n1odernisin COLL P B'b60 FABIAN SOCIEYY BLPES FABIAN PAMPHLET 585 21 0453621 2 This pamphlet, like all publications of the Fabian Society, represents not the collective views of the Society but only the views of the authors. This publication may not be reproduced without express permission of the Fabian Society. Published March 1998 by The Fabian Society Printed by The College Hill Press Ltd 37 Webber Street, London SE1 8QW ISBN 0 7163 0585 2 ISSN 0307 7535 The Fabian Society, 11 Dartmouth Street, London SW1 H 9BN Telephone 0171 222 8877 Fax 0171 976 7153 Email fabian-society@geo2 .poptel.org.uk FABIAN SOCIE'I'Y The Fabian Society The Fabian Society is Britain's senior think tank. Concerned since its foundation with evolutionary political and economic reform and progressive social change, the Fabian Society has played a central role for more than a century in the development of political ideas and public policy on the left-ofcentre. The Society is affiliated to the Labour Party but is editorially and organisationally independent. In recent years the Society's work on the modernisation of the Labour Party's constitution and its analysis of changing political attitudes have played a significant part in the renewal of the party's public appeal. Today the Fabian Society seeks to help shape the agenda for the medium and long term of the new Labour Government. Analysing the key challenges facing the UK and the rest of the industrialised world in a changing society and global economy, the Society's programme aims to explore the political ideas and the policy reforms which wi ll define the left-of-centre in the new century. Through its pamphlets, discussion papers, seminars and conferences, the Society provides an arena for open-minded public debate. The Fabian Society is unique among think tanks in being a democratically- constituted membership organisation. Its five and a half thousand members engage in political education and argument through the Society's publications, conferences and other events, its quarterly journal Fabian Review and a network of loca l societies and meetings. The Modernising Britain Programme Th e Prime Minister has made the modernisation of Britain a key theme of the new Labour administration. The concept of 'the modern ' has a long and complex history. The Fabian Society's programme seeks to examine what modernisation might mean -politically and culturall y -as Britain enters a new century. In some fields the idea of modernisation is clear. In key areas of British li fe there are many old institutions which no longer meet the chall enges of today's society. One element of the Fabian Society's programme focuses on how such institutions can be reformed. But in other fields the idea of modernisation remains open for debate and challenge. The programme seeks to ask what kind of country Britain should be in the 21st century, exploring our cultural identities and social structures. it aims to explore different visions of modernity and their implications for public policy. High Time or High Tide for Labour women Contents 1. 101 is not enough 2. From Quotas to all-women shortlists 5 3. The current context 11 4. Selection mechanism to promote women's representation 1 7 5. Equal opportunities procedures 24 6. Conclusion 29 Footnotes 31 joni Lovenduski is professor of politics at Southampton University and the author of numerous book on women and politics. Maria Eagle was elected a Labour Member of Parliament for Liverpool Garston in 1997. 1. 101 is not enough The 1997 general election was as notable for its effect on the representation of women as it was for the fortunes of the Labour Party. labour won 418 of 659 seats in the House of Commons and achieved the largest majority it has ever had, better even than the 1945 landslide. The results for Labour women were as dramatic as those of their party. As constituency after constituency declared its results, the election of one Labour women after another was announced. The total number of women MPs doubled from 60 to 120. The number of Labour women MPs almost trebled, rising from 38 to 101. With a women's representation of 18 per cent Britain for the first time moved into the top 25 of the world league table of women's parliamentary representation. The scale of the change was captured a few days later in the widely published photograph of T ony Blair surrounded by Labour women MPs, a photograph that provided one of the most vivid (and probably most enduring) images of the election. Table 1: Women in Parliament 1979-1997 Ye ar Lab o ur C o n. Lib Oth e r T o ta l % D e m * 19 79 1 1 8 1 9 2 .9 198 3 10 1 3 23 3 .5 198 7 2 1 1 7 2 41 6.2 1992 3 7 2 0 2 6 0 9.2 1997 1 01 13 3 3 12 0 18 .2 * including Liberai/ SDP/SLD in appropriate vears The increase in women's representation did not happen by accident. lt was a Labour phenomenon, the result of Labour Party efforts. The minor landslide of the Liberal Democrats returned only three women out of 46 while the Conservative complement of women dropped from 20 to 13. On the morning after the election Labour became the first party in British history to offer a respectable pari iamentary representation of women, some 24 per cent of the new Parliamentary Labour Party (PLP). This result was greeted with some approval by the media and the British public. The party looked more modern and representative, the long overdue women's breakthrough was thought to have arrived and many were confident that the increase of women would continue and with it the face of British politics would change. Today the presence of Labour women in Britain's elected assemblies stands at record levels. Women are 24 per cent of the PLP, 25 per cent of Labour councillors and 21 per cent of Labour MEPs. But these figures sti 11 fall well short of the party's goal of fair representation for women and men at all levels. The challenge ahead Labour's achievement was not an indication of a popular enthusiasm within the party for the selection of women candidates. lt was largely the result of a se lection mechanism-compulsory all-women shortl ists-which gave local parties no choice but to select women. This policy began in 1993, was controversial throughout its lifetime, and was ultimately dropped in January 1996. These two years saw the selection of many women candidates. In this pamphlet we argue that the party would never have made such progress without the mechanism of all-women shortlists. Moreover the representation of Labour women may well fall over the next electoral cycle. Current levels of women's representation were achieved when Labour was at a peak of electoral success. Not only was the 1997 lands I ide at Westminster a record victory, but the 1994 European elections were Labour's best ever and Labour has a record number of local councillors, whilst the Conservatives local government base virtually disappeared during the 1980s and 1990s. Likely electoral developments mean that opportunities for new candidates at Westminster, in Europe and at local level, will arise only where retirements take place, a prospect that suggests slow change even if Labour remains successful. Historically there are few Westminster retirements and seats replacing retiring MPs are the hardest fought because they are the safest. At the next general election, by almost all predictions, Labour will lose seats and there wi ll be fewer retirements because of the high proportion of youthfu l Labour MPs. The introduction of a new electoral system for the European Pari iament will mean a substantially reduced proportion of Labour seats. At local level, reaction against the governing party commonly reduces its share of council seats. Only in the newly constructed assemblies in Scotland, Wales, London and elsewhere wi 11 Labour have the opportunity to place new women representatives in a substantial number of vacant seats. All of these factors suggest that only with full commitment from all levels of the party will it be possible to continue Labour's progress toward the fair representation of women. The challenge facing Labour is to devise the mechanisms to secure that commitment. In this pamphlet we consider the implications of this challenge. We first discuss the fundamental question of why Parliament, or any other elected assembly, needs women. Second, we look at the recent history of Labour policy on women's representation. Third, we review the context in which the party needs to meet the new challenge, in terms of the law, electoral systems and the party's culture. Finally, we set out a series of proposals for policies to consolidate and develop current levels of women's representation and argue that a change in the law is necessary if progress is to continue. Why equal women's representation? Why should women have equality of political representation? How would society benefit from electing a balance of women and men to the legislature? And why should a pal itical party benefit from seeking to achieve this? The most important benefit would be the knowledge that the party was behaving in a just manner. lt is no longer acceptable to exclude women from any area of public life. At one time it was thought that women were adequately represented by the male head of household who was vested with the authority to rule and represent his family. The notion that women had interests separate from their families was a controversial one. Today it is better appreciated that women have rights as citizens and that such rights entitle them to participation in government. The air of excitement around the election of Labour's 101 women MPs came partly from the widespread hope that politics would change. This applies both to the types of policies which are implemented and to the culture, reputation and practices of politics itse lf. Many people believe that increasing the presence of women in parliament will, once a substantial minority has been achieved, help build a political process which is more constructive and less adversarial. Crucially for women in the population, it should also bring changes in policy which benefit women, as decision makers have more knowledge of women's issues and perspectives. Research in countries where women have built up a significant political presence has demonstrated changes in political style, discourse, decision making and efficiency, adding up to a greater 'women-friendliness' of public life. Such enhanced 'women-friendliness', which offers a politics that is more people friendly in general, may be particularly important in Britain. Public distaste for the adversarial conventions of pal itics and distrust for pal iticians is at record levels. As shown in the Fabian pamphlet of 1987 by Deborah Mattinson and Patricia Hewitt-herself one of the new 1997 intake-voters are more likely to trust a woman politician than a man. 1 Allegations of sleaze against women politicians have been rare. A balanced legislature is one which is more likely to gain the trust of the population. lt is now quite widely accepted that in a modern democracy parliament should properly represent the whole population. Few people would accept that a parliament which was 90 per cent male could adequately do this. Yet until 1 May this was the situation in Britain, and even now 82 per cent of parliamentarians (far more when the House of Lords is included) are male. For many years British political parties have paid lip service to the idea of improving women 's representation, but that rhetorical commitment led only to slow progress. Although women, both inside and outside the parties, have long argued for better representation in politics-on the basis of equity or in order to see different policies delivered-it was these more pragmatic arguments which finally pushed Labour into taking action. Research by Labour's Shadow Communications Agency carried out after the 1987 defeat showed that voters found masculine party images old fashioned and unattractive, and they believed that Labour was more male dominated than other parties.2 This was particularly prevalent amongst women voters. Such images may have been linked to women's resistance to vote Labour. Pippa Norris showed that between 1945 and 1983 a gender gap existed in the British electorate whereby women were more likely than men to vote Conservative. The gender gap was so pronounced that, had women not had a vote, there would have been no Conservative governments between 1945 and 1979. After 1979 the gap steadily closed to insignificance in 1987, only to reopen in 1992.3 Following the 1992 defeat, senior party figures were more wi 11 ing to concede greater women's representation as part of a strategy to win the next election. lt was widely perceived that party reform entailed the adoption of new images and the construction of a wider popular appeal. Women were crucial to both objectives. A concerted effort was made to feminise the party's image and to attract women voters. At the 1 997 election the gender gap once again disappeared in all but the oldest age groups. Among younger women a 'reverse' gap in favour of Labour was apparent. Maintaining its appeal to women is one of the most important challenges Labour faces as it plans its strategy for future elections. Initiatives in the run up to the 1997 election caught other British political parties unaware, giving Labour an advantage that will not be repeated. The support of women cannot be taken for granted. Other parties can be expected to compete vigorously for the prize of women's votes. 2. From quotas to all-women shortlists: 1988-98 The Labour Party and quotas By the end of the 1980s a number of feminist voices in different parts of the Labour Party were campaigning for power for women . Similar campaigns were conducted in socialist parties around the world, co-ordinated by Socialist International Women. In many countries the parties operated quotas for women, both within their party structures and in the selection of pari iamentary candidates. The first step on this route was taken by the British Labour Party in 1989 when, following a proposal from the Tribune group, a quota was introduced in the Shadow Cabinet. The quota required all MPs to cast votes for three women in order for their ballot paper to be valid . The size of the shadow cabinet was increased by three from a membership of 15 to 18 so that no male ' losers' were created by the new policy. In 1989 Labour Party conference also accepted the principle of quotas to improve women's representation at every level of the party. A Fabian pamphlet the following year by Rachel Brooks, Angel a Eagle and Clare Short called for compulsory quotas throughout the party, including all-women shortlists for parliamentary selection.4 The 1990 conference began the implementation of many of their proposals, agreeing that quotas should be set at 40% throughout the party and calling on the NEC to bring forward proposals to achieve this. This conference also set a time limit, of three general elections, to bring the proportion of women in the PLP up to 50 per cent, but no means to achieve the target were established . In 1991, rule changes were passed by conference to achieve internal party quotas. The changes included a 40 per cent quota for women on the NEC, to be phased in by 1995. There would be a quota of conference delegates which ensured that local parties sent equal numbers of men and women to national and regional conferences. Within local parties quotas would ensure that roughly half the officers at branch and constituency level were women . Response to the policy varied. There was great support in many parts of the party but considerab le opposition in traditional Labour areas. The significant exception to a widespread initial acceptance of quotas was in the selection of pari iamentary candidates. On this matter policy remained weak, consisting of a rule-dating from 1988-that, where a woman had been nominated, at least one woman had to be included on the short! ist for constituency selection . The rule was accompanied by regular exhortation by party leaders and officials that more women shou ld be selected. More women were selected, but as Table 2 shows, while the number of women nominated increased at every election, most nominations were in unwinnable or marginal seats. In 1992 Labour returned only 37 women, some 14 per cent of the PLP, despite fielding 138 women candidates. Of the 20 Labour MPs who retired in safe seats that year, 18 were replaced by men. Table 2: Labour women MPs and candidates 1979-1992 1979 52 11 21.2 1<:ID 78 10 128 1~7 92 21 228 1CJ92 138 37 26.8 lt was clear that, without some mechanism to secure improvement, progress in women's representation would continue only slowly. But designing a suitable mechanism proved difficult. The British first past the post, single member constituency electoral system is characterised by the predominance of safe seats. lt awards a huge advantage to incumbent MPs. At each election only a few retirements take place and only a few vacancies occur. Most of the turnover occurs in marginal seats, a few of which change hands at each election. All-women shortlists Attention therefore focused on the idea of all-women short! ists. This proposal already had a long history. For many years the Labour Party Women's Conference had voted that progress should be achieved by insisting that all- women shortlists were required in all seats which Labour was likely to win with a small electoral swing or where Labour MPs were retiring. These proposals were repeatedly voted down at Annual Conference. In the early 1990s a compromise was found: to apply all-women shortlists in half of all winnable seats. The policy would apply both to seats where Labour MPs retired and to target marginals. Because it allowed men half of the new seats which came available, ensuring that popular male candidates who had fought seats previously were protected, the decision was felt to be fair. Nevertheless it was still controversial and it was won only with the full backing of John Smith, then Leader of the Party. In one of the most hard- fought debates of recent years the policy was approved by conference in 1993. The mechanism was part of the package of parliamentary selection reform which included the move to one member one vote (OMOV). The means of deciding which half of the seats shou ld have all-women shortlists was critical and extremely sensitive. The method chosen was to require half the seats in each region to adopt all-women shortlists, with the decision on seats to be sought by consensus at regional level meetings of officers from all the constituencies concerned. Crucial to the success of the policy was that internal party quotas already required half the constituency officers to be women . There was hosti I ity from many men in the party, with some-i ncl ud i ng members of parliament-taking up the frequent opportunities offered by the press to 'rubbish' the policy. Ann Carlton, the wife of Labour MP Denzil Davies, set up the campaign group 'Labour supporters for Real Equality', which was a small group but received considerable press attention. There were problems reaching consensus in the party's North West region. In the south the Slough constituency was forced by the NEC to accept an all-women shortl ist when the local party failed to agree. Nevertheless, by jan uary 1996 consensus meetings had been held in all regions and most parliamentary candidates in key target seats were in place. Half of them (33) were women chosen from all-women shortlists. In addition two women were selected in this way to replace retiring Labour MPs in 'safe' seats. However, the mechanism was brought to an abrupt end when two disgruntled male aspirants took the party to court. On 8 January 1996 the Leeds Industrial Tribunal Uepson and Dyas-EII iott v The Labour Party and Others) declared that selection procedures of political parties (normally exempt from the provisions of the Sex Discrimination legislation) faci I itate access to employment and that all-women shortlists contravened the Sex Discrimination Act. Anxious to compiP.te its selections in good time for the general election and concerned not to jeopardise the positions of women already selected under the policy, the NEC decided not to appeal against the decision . Instead a working party was established to identify effective and legal ways to maximise the number of women candidates nominated and selected in the remaining vacant Labour seats. Women did continue to be selected from open I ists after the tribunal decision, but the numbers were noticeably reduced . The success of all-women shortlists Although controversial, all-women shortl ists were highly successful. The real effect of the policy was not so much that it led to an increase in the number of women candidates-they had been increasing steadily for the last 5 elections-but that women were selected for winnable seats. Thus in 1997, Labour women candidates increased their success rates to a dramatic 64 per cent as 101 of 158 women candidates were elected. Women were 24.6 per cent of candidates and 24.2 per cent of elected Labour MPs. To see the extent to which all-women shortlists were responsible for this, it is necessary to distinguish between five types of seats: Labour incumbent seats, Labour retirement or vacant seats, key seats, unexpected or surprise Labour gains and unwinnable seats. As Table 3 shows, the party's biggest success-which was responsible for the largest number of gains-was in selecting women as candidates in half the 'key' or marginal seats, resulting in 43 new women MPs. In total, 33 of them were selected from all-women shortlists. Of th e remainder, 'three constituencies had already agreed to use all-women shortlists at the time of the tribunal ruling, and went on to select women candidates. Three constituencies-Gravesham, Great Yarmouth and Leeds North East-had initially agreed to all-women shortlists but went on to se lect men. Table 3: Labour candidates by type of seat women men total %women Returned Labour incumbents Labour retirements Key seats Unexpected gains Total MPs Unwinnable seats 36 11 43 11 101 57 199 21 42 55 317 166 235 32 85 66 418 223 15.3 34.4 50.6 16.7 24.2 25.6 Total candidates 158 483 641 24.6 Labour did rather less well at electing women to replace retiring MPs. In these safest of seats only 11 women were elected from amongst the 32 seats available. Before January 1996, two of these constituencies se lected candidates from all-women shortlists. Two constituencies which had opted for all-women short! ists went on to select women after the tribunal. Five women were selected from short! ists drawn up by the NEC by-election panel, which in two cases itself drew up an all-women shortlist. Therefore only two women selected in Labour-held seats were chosen from totally open shortlists -no more than in 1992. The increase in Labour women MPs resulted largely from the party's remarkable success in winning 85 key seats. it is true that an additional 66 seats which were not on the target seat I ist were won. But as Table 3 shows, it is a myth that this is where Labour women won. In fact only 11 out of 66 unexpected gains (16.7%) were made by women, the worst success rate for se lecting women of any of the types of seats. In these seats no quota mechanism applied but selections were still hard fought because local parties thought they had a chance of winning. Ironically, if Labour had won only the Table 4: Women selected by type of seat and means of short! isting key seats, the proportion of women would be higher, at 90 MPs out of 352 V\Dmen selected through Type of seat all V\Dmen short! ist open short! ist byOO:: tim L'l'I"E total V\Dmen selected total men selected % V\Dmen candidates f\.APs rv'a rv'a rv'a 36 235 15.3 Safe 2 4 5 11 21 34.4 Target 33 10 0 43 43 so Surprise gain 0 11 0 11 55 16.7 Unwinnable 0 57 0 57 166 25.6 Total 35 82 5 158 483 24.6 (25.6%). lt is also noteworthy that only 57 out of 223 of Labour's unsuccessful candidates were women. This is a good result in one sense-that the vast majority of Labour's women candidates were elected. This compares starkly with the Liberal Democrats who selected 122 women, of whom only three won seats. However, it is also an indication of the reluctance of local parties to select women candidates even for unwinnable seats when they are not required to do so. Table 4 summarises how women candidates were selected for all types of seat, and includes also women MPs who stood again. Labour's hospitality to women varies by region . The seats that were required to select from all-women short! ists were determined at consensus meetings organised on the basis of party regions. Compulsion was more sparingly used than press coverage suggested and resistance was greater in some regions than others. Table 5 shows the number of women MPs and new women MPs and their proportion of the total by region The South East region offers a particularly stark example of what happened with open short! ists. Selections were not complete in the South East region by the time of the industrial tribunal, by which time only three all-women shortlist selections were complete. Two constituencies dropped their all-women short! ists after the tribunal and went on to select men. Limited implementation of the all-women short! ist policy in the region, combined with the fact that there were many surprise gains in the South East outside the key seats, resulted in only four women being elected from a total of 23 new Labour MPs-three of them from all-women shortlists. This represents just 17.4 per cent of the total. Other evidence suggests a similar pattern. The number of women serving as Table 5: Labour MPs by region Labour councillors is an indication of local parties' wi 11 ingness to select women . Local government selections are made from panels of candidates, with ward Labour parties selecting their candidates from the panel drawn up by local parties. Despite exhortation over many years to select more women as councillors, and despite guidance that panels and local shortlists should contain 50 per cent women, the proportion of women Labour councillors remains low. There are many authorities where there are no women Labour councillors, or where the proportion of women is below 10%. Aside from its intrinsic importance in reflecting the community, women's representation at local government level is also an indication of the size of the pool of women who are available to stand for office at more senior levels, At least 65 per cent of Labour MPs and MEPs have been councillors at some point in the past. In summary, the available evidence supports the view that constituency selectors are unlikely to nominate enough women to make a difference to the proportion of women in elected office, unless there is some compu lsion to do so. However, although quotas remain in force for internal Labour Party positions, the industrial tribunal decision in 1996 brought an abrupt end to mechanisms to promote women at Westminster. Uncertainty about the legal position has been allowed to halt the party's programme. Region All MPs New MPs women Itotal 1% women Itotal South West 10 21 48 8 16 North West 18 60 30 12 28 Central 13 44 30 12 29 Greater London 17 57 30 7 26 West Midlands 12 45 27 7 16 South East 5 27 19 4 23 North and Yorkshire 13 75 17 7 23 Scotland 9 56 16 5 13 Wales 4 34 12 3 8 3. The current context Before exploring what action the party might take to continue progress on women's representation , it is necessary to examine briefly the context in which such policy must operate. The context is characterised by continuous change. it is now nine years since quotas were first accepted in principle by the Labour Party. The intervening years of I ivi ng with the policy of quotas on internal positions has affected both culture and decision making in the party. The Leeds industrial tribunal decision has also set a new context-one where parties need to consider the possible legal implications of any action under employment law. Finally, existing selection procedures are being overtaken by changes in the electora l system which require new procedures to be found. Party culture it is clear from the evidence above that women's representation at Westminster did not improve as a result of a spontaneous change in party culture which made selectors more open to women candidates. it is impossible to know what the result of the election wou ld have been without all-women shortlists, but it is safe to assume that fewer women wou ld have been selected. In safe seats where Labour MPs retired there might have been no more than tWo new women se lected, which wou ld have resulted in no net gain as two of the retiring MPs were women. In seats where all-women shortlists did not apply, the decisions taken indicate that men remained the preferred candidates. However, the all-women shortlists process itself may have had some impact on party culture. In seats which adopted all-women shortlists, selectors were pleasantly surprised by the quality of women candidates, with many commenting that they were of higher calibre than the men. The controversy surrounding the policy generated debate about the need for more women candidates, and the election of the 101 women may have normalised the image of women as potential politicians. In recent years conference debates on the need to improve women's representation have become uncontroversial, in stark contrast to those of the early 1990s. Quotas have become a well established part of the party culture, at least as far as internal party positions are concerned. And the fact that so man y women now hold positions throughoutthe party-as constituency officers, conference delegates and on the NEC-affects decisions about women's representation . The recent decisions on selection processes for the Scottish parliament and Welsh assemb ly-described below-are a case in point. These appear to have been uncontroversial for an NEC on which half the members are women. However, behind this pub I ic face, at local level, a different story is sometimes told. Already constituencies in Wales are resisting the decision Table 6 Prop ortion Of Labour Councillors Who Are Women, By L abour Party Region Region %wome n councillors Sou th West London South East Central North West North and Yorkshire West Midlands Scotland Wales 2 7.2 2 6 .3 2 5.4 2 5 .2 2 4 .8 2 4 .2 2 3. 7 2 3 . 7 2 0.5 TOTAL 24.7 that women must be selected for the Welsh assembly. In some areas of the country internal party quotas are not filled. While the public voice of the party appears to be in favour of women, many local parties have failed to implement the policy. The legal position The Leeds industrial tribunal changed the lr~ndscape by ruling that the role of a pari iamentary candid ate could be considered as employment for the purposes of the Sex Discrimination Act. The judgement disregarded the fact that political parties are explicitly exempted from the scope of the Act, and that lega lly the means of becoming an MP is in the hands of the electorate and not the party. In broad terms, the implications of the Leeds decision are that any selection procedure undertaken by a political party for elected and salaried office could be subject to complaint if equal opportunities procedures are not followed. Although strongly advised, even after the case, that its procedures were within the law and against lega l advice that it was likely to win an appeal, the NEC agreed not to appeal against the Leeds judgement. This was a political decision, taken to minimise disruption as the General Election drew near. The Leeds ruling itself did not prevent Labour from taking further action, on parliamentary or other selections. The ruling applied only to the specific case of the individuals who brought it, and did not create a formal legal precedent which the party was required to follow. However the decision did create considerable confusion. Most important it has enabled opponents of all-women shortlists to assert that positive action is illegal. Thus the current lega l situation is complicated and uncertain. The high profile of the Leeds tribunal opens the possibility offurther challenges from disgruntled men (or women) aspirants whenever they sense unfair treatment in selections. The implications may be quite far reaching as they include indirect and perhaps unintended discrimination as we ll as any further positive action measures taken by parties which exclude candidates because of their sex. The safest course of action for Labour-now in government-wou ld be to amend the law explicitly to al low positive action as a means of achieving fair selection procedures. Such action might take the form of an amendment to the Sex Discrimination Act itself, or appropriate clauses might be included in other relevant Acts such as the Scotland Act, Wales Act and any local government Act. Legal change of this kind would probably be safe within European law, it would reflect the mood of the newTreaty of Amsterdam- due to come into force in 1999 -and recent decisions in the European Court of justice which have been supportive of positive discrimination. At the time of going to press the government appears unwi 11 ing to sanction such an amendment.5 Current election and selection systems Britain now has a number of different electoral systems, and these have different implications for candidate selection. Westminster Although various systems of se lection operate in the Labour Party it is the Westminster system, in single member constituencies, around which most discussion takes place. lt currently operates as follows: Candidate se lection takes place within local constituency parties. Aspirant candidates must learn of vacancies, then apply to party branches, trade union branches and women 's sections which each have the right to nominate one person. These nominees are then considered by the Constituency Executive Committee which draws up a shortlist of potential candidates from the branch nominees according to rules established in the party constitution. The shortl ist is then presented to all constituency members who vote at a specia l meeting for the candidate of their choice. Provision is also made for a postal ballot for those who cannot attend. At the meeting candidates present themselves at 'hustings' in a format devised by the constituency. Many constituencies require short speeches from each candidate, but other formats such as panel discussions are sometimes used. The prospective pari iamentary candidates so selected must be endorsed by the NEC which has, but seldom uses, powers to overturn a local choice. This system has various disadvantages, both for members involved in selection and for aspirant candidates. For party selectors: • the process is laborious and time consuming • there is limited choice of candidates at constituency level • selectors have I im ited knowledge about candidates because of disapproval of canvassing, poor provision of information and the nature of typical selection meetings • it is difficult to agree sensible qualifi cations, the old fashioned notion of a 'good' MP often prevails • selection approval by NEC at the end of the process is potentially embarrassing to the party • nomination and shortlisting decisions are opaque • the advantage to locals often rul es out good candidates. For aspirants: • non standard candidates find it hard to get selected • it is difficult to get reliable advance information and advice • finding a seat is costly and time consuming • there is a confusing array of formal and informal rul es. The method has been shown to be unfair to women not least because selector resistance to women is easy to conceal in the labyrinthine politics of Westminster selection. Women are less I ikely to have access to the information and alliances necessary to make the process work for them. The all-women shortl ist formula was devised as the only sure way of getting reasonable numbers of women selected under this system. local government A similar system exists for local government. In some areas there is one councillor elected per electoral ward at annual elections, and in others groups of two or three councillors are elected together at four yearly elections. This system therefore presents similar problems of devising appropriate mechanisms to promote women's representation. In areas where two or three councillors are elected at once the situation is potentially easier. The selection of councillors is quite similar to Westminster selections but- by law-involves local candidates onl y. Selection takes place in the year preceding the elections. All the party members living in the ward are entitled to take part in shortlisting and selection (but with no postal votes). Selection meetings follow a similar 'hustings' format to that for pari iamentary selections. An important difference to parliamentary selections is that a 'panel ' of candidates is devised at local authority level , with an initial screening before aspi rant candidates get onto the panel. Ward members are bound by rule to select candidates from those on the panel . Prior to the Leeds decision in 1996 the party was considering mechanisms to ensure that one councillor per branch was a woman, but these plans were abandoned in fear of another legal challenge. Although women 's representation at local level is only 25 per cent there is currently no mechanism in place to produce the improvements the party claims it wants. Scottish parliament and Welsh assembly Candidates for the Scottish parliament and Welsh assembly are being selected for the first time. There are no incumbents, hence there is an opportunity to create bodies where women are represented equally-at least on the Labour side-from the very start. Both assemblies will be elected using an additional member system which combines constituency representation with an element of proportionality through a 'top up' list. lt will be the first time that such a system has been used in the UK. However, these are two areas of the country where women have historically found it particularly hard to be selected. In Wales, only four of Labour's 34 MPs are women, and three of them were selected from all-women shortl ists. Wales is also the area with the lowest proportion of Labour women councillors-at 20.5 per cent. In Scotland, nine out of 56 Labour MPs are women. Five of these were elected for the first time in 1997, only one from an open shortlist. In recognition of these problems and the importance of selecting women, the NEC decided in January 1998 to pilot a system of 'pairing' constituencies for selection of the constituency representatives, with each pair selecting one man and one woman. This system is described in more detail below. The European Parliament In the summer of 1997 Home Secretary Jack Straw announced that Britain will come into line with the rest of Europe and elect its MEPs under a system of proportional representation using 'closed' regional party I ists, a major change from the previous system of single member constituencies. Electors wi 11 in future vote for a party, rather than a specific candidate. The number of MEPs elected by each party in the region will reflect that party's share of the vote within the region. The new electoral system offers opportunities for the representation of women but, at least in the short term, it will put pressure on Labour to maintain something as near as possible to the status quo. From its high complement of 62 MEPs elected in 1994, Labour was already expecting a drop in numbers as is usual for a governing party at this stage of the electoral cycle. The introduction of proportional representation wi 11 reduce Labour's share of MEPs to approximately 43, if patterns of voting in the general election are repeated. Of course by 1999 Labour's popular vote may have fallen well below the May 1997 level. Thus up to half of sitting MEPs may lose their seats. Prior to 1997 candidate selection for MEPs was managed by the European Constituency Parties. The final decision was taken by a one member one vote ballot, with Westminster constituencies within the area each entitled to nominate one candidate. Candidates would be chosen at 'hustings' meetings. For the transitional year between the two electoral system the NEC has proposed a mixture between the old method and a more centralised system based on a national panel. One reason given for this is the need to protect the 13 existing women MEPs and ensure they receive places high on the regional lists. The Liberal Democrats, who are likely to take far more seats in Europe than under the old system, have stated their determination to nominate more women and are in a position to make Labour look extremely backward at these elections if party women do not do well. 4. Selection mechanisms to promote women's representation In October 1998 the Labour Party Annual Conference is due to consider procedures for parliamentary and local government selections. But the selection of candidates for the European pari iament, Scottish pari iament and Welsh assembly must begin by Spring 1998. Following the Leeds industrial tribunal ruling, and with selection systems at all levels of government due for review, the party needs to consider, with urgency, what form of selection procedures it will put in place to ensure that women are fairly represented at all levels. This will not be easy-some difficult decisions may have to be taken to ensure that the goal of fair representation of women is met. With determination and imagination Labour can devise new selection methods to suit different electoral systems, and for each system ensure that a fair proportion of women are selected. Inevitably the different electoral systems offer different prospects for success and different legal implications. All-women shortlists As described above, all-women shortlists were adopted, after much negotiation, in 1993 as the party's solution to its historic inability to select fair numbers of women. lt was felt to be the only method in a single member system that could be guaranteed to deliver a better proportion of women in parliament. All-women shortlists can operate within any single member constituency system-i ncl ud ing the additional member system. The policy led to a dramatic increase in the number of women MPs. lt would take only a relatively simple change in the law to ensure it was legally safe. At the next general election there will be at most only a handful of retirements by Labour MPs, and Labour is unlikely to gain any seats which it does not currently hold. The seats vacated by retiring Labour MPs are always the hardest contested. In 1997 women ' inherited ' 11 out of 32 vacant labour seats, up from two of 20 in 1992. Pressure to select women for one-half of its vacant seats will therefore take place in a context of fierce competition. Only with great determination will Labour be able to continue its progress toward fair representation of women. lt is difficult to see how the selection of women can be assured without recourse to the device of all-women shortl ists. All-women shortlists should also be considered for local government selections. Here the problems are basically twofold. First, there are problems of supply. lt is necessary to ensure that sufficient candidates come forward, even in areas where Labour is un I ikely to win. Second, there are problems of demand. Women are often refused or blocked from selection in areas where Labour holds control. A system of all-women shortlists for half of all Labour held seats which become available, operated by local authority area on the basis of consensus between wards, just I ike the old Westminster system, could help deal with the demand problem. Clearer equal opportunities procedures-which might break the pattern of selecting less effective male candidates-w ill enhance demand. Finally, encouragement for women is needed in order to address problems of supply. Party experience of quotas suggests that as genuine demand for women candidates increases, so does the willingness of women to come forward. Quotas on shortl ists At the time of writing in early 1998, the only positive action measure for Westminster selection remaining in Labour Party rules is the 'one woman on a shortl ist rule'. Adopted in 1988, the rule specifies that where any women are nominated by constituency branches or sections, at least one of them must be included on the shortlist. This rule applies to any system requiring a shortl ist to be drawn up locall y. Selections for the 1992 general election however showed that this minimalist measure was insufficient to deliver fair representation for women. Women did get shortlisted as a result of it, but often found themselves in the position of tokens, not taken seriously by constituency se lectors. Indeed it was the experience of such tokenism that prepared opinion to accept the all-women shortlist policy in 1993. Though I imited, the policy was at least a start. However it was never adopted for any other form of selection. Despite its disappointing results the rule did make women potential candidates available to selectors and gave women aspi rants experience of the selection process. lt might work better under OMOV which was introduced more recently. Therefore, as an absolute minimum, the requirement that there should be at least one woman on a shortlist should be applied to all other selections, including local government (at ward level), the European pari iament, Scottish pari iament and Welsh assembly. Although better than nothing, the one women on a shortlist rule is unlikely to break down entrenched selector resistance to women candidates. A more effective option might be to adopt a rule that shortlists must include a certain proportion of women. A quota of 25 per cent women on parliamentary shortl ists was discussed, but not adopted, by the Conservative Party in its recent review of internal organisation. Labour might prefer to opt for 50 per cent. Men and women would then be included in equal numbers and-at least in theory-have equal chance of success. Then, selectors who may wish to have a candidate of a particular sex but also have other priorities could be accommodated. For example selectors with a preference for a woman candidate may also wish to choose a candidate who is local, or who is a member of a particular minority group, or who has particular political interests or views. Even with a quota of SO per cent, there is no guarantee that women will be selected. If the culture of the party is such that the standard and ideal candidate is a man, then the net result could be no women selected at all. There is abundant anecdotal evidence that in the last parliamentary selection round some constituencies deliberately overlooked well qualified women and shortlisted 'token' women who were clearly not of sufficient calibre to threaten favoured male candidates. Often highly qualified women were rejected in favour of unqualified women in ruthless attempts to resist women candidates. Such sharp practice is usually the result of efforts to 'stitch up' a seat for a favoured candidate by local party elites and is particularly acute in win nab le and safe seats and in areas of the country with poor track records in selecting women. A further disadvantage of quotas on shortlists is that they are as vulnerable to legal challenge as all-women shortlists. Without legal change, disgruntled male aspirants could claim that they would have been short! isted and might have been elected if 'reserve' seats had not existed for women. If the law is to be changed, it may as well be amended to allow a policy that will guarantee results. Pairing More freedom exists for newly created legislative bodies, because there are no incumbent members who have to be removed to make room for women. For new bodies such as the Scottish parliament, Welsh assembly and London authority the parties start with a clean slate. Thus sex equality is a real istic immediate prospect. The electoral systems for these new assemblies also offer scope for new mechanisms to be deployed to ensure fair representation. After the Leeds tribunal decision, women activists in Scotland-who had long anticipated a Scottish parliament when Labour came to power -looked for another means to achieve fair representation whilst staying within the law. They proposed a system of 'pairing' in which, constituency parties are matched. They then select one male and one female candidate between them . To ensure equality of outcome and to acknowledge the importance of local factors, the pairing must be organised on the basis of geographical proximity or similarity, and winnability. Once the list of pairs is agreed, the two constituency parties in each pair are asked to nominate a certain number of women and men. The constituency short! ists are merged to provide a common short! ist for the pair of constituencies. Members are then invited to vote for one man and one woman, with the outcome-if male and female lists are kept separate- necessarily being one male and one female candidate. The candidates then negotiate who stands for election in which of the seats. If the pairing process is carried out scrupulously at the start in terms of winnability, the system should result in the certain outcome of an equal number of men and women elected. it is as foolproof in this sense as the old system of all-women shortlists in SO percentofwinnableseats. In January 1998, following agreement by the Scottish and Welsh party executives, the NEC agreed that the mechanism of pairing would apply for the constituency element of the Scottish pari iament and Welsh assembly selections. This was a hard fought victory for Labour women in Scotland. Whilst it is felt to be legally safer than the all-women shortlist policy, since nobody is excluded from any selection on the basis of their sex, a challenge could still be disruptive and embarrassing to the party, and might even result in another legal defeat. Once again, the only safeguard against this is a change in the law, which might be achieved through an amendment to the Scotland and/or Wales Act. Whilst pairing may offer an ideal solution for new legislative bodies, and could successfully deliver equality in these bodies, it does not offer any help in increasing the numbers of women in established bodies in which incumbency protects male domination. Labour seats in the House of Commons normally become vacant on ly as incumbent MPs retire. Other opportunities occur only as marginal seats are fought. Both these conditions are generally too thinly spread geographi cally to allow any viable system of pairing seats. A similar situation exists in Europe and in local government. Zip lists As proportional representation is gradually introduced, new challenges and opportunities are provided by the need to have women fairly represented on party I ists. The electoral systems adopted for the Welsh assembly and Scottish parliament require a 'top up' list of candidates to be provided by each party. The revised electoral system for the European parliament is organised solely around party I ists. List systems are general ly far easier to adjust than the traditional Westminster system. For many years Labour's sister socialist parties have successfully employed quotas which rely on ensuring women are represented fairly on party I ists. Their experience shows that prejudices against women candidates on party lists are difficult to hide and electors can easily see if party 'slates' are balanced. Once one party began to compete on the basis of its representation of women, others were under pressure to follow. This pattern has been seen, for example, in Sweden and Germany.6 Such party competition is already apparent in Britain. Having fai led so dismally to elect women amongst their new intake of MPs in 1997, the Liberal Democrats were quick to agree action to ensure that they wou ld improve women 's representation at the European level. Their conference in autumn 1997 agreed a policy of 'zipping' for the European lists-meaning that each of their lists would compri se male and female candidates alternately. Thus, roughly half the candidates elected would be women. However, the Liberals have been advised that this policy is illegal. lt is likely to run into the same problems as the all-women shortl ist policy, given that placing a person in the first or second position on a I ist for one of the three main parties makes it almost certain that person wi 11 be elected . Party rules specifying that these positions must be restricted by sex are potentially in breach of the law. In the European elections for 1999, the Labour Party is faced with far more practical problems in adopting a system of 'zipping' because it has a large number of incumbent MEPs. (This is not a problem for the Liberals!) Currently women are 13 of the Labour Party's 62 MEPs. Labour's women are not geographically spread but mostly concentrated in two regions, with no women in Scotland or the north of England. The transitional sol uti on proposed by the NEC aims to improve women's representation in order to protect sitting women MEPs but not in a systematic way. However, Labour will be able to move to full zipping for the 2004 elections, should there be a suitable amendment to the law. In the elections for the Scottish Parliament and the Welsh assembly, additional members will be elected from lists in order to balance the total membership of the pari iament and assembly to reflect the balance of votes. Si nee Labour wins a higher proportion of first past the post seats than its share of the vote would suggest it is likely to gain few, if any, additional members. Thus Labour cannot address its need for balance in these bodies solely by 'zipping' these lists (or even by ensuring that they are women only). In principle however, 'zipping' offers possibilities for parties wishing to achieve a balance between the sexes. lt is likely to prove particularly useful to the Liberal Democrats and the smaller parties, who are liable to gain most of their seats via the additional members' lists. Balanced panels A panel is a list of approved candidates. Its use and possible form has not been widely debated in recent party discussions so it is necessary here to describe how a panel system might work. A panel of candidates is compiled at the level for which the legislature exists (i.e. nationally for national government, regionally for regional government, etc.) and used by local parties to select their candidates. Because all candidates go through a central assessment process, rather than purely being seen by constituency or local parties, elements of consistency and equal opportunity may be built in, at least at the initial stage. Panels currently exist for local government selections and have recently been approved as part of the selection process for Scotland and Wales. The new process for selecting the regional I ists for European elections now necessitates a regional panel, and it is proposed that for 1999 this would have national co-ordination. The suggestion of a panel for Westminster has been considered by the Labour Party at various times, including in 1997. However it did not go to conference on that occasion because it was too controversial. The trade unions were uneasy about it, fearing that panel -criteria would favour middle class barristers instead of working class trade unionists. Such fears are not unreasonable. Nevertheless the establishment of a national panel could offer many advantages. If criteria and procedures are clear they should not be biased in favour of one particular type of candidate. Under the panel system at local government level, potential candidates must first apply to the panel and be accepted, with the I ist of names on the panel then being circulated to local branches for selection. However, party rules do not specify what form of assessment must take place before candidates are allowed onto the panel. There is no standard application form and there is no requirement for candidates to be interviewed. There are also no rules about who should conduct the interview. In some areas all applicants are accepted to the panel without any assessment at all, and in many areas sitting councillors are accepted without assessment. These rules are currently under review as part of the recent drive to improve the quality of Labour local government candidates. Procedures agreed for selection to the Scottish parliament and Welsh assembly build on, and formalise, this type of system. Before consideration by constituencies all potential candidates must first be accepted onto a panel compiled at the relevant level. The assessment boards wi 11 be made up of a representative group of senior party figures, who must evaluate candidates against an agreed set of criteria. If a national panel were to be established, this could be administered by the Labour Party's national Candidates Unit-which is in any case an essential element of a more extensive equal opportunities process (see below). The unit would receive and process applications to the panel and weed out any applicants who did not meet basic criteria, such as sufficient length of party membership. Aspirants would then be given further consideration by a representatively constituted assessment board. The form of assessment and the time taken over it would vary according to the level of office candidates were applying for. For local government selection the process for each candidate might be little more than a 20 minute interview. For Westminster and European selections it might take up a full weekend, with groups of candidates assessed together. A selection weekend might require potential candidates to undertake various kinds of recruitment 'tests'. For example, participating in mock debates, undertaking radio/tv interviews, holding surgeries, chairing a committee, writing press releases, devising campaign strategies, etc. These exercises would be observed by members of the assessment board. Aspirants who did well at a good proportion of these exercises would then be recommended for approval. Aspirants who did badly would not be approved, but would be given sympathetic advice on how to proceed . For instance they may be promising individuals who simply need more political experience after which they should be encouraged to reapply. All members of the panel would then be circulated regularly with information about vacancies and could submit their CV to local parties, who would be required to choose their candidate from applicants who are on the panel. The process of campi I i ng a panel could be an exercise in party building, with members from different levels and parts of the organisation coming together to select aspi rants for the panel. The panel of candidates should be compiled at the level for which the legislature exists (i.e. nationally for national government, regionally for regional government, etc.) and used by local parties to select their candidates. Clearly all panels of candidates drawn up by the party should aim for a balance of men and women. A balanced panel is an important prerequisite to a fair selection process, as it facilitates the availability of a substantial and diverse pool of women from which selections could be made. However, in the Labour Party, it has long been accepted that the problem of women's under-representation is not a problem of supply, but one of demand. lt results not from a shortage of suitable women, but from barriers in the selection process itself. The least controversial way to achieve a balance on panels is to specify targets, and require that panels remain open to women whilst they are underrepresented . A target of 40 per cent (which might be set for both sexes) appears most appropriate. Party rules should also require that where targets are not met, active measures are taken to encourage more women to come forward. This is particularly relevant at local government level, where local parties could, for example, be specifically required to mail all women members in the area individually. In our view such a rule might be necessary but would not be sufficient to ensure the fair representation of women. Balanced panels in themselves cannot guarantee that women wi 11 be selected, although certainly women are less likely to be selected without them . Balanced panels might make it more likely that selectors considered well qualified women, but they will not guarantee it. 5. Equal opportunities procedures The Leeds tribunal ruling on all-women shortlists required party managers to look at the pari iamentary selection process in an entirely new I ight. The ruling implied, for the first time, that selecting a parliamentary candidate was equivalent to appointing to a job. Therefore, equal opportunity procedures needed to apply. The reasoning is straightforward. If parliamentary selection within parties is covered by employment law, then a considerable tightening up of procedures is necessary, if the possibi I ity of legal challenge is to be avoided. Immediately following the industrial tribunal ruling, the Labour Party National Executive Committee set up a subcommittee to examine the implications for the parliamentary selection process. The Women's Representation Working Party drew up a set of equal opportunity guidelines to be circulated to all constituencies which had not yet completed their selections. These represented a break from the past in recommending a new openness and clarity to the selection process, bringing in some procedures- such as circulation of job descriptions -which seemed novel in political selection but which are standard in employment practice. Now, with more time available before the next round of parliamentary selections begins, the opportunity exists to build on these guidelines. Clearer procedures could enable Labour to benefit from a bigger, more diverse pool of potential candidates. The reforms would not only benefit women, but other underrepresented groups including ethnic minority and working class candidates as well as other groups of party members who are excluded from current party elites and informal networks. The procedures could be applied at all other levels of candidate selection-from local government through to European parliament. Equal opportunities procedures require a clearer selection process including suitable explicit candidate criteria, widespread availability of information, a means of self nomination, targeted training provision and fair procedures. A clear selection process In order to attract the broadest possible range of candidates, it is essential that party members are made aware of opportunities to stand, and of procedures and timetables. Current selection procedures are opaque, with a minority of party members-most of them are men-fully aware of opportunities to come forward. Traditionally Westminster selections have been advertised in a patchy fashion. Often upcoming selections have been notified mainly by advertisements by individual constituencies in the Tribune newspaper. The national! ist of interested candidates has never been used systematically to inform aspirants of new vacancies. This is inadequate. Good equal opportunities practice would require the party to ensure that all m mbers are informed of upcoming selections. Members should be notified, by the national party through its publications (and, if necessary, through the press) of the timetable for selection and how to take part. There should be a rei iable central contact point from which interested members can get detai Is of the relevant timetable and procedures, including details of which party unit (e.g. region, local government party) controls the procedure locally. Where national office has a key role in the procedure, members should be invited to join a mailing list to be kept in touch with vacancies and opportunities for training, etc. For Westminster and Europe this would be best co-ordinated by a well-resourced Candidates Unit based at head office. A good information source on procedures has been produced for many years by Labour Women's Network. Their series of Uphill All the Way booklets explain the role of elected representatives at different levels and details the selection process in clear and straightforward language. The booklets are directed to potential women candidates. The Labour Party has produced a booklet for potential local government candidates in conjunction with Labour Women's Network entitled A woman's place in on the council. They are models of clarity. it should be standard procedure for the national party to provide information of this quality to members, and advertise it widely. Candidate criteria and information Linked to clear information about the process is the need for clear information for both candidates and selectors about what qualities are needed in an elected representative. Part of the information which prospective candidates need is a 'job description of the work involved in the 'job' for which they are considering applying. The work undertaken by representatives is diverse, with much scope for individuals to tailor the role, but in all cases it includes a balance of key elements: • representing constituents, both individually and collectively, • holding advice surgeries, taking up individual cases, taking up local causes and reporting back on progress, • being part of a legislature-devising policy and strategy, either as a backbencher or frontbencher, taking part in debates, committees etc. • being a party advocate-supporting and propagating the values of the party and raising awareness of the party's achievements: nationally, regionally or amongst the local community, in partnership with party members. Clear guide! ines on what the job of representative involves at each level, including practical considerations such as core hours and overall time commitment would help members to decide whether to put themselves forward. The same guidelines would provide selectors with a yardstick against which to measure the aspirants who do come forward . The guidelines should be produced locally, using a national model or template. Thus local needs _6 could be taken mto account at the same t1me as the benefits of expenence throughout the party are max1m1sed. In equal opportunities recruitment, a job description is accompanied by a person specification, outlining the qualities (knowledge, skills, etc.) which are sought in the successful candidate. These follow from the job description, and for an elected representative would therefore be diverse, ranging from commitment to the values of the party through to listening skills, analytical ski lis and pub I ic speaking ski lis. Local parties may want to prioritise some skills over others, but 1t is Important they are aware of the range of skills requ1red of an MP, MEP or councillor. The guidelines circulated for pari iamentary selections following the Leeds tndustnal tn bunal ruling set down some good practice in this area, upon which the party urgently needs to build. The guidelines stated that parties should agree the job description and person specification at the beginning of the selection procedure. These comprise the selection criteria. To be effective, such criteria must be scrupulously devised in order to ensure that they are not themselves discriminatory. lt would be easy to draw up criteria which continued to discriminate against women-for example by requiring expenence at sen1or level outside the party. Constituency parties should prepare a pack to be sent to all potent1al candidates giving details of the constituency, the party and 1ts members (the job descnption) as well as the cnt na wh1ch were bemg sought in the candidates (the personal sp C1ficat1on). The local party should then make refer n e to the selection cntena when dec1d1ng wh1ch andidates to shortlist and ult1mately select. Candidate panels The 1ntroduct1on of a panel of candidates can be seen as part of an equal opportun1t1es procedure, 1f prop rly admm1stered. To ensure equal opportunity and guarant that the panel has the full support of local part1es, any a ses ment board must be representative of d1fferent parts of the party and mclude repre entat1ve of const1tuenc1es, un1ons, tc. as well a xp nenced councillors/ MPs/MEPs as appropnate. Women and mmont1es should be fully repre ent d. B for elect1ons b gm th m mbers of the board must be properly tra1n d. A part of an equal opportun1t1 s el t1on pro edure, a prop rly operated pan I of cand1date ha a numb r of potential advantag : • The tran par n y of the pro dur s would b 1mproved. • A ad fin1t1v l1 t of pot nt1al cand1dat , the pan I an be u ed to ke p i:hplrant mform d automatically of upcommg s I opportun1t1e~. • Th e penenc oi gomg through the proc ~gives ao;p1ranb a real1 ti v1 w of what is expected of an elected representative. • Because selectors have access to the whole panel, constituency/ward selectors have a the widest possible choice of candidates. • The assessment board can be required to pay special attention to the need for more women and minority candidates. The panel could have quotas of various kinds to ensure a more balanced composition (possibilities for a gender quota were examined above) . The implication of the Leeds ruling is that all parties should apply rigorous equal opportunities procedures including clear candidate criteria which are used by selectors-preferably through a scoring system-in making their choice. These are necessary both to ensure fairness and to protect the party from further legal problems. With an unwritten, informal and thus possibly discriminatory process, a further legal challenge by an unsuccessful candidate -most likely female-remains likely. Although necessary, the use of Equal Opportunities selection is un I ikely to be sufficient to ensure the increased representation of women. Self nomination procedures Labour is increasingly moving away from formal nomination procedures for candidates, traditionally carried out by branches, women 's sections and trade unions. For the new Scottish parliament and Welsh assembly it has been agreed that individuals should be able to nominate themselves. This system has recently been agreed for local government selection. The removal of the formal nomination procedure may be beneficial to women and other under-represented groups, who have frequently found themselves blocked at the first hurdle through not being able to get nomination from male dominated party units. However, a 'self nomination' system can only succeed if coupled with proper publicity, reaching all members of the party, about the opportunity to stand . lt also depends upon screening through a panel system of some kind, so that interested members continue to be given a form of endorsement by their peers. lt is noteworthy that replacing formal nomination with individual applications to a screened panel remains the only recommendation in the 1990 Fabian pamphlet by Brooks, Eagle and Short which has not been implemented. Training A final essential element of an equal opportunities system is training. But training is not just for aspirant candidates. Training is one of the most effective ways to increase the supply of qualified women who come forward as potential candidates. lt should also be made available to selectors whose ability to make the best choice of candidates relies on their understanding both of what is needed from a candidate and of what their contribution as selectors should be. The selection system mu~t be demystified not just for the candidates, but also for the local members, who want to play an effective part in selecting a good loca l representative. Frequently members are confused by the process and may not participate. Pre-selection training and briefing on the process should be a standard first step of the selection process. Short! isti ng is a particularly critical gateway to candidacy and there is a case for requiring this to be done by a representative and accountable committee who are trained in equal opportunities and other good recruitment techniques. Those involved in shortlisting must be aware of any candidate criteria which have been agreed and must have a clear understanding of the type of candidates sought. Safeguarding fair procedures The short! isted candidates are invited to interview. The current system, involving all members of the constituency I branch on an OMOV basis, is designed to maximise fairness. However, to safeguard this, postal votes shou ld be tightly restricted to those who can demonstrate that they are unable to attend meetings. Selectors should be expected to attend a meeting and see the candidates perform . They should be provided with a copy of any agreed candidate criteria, as well as candidates' supporting statements, in a standard format. The design of meetings is a matter for local parties, and there is room for experiment beyond the standard 'hustings' type of meeting which may offer on ly a limited insight into candidate quality. Local parties may benefit from using some of the forms of exercises recommended for the panel selection process (above). As a general rule, local parties should devise their selection system with close reference to the local job description which they have drawn up, and seek to evaluate the full range of candidate ski lls, abi lities and interests, including campaign skills, commitment and ability to represent constituents, as well as public speaking abi lity and political views and interests. The more diverse the range of qualities tested by the selection process, the better choices constituencies wi 11 make. 6. Conclusion The Labour Party has made great strides on women's representation . However, the biggest achievement-that of electing 101 women MPs to Westminster-was brought about largely because of all-women shortl ists. Thirty five of the sixty-five new women elected in 1997 were chosen from all-women shortlists. In declaring in favour of Jepson the Leeds tribunal empowered opponents not only of positive action but also of women's representation. Legal experts continue to be divided about the judgement with the result that opponents of positive action prevai I. While the Leeds decision goes unchallenged and the law unchanged it will prove very difficult to devise an effective mechanism to replace compu lsory all-women shortlists. All-women shortlists would not be needed if there were no systemic resistance to women in the party. A traditional masculine culture coupled with a set of entrenched male networks has long operated against women. There is no evidence to suggest that the culture of the party has changed in favour of selecting women-indeed the pattern of selections for the 1997 general election, once all-women shortl ists are taken out of the equation, suggests otherwise. In order to continue progress in this area, the party therefore needs to adopt new mechanisms and procedures. As the party of government Labour shou ld change the law to allow positive action. As a party of fair representation it shou ld adopt new, fairer selection procedures. These will vary for different bodies and for the different electoral systems now being introduced. In addition to specific mechanisms to promote women's representation, we have proposed some ways in which party se lection procedures cou ld be adapted to be fairer to both sexes and improve thecal ibre of candidates se lected. These changes can be introduced alongside quota mechanisms, and once fair gender representation is achieved, should help to maintain the balance. Table 7 summarises the recommendations in this paper, setting out what action the party should take for each level of election. In order to reintroduce all-women shortlists for Westminster and local government, to allow Zip lists for Europe and the new assemblies and to implement pairing without fear of a legal chal lenge, a change in the law will be necessary. Under the current interpretation of the Sex Discrimination Act a reliance upon weaker (a nd possibly ineffective) mechanisms such as compulsory quotas on shortl ists could be subject to challenge. Ideally legal change would permit, but not require, a widespread use of positive action strategies to achieve sex equality. Once positive action is permitted, then parties can decide what, form if any they wish to adopt. All-women shortlists are the most effective means of securing fair representation of women. But narrower measures would also enable progress to be made. Clarification of the political parties exemption from the Sex Discrimination Act would ensure that this exemption applies to the selection of candidates. Where legislation is required before the appropriate election can take place-as in Scotland, Table 7: Recommended procedures for fair representation of Women level of government & electoral svstem At next elections Thereafter Westminster single rrerrber consti tuencies 2001/2 All INOrren shortI i sts AIIINOrren shortl ists for safe and winnable seats, SO-SO shortli sts elsewhere until equal representati on achievoo, then eaual oooortunities selection European parliament regional Iists 1999 Corrpromise arrangerrents due to new electoral svstem 2004+ Zip lists Scottish pari iament and Welsh asserrbly: single rrerrber constituencies addi tional rrerrber I ists 1999 Pairing Zio lists 2003+ Equal opportunities selection Zio lists Local government single'multi rrerrber constituencies Ongoing All 'NOmen shortlists Ongoing All 'NOmen shortlists or safe and winnable seats, SO-SO shortlists elsewhere until equal representation achievoo, then eaual oooortunities selection Wales and Europe-this clarification can be inserted in the relevant Bill . In the longer term-particularly if the Sex Discrimination Act is overhauled, as Labour promised in opposition-the Act itself should be amended. Such change would be within the spirit of European law, which has allowed socialist parties in Europe to apply quotas to candidate lists for many years. Our proposals promise equality in the Labour groups in the Welsh Assembly and the Scottish Pari iament. The party has the chance to ensure th at each new elected body it creates has equa l representation of women and men on the Labour side from its inception. it should be possible to implement Zip lists and achieve equality in the European parliament by 2004. In local government and at Westminster, progress wil l be slower. Even with ompulsory all-women shortl ists in half the seats which become vacant it will probably take a generation to deliver equality. Nevertheless, if the Labour Party keeps its nerve and sticks to its commitment to achieving equal representation, women will at last be fairly represented throughout the British structure of government. Footnotes 1. Patricia Hewitt and Deborah Mattinson, Women 's Votes : the Key to Winning, Fabian pamphlet, 1987. 2. Women and the Labour Party, Clare Short, Parliamentary Affairs, 1996. 3. P. Norris and J. Lovenduski, Political Recruitment: Gender, Race and Class in the British Parliament, Cambridge University Press, 1995. 4. Rachel Brooks, Angel a Eagle and Clare Short, Quotas Now: Women and the Labour Party, Fabian pamphlet, 1990. 5. The Guardian, 4 March 1998. 6. j. Lovenduski and P. Norris (eds.) Gender and Party Politics, SAGE 1993. The authors are grateful to Meg Russel who supplied information for this pamphlet. FABIAN SOCIETY High Time or High Tide for Labour women ? The photograph of Tony Blair surrounded by Labour's women MPs provided one of the most vivid and enduring images of the 1997 General Election. The number of Labour women MPs had almost trebled to 101 . This was brought about largely as the result of the policy of all-women shortlists. However the use of this mechanism was brought to an abrupt end when an Industrial Tribunal in 1996 declared that all-women shortlists contravened the Sex Discrimination Act. The authors of this pamphlet argue that there is no evidence that the traditional male culture of the party has changed In favour of selecting women. They argue that positive action is still needed to ensure proper equality of male and female representation . Since forming the Government the Labour Party has introduced a number of new representative bodies and electoral systems. This pamphlet examines each and suggests ways in which selection procedures can be adapted to be fairer to both sexes and to improve the calibre of candidates selected. Their proposals for a national candidates panel , 'zip lists' and equal opportunities selection procedures promise over time to achieve equality in the Labour Groups in the European and Scottish Parliaments, Welsh Assembly, and local government and continued increase in representation in the House of Commons. joni Lovenduski is professor of politics at Southampton University. Maria Eagle was elected Labour Member of Parliament for Liverpool Garston in 1997. This pamphlet is part of the Fabian Society's Modernising Britain programme, which seeks to examine what 'modernisation' might mean-politically and culturally-as Britain enters the next century. Previous publications have examined the armed services (Arms & the Man by Eric joyce) and the monarchy (Long to reign over us? by Paul Richards). Fabian Pamphlet 585 £5