendency to rely upon ‘softer’ ap- tive action initiatives covering dif- provide any examples of measures re- proaches to evaluation including ad ferent grounds, albeit following the lated to religion or belief.

7.2. Key messages

A number of key messages can be drawn from the research presented in this report which in turn can be summarised under the main headings corresponding with the findings of the study discussed in earlier chapters. Context of equality and diversity

• Positive action is generally undertaken within the framework of written equality policies, mission statements and annual reports, which reflect a commitment to equality.

• Monitoring and specific target setting relating to equality and diversity are widespread, with gender being the most widely monitored and sexual orientation the least monitored ground.

• Lack of disaggregated data in key sectoral fields means effective positive programmes cannot be comprehensively developed and implemented.

• Attracting diverse communities into the organisation may not necessarily mean a change of culture and attitude towards positive action.

• Definition and understanding of positive action

• There is confusion and inconsistency in the terminologies used to describe positive measures across the study countries.

• There is a lack of common understanding in Europe, within countries and across sectors on the meaning of positive action.

• A clarification of the nature and purpose of positive action measures can promote a better understanding of the actual nature of the measures and in turn generate a more positive attitude and perception of the utility of positive action.

• The historical and political context of the respective countries influences the formulation and adaptation of positive action. Drivers for positive action

• Legislation remains the main driver for positive action. Other key drivers include altruistic reasons, moral/ ethical considerations, business reasons, demographic changes, corporate social responsibility, organisational policy and grassroots efforts.

• Negative factors are seen to drive positive action in some instances where organisations set up programmes for political and financial gains with little genuine interest in the essence of positive action.

64 7. Conclusions and recommendations

Barriers to positive action

• Lack of resources (human, financial and time) is the most frequently cited barriers to positive action, particularly in the European countries. Lack of senior management buy-in, continued support and commitment remain major impediments to successful and sustained positive action.

• Legal frameworks on positive action lag far behind social policy, and conflicting data protection arrangements in some countries create serious barriers to implementing positive action.

• There are difficulties in ensuring that the legislative framework is consistently applied in practice. Further- more, there are differences among countries in implementing sanctions against organisations that do not implement positive action or achieve equitable practice.

• Lack of awareness of the benefits of positive action measures within the workforce and in the wider society, and the role of the media in problematising these measures and rendering outcomes as tokenistic, are seen as major barriers to the acceptability of positive action. Support for positive action

• Support of the wider society is essential for the success of position action programmes. Strong individual commitment, support from colleagues, leadership and senior management buy-in are necessary to sustain positive action activities.

• There are differences between all sectors in their enthusiasm and implementation of positive actions. There is evidence of commitment of some public sector organisations to push the boundaries of policy, to develop a much broader application of positive action.

• Positive action programmes are most successful with the inclusion of meaningful involvement of the target groups in design, planning, implementation and evaluation Outcomes and impacts

• There is lack of systematic monitoring of the effectiveness of positive action in terms of outputs and outcomes. Organisations are struggling to develop robust evidence, and tend to rely on ‘soft’ measures to assess the impact of positive action.

• Whilst on the whole, positive action measures are seen to be effective in raising awareness of equality issues in organisations, having real impact on minority groups, and improving an organisation’s image and reputation, their potential contribution to business success is less well recognised and is not always deemed to translate into better financial results.

• Positive action initiatives are largely time limited and not seen as long-term measures. Generally, the groups that benefit most from positive action initiatives are minority ethnic groups and women, and the least likely to be beneficiaries are LGBT and disabled people. Positive action in practice

• Examples of practice in organisations confirm confusion about the scope of positive action measures and the overlap with other complementary measures such as equality and diversity monitoring and impact assessment.

65 International perspectives on positive action measures

• There is an incredibly diverse range of activities that fall under the heading of positive action. Many countries focus on specific groups, perhaps at the expense of others, which might be a reflection of the particular context or ‘politics’ of that country.

• In practice, positive action measures tend to focus more on training and improving employment opportunities rather than service delivery.

• The introduction of positive action within organisations can produce some negative consequences or backlash- es such as negative stereotypes, stigmatisation, lack of proper oversight, dishonest behaviour and malpractice.

7.3. Recommendations

7.3.1. Research on positive action

• Research should be undertaken in an attempt to map the current situation of ‘disadvantage’ with regard to the different fields in which positive action can be applied, e.g. employment, education, housing, healthcare, etc. relating to these different grounds. This is essential and may need to use tools established in one field or ground of equality/discrimination, to examine parallel effects caused by a different ground of social exclusion.

• This research should be carried out at the national level, as it can also be expected that the situation with regard to different discrimination grounds will also vary across the Member States. These national mapping studies should provide the basis for any further policy review focusing on where (which fields and grounds) to allow positive action, and what (public) resources to direct towards particular forms of positive action.

• There needs to be research to assist courts which are called upon to establish whether a prior situation of disadvantage justifies the use of a particular positive action measure. The research should assist courts to identify the relevant questions or issues which should be resolved, and could result in a series of model questions (which could be adapted on a case-by-case basis) which the court would need to address. This research should be carried out on a European basis. The model questions could then provide a basic EC law framework for assessing ‘disadvantage’, but could also be added to in light of further national law requirements. An approach analogous to ‘cost-benefit analysis’ based on some measure of justice might be developed.

• Given that an organisation’s equality and diversity objectives can be hampered by a lack of understanding of the rationale for positive action (and the need to continue to make the case for and measure the impact of positive action), undertaking research around the economic advantages of positive action may prove beneficial.

• A comprehensive intervention study is necessary to develop a coherent model/theory for measuring success which would guide the type of positive action measures implemented.

• In fact, there is so little evaluation of ‘good practice’ in positive action, that development of a model for identifi- cation, evaluation and dissemination of ‘best practice’ would be advantageous.

• In light of the centrality of monitoring in promoting sustained positive action measures, organisations need to gather disaggregated data in key sectoral fields on all grounds of discrimination. This may not appear to be a radical or innovative suggestion, but in the light of the poverty of practice, we feel that while listed last in this catalogue of actions, it is of primary importance and would, for the majority of organisations and agencies, be innovative in practice.

66 7. Conclusions and recommendations

7.3.2. Law and policy development

• In order to foster a shared understanding of what is meant by positive action within the European Union, the European Commission should promote dialogue with civil society organisations and the social partners.

• Based on such dialogue, EU-level guidance on the meaning of positive action should be developed. This could be in the form of a non-binding legislative instrument, such as a Commission recommendation or a Council resolution. Alternative mechanisms could include a joint declaration by the social partners.

• In the introduction and revision of EC anti-discrimination legislation, it should be ensured that public, private or voluntary organisations who wish to engage in positive action are entitled to do so.

• Member States should revise national legislation where this prohibits or restricts the opportunity for public, private or voluntary organisations to take positive action.

• In order to evaluate the need for, and effectiveness of, positive action, data collection is required. Whilst respecting data protection legislation, Member States should ensure that organisations may engage in data collection where this is designed to facilitate and analyse positive action measures.

• The EU institutions and Member States are recommended to introduce legal duties to implement positive action measures where necessary to achieve full equality in practice.

7.3.3. Practice

European and national levels

• Create a European-level framework of understanding of positive action measures and define specific indicators of success in the implementation of these measures. The EC needs best-practice networks to support Member States in dealing with uncertainties and ensure parallel translation and application of the EU approach to positive action. These networks should operate at national and cross-sector levels to enable the sharing of ideas, approaches and activities, and encourage organisations to move from intention to action. This may help to move the focus from rhetoric to outcomes.

• Governments should undertake to educate the general public through ‘social marketing’ about positive action, in order to address widespread misunderstandings that appear to exist, and to facilitate the linking up of various stakeholders already engaged in such measures. Widespread awareness-raising campaigns of both the need for positive action measures for disadvantaged groups and the benefits of such measures for wider society will promote a wider acceptance and positive attitudes towards positive action.

• Develop clear strategies for identifying and managing the negative consequences of positive action (or its misuse and abuse). The role of the media and other robust communication approaches should be considered. Establish educational forums and networks to promote understanding and dialogue in relation to positive action.

• Government bodies at all levels should actively be encouraged to implement positive action programmes for disadvantaged groups in order to set an example for the rest of society in overall attitude and approach to such measures. Minimum operating standards for positive action application should be set by the EC with appropriate arrangements for reporting successes and challenges on an EC-wide basis. This might be underpinned by selective punitive or enforcement action against bodies failing to meet existing minimum targets for compliance with equality and human rights duties.

67 International perspectives on positive action measures

• Make available adequate financing through national government or EU funds to support complex programmes required to ensure effective implementation and evaluation of positive action. Such programmes might include those that promote intersectional and intersectoral approaches to non-discrimination.

• Develop tools to assist organisations to establish baseline data to facilitate positive action implementation and design robust strategies to support the evaluation of the effectiveness of measures taken. Adoption of an EU and a national action plan that identifies systems that need to be in place to ensure efficient and robust monitoring. Impact assessment tools may be deployed for this purpose.

Organisational level

• Maintain positive action as part of a broader normative change, supported by institutions with proper mentoring and training. Increase internal and external acceptability of positive action by raising awareness of the nature and benefits of positive action. Increase knowledge and practice of positive action, its benefits and its role within diversity strategies through a programme of education and training including seminars and events.

• Address positive action as an integral part of a wider organisational corporate mission, workforce planning and service development, working closely with the relevant governmental bodies. Integrate positive action within talent management, succession planning frameworks and wider employment and service development practices. This may require cross-departmental working in order to ensure a more coordinated approach.

• Adopt a more coherent and collaborative approach to the introduction of positive action between organisations. This collaboration will not only help to increase the acceptability of the programmes but may also help convince managers of the likely benefits of positive action, not least if other organisations are competitors. A strategy found effective in the United States is to encourage organisations to compete for recognition in equality and diversity (‘justice’) fields: if bodies are competing to excel in positive action, this creates healthy competition. Awards might be created and publicly presented to encourage this.

• Ensure involvement of members of minority groups in the development and evaluation of positive action measures. Individuals who have benefited from various positive action initiatives should be encouraged to work within the extension of such programmes, in order to increase representation amongst positive action implementers.

68 Bibliography

Bibliography Band, S. and Parker, A. (2002), Executive summary: ethnic minority undergraduate scheme (EMUS), available at: http://www.uel.ac.uk/nmc/mentoring/documents/CEDARReportUniversityofWarwick.doc Adam, K. (1997), ‘The politics of redress: South African style affirmative action’, The Journal of Modern African Baxter, C., Archibong, U., Giga, S. and Kular, R. (2008), Studies, 35(2), pp. 231–249. ‘Critical success factors in the implementation of positive action in the NHS UK’, International Journal of Diversity Adamson, J., Gooberman-Hill, R., Woolhead, G. and Do- in Organisations, Communities and Nations, 8, available at: novan, J. (2004), ‘“Questerviews”: using questionnaires in http://www.Diversity-Journal.com qualitative interviews as a method of integrating qualitative and quantitative health services research’, Journal of Health Bell, M. (2005), ‘Positive thinking: duties to promote equal- Services Research and Policy, 9(3), pp. 139–145. ity and the European Union’, in: Cinar, D., Gürses, H. and Strasser, S. (eds), Auf dem Weg zu einer Politik der “dif- Agocs, C. (2002), ‘Canada’s employment equity legislation ferenzierten Gleichheit”: Theoretische Reflexionen über die and policy, 1987–2000: the gap between policy and prac- Anti-Diskriminierungspolitik der Europäischen Union — tice’, International Journal of Manpower, 23(3), pp.256–276. Endbericht, Vienna: Europäisches Zentrum für Wohlfahrt- spolitik und Sozialforschung, pp. 110–128. Anderson, V. (2004), ‘Women managers: does positive action training make a difference?’, Journal of Management Blom, J., Fitzpatrick, B., Gregory J., Knegt R. and O’Hare, U. Development, 23(8), pp. 729–740. (1995), ‘The utilization of sex equality litigation procedures in the Member States of the European Community: a com- Archibong, U. and Burford, B. (2007), ‘The cultural un- parative study’, Brussels: European Commission, DG-V. derstanding in leadership and management — Phase I’, in: Marshall, S. (ed.) Strategic leadership of change in higher Bowes, A. and Sim, D. (n.d.), ‘Reviewing positive action education: what’s new? London: Routledge. programmes in the UK’, retrieved on 30.5.2008 from: http:// www.urbanicity.org/FullDoc.asp?ID=470 Archibong U., Bucktrout, A., Giga, S., Ashraf, F., Baxter, C. and Johnson, M. (2006a), ‘A concept analysis of positive ac- Brew, K. and Garavan, T. N. (1995), ‘Eliminating inequaltiy: tion in health and education’, Diversity in Health and Social women-only training’, part 2. Journal of European Industrial Care, 3(4), pp. 223–243. Training, 19(9), pp. 28–32.

Archibong, U., Giga, S., Ashraf, F., Bucktrout, A., Jack- Burrows, N. and Robinson, M. (2006), ‘Positive action for son, H., Baxter, C. and Johnson, M. (2006b), Positive ac- women in employment: time to align with Europe?’, Journal tion in education and health project report, University of of Law and Society, 33(1), pp.24–41. Bradford. Carter, J. (2000), ‘New public management and equal op- Archibong, U., Giga, S., Ashraf, F., Bucktrout, A., Jackson, portunities in the NHS’, Critical Social Policy, 20(1), pp. H., Baxter, C. and Johnson, M. (2007), ‘Representational 61–83. diversity: the experience and interpretation of positive action, International Journal of Diversity in Organisations, Centre for Strategy and Evaluation Services (2007), `Non- Communities and Nations, available at: http: //www.diver- discrimination mainstreaming - instruments, case studies sity_journal.com and way forwards’ (Brussels: Commission).

Bacchi, C. (2004), ‘Policy and discourse: challenging the Chater, R. E. J. and Chater, C. V. (1992), ‘Positive action: construction of affirmative action as preferential treatment’, towards a strategic approach’, Women in Management, 7(4), Journal of European Public Policy, 11(1), pp.128–146. pp. 3–14

Bagilhole, B. and Stephens, M. (1999), ‘Management re- Dainty, R. J. A., Neale, H. R. and Bagilhole, M. B. (1999), sponses to equal opportunities for ethnic minority women ‘Women’s careers in large construction companies: expec- within an NHS hospital trust’, Journal of Social Policy, 28(2), tations unfulfilled?’, Career Development International, pp. 235–248. 4(7), pp. 353–357.

69 International perspectives on positive action measures

Davey, L. and Davidson, J. (2000), ‘The right of passage? Gijzen M.H.S. (2006), Selected Issues in Equal Treatment The experiences of female pilots in commercial aviation’, Law: A multi-layered comparison of European, English Feminism & Psychology, 10(2), pp. 195–225. and Dutch law, Thesis, Intersentia.

David, M. and Sutton, C. D. (2004), Social research: the ba- Glastra, F., Schedler, P. and Kats, E. (1998), ‘Employment sics, London: Sage. equity policies in Canada and the Netherlands: enhancing minority employment between public controversy and Davies J. and Davies W. (2000), “Reconciling Risk and the market initiative’, Policy and Politics, 26(2). Employment of Disabled Persons in a Reformed Welfare State”, 29 Industrial Law Journal, 347-377. Government Equalities Office (2008),Framework for a fairer future — the Equality Bill, London: The Stationary Office. De Vos, M. (2007), Beyond formal equality: positive action under Directives 2000/43/EC and 2000/78/EC, Brussels: Groschi, S. and Doherty, L. (1999), ‘Diversity management Directorate-General for Employment, Social Affairs, and in practice’, International Journal of Contemporary Hospital- Equal Opportunities. ity Management, 11, pp. 262–268.

Department of Communities and Local Government Heaton, N. and Teague, P. (1997), ‘Towards fair employ- (2007) Discrimination Law Review - A framework for fair- ment in Northern Ireland?’, International Review of Applied ness: proposals for a single equality bill for Great Britain. Economics, Taylor and Francis Journals, 11(2), pp. 263–85. London: The Stationery Office Hepple, B., Coussey, M. and Choudhury, T. (2000), Equal- Department of Labour Republic of South Africa (2008), ity: a new framework. Report of the independent review ‘Basic guide to affirmative action May 2008’, accessed from: of the enforcement of UK anti-discrimination legislation, http://www.labour.gov.za/legislation/acts/basic-guides/ Oxford: Hart Publishing, 2000. basic-guide-to-affirmative-action Holzer, H. and Ihlanfeldt, K. (1998), ‘Customer discrim- Dhami, R. S., Squires, J. and Modood, T. (2006), Developing ination and the employment outcomes of minorities’, positive action policies learning from the experiences of Eu- Quarterley Journal of Economics, 113(3), pp. 833–65. rope and North America, Leeds: Department of Work and Pensions. Holzer, H. and Neumark, D. (2000), ‘Assessing affirmative action’, Journal of Economic Literature, 38, Septem- Disability Discrimination Act (2005), accessed on 30.5.2008, ber, pp. 483–568. available at: http://www.opsi.gov.uk/Acts/acts2005/ukp- ga_20050013_en_1 Iganski, P., Mason, D., Humphreys, A. and Watkins, M. (2001), ‘Equal opportunities and positive action in Echiejile, I. (1994), ‘Empowering disadvantaged employ- the British National Health Service: some lessons from ees’, Empowerment in Organizations, 2(1), pp. 31–37. the recruitment of minority ethnic groups to nurs- ing and midwifery’, Ethnic and Racial Studies, 24(2), European Commission (2007), `The application of Direc- pp. 294–317. tive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of Iles, P. and Hayers, K. P. (1997), ‘Managing diversity in racial or ethnic origin’ COM (2006) 643. trans-national project teams: a tentative model and case study’, Journal of Managerial Psychology, 12(2), pp. 95–117. European Commission, `Comparative study on the collection of data to measure the extent of discrimination Johns, N. (2005), ‘Positive action and the problem of within the United States, Canada, Australia, the United merit: employment policies in the National Health Serv- Kingdom and the Netherlands’ (Luxembourg: Office of ice’, Critical Social Policy, 25(2), pp. 139–163. the Official Publications of the European Communities, 2004). Kottke, J. L. and Agars, M. D. (2005), Understanding the processes that facilitate and hinder efforts to advance wom- Fielding, N. (1993), ‘Interviewing’ in: Gilbert, N. Research- en in organisations, Career Development International, 10(3), ing social life, London: Sage. pp. 190–202.

70 Bibliography

Liff, S. (1999), ‘Diversity and equal opportunities: room PATH: Yorkshire Positive Action Training, accessed on for a constructive compromise?’, Human Resource Man- 30.5.2008, available at: http://www.pathyorkshire.org.uk/ agement Journal, 9(1), pp. 65–76. case_studies.html

Makkonen T. (2006), `European handbook on equality Payne, S. C. and Huffman, A. H. (2005), ‘A longitudinal ex- data’ (Luxembourg: Office of the Official Publications of amination of the influence of mentoring on organisational the European Communities, 2006). commitment and turnover’, Academy of Management Jour- nal, 48(1), pp. 158–168. Mathur-Helm, B. (2005), ‘Equal opportunity and affirmative action for South African women: a benefit or bar- Prior, L. (2003), Using documents in social research London: Sage. rier?’ Women in Management Review, 20(1), pp. 56–71. Refugee Council (2006), ‘Training for refugees and asylum- McCrudden, C. (2007), ‘Positive action: definition, types, seekers’, available at: http://www.refugeecouncil.org.uk/ aims and justifications’, keynote presentation. ‘Equal op- refugeecouncil/training/tes.htm portunities for all: what role for positive action?’, 2007 annual conference of the action programme to combat Secker, J. (2001), ‘Athena report and action plan’, available discrimination, Rome, 23 and 24 April. at: http://www.admin.ox.ac.uk/eop/gender/athena.shtml

McCrudden, C. (1986), ‘Rethinking positive action’, In- Sheffield, J., Hussain, A. and Coleshill, P. (1999), ‘Organisa- dustrial Law Journal, 15(1), pp. 219–243. tional barriers and ethnicity in the Scottish NHS’, Journal of Management in Medicine, 13(4), pp. 263–284. McCrudden, C., Ford, R. and Heath, A., (2004), ‘Legal regulation of affirmative action in Northern Ireland: an Shields, M. A. and Wheatley Price, S. (2001), ‘Language flu- empirical assessment’, Oxford Journal of Legal Studies, ency and immigrant employment prospects: evidence from 24(3), pp. 363–415; doi:10.1093/ojls/24.3.363. Britain’s ethnic minorities’, Applied Economics Letters, 8, pp. 741–745. Miller, J. and Rowney, J. (1999), ‘Workplace diversity management in a multicultural society’, Women in Management Shifrin, T, (23.6.2004) Prem Singh interview, available at: Review, 14(8), pp. 307–315. http://societyguardian.co.uk/NHSstaff/story/0,,1244697,00. html Murphy, P. (1993), ‘Positive action for women in Northern Ireland’, Women in Management Review, 8(1), pp. 15–27. Silverman, D. (2001), Interpreting qualitative data: methods for analysing talk, text and interaction (2nd ed.), London: Sage. Museums Association: ‘Setting up a traineeship’, cited on 30.5.2008, available at: http://www.museumsassociation. Spencer, L. (1989), Winning through participation: meeting org/14276 the challenge of corporate change with the technology of participation, Iowa: Kendall/Hunt. NHS Employers (2005), ‘Positive action in the NHS’, available at: http://www.nhsemployers.org/excellence/excel- Stanfield, R. B. (2002), The workshop book: from individual lence-494.cfm creativity to group action (ICA series), Toronto: Canadian Institute of Cultural Affairs. Office of the First Minister and Deputy First Minister in Northern Ireland (OFMDFMNI) (2006), ‘Affirmative ac- Stephanopolous, G. and Edley, C. Jr. (1995), ‘Affirmative action’, available at: http://www.ofmdfmni.gov.uk/index/ tion review’, report to the President, Washington, DC. equality/fair-employment/affirmative_action.html Tashakkori, A. and Teddlie, C. (1998), Mixed methodology: Osborne, R. and Shuttleworth, I. (eds) (2004), Employment combining qualitative and quantitative approaches. London: equality: a generation on, Belfast: Blackstaff. Sage.

Parker, S., Pascall, G. and Evetts, J. (1998), ‘Jobs for the girls? Thomas, A. and Jain, H. (2004), ‘Employment equity in Cana- Change and continuity for women in high street banks’, da and South Africa: progress and propositions’, International Women in Management Review, 13(4), pp. 156–161. Journal of Human Resources Management, 15(1), pp. 36–55.

71 International perspectives on positive action measures

Thomas, D. and Ely, R. (2002), ‘Making differences mat- Waddington L. and Bell M (2001), “More Equal than Oth- ter: a new paradigm for managing diversity’, in: Harvard ers: Distinguishing European Union Equality Directives”, Business Review on Managing Diversity, Harvard Business Vol. 38, Common Market Law Review, 587-611. School Press, pp. 33–66. Ward, J. (2006), ‘Tomorrow’s Planners initiative promotes Vries, S. and Pettigrew, T. (1994), ‘A comparative perspective diversity and equality’, Planning, 1676, p. 22. on affirmative action: Positieve Aktie in the Netherlands’, Ba- sic and Applied Social Psychology, 15(1–2), pp. 179–199. Welsh, C., Knox, K. and Brett, M. (1994), Acting positively: positive action under the Race Relations Act 1976, Research Waddington L. (1996), “Reassessing the Employment of Peo- Series No 36, London: Employment Department. ple with Disabilities in Europe: From Quotas to Anti-discrimination Laws”, 18 Comparative Labor Law Review, 62-101.

Waddington L. (2000), “Tweede-generatie richtlijnen Gelijke Behandeling: de nieuwe Richtlijn inzake gelijke behandeling ongeacht ras of etnische afstamming en de Kaderrichtlijn gelijke behandeling in arbeid en beroep”, 12 Sociaal Recht, 357-362.

European Commission International perspectives on positive action measures A comparative analysis in the European Union, Canada, the United States and South Africa

Luxembourg: Office for Official Publications of the European Communities 2009 — 72 pp. — 21 x 29.7 cm

ISBN: 978-92-79-11156-3 DOI: 10.2767/13963

This study was undertaken to help the European Commission develop a framework for better understanding the role that positive action measures can play in practice in preventing or remedying discrimination. In addition, it offered insight into the kind of practical positive action measures already being taken in the EU (and in the EFTA-EEA countries), as well as the possible costs and benefits of the positive action measures. The study also sought to examine how legal frameworks, policies and practices of positive action in the EU compared with Canada, the USA and South Africa.

This brochure is available in printed format in English, French and German.

How to obtain EU publications

Publications for sale: • via EU Bookshop (http://bookshop.europa.eu); • from your bookseller by quoting the title, publisher and/or ISBN number; • by contacting one of our sales agents directly. You can obtain their contact details on the Internet (http://bookshop.europa.eu) or by sending a fax to +352 2929-42758. Free publications: • via EU Bookshop (http://bookshop.europa.eu); • at the European Commission’s representations or delegations. You can obtain their contact details on the Internet (http://ec.europa.eu) or by sending a fax to +352 2929-42758. KE-81-09-550-EN-C

Are you interested in the publications of the Directorate-General for Employment, Social Affairs and Equal Opportunities?

If so, you can download them at http://ec.europa.eu/employment_social/emplweb/publications/index_en.cfm

or take out a free online subscription at http://ec.europa.eu/employment_social/sagapLink/dspSubscribe.do?lang=en

ESmail is the electronic newsletter from the Directorate-General for Employment, Social Affairs and Equal Opportunities.

You can subscribe to it online at http://ec.europa.eu/employment_social/emplweb/news/esmail_en.cfm

http ://ec.europa.eu/social