Trade Union Recognition in Britain – An Emerging Crisis for Trade Unions?

Professor Gregor Gall, Centre for Research in Employment RelationsUniversity of Hertfordshire

Introduction 

The statutory provisions for gaining union recognition in Britain, contained in the Employment Relations Act 1999 (ERA), came into force on 6 June 2000 and represent the third such statutory mechanism for gaining union recognition to be introduced (Gall 2003a:15-16). The provisions enable independent trade unions to apply for statutory union recognition where they have at least ten per cent of the proposed bargaining unit in membership, where a majority of workers in the proposed bargaining favour their terms and conditions of employment being determined by collective bargaining (usually evidenced through a petition), and where existing voluntary recognition of another independent union does not exist (see also Gall 2003a:13-14).

An application resulting in statutory union recognition can be gained through two routes after meeting the aforementioned preliminary conditions. The first concerns simple majority membership through an independent audit of union membership. The second concerns a ballot where the union must win not only a simple majority of votes cast but this number must also be equivalent to at least 40% of all those entitled to vote in the bargaining unit. Consequently, reluctant or resisting employers can now be compelled by law to engage in representation and negotiation with trade unions, allowing an elementary level of workplace co-determination of the employment relationship and the terms and conditions of the wage-labour exchange to emerge (Gall 2003a:6-8).

But despite the existence of the statutory provisions, the creator of the legislation, the ‘new’ Labour government, made it clear that it wished to see the use of the provisions as a recourse of last resort. This position was entirely in keeping with ‘new’ Labour’s neo-liberal induced minimal interventionist-cum-voluntarist predilection to industrial relations. In tandem with this, trade unions have been vociferous critics of the overly ‘business friendly’ nature of the statutory provisions, where reluctant and resisting employers have been able to engage in lawful ‘union busting’ activities with the statutory framework.  The creation of the third statutory union recognition provisions resulted from a long process of union lobbying. This began with a small number of unions winning the union movement, through its peak organisation, the Trades Union Congress, to the policy of advocating a statutory mechanism.

From here, the Labour Party was then won to this policy position. The dynamic compelling trade unions to espouse such a policy, and thus overturning their traditional advocacy of collective laissez-faire, was their industrial weakness as a result of emboldened and unilaterally-minded employers. Consequently, union efforts to resist the withdrawal of recognition – called derecognition – in existing workplaces and to gain recognition in new and greenfield workplaces was very limited. The wider backdrop to this was the disorganisation and demobilisation of trade union presence and power. Across the measures of union membership, collective bargaining coverage and strike activity, unionisation weakened and atrophied, particularly in the private sector of the economy and especially in the private services sub-sector of this.       

This article critically examines the ability of the trades unions to take advantage of the relatively more favourable legal and public policy environment indicated by the ERA. Prima facie, there is much evidence that unions have been able to take advantage of this new situation and benefit from it. Since 1995, when it became likely that the Labour Party would win the next general election and legislate on its policy to establish a statutory mechanism for gaining union recognition where there was majority support amongst the workforce, the industrial relations climate began to change, influenced by the imminence and then presence of the statutory mechanism. Moreover, from 1995, trade unions in Britain also began adopting ‘union organising’ strategies to organise previously unorganised workers and workplaces. Consequently, just over 3,000 new union recognition agreements have been signed covering around 1.2m workers between 1995 and 2005 (see note to Table 1) as a result of the enmeshing of the two processes. Prior to 1995, the seven years between 1988 and 1994 yielded some 435 new union recognition agreements, covering some 42,000 workers (Gall and McKay 1994:444-445).

Thus, the annual average for the period 1995-2005 of 273 new agreements per year (with just under 80,000 known new workers covered) compares to the average of 62 new agreements per annum (with just under 6,000 know new workers covered) for the period 1988-1994.  This article examines the significance of the growth in the new recognition agreements and considers the environment and contexts of the campaigns from which these agreements were gained. It concludes by suggesting that despite the evident union success in gaining the new agreements, using a wider lens to judge this achievement indicates that there exist a number of important limitations to current union organising capacity. Union organising capacity refers to the quantitative and qualitative aspects of a union’s ability to unionise and organise workers and represent their interests through collective bargaining after having gained union recognition. This ability centres around the financial, ideological, organisational and personnel resources and expertise that a union has or can bring to bear on a particular situation. Of particular importance are the aspects of the relative numerical preponderance and motivational drive of mileux of workplace (lay) activists to the carrying out of the recruitment and organising, and the number of union full-time officers (FTOs) which are dedicated to, or spend a considerable amount of their time, recruiting and organising. 

For trade union-orientated audiences outside Britain, particularly those in Australia, Canada, New Zealand and the United States of America, the salience of this examination of current developments in trade union recognition in Britain lies in three aspects. The first concerns the efficacy of the outcome of the implementation of ‘union organising’ approaches, strategies and techniques, given the parallel existence of a turn to ‘union organising’ in these other countries.

The research reported here suggests that on a national scale and despite the not inconsiderable human and financial resources devoted to ‘union organising’, the goals of union renewal and growth still remain elusive, a conclusion which has much in common with the renewal and growth projects in these other countries.  The second concerns the evaluation of the benefit to trade unions of a new statutory provision for gaining union recognition where, broadly speaking, similar mechanisms exist in these other countries but where the wider socio-political environments are markedly different (see Gross 2003, McArthur 2004). The salient point to emerge here is that not only are the nature and complexion of statutory union recognition mechanisms influenced and determined by contending social interests and dynamics, and the balance thereof between them, but that statutory union recognition mechanisms are constrained by the environmental contexts into which they are inserted. The third revolves around whether, and to what extent, such union activity and components of the terrain on which unions operate are helpful in stemming union decline and revitalising union movements in these countries. Previous debates on union fortunes have discerned a dialectical relationship between union agency and societal structuration. At present, union agency may be seen to remain sufficiently weak and constrained as to be unable to overcome the inhospitable forces and environment in which it finds itself.  

The data for this research are derived from a number of sources. The first is material from the LRD-TUC Trade Union Trends surveys (1996-2005) on union recognition. Starting in 1995, these surveys, compiled from questionnaires sent to unions, cover on average 81% of the TUC’s affiliated membership (see Table 17 in Appendix). Second, over eighty semi-structured interviews with regional field and national FTOs from the fifteen major unions in Britain, with a membership of 5.4m workers or 73% of union members. The FTOs from these unions were interviewed in 1999, 2000, 2001 and 2002. Those interviewed in 1999 were re-interviewed in subsequent years while new ones were interviewed in 2000 and 2001. Each new ‘interviewee’ was re-interviewed in subsequent years. The third is information provided by the major employers’ federations, through either interview or correspondence. The fourth is the monitoring of the journals of the largest thirty unions and of secondary industrial relations sources for the last nine years. Lastly, the determinations of the Central Arbitration Committee (CAC) and the Labour Court in Northern Ireland, the state bodies which adjudicate on applications for statutory union recognition, were used – 485 applications had been made by the end of 2005.

From these sources, data was collected on the number and character of new recognition agreements as well as the processes by which, and contexts in which, they were signed. This data is reported and analysed. The data is, thus, extensive in its coverage but not inclusive of all new agreements and relevant developments because of poor information gathering and record keeping (as well as disseminating and publicising the data) within the different levels of unions, particularly at the national/headquarters level. It is, however, more inclusive of actual developments than any other data set. For example, the Trade Union Trends surveys report on only those agreements and campaigns that were identified to it through its surveys[1]. Other deals, which may be known of, are thus not reported. Further, although the percentage of unions that respond to the Trends surveys by affiliated membership is high, this still means that many unions have not responded to each survey.

The effect of these characteristics is offset by the use of the other data collection methods for this research, in particular interviews and monitoring union journals. However, again the poor record keeping within unions means that it can be said with certainty that not all agreements and campaigns are known of. Similarly, not all the details of reported agreements and campaigns are known of. What the exact degree of overall ‘underreporting’ exists is impossible to state. But this research’s data is, nonetheless, the most inclusive of any existing data on such developments (cf. Moore et al. 2004:13 and Blanden et al. 2006).

It is clear from Table 1 that there has been a significant increase in the last nine years in number of new recognition agreements being signed by trade unions. Concomitant, the phenomenon of union derecognition appears to have been almost eclipsed. This stands in the context of cases of derecognition being of greater significance, despite being matched by cases of new recognition agreements, in the early 1990s (Gall and McKay 1999).

Table 1 also indicates that the vast majority of new agreements are voluntary agreements, signed in the shadow of the ERA, but without recourse to it. However, alongside the recent considerable growth in new agreements, it is also apparent that the highpoint in terms of the number of agreements, and represented by the year 2001, is now over. The post-2001 level of annual agreements shows indications of settling down to the 1998 level, that is, before the statutory mechanism came into force and before the ERA was put on the statue book. Of note is also that the level of reliance of trade unions on the CAC to secure recognition has increased from 6% in 2001 to 28% in 2005, particularly where most of this concerns statutory awards (see Table 1).

This suggests the terrain on which union recognition may be gained is getting harder, that is, the targets of union recognition are increasingly less receptive and increasingly more hostile. The trend in the annual rate of signing of new union recognition agreements is matched by the picture emerging from the state body, the Advisory, Conciliation and Arbitration Service (ACAS), through its casework in terms of volume of cases and successful resolution of cases for trade unions with regard to union recognition (see Table 2). ACAS can arrange to test the union density via membership audits or the support for union recognition by organising ballots in a way that is similar to the CAC.

Proportionately, and from 1999, the extent of use of ACAS to help secure union recognition rose from 21% to 31% in 2001 before falling back to 21% in 2003, suggesting that there are increasingly fewer relatively more receptive employers to union recognition, for accepting ACAS involvement is voluntary and dependent upon employer agreement. 

The data in Table 1 represents only one aspect, albeit an important one, of how the current level and coverage of union recognition can be assessed. Table 1 represents an ‘input’ to the aggregate level of union recognition while closures of establishments with union recognition as well as redundancies amongst workers covered by union recognition represent ‘outputs’. Meanwhile, growth in the number of new establishments (and their workers) without union recognition and the recent increase in the number of workers in the heavily unionised public sector constitute further ‘inputs’. Between 1997 and 2005, 1.1m jobs have been ‘lost’ in manufacturing (Guardian 17 March, 5 December 2005), one of the two heartlands of trade unionism, the other being the public sector. In the public sector, the level of employment grew by 0.719m between 1998 and 2005 (NSO 2005).

Far less common have been cases, like Virgin airlines, of employing organisations which have recently granted union recognition expanding the number of workers they employ (cf. Moore et al. 2005:31). On this basis, data from the Labour Force Survey (Table 3) indicates that the impact of the new union recognition agreements is likely to have been marginal to the aggregate level in the period 1999-2005. Another means of assessing the impact of the tranche of recent union recognition agreements upon aggregate levels of collective bargaining coverage can be derived from a breakdown of the LFS data. Between 1999 and 2003, the number of union members covered by collective bargaining over their pay and conditions rose from 5.4m to 5.5m, whilst the number of union members not covered by collective bargaining over their pay and conditions stayed still at 1.6m (Metcalf 2001:32, 2005:21).

Meanwhile, the number of non-union members covered by collective bargaining over their pay and conditions rose from 3.1m to 3.3m and the number of non-union members not covered by collective bargaining over their pay and conditions rose from 13.6m to 14.0m (Metcalf 2001:32, 2005:21).  Thus, and overall, union recognition coverage is like a set of revolving doors, where as some workplaces and workers ‘enter’, others workplaces and workers ‘leave’. In the period of (relative) substantial growth in union recognition agreements (1999-2004), the impact of the number of unionised workers affected by redundancy and the number of new non-union workplaces established has been a greater influence on the aggregate coverage of union recognition than the former. Indeed, as big a positive impact on the input side of the equation of union recognition coverage is likely to be recorded by the increase in the number of public sector workers.

In addition to this, other data such as the time series the Workplace Industrial Relations Survey/Workplace Employment Relations Survey (Table 4) and the annual British Social Attitudes surveys (Table 5), indicate the considerable decline in both union recognition and collective bargaining coverage over a longer period. Despite the slight increase in the British Social Attitudes survey data between 2000 and 2002, collectively the two datasets suggest that the extent of union organising and the impact of the statutory mechanism on the gaining of union recognition would have to have been far greater than has been the case to exert a sizeable positive impact on the overall coverage of union recognition (workers covered, workplaces covered). Finally, the British Social Attitudes survey data also indicated that the proportion of employees who reported that unions were not recognised for negotiating on pay and conditions in their workplace rose from 33% in 1983 to 46% in 2003 (Kaur 2004:18).  

While the union movement has striven to take advantage of the new legal remedy to cajole and force employers to grant union through a significant increase in both union organising and union recognition campaigns (Heery et al. 2000a, 2000b, 2000c, 2003), the extent to which unions have been able to make progress can be assessed in a number of ways. Setting aside any legitimate criticisms that may exist of the implications for unions of the restrictive nature of the statutory union recognition provisions (see later), a number of initial observations are useful to contextualise an assessment.  

First, out of 6,729 recognition campaigns run between 1995 and 2005 which were identified from the research undertaken for this paper, the success rate for campaigns to date has been only 40%. The measurement of ‘success’ is based on the winning of union recognition within the time period, for merely gaining members in a non-recognised workplace severely limits the extent to which a union can represent its members’ interests vis-à-vis the employer without union recognition. For example, in the Liverpool, North Wales and Irish region of the GMB union, of 63 recognition campaigns run between 1999 and 2001, only seven gained recognition by 2002 and only four remained as campaigns thereafter. One inference that could be drawn from this is that the number of new agreements signed is merely a factor of the much larger number of campaigns run where the unions must, on average, run five campaigns to gain success in every two campaigns (Gall 2003b). However, one caveat needs to be added. In cognisance of the considerable time that is often required in which to gain recognition (see Table 13), some of the campaigns, particularly those begun in the latter part of the period 1995-2003, are likely to be successful after 2003.  

Second, the Confederation of British Industry’s Employment Trends Surveys (Table 6) recorded that employers, both in relative and absolute terms, have anticipated considerably fewer recognition claims from trade unions after 2001[2]. Similarly, the Trade Union Trends surveys also indicate a quantitative decline in the number of recognition campaigns run and the number of workers covered by these campaigns (see Table 11). The data is consistent with the rise then fall from a minority to a majority and then back to a minority of unions in the Trade Union Trends surveys reporting that they had signed more than one new union recognition agreement between 1997 and 2005 (Table 17, Column D). Employment lawyers Dibb Lupton Allsop (2005:6) also commented in its annual industrial relations survey that ‘Trade union activity in relation to the seeking of and achieving of new recognition agreements … appears to be faltering.’ 

Third, the increasing use of the statutory procedure by unions to gain recognition has become more marked (see Table 1, Columns F, G and H). This is noticeable for unions have been widely critical of its challenging aspects such as the high membership and worker support required thresholds, the complex and long and drawn out procedure, and the three-year bar on rejected applications. As a result, unions did not anticipate using the statutory procedure to any great extent, and only so as a last resort. The growing usage, relative to the number of new recognition agreements gained, suggests that unions are increasingly facing more resistant employers, and following this require extra-workplace national union help like legal support to gain union recognition from such employers.

Fourth, the industrial sector distribution of the new union recognition agreements remains heavily slanted, by number of agreements and workers covered, to the manufacturing sector (see Gall 2004a, 2005a, 2005b). Union targets in the ‘new economy’ and the private services sector like large, multi-site and well-known employing organisations such as T-mobile and Vodafone telecommunications, and retailers like Marks and Spencer, high-street betting shops, and Egg internet bank remain elusive to the gaining of union recognition despite long, on-going and resource-intensive campaigns. Although the picture concerning the extent of union recognition in call centres (see Table 7) looks to be relatively favourable towards trade unionism, the nature of the Incomes Data Services (IDS) surveys compromises this. Thus, the IDS’s surveys predominantly cover the larger call centres and those within larger organisations, and its respondents are primarily concentrated in long-standing employing organisations in already well-union operations in telecommunications, former public utilities, banking and finance, and the public sector. Consequently, the picture for unions here continues to be one where they have scored relatively little success breaking into the call centre part of the ‘new economy’ or the private services sector.

The same broad picture emerges from the Industrial Relations Services (IRS) surveys (see note to Table 7). An examination of further data from the LRD/TUC Trade Union Trends surveys (Tables 8 and 9) suggests, inter alia, that employers’ willingness to grant union recognition is increasingly concerned with specific union recruitment campaigns and higher membership density rather than the innovative and legal influence of the ERA, and that trade unions recognise this. But unions’ willingness to launch union recognition campaigns is derived largely from self-initiated increases in union membership in the workplace (Table 9). This, in tandem with Tables 10 and 11, suggests that employers are increasingly insisting on unions meeting the stipulations of the statutory process within the voluntary area, that is, 50% plus union density in membership audits and, where ballots are conducted, attaining 50% plus of the votes which also equates to at least 40% of all those entitled to vote. Indeed, the use of the ERA thresholds by employers in the voluntary arena has led unions to aim to gain substantial majorities above these minimum levels in order to counter any employer antipathy or ambivalence. Tables 10 and 11 also indicate that substantial proportions of successful campaigns (respectively 55%, 45%) comprised over 60% union density or 60% worker support in union recognition ballots.

Overall, it can be suggested that the influence of the ERA has been to raise the bar for recognition in the voluntary arena, where previously there was no one standard threshold. Nonetheless, it is noticeable that compared to data for the density at which union recognition was gained between 1995 and 2002 (Gall 2004a:263), the data for 1995-2004 (Table 10) shows that the proportion of new agreements gained with high densities (>70%) has declined from 34% to 28%, and the proportion of new agreements gained with low densities (10%-49%) has risen from 25% to 31%. With the addition of very little retrospective new data for 1995-2002, the changes reflect the increasing reliance of the trade unions on the statutory procedures for gaining new recognition agreements. In  successful statutory recognition ballots, union density need only be 10% at a minimum so long as there is evidence of majority worker support (through a petition, for example) for collective bargaining.  

Tables 12 and 13 indicate that not only are the vast majority of union recognition campaigns conducted amongst small to medium-sized employing organisations but that the vast majority of new union recognition agreements are also won amongst small to medium-sized employing organisations. Indeed, Table 13 shows that the new recognition agreements per year are now covering fewer numbers of workers not just because there are fewer agreements but because the numbers of workers covered by the agreements are getting fewer. The pattern of proportionately more smaller and fewer large new union recognition agreements by the number of workers covered found in the mid-1990s has been returned to by the early 2000s after a period of proportionately more larger and fewer smaller agreements. While these trends might be expected, given that the numerical majority of employing organisations are of a small to medium workforce size (<250 employees), that a greater number of larger employing organisations have not been targeted to date and that union recognition has not been won amongst a greater number of larger employing organisations suggests that unions will find it difficult to significantly increase the aggregate coverage of union recognition. Winning recognition in large (500> employees) employing organisations is generally more of a challenge for trade unions by virtue of the greater number of sites and shifts to be organised across.

Another indication of the potential difficulties unions are experiencing concerns the length of time taken to gain union recognition, with the implication that, on average, longer successful campaigns will require greater amounts of resources than shorter, successful campaigns where unions resources are relatively limited. Table 14 highlights that where recognition was won, 58% of campaigns were successful within two years and 42% of successful campaigns took more than two years to fruition, with 30% taking more than three years. One ‘positive’ reading of this is that recognition was gained in a short period of time in a majority of cases. However, and taken with the data on the success rate of recognition campaigns, another reading is that if recognition is not gained within two years, the prospects of doing so thereafter are very poor. Nonetheless, it is noticeable that compared to data for the length of time taken to gain new union recognition agreements between 1995 and 2002 (Gall 2004a:262), the data for 1995-2004 (Table 14) shows that the proportion of new agreements gained within a year has declined from 26% to 25%, while the proportion of new agreements gained after between one and four years has risen from 46% to 50%. With the addition of very little retrospective new data for 1995-2002, the changes reflect the increasing length of time unions are expending on gaining new recognition agreements. Moreover, and where recognition was achieved after two years, substantial union lay and FTO resources have been necessary to sustain the campaign and to secure a return on the initial resource investment.  

The frequency of incidents of employer anti-unionism and the number of employers using anti-union tactics where employers have faced union recognition campaigns has increased significantly over the period 1995-2005 (see Table 15). Incidents of employer anti-unionism comprise tactics of union suppression like victimisation and sackings and tactics of union substitution like works councils and higher-than-expected pay rises. The level of employer anti-unionism falls from its highpoint in 2002 to a level commonly found in the mid-1990s by 2005. This is explicable by the increasingly fewer number of recognition campaigns run by trade unions. Of critical importance is that union recognition campaigns were successful in less than 30% of the employing organisations which deployed these tactics (see also Gall 2004b).

Furthermore, of the 836 recognition campaigns identified in the Trade Union Trends surveys run between 1997 and 2003, employers in 37% of cases where membership density was 50% or above, which would permit a statutory award of recognition, refused recognition, accounting for 100,000 workers out of 500,000 covered by recognition campaigns in this period (Gall 2004b). However, one countervailing point is that the Dibb Lupton Allsop (1998-2004) annual industrial relations surveys of non-union employers indicate that employer hostility to trade unionism continues to fall (see Table 15). This suggests that a normalising impact, resulting from the recent growth in the number of new union recognition agreements, may be taking place. But because the Dibb Lupton Allsop data is derived from relatively small samples of non-union employers (see Table 15, Row D), considerable caution must be taken in making a definitive judgement here.

For heuristic reasons, the implicit position underlying the preceding assessment of both union organising activity and the utility of the ERA to trade unions in Britain has been that the union movement should, by deploying both, to be able to halt, if not reverse, the decline in the coverage of union recognition. Indeed, the union movement should be seeking to generate the conditions for a mutually reinforcing ‘virtuous circle’ between union recruitment and union recognition (Bain and Price 1983:18), where an upward spiral is created. Rather, and in practice, the most appropriate characterisation of trade unions in Britain vis-à-vis union recognition and union membership (see also Column D, Table 16) between 1995 and 2005 would be ‘running very fast merely to stand still’. In 1995, union membership stood at 7.07m (29.0% density). In 2005, it stood at 6.67m (26.2% density) (Grainger 2006:15). More tellingly, union density in this period in the private sector fell from 24.0% to 17.1% (Grainger 2006:17), meaning that the private sector has increasingly become non-union (as in the United States).

Only the public sector union membership recorded a small absolute increase in the period (although it experienced a decline in density). Indeed, Tony Woodley, Transport and General Workers’ Union general secretary and a leading exponent of ‘union organising’, recognised this when he judged that despite the number of organising initiatives the union movement was ‘struggling to stand still at present’ (Morning Star 13 February 2004, see also Metcalf 2005:23). Thus, the resource inputs have been large but the output gains relatively meagre. The little comfort that the unions can take from this is to counter-factually ask how much less would be the extent of union recognition coverage be if union organising had not been carried out. There are three aspects to such an assessment, concerning the trade union resources, employer opposition and the pro-union strength of the statutory provisions. All have a direct bearing on union organising capacity, and their present complexion and collective configuration suggest significant limitations on capacity.

One environmental  Trade unions have organised more recognition campaigns in the last ten years than at any time since 1979. On the one hand, the relatively low success rate suggests that they may need to consider more strategically which campaigns they mount with a view to raising their ‘strike rate’. However, the genesis of recognition campaigns militates against a high degree of strategic choice for most recognition campaigns develop from (aggrieved) workers approaching unions and not vice-versa (Gall 2004a, and Table 8). On the other hand, and despite the increased resources given over by unions in recent years to union organising of non-union workers and recognition campaigns, the relatively low success rate suggests that if this absolute success rate is to remain unchanged then unions will have to organise many thousands more recognition campaigns to win hundreds more recognition agreements. Here, a key resource weakness for unions is not only the quantitative decline in the size of the lay activist milieu from around 500,000 in 1980 to 230,000 in 2003 (TUC 2004b) but also the qualitative decline in the level of activism and motivation amongst lay activists.

Activists are now less able to be active because of reductions in employer-provided at work facility time for conducting union business in and less inclined to be active in their own non-work/out-of-work time. The specific salience of this decline is fourfold. First, there is a limited amount of organising and recognition campaigning that the 3,000 FTOs in Britain can carry out and there are limited amounts of union financial resources that are available to increase the number of FTOs for this purpose. So far only the TGWU appears to be prepared to spend significant sums on ‘union organising’ in the way that the Service Employees’ International Union in the US has done. Second, there is the unsatisfied demand for union representation through union recognition judged by the 2.8m non-unionised workers who desire union representation or would be very likely to join a union if one were available to them and the 1.6m workers who are union members but are not covered by recognition (Metcalf 2003:38). The pertinence here is that union activists are necessary to locate, recruit and organise these potential members. Third, the cost of recruiting and organising new members in non-union workplaces is sufficiently high (Heery et al. 2000b), because it is often done by FTOs, that additional surplus from these new members is either low or non-existent. This means that inflows of new resources are not being created to recruit further members in non-union workplaces. Lastly, the fragmentation and now decomposition of the social democratic and socialist projects have both reduced the size of the traditional pool from which lay union activists are traditionally drawn and lessened the motivational drive of those that remain as activists. Employer opposition to granting further union recognition now appears qualitatively more important than at any time in the last ten years. This situation arises because the initial increase in new union recognition agreements reflected the relative ease unions had in signing new agreements in the light of the imminence and immediate presence of the new legal provisions by using their existing stock of ‘strong cases’.

Strong cases are defined as those workplaces where union membership was/is in excess of 60% density, membership attachment to the union high and membership participation significant. However, and having used up their strong cases, unions are experiencing difficulty in creating further strong cases and are encountering entrenched anti-unionism amongst the remaining non-union employers which they target. In essence, the potential for new agreements with the less hostile employers has been largely expended while the remaining and reduced scope for new agreements is to be found amongst the more hostile employers[3]. Indeed, although the extent of cases of such employer anti-unionism vis-à-vis union recognition campaigns has declined recently (Gall 2004b), this is a factor of fewer campaigns being run, which itself reflects the importance of underlying of anti-unionism amongst those employers where trade union members exist. Setting aside the validity of any union criticism of the ERA[4], the third statutory recognition mechanism in Britain has had a considerable positive impact on the number of new agreements signed, but only a marginal effect on the aggregate coverage of union recognition.

This assessment of the overall impact of the third statutory recognition mechanism is consistent with the limited impact of the previous statutory recognition mechanisms which operated between 1971-1974 and 1976-1980 (Gall 2003a). The common characteristic of the three statutory mechanisms, notwithstanding significant differences, is their insertion into an industrial relations system of voluntarism or ‘collective laissez faire’ and a stated preference by lawmakers for parties to reach voluntary settlements to recognition issues before taking recourse to the statutory mechanism. Indeed, the CAC will only accept an application where the voluntary route has failed. Calls by the trade union movement for significant reform of the statutory mechanism have concentrated on reducing the exacting nature of a number of the membership and support thresholds and removing a number of barriers as well as creating the offence of an ‘unfair labour practice’. Whilst these measures would be of some benefit in gaining further new recognition agreements and in the ease of doing so, it can be seriously doubted whether they would allow further significant numbers of new recognition agreements to be gained because of employer anti-union and non-union policies and practices.

Even the creation of the offence of ‘unfair labour practice’ following the review of the ERA (Labour Research Department 2004) only pertains to cases subject of accepted CAC applications and, thus, not employer actions in the voluntary arena. Consequently, employer use of unfair labour practices may prevent unions being in a position to make applications to the CAC. Whilst the project of ‘union revitalisation’ in Britain, and embodied in ‘union organising’, has become (relatively-speaking) increasingly more widespread in recent years and has witnessed innovation in union form and practice (Heery et al. 2000a, 2000b, 2000c, 2003), renewal in the manifest fortunes of unions has not been so marked.

Absolute and relative membership levels, along with aggregate union recognition and collective bargaining coverage, have not experienced any form of ‘resurgence’. Rather they have shown a degree of stability at vastly lower levels following many years of substantial annual decline. What wider conclusions can be drawn from this? Under the broad internationalised project of renewal, trade unions in Britain have probably performed in relative terms at least as well as their contemporaries in the US (see Clawson 2003:15, Fiorito 2003:202-207, Jordan and Bruno 2006:182, Katz 2001:346) who began their turn to ‘union organising’ in advance of unions in Britain. One comfort that British unions can draw is that their union renewal project started (and remains) at a higher baseline than that found in the US. Such an overall trajectory for unions in Britain would not seem to equate, for the time being, to either ‘dissolution’ or ‘perdition’ (cf. Metcalf 2001, 2003, 2005). Indeed, by again posing the counter-factual question, ‘How much worse would the situation be in Britain if unions had not engaged in ‘union organising’?’, we may get a better sense of the positive, if nonetheless, limited impact of recent union revitalisation work in Britain.

Nonetheless, the technical and bureaucratic rationalist bent of ‘union organising’, as a ‘top-down’, centralised exercise employing full-time officials as the fulcrum of renewal, and as practiced in Australia, Britain and the US, raises the issue of whether such an approach complements or substitutes itself for the primary underlying dynamic of trade unionism, that of being a social movement (see also Carter 2006, de Turberville 2004, Gall 2006). So, a mainstay of evaluating ‘union organising’ may not be just whether it is appropriate by virtue if the size and scale of the human and financial resources ploughed into it. Rather, it may be to suggest that renewing and growing the networked grassroots activist base of trade unionism is at least as important and this may be usefully approached through deploying social democratic or socialist ideologies which seeks to attain social justice within and without the workplace. This may be one route to ‘squaring the circle’, namely, for unions to engage not just in ‘organising’ renewal but also ‘political’ renewal so that union activists as the key milieux and agency have both the practical and ideological tools to produce widespread, secure and deep-seated ‘union renewal’.      

 ACAS (2000-2005) Annual Report, ACAS London.

Bain, G.S. and Price, R. (1983) ‘Union growth: dimensions, determinants, and destiny’ in Bain, G.S. (ed.) Industrial Relations in Britain, Basil Blackwell, Oxford, pp3-33.

Blanden, J., Machin, S. and van Reenen, J. (2006) ‘Have unions turned the corner? New evidence on recent trends in union recognition in UK firms’ British Journal of Industrial Relations, 44/2:169-190.

Carter, B. (2006) ‘Trade union organizing and renewal: a response to de Turberville’ Work, Employment and Society, 20/2:415-426.Central Arbitration Committee (2002) Annual Report, CAC, London.

Clawson, D. (2003) The Next Upsurge: Labor and the new social movements, ILR/Cornell, Ithaca. Confederation of British Industry (1998-2004) Employment Trends Survey, CBI, London.

Cully, M., Woodland, S., O’Reilly, A. and Dix, G. (1999) Britain at Work, as depicted by the 1998 Workplace Employee Relations Survey, Routledge, London.

de Turberville, S. (2004) ‘Does the ‘Organizing Model’ represent a credible union renewal strategy? Work, Employment and Society, 18/4:775-794.Dibb Lupton Allsop (1998-2005) Annual Industrial Relations Survey, DLA/Gee, Leeds/London.

Fiorito, J. (2003) ‘Union organising in the United States’ in Gall, G. (ed.) Union Organizing: campaigning for trade union recognition, Routledge, London, pp191-210.Gall, G. (2003a) ‘Introduction’ in Gall, G. (ed.) Union Organizing: campaigning for trade union recognition, Routledge, London, pp1-18.

Gall, G. (2003b) ‘Trade union recognition in Britain: developments, issues and prospects’ Human Resources and Employment Review, 1/4:196-202

Gall, G. (2004a) ‘Trade union recognition in Britain: turning a corner?’ Industrial Relations Journal, 35/3:249-270.

Gall, G. (2004b) ‘Management Control Strategies and Union Recognition in Britain’ Human Resource Management Journal, 14/2:36-53.

Gall, G. (2005a) ‘Organising non-union workers as trade unionists in the ‘New Economy’ in Britain’ Economic and Industrial Democracy, 26/1:43-65.Gall, G. (2005b) ‘Union organising in the ‘new economy’ in Britain’ Employee Relations, 27/2:208-225.

Gall, G. (2006) ‘Conclusion: issues and prospects’ in Gall, G. (ed.) Union Recognition: organising and bargaining outcomes, Routledge, London, pp232-237

Gall, G. and McKay, S. (1994) ‘Trade union derecognition in Britain 1988-1994’ British Journal of Industrial Relations, 32/3:433-448.

Gall, G. and McKay, S. (1999) ‘Developments in union recognition and derecognition in Britain 1994-1998’ British Journal of Industrial Relations, 37/4:601-614.

Grainger, H. (2006) Trade Union Membership, DTI/National Statistics, London.

Grainger, H. and Holt, H. (2005) Trade Union Membership, DTI/National Statistics, London.

Gross, J. (2003) ‘Recognition of labor unions in a comparative context: has the United Kingdom entered a new era?’ Chicago-Kent Law Review, 78/1:357-380.

Heery, E., Simms, M., Simpson, D., Delbridge, R., and Salmon, J. (2000a). ‘Organising unionism comes to the UK’ Employee Relations, 22/1:38-57.

Heery, E., Simms, M., Delbridge, R., Salmon, J. and Simpson, D. (2000b) ‘The TUC’s Organising Academy: an assessment’ Industrial Relations Journal, 31/5:400-415.

Heery, E., Simms, M., Delbridge, R., Salmon, J. and Simpson, D. (2000c) ‘Union organising in Britain: a survey of policy and practice’ International Journal of Human Resource Management, 11/5:986-1007.

Heery, E., Simms, M., Delbridge, R., Salmon, J. and Simpson, D. (2003) ‘Trade union recruitment policy in Britain: form and effects’ in Gall, G. (ed.) Union Organizing: campaigning for trade union recognition, Routledge, London, pp56-78.

Income Data Services (1998-2003, 2005) Pay and Conditions in Call Centres, IDS, London.

Industrial Relations Services (2001) ‘Call centre rewards’ IRS Employment Review, 733:4-10.

Industrial Relations Services (2003) ‘Call centre rewards and bonuses’ IRS Employment Review, 786:27-36.

Industrial Relations Services (2005a) ‘Call and contact centre pay and benefits’ IRS Employment Review, 820:27-32.

Industrial Relations Services (2005b) ‘Beyond dispute? Prospects for workplace peace in our time’ IRS Employment Review, 824:9-13.

Industrial Relations Services (2006) ‘Waiting for the backlash’ IRS Employment Review, 848:12-16.

Jordan, L. and Bruno, B. (2006) ‘Does the organising means determine the bargaining ends?’ in Gall, G. (ed.) Union Recognition: Organising and Bargaining Outcomes, Routledge, London, pp181-197.

Katz, H. (2001) ‘Afterword: Whither the American labor movement?’ in Turner, L., Katz, H. amd Hurd, R. (eds.) Rekindling the Movement: Labor’s quest for relevance in the twenty-first century, ILR/Cornell, Ithaca, pp339-350. 

Kaur, H. (2004) ‘Employment attitudes: Main findings from the British Social Attitudes Survey 2003’ Employment Relations Research Series, No. 36, DTI, London.

Kersley, B., Alpin, C., Forth, J., Bryson, A., Bewley, H., Dix, G., and Oxenbridge, S. (2005) Inside the Workplace: First findings from the 2004 Workplace Employment Relations Survey, DTI, London.

Labour Research Department (2004) ‘Government promises to act on union recognition rights’ Labour Research, 94/5:27.

LRD/TUC (1995-2005) Trade Union Trends/‘Focus on recognition’ Trade Union Trends, Trades Union Congress, London.

Metcalf, D. (2001) ‘British unions: dissolution or resurgence revisited’ Centre for Economic Performance Discussion Paper No. 493, London School of Economics.

Metcalf, D. (2003) ‘British Unions: Resurgence or Perdition?’ London School of Economics, mimeo.

Metcalf, D. (2005) British Unions: Resurgence or Perdition?, Provocation Series, 1/1, The Work Foundation, London.

McArthur, B. (2004) ‘The efficacy of statutory union recognition under new Labour: a comparative review’ International Journal of Comparative Labour Law and Industrial Relations, 20/3:399-421.

Millward, N., Stevens, M., Smart, D. and Hawes, W. (1992) Workplace Industrial Relations in Transition, Dartmouth, Aldershot.

Millward, N., Bryson, A. and Forth, J. (2000) All Change at Work? British employment relations 1980-1998, as portrayed by the Workplace Industrial Relations Survey series, Routledge, London.

Moore, S. and McKay, S., and Bewley, H. (2004) ‘The content of new voluntary recognition agreements, 1998-2002: Volume one – an analysis of new agreements and cases studies’ Employment Relations Research Series, No. 26, DTI, London.

Moore, S., McKay, S. and Bewley, H. (2005) ‘The content of new voluntary recognition agreements, 1998-2002: Volume two – Findings from the survey of employers’ Employment Relations Research Series, No. 43, DTI, London.

National Statistics Online (1995) ‘Public sector employment growth, 1998-2005’ available at www.statistics.gov.uk/cci/nugget.asp?id=407.

TUC (1996-2004a, 2005, 2006) TUC Directory, TUC, London.TUC (2004b) Union reps website, accessed 12 August, TUC London.

 Table 1: New Trade Union Recognition Agreements, 1995-2005 

Source: See methodology.Note: By taking the known number of workers covered from 71% of the new agreements and averaging up from this for the remainder, around 1.2m workers are covered by all the new agreements.

Table 2: Trade Union Recognition Claims Involving ACAS, 1999-2004 

Source: ACAS Annual Reports (2000-2005) and communication with ACAS.Note: n/a = not available.

Table 3: Union Recognition Coverage from the Labour Force Survey, 1993-2005

Source: Labour Market Trends (various), Grainger (2006:39) and Grainger and Holt (2005:36).Note: After 1998 the LFS questions were changed so continuing data on the 1993-1998 questions on union recognition is not available. Data on the number of employees whose pay is affected by collective agreements  is the best available proxy to those covered by union recognition.

Table 4: Coverage of Union Recognition and Collective Bargaining 

Sources: Column B-Millward et al (2000:96), Kersley et al. (2005:13), Column C- Millward et al (1992: 107), Cully et al (1999:92), Column E- Cully et al (1999:242).

Note:  The first figures for 2004 are those which are commensurate with the previously used question. The second figures for 2004 are those where managers were now asked if unions were recognised even where there were not believed to be any union members in the workplace (resulting from industry-level agreements or extension of recognition to new areas of an employing organisation).

Table 5: British Social Attitudes surveys: percentage of workers in workplaces with recognition

Sources: Alex Bryson, Policy Studies Institute, personal e-mail communication run from British Social Attitudes surveys, 24 March 2004 and Harjinder Kaur, Department of Trade and Industry, personal e-mail communication run from British Social Attitudes surveys, 20 July 2005.Note: The question on union recognition was not asked in either 1988 or 1992.

Table 6: Employers’ Perception of Recognition Claims 1998-2004  

Source: CBI (1998- 2004).  

Table 7: Union Recognition in Call Centres, 1998-2005 

Table 8: Reason for Gaining Recognition 1996-2004 

Source: Trade Union Trends (1996-2005).Note: Respondents can enter one or more reasons or none, producing totals greater or lesser than 100%.

 Table 9: Reason for Recognition Campaigns, 1996-2003  

Source: Trade Union Trends (1996-2004).Note: Respondents can enter one or more reasons or none, producing totals greater or lesser than 100%.

Table 10: Union Recognition Granted 1995-2005: known union density level (n=523)

                                 
Source: Data gathered from fieldwork (see methodology).

Table 11: Known Ballots Results where Recognition Gained, 1995-2005

Source: Data gathered from fieldwork (see methodology).Note: Overall support level results from ‘turnout’ multiplied by ‘vote for’ where data exists or where just overall level of support known. The number of ballots (voluntary and statutory) held annually since 2000 which have led to new union recognition agreements runs as follows: 46 (2000), 67 (2001), 44 (2002), 33 (2003) and 24 (2004). This means that the number of new agreements gained per annum by ballot has varied between 9% and 13%. In addition some forty three, largely statutory ballots, were held in the period that did not lead to union recognition.  

Table 12: Number of workers in each campaign, 1996-2005

Source: Trade Union Trends (1996-2006).Note: Figures for 2002-2004 are not available from the surveys because only those campaigns with in excess of 50% membership were listed.

Table 13: Number of workers covered by each new agreement, 1995-2005 (n=2133)