Trade Unions: Back from the brink or still on the margins?

Issue: 105

Gregor Gall

Socialists recognise unions’ central role and workers’ collective power at the workplace in not only the socialist project but also in advancing workers’ economic and political interests, and the link between the two (revolution and reform). The argument presented here is that so significant is the contemporary disarticulation in union presence and power in Britain that this must temper expectations of what trade unionism can currently achieve. It must, therefore, also temper the expectations of the returns on socialists’ union work. Consequently, this article seeks to provide an analysis of the current union movement and make a critical examination of the ‘industrial perspective’ put forward by the Socialist Workers Party (SWP)1 over the last decade.

The SWP’s position can be characterised as one of the industrial struggle ‘being on the cusp’ of a significant upturn in workers’ combativity and political consciousness, although not necessarily of an ‘upturn’ in toto. The language is replete with ‘new mood’, ‘revival’, ‘explosive situations’ and ‘magnificent’ struggles in a ‘revival of militancy’ with a growing rank and file attitude in unions.2 Despite frequent caveats and qualifications, the consistent message is one that can be said to be highly optimistic, positive and confident. This article also puts forward a contrasting analysis with a view to guiding the perspectives of socialists’ and militants’ union work. It characterises the early 1990s onwards as neither an era of ‘downswing’ nor ‘upswing’ in union struggle but one of ‘disarticulation and disorganisation’, where levels of union organisation and combativity are low but where within these parameters the results of that organisation and activity are not completely unsuccessful. Thus the question addressed is, are unions in Britain back from the brink or still on the margins?

The current state of play: the upside

The most obvious sign is the depletion of the right’s hold on union leader-ships. Together with pre-existing left leaderships, the vast majority of major, high profile and strategically placed unions are now left-led. Recent lay NECs elections have followed this trajectory. Part of this trajectory has been the questioning of unions’ political allegiance and that of their polit¬ical funds. Where democratisation and disaffiliation have been rejected, unions have implemented reduced funding and more exacting policies of what their money buys. Developments in the FBU and RMT are the starkest aspects of this general phenomenon. Politically, unions are now more popular than at any time since 1979. Successive polls have demonstrated that clear majorities believe employers are too powerful, unions are ‘good’ in wielding the ‘sword of justice’ (as opposed to representing ‘vested interests’) and that unions are not powerful enough. Moreover, unions have constituted the most effective opposition to Labour, marginalising the official opposition parties.

The decline in membership has been halted, with increases recorded in 1998, 1999 and 2000, then falls in 2001 and 2002, and another increase in 2003 (Table 1). Since 1995, and with the imminence of a statutory union recognition mechanism, over 2,500 new recognition agreements have been signed, covering 1.1 million workers.3 Unions have increasingly adopted an ‘organising’ approach based on members setting the agenda in recognition campaigns and the new agreements are overwhelmingly not ‘partnership’ agreements. These new agreements and the increase in public sector workers have countered further retrenchment of unionised manufac¬turing jobs so that the absolute number of workers whose pay is covered by collective bargaining has also witnessed a similar pattern to membership (Table 1). Allied to these, the decline in the number of days ‘lost’ in strikes since the late 1980s has bottomed out (Table 2). The proportion of strikes that were unofficial has increased to around 40 percent (Table 2), indicating some resilience in workplace organisation. Of significance is that the results of industrial action have not continued the rout and retreat of the 1980s. The FBU’s strikes of 2002-03 and the Scottish nursery nurses’ strike of 2003-04 represented relatively rare events of the last decade, that is, defeats for major strikes. The dominant outcome has been of favourable compromises (eg medical secretaries, and postal, hospital, bus and railway workers).

Table 1: Union density, recognition coverage and bargaining coverage, 1993-2003

Source: TUC and the Labour Force Survey in Labour Market Trends

Year TUC members total union members union density workers covered by recognition recognition density workers covered by cb * cb density
1993 n/a 8.001m 35.1% 10.420m 48.9% n/a n/a
1994 n/a 7.756m 33.6% 10.374m 48.2% n/a n/a
1995 6.898m 7.543m 32.3% 10.226m 46.8% n/a n/a
1996 6.799m 7.472m 31.5% 10.141m 45.8% n/a n/a
1997 6.756m 7.372m 30.4% 10.032m 44.3% n/a n/a
1998 6.754m 7.396m 29.9% 10.081m 43.5% n/a n/a
1999 6.746m 7.498m 29.6% n/a n/a 8.771m 36.1%
2000 6.816m 7.580m 29.5% n/a n/a 8.924m 36.2%
2001 6.721m 7.550m 29.1% n/a n/a 8.869m 35.6%
2002 6.685m 7.390m 29.0% n/a n/a 8.700m 35.6%
2003 6.690m 7.420m 29.1% n/a n/a 8.750m 35.8%

* collective bargaining

Table 2: Strike activity in Britain, 1992-2003

Source: Labour Market Trends and G Gall (2004)6

Year No of strikes Workers involved (m) Days ‘lost’ (m) % strikes which were unofficial % all workers involved in unofficual strikes % all days not worked in unofficial strikes
1992 253 0.148 0.528 40% 19% 9%
1993 211 0.385 0.649 36% 5% 7%
1994 205 0.107 0.278 48% 48% 32%
1995 235 0.174 0.415 39% 28% 21%
1996 244 0.364 1.303 47% 9% 4%
1997 216 0.130 0.235 36% 50% 41%
1998 166 0.093 0.282 49% 51% 58%
1999 205 0.141 0.242 45% 33% 26%
2000 212 0.183 0.499 80% 20% 17%
2001 194 0.183 0.525 59% 43% 15%
2002 146 0.943 1.320 38% 3% 4%
2003 133 0.151 0.499 32% 41% 26%

Alongside these developments there has been the relative decline in the pervasiveness of ‘partnership’ as an ideological current and the emergence of oppositional networks around newspapers in the post, fire service, NHS and rail transport. Finally, national unions have begun to engage in extra-workplace trade unionism such as community unionism (eg TELCO) and social movement unionism (eg the anti-capitalist and anti-imperialist movements). Now a new space has been opened up for more critical and progressive ideas to circulate more widely concerning some basic questions such as what the role of unions should be (defenders of workers’ interests, not of employers and Labour governments) as well as more developed ques¬tions about the nature of capitalism (over globalisation, war, imperialism).

The current state of play: the downside

So there have been a number of forward moving developments in the last decade. The salient issue becomes what significance and portent to attribute to these. In doing so, recognition of a number of other phe¬nomena is necessary for full contextualisation. While the ‘sensible squad’ leads ASLEF, TSSA and USDAW, of greater significance is the ‘awkward squad’s’ political and industrial diversity over affiliation to Labour, political funds and European integration and how much pressure to exert on Labour and how to do this. While withering criticism and behind the scenes lob¬bying are widespread, only a minority advocates mass mobilisation and extra-parliamentary campaigning. Consequently, and despite coordination among the left unions at Labour and TUC conferences, there is little evidence of these left leaders working together in a collective and effective way outside ‘forums of the few’. And therefore, there is also no significant evi¬dence of the unions being able to lever the concessions they want on revisions to the statutory recognition provisions, abolishing the anti-union laws or ending PFI/PPP.

There exists a considerable gap between the political and industrial struggles within unions, where political advances are not in the main the result of industrial advances. Indeed, they are, in the main, an expression of a relatively passive rejection of collaborationist strategies, not an active endorsement of militant strategies, and they emanate, in large part, from developments outside workplaces and trade unionism itself. Another notable feature of rejecting ‘new’ Labourism is the absence of a credible left wing alternative. There has been no new Alternative Economic Strategy. This is not to bemoan these characteristics of ‘stunted’ political develop¬ment but to suggest they have implications for the pervasiveness of the politicisation and how it can be operationalised within workplaces.

Despite developments with the FBU and RMT, nine of the biggest ten unions are led by general secretaries who are of the ‘Reclaim Labour’ position, have never questioned their unions’ existing affiliation to Labour or are not affiliated to Labour. They comprise 80 percent TUC affiliated members. PCS is the only exception by dint of its leading officers but it has yet to determine its political affiliation, if any. The RMT has 71,000 members while the FBU has 52,000. Both were predisposed to the moves they made, namely, being small, relatively homogeneous unions with public sector/ex-public sector based memberships and having already been left-leaning for a considerable period of time. The stimulus of recent experiences (reversal of renationalisation pledge, strike defeat) has coalesced around these longer standing characteristics. The prospect of major unions moving towards RMT/FBU positions is not likely in the short or medium term. Consequently, the emergence of small-scale union support for the independent left is likely to remain as such.

Turning to the industrial struggle, the annual number of strikes and workers involved continues to fall, and where strikes occur the over¬whelming majority are in the public and ex-public sectors (post, railways, buses, councils, hospitals and fire service). Sizeable strikes in the private sector are becoming increasingly rare. Strikes generally are predominantly characterised by being discontinuous one-day actions and by selective groups of workers. Consequently, annual movements in strikes (days ‘lost’ and workers involved) are heavily affected by single strikes as in 2000 and 2002. Strikes also continue to be predominantly defensive (against management’s demands, against wages offers etc) rather than offensive (for workers’ demands, for higher pay rises, for restricting management control) action.

The location of strike activity reflects the distribution of membership. Most union members are public sector workers (circa 60 percent), with public sector density being 61 percent and private sector density being 19 percent.4 The new recognition agreements, most of which are in the private sector, have not altered this. Atrophy in the private sector is unlikely to change as the annual number of new recognition agreements has declined since 2001 and unions are now meeting stiffer employer resistance to granting recognition. Post-recognition, many employers who reluctantly conceded recognition are attempting to limit its worth. Currently, unions have to run five recognition campaigns to win every two. In sum, unions’ organising capacity (financial, ideological and organisational resources) is far less than the scale of the task facing them. Finally, the union ‘mark-up’ (the positive difference between wages and conditions in unionised and non-unionised workplaces) has declined from 13 percent in 1994 to 3 percent in 2001.5

These ‘negative’ features provide a sobering downside to the ‘posi¬tive’ developments outlined earlier. This indicates that the positive features are contested and counter-balanced by other developments. Of course, the key point for socialists concerns how to understand the portent of these positive developments. But before this can be attempted, existing, longer-term trajectories and tendencies must also be considered. This adds another important corrective.

The dimensions of decline

The 1970s constitute the seminal reference point for a socialist analysis of the state of trade unionism today for they represent the highpoint of post¬war struggle before the employer and state offensive against organised labour. They are, without being overly mechanical and notwithstanding some characteristics like sectionalism, a high tide that socialists aspire to recreate in terms of workers’ collective self-confidence and combativity. Taking the year 1979 as the historical turning point allows the last decade to be put in a historical context.

Membership stood at 13.5 million (55 percent density) in 1979. In 2003 it stood at 7.4 million (29 percent density). Now nearly 50 percent of workers have never been union members at any time, and only 11 percent of workers aged 16 to 24 and only 25 percent of workers aged 25 to 34 are members,7 signalling an ageing membership. In 1979 there were 500,000 union workplace reps like shop stewards. In 2003 there were some 230,000. Workplace union organisation has been severely weakened in most places and atrophied elsewhere. The vibrant networks of inter- and intra-industry shop stewards no longer exist in any meaningful way. The major exceptions to this pattern are a number of workplaces in the public and ex-public sector that can support strikes (see before). Membership participation in unions is low. Membership is seen in a passive way and as a form of instru¬mental transaction for insurance: paying dues and expecting service in return without participation and activity in protecting interests.

The percentage of workplaces covered by recognition fell from 64 percent (1980) to 42 percent (1998) while the percentage of workers in workplaces with recognition fell from 66 percent (1983) to 47 percent (2001).8 Bargaining coverage has fallen from 70 percent (1984) to 41 percent (1998).9 Some 3 million workers are ‘free riders’—benefiting from bargaining without being members—while 1.7 million members are not covered by recognition. Strike activity has fallen from 2,125 strikes in 1979, with 4.6 million workers involved and 29.4 million days ‘lost’, to 133 in 2003, with 0.151 million workers involved and 0.499 million days ‘lost’. The only major blip was the 1984-85 miners’ strike. Occasional large strikes (eg postal workers, 1996, and council workers, 2002) have merely and temporarily made a slight dent. Furthermore, solidarity strikes, the hallmark of combative trade unionism, are almost unheard of now, apart from those in the post. And unofficial strikes have fallen from 95 percent of all strikes prior to 1980 to 40 percent today.

This retrenchment in union presence and power means that employer ability (private or state) to determine employment conditions, as well as wider issues in society, has increased dramatically. What makes this decline particularly acute is that unlike previous periods of low unemployment, like 1988-89, workers have not been able to take advantage of the tightened labour market.10 Likewise in periods of growing levels of employ¬ment union membership would have been expected to grow substantially. Rather there appear to be contradictory routes for small membership increases for individual unions whether by oppositionalism (eg PCS) or partnership (eg USDAW).

Disarticulation and disorganisation

The historical and contemporary contextualisation suggests three points: the positive developments are built on weak foundations; the positive developments are themselves quite frail and potentially superficial; and unions are still on the margins. Consequently, disarticulation and disorganisation are the most appropriate characterisation for trade unionism in Britain in the last decade. Disarticulation refers to the quantitative state of unions while disorganisation refers to the qualitative state of unions. The quantitative aspect concerns the large gap between the number of unionised workers and the desire of worker representation while the quali¬tative aspect concerns the relatively poor health and lethargy of existing union membership and organisation.

The purchase of this characterisation is that it analytically helps unravel a period that is neither one of ‘downswing’ (eg 1921-35, 1979-89) nor of ‘upswing’ (eg 1910-14, 1917-20, 1968-74). The measures to judge any period, but in particular whether a period can be characterised as one of downswing or upswing, are those of process and outcome. Broadly speaking, ‘process’ denotes the level and nature of membership, organisation, bargaining and strikes while ‘outcome’ denotes the results gained by workers’ actions and organisations in terms of wages and conditions, control of the employment relationship and consciousness. 1993-2003 has shown some relatively modest signs of continuing resilience and some small indications of recovery in process and outcome but these are evident within an era following a heavy downswing that still sets the broad ideological and organisational parameters of what can be achieved.

The importance of the characterisation of disarticulation and disor¬ganisation is that it suggests there are there intermediate periods of transition between the two types of ‘swing’ which are highpoints in open and intense class struggle. Moreover, ‘swing’ periods do not directly determine what unfolds thereafter for there is not a direct and close correlation between political and industrial struggles. For the last decade the manifesta¬tions of a higher degree of political and social discontent have not been paralleled by those of the (lower) level of industrial discontent. How does this differ from the period of downswing preceding it? Successful state and employer offensives are not the hallmark of the last decade as they were of 1979-89. Workers’ grievances are widespread but are underlying and sullen without significant mobilisation to secure redress. Between 1979 and 1989 these were open and stark through mobilisation.

A number of categorisations of the last decade and the next few years for the state of trade unionism are then most credible. The main ones can be outlined as follows.

First, an unevenness in a tentative and protracted process of industrial and political union revitalisation, where there is diversity across, and within (horizontally, vertically), unions. There is the unsatisfied demand for repre¬sentation through union recognition judged by the 2.8 million non-union workers who desire union representation or would be very likely to join a union if one were available11 and the union members not covered by recog¬nition. However, ability to recruit and organise these groups is made difficult by the depleted numbers of shop stewards, the small number of full time officers (around 3,000), and the high costs of recruitment through full time organisers.12 Lessons from, and movements in, strikes, bargaining set¬tlements, internal elections and union organising permeate slowly across unions, with only a sizeable and cumulative positive effect being noticeable after a number of years. Politically, the further growth of the still small inde¬pendent left milieu in the unions is held in check by the dominant ‘Reclaim Labour’ union leaderships and the absence of any other large and credible alternative. The independent left is able to address but unable to make a firm connection with a growing audience because it lacks the resources, opportunities and credibility. This could be characterised as ‘slow one-way traffic’.

Second, the opening up and deepening of contradictory tendencies, where there continues to be no dominant overall response to the weakening of trade unionism and no overall strategy towards revitalisation, reassertion and growth. The tendencies towards assertive actions, political radicalisation and industrial conflict (eg PCS, RMT, TGWU) are matched by those ten¬dencies towards passive reactions, neutered political criticism and conservatism, and industrial cooperation (eg TUC, Amicus, GMB, UNISON). Overall the continuing resilience, though inactivity, of social democracy is a major feature. Social democracy, being broad and heteroge¬neous, allows a) micro- and macro- ‘social partnership’ with employers and government respectively, b) ‘organising’ approaches to increase bargaining power with employers and government to gain better compromises, and c) consistent attempts to mobilise members and wider forces to secure political and economic victories for workers, to simultaneously exist.

Third, current trends merely indicate the bottoming out of the downward path of decline and subsequent stabilisation of union presence, organisation and activity at this lower level. Here the main tendency is to manage the decline while waiting for extraneous developments to lead to growth and renewal or to accept the historical weakening of trade unionism as a secular trajectory without acknowledging the cyclical nature to indus¬trial struggle and trade unionism. If this path continued, and was neither challenged nor interrupted, the prospect is of an ‘American nightmare’: unionisation fell from 33 percent in 1965 to 15 percent in 2000, of which 70 percent is now found in the ‘public’ services, and the private manufac¬turing and service sectors are effectively de-unionised and non-union.13 As in the US and no matter how ineffectually, the political representation of organised labour continues to be provided by the more progressive of the main neo-liberal parties, eg Labour.

Fourth, current developments represent the rearranging of existing union forces to the left without a significant growth in the absolute or rela¬tive size of these. Consequently, the left dominates but in a period of depleted power and action, suggesting that if unions under left leadership cannot break out of encirclement by hostile forces then the window of opportunity for the left and union revitalisation will be lost for some con¬siderable time to come.

The fifth scenario that socialists and most trade unionists would put at the top of their wish list is in all probability just that: a wish. This com¬prises significant and short term upturns in the strength and level of membership, organisation, and industrial struggle on the one hand, and considerable material benefits and confidence-raising outcomes to strikes and disputes, and a leftward moving mass politicisation, on the other. While upturns have to, and do, come from somewhere, no matter how great workers’ weakness at the time, it is not credible to suggest they come out of nowhere. The signs and reverberations can be detected and not just in retrospect. Posing the counter-factual, ‘What would an upturn look like?’ would help. The scenarios highlight the current complexity and not the unidimensional formulation of, ‘Either trade union militancy will have to rise to the level of the political movements or the political movements will fall back to the level of the class struggle’.14

Union tasks today

Far greater financial and physical resources need to be put into recruitment, retention and organising so that new members and new recognition agreements can be gained in droves, not dribbles. Currently the ‘Organising Academy’ and the spread of the ‘organising culture’, though welcome, are insufficient to the task. Unions need to spend 30 percent, not 5 percent, of their income on these. Members must participate more. Open, devolved and transparent structures need to be fought for. Members must not view their union as an insurance policy, but an active, living being. Unions still remain too ‘male, pale and stale’. Internal education about trade unionism is needed to raise members’ consciousness and understanding. Workers and unions need to understand mobilisation is the critical factor in successfully pursuing relatively ambitious demands. Unions also desperately need some high-profile victories to ‘sell’ their wares. Whether affiliated to Labour or not, unions need to engage in extra-parliamentary mobilisation for this is where their strength lies. Only if done in this way can unions benefit from the changes they have forced for neither repealing the anti-union laws nor improving the recognition will in themselves lead to more successful strikes or more union recognition. Workers need to win back their collective con¬fidence for this to happen.

Probabilities and possibilities

While there is broad consensus on such tasks, there is disagreement on the prospects of achieving these. The disagreement stems from problematic assessments that are based on the ungrounded futurology of possibilities and potential and not the more exacting measures of probabilities and prospects. The source of this weakness is the difficulty of setting out a socialist perspective that is also sufficiently favourable to be able to motivate to the relatively small forces at hand. This commonly leads to exaggeration of the favourableness of conditions and, based on this, exhortations to action and activity. Socialists must have motivation to become an agency within unions for socialism but this must be one that is not corrosive of that motivation by becoming bombastic. Achieving a grounded perspective which contains an analysis of the ‘state of play’ and a guide to action is based on making grounded abstractions.

Industrial generalisations

While socialists must generalise from the experience of the most militant workers and seek to apply this elsewhere, they must also be aware of the specificities of certain unions, industries and struggles. The best example to illustrate this is the postal workers. Arguably, they have been the most com¬bative group of workers in the last decade. But in responding to the marketisation and de facto privatisation, a trajectory experienced elsewhere by other workers, postal workers have operated in a specific context of de facto monopoly provision, a service for which there is no ready substitute and a highly integrated business (internally and externally) where strike action is highly visible and makes an immediate impact.15 While it is right to point to postal workers and right that other workers should try to emulate them, this cannot be done without being acutely sensitive to their idiosyn¬crasies. Moreover, their strikes have accounted for less than 1 percent of all the days worked in the last decade in Royal Mail.16 The salience is that different exemplars are appropriate in different cases and generalisation from one group of workers to all workers needs to be done with extreme caution.

Political generalisations

Another instance of inappropriate generalisation concerns the degree of imputed and alleged overt politicisation that is manifest in strike activity. Most workers’ consciousness continues to exhibit relatively weak and indi¬rect links between experiences of work and experiences of the party in office and capitalism in general. Striking in the current period does not lead to extensive political generalisation and does not even indicate great overall hostility to the government, even where the government is the ultimate employer. Why is this so? Among the most salient factors are that strikes are temporary affairs and predominantly concerned with specific workplace issues, and current strikes are of a limited duration and mobilisation. Moreover, workers’ ‘anger’ does not represent advanced political under¬standing for it can be unstable and lead in different directions. Where leftward politicisation has taken place in recent years (eg firefighters, nursery nurses and railway workers), this indicates that bitter, protracted strikes by a mass of workers are necessary along with the bargaining focus being a public authority where the assumed party of organised labour refuses to intervene or accede for clear neo-liberal political reasons, eg to gain ‘modernisation’ or because wages ‘cause’ inflation. In this situation, socialists have to not only recognise that those workers who have made the requisite generalisations are a minority but also quantify this minority so as to gain a better understanding of the size of socialist forces and the periphery around them. If the size of the left vote in recent union elections is used as a yardstick, it must be recognised that many voting left will be voting out of rejection for other candidates and for people other than themselves to carry out the activism.

Historical generalisations

The forcefulness of national union leaderships’ tendency to act in cooperative ways as a result of their intermediate class position is overstated for little thought is given to the impact of contemporary processes. If workers have experienced demoralisation in the Tory era, then union leaderships are not immune either. Consequently, the cautious nature of what union leaders are prepared to call has related to their analysis of the state of their members’ combativity and confidence. Leaders can campaign and motivate but they cannot magic away the effects of disarticulation and disorganisation.

What are the practical conclusions to this? First, expectations of what unions and workers are capable of must be guided by grounded analysis that is sensitive to nuance. This means not collapsing objectivity and subjectivity into each other: two swallows do not make a summer. Consequently, less hyperbole should be used, tone should be moderated, and inflection added. Doing otherwise is to risk disorientating activists and alienating them from other new and seasoned forces. Second, previous predictions of improve¬ments in industrial struggle which did not transpire need to be critically re-examined. Third, so serious is union decline that socialists must give union work particular attention. The decline will either hold back any advances workers make generally or they will be built on shaky founda¬tions. In order for this to happen socialists must work hard with others to increase membership and (re-)establish organisation and so on but with rather more limited expectations. Importantly, this means initiating cam¬paigns and struggles that are not ‘party’ campaigns in order to provide leadership, ideas and organisation to champion members’ interest within unions. In this way, socialists can win respect by being among the best, if not the best, trade unionists. ‘Party’ campaigns can then be run from a solid base and socialists can then without reproach raise the issue of democratisa¬tion of the political fund and challenge the grip that Labour has on the unions.


1: See the writings of Martin Smith (The Awkward Squad: New Labour and the Rank and File (SWP, 2003); Socialist Review, December 2003, June 2003, September 2004; and ‘The Return of the Rank and File?’ in International Socialism 94, Spring 2002) and those of other leading members of the SWP in Pre-Conference Discussion Bulletins (2002, 2003), Socialist Worker (eg 8 November 2003, 17 January 2004, 21 and 27 February 2004) and Socialist Review (eg September 2003, November 2003, December 2003, September 2004). These are the most recent examples but the same position has been adopted for a number of years.

2: Taken from M Smith’s The Awkward Squad and International Socialism article, as above.

3: G Gall, ‘The Fruits of Our Labour? Outcomes of the New Union Recognition Agreements’ (TUC, 2003).

4: Labour Force Survey in Labour Market Trends, July 2002.

5: A Bryson and R Gomez, ‘Marching on Together? Recent Trends in Union Membership’, in British Social Attitudes: the 19th Report (London, 2002), p61.

6: Labour Market Trends’ annual article on industrial disputes (various years) and G Gall, ‘Unofficial Strikes in Britain Twenty Years on from the Miners’ Strike: Challenges and Changes’ (Newcastle, 2004)

7: Labour Force Survey in Labour Market Trends, March 2004

8: N Milward, A Bryson and J Forth, All Change at Work? (London, 2000), and data from the British Social Attitudes survey from Alex Bryson, Policy Studies Institute, London.

9: N Milward, et al, as above

10: Compare Martin Smith, ‘The return of the rank and file?’, p54, The Awkward Squad, p12, as above.

11: D Metcalf, ‘British Unions: Resurgence of Perdition’, LSE 2003.

12: E Heery and others, ‘The TUC’s Organising Academy: an assessment’, Industrial Relations Journal 31/5.

13: B Towers, ‘Comparisons and Prospects’ in G Gall, Union Organizing (London, 2003)

14: M Smith, ‘The Return of the Rank and File’, p53. See also, the Awkward Squad, p34.

15: G Gall, The Meaning of Militancy? (Aldershot, 2003)

16: as above.