XI. LERA REFEREED PAPERS: LABOR AND EMPLOYMENT LAW, LABOR UNIONS/LABOR STUDIES, INTERNATIONAL/COMPARATIVE & NAFTA

Trade Union Recognition in Britain: Is a Corner Being Turned?

Gregor Gall
University of Stirling

Introduction

     Like many other union movements throughout the Western world, the trade union movement in Britain has experienced significant decline in terms of union recognition coverage; however, the considerable recent growth in new trade union recognition agreements in Britain is now well established. More than 2,600 new recognition agreements, covering some 1.1 million workers, have been signed since 1995 (Table 1). The annual growth rate has increased dramatically since 1999. Broadly speaking, this is attributable to increased union organizing activity seeking to benefit from the more favorable legal environment as a result of the statutory union recognition provisions of the Employment Relations Act 1999 (ERA). Only a small part of this growth, however, comprises the 330 applications under ERA, which have led to eighty-nine statutory and sixty-eight voluntary agreements since June 2000. The demonstration and shadow effects of ERA union recognition provisions in terms of their imminence, presence and usage, along with considerable union campaigning activity, thus account for

the vast majority of this growth in what are voluntary agreements. Previously, around seventy new deals were being signed annually in the 1990s, matched by an equal number of cases of derecognition. This paper asks, inter alia, a number of questions about this growth: Does it actually make a difference to the contours of industrial relations in Britain? What kind of growth would be necessary to take the coverage of recognition back to former levels? What are the probabilities of current growth levels being maintained? And what kind of public policy and legal framework would be supportive of increased and prolonged levels of growth?

Methodology

     The data for this research are derived from a number of sources. First, material from the Labour Research Department¿Trades Union Congress (LRD-TUC) Trade Union Trends (1996¿2004) surveys on union recognition. Beginning in 1995, they are compiled annually from questionnaires to unions that cover on average 75 percent of the membership of TUC's affiliates. Second, eight semistructured interviews with regional field and national fulltime officers from the fifteen unions who were involved with or responsible for union recognition campaigns. These fifteen officials were interviewed in 1999, 2000, 2001, and 2002. Third, the determinations of the CAC, the body charged by the ERA with adjudicating on applications for union recognition, were also utilized. These were supplemented by the monitoring of more than thirty union journals and newsletters from 1995 to 2003, information gained from various union-orientated conferences on union recognition, the survey of union-orientated publications.

Question 1: Does the Growth Alter the Existing Industrial Relations Landscape in Britain?
     Does the growth in new union recognition agreements mean that industrial relations, defined as the institutions and processes of joint-regulation based on trade unionism and collective bargaining, is becoming any more pervasive and relevant? The gains in themselves are impressive given the annual increase in agreements, as Table 1 makes clear. This number of new agreements is testament to both the hard work of members, activists and full-time organizers (FTOs) alike and the increased amount of resources devoted to organizing and recognition campaigns by trade unions. But a number of aspects cast doubt on the longer-term significance of this growth, where it is assessed using the criteria of having the potential to significantly increase union recognition and collective bargaining coverage (by establishments and workers) throughout the economy.

     First, the persistence of considerable numbers of redundancies, and even closures, among employers that are unionized in the period, particularly in manufacturing, means that the potential increase in union recognition coverage is offset by this countervailing process. The situation is akin to a set of revolving doors„as some enter, others leave. It is unclear what the exact countervailing impact is, where the Labour Force Survey (LFS) no longer investigates recognition and bargaining coverage in the way it formerly did; however, using the absolute and relative numbers of workers whose pay is affected by collective bargaining as a proxy, the LFS records a marginal rise in 1999¿2000 and then marginal falls in 2000¿2001 and 2001¿2002 (Brook 2002, 2003).

     Second, the continuing overall growth in employment in the non-unionized private service and "new economy" sectors suggests that even if all the growth in new recognition agreements was not offset by redundancies, the relative coverage of union recognition may still not increase because of the growth of non-unionized employment. Third, and related to this, the success of unions in winning new agreements in the private service and "new economy" sectors has been limited. By contrast, unions have had much greater relative and absolute success in winning new agreements in their traditional areas of strength such as general manufacturing. Because of this, redundancies among organizations that have recently signed (new) union recognition agreements mean that another countervailing process exists. Far fewer cases of increases in the size of workforces employed by organizations that have signed new recognition agreements have been recorded. Finally, the significant fall in the number of new agreements in 2002 and 2003 raises starkly the prospect of, inter alia, a decline in union-organizing capacity, an increased reliance on use of the statutory provisions and the meeting of greater employer resistance (see Question 3).

     The growth in new agreements must also to be put in the context of over two decades of continual decline in union recognition coverage since 1979. The Workplace Industrial Relations Survey (Millward, Bryson, and Forth 2000, 96) records union recognition by establishment falling from 66 percent in 1984 to 53 percent in 1990 to 42 percent in 1998, whereas Bryson (A. Bryson, personal communication), from the British Social Attitudes surveys, records the percentage of workers in workplaces, with union recognition as falling from 66 percent in 1983 to 58 percent in 1990, 50 percent in 1997 and to 47 percent in 2001. Therefore, in terms of the effort being put in by unions, it maybe that they are running very fast merely to stand still, and this may support the contention of the skeptics of union revitalization„e.g., Machin (2000a, 2000b) and Millward, Bryson, and Forth (2000).

     A broadly similar picture appears to be the case with union density. Union density is closely linked to the extent of union recognition because it is only with the latter that the full benefits of the former may be gained. Membership density has, according to the LFS (in Labour Market Trends, various issues), fallen from 55 percent in 1979 to 29 percent in 2002 with the first halt in decline in 1998 followed by the first increases in 1999 and 2000 since 1980 but declines in 2001 and 2002. But again, these are small„an aggregate increase of around 160,000 members, and this was followed by a fall of 100,000 in 2001. Arguably, similarly "slim pickings" again for all the efforts and resources expended by the unions.

Question 3: How Achievable Is Returning Back to the 1979 Level of Union Coverage, and What Are the Necessary Conditions?
      At present, hundreds of campaigns covering nearly 570,000 workers are underway, in which around 60,000 workers are in workplaces with in excess of 50 percent density. There is also scope for further campaigns that do not have to start from scratch. WERS98 (Cully, Woodlands, O'Reilly, and Dix 1999, 93) found 8 percent of workplaces had a union presence but no recognition. In these, the average density is 23 percent with "only" 44 percent being below 10 percent (one of the thresholds for acceptance of a CAC application). Some 85 percent of these are in the private sector. Confederation of British Industry (1998, 23) data shows a similar picture. Hope for the unions might come from the influence of a normalization effect of increasing numbers of new union agreements on the attitudes and behaviors of employers subject to campaigns now and in the future. Thus, these employers would see signing union recognition agreements as not out of the ordinary, but rather mainstream acts. In other words, there would be a bandwagon effect. This bandwagon would also stimulate further union recruitment and recognition campaigning, creating the kind, if not the level, of continuous growth that unions experienced from 1960 to 1979 in Britain via the "virtuous circle of cause and effect" (Bain and Price 1983, 18¿19). Thus, the interpretation might be "the future looks bright."
     The rate of signing of agreements, however, has slowed dramatically from its peak in 2001. This suggests the annual rate of new agreements being signed may return to around its pre-1999 level. In 2002, ACAS suggested the initial bulge in its cases of collective conciliation on recognition was returning to its former pre-1999 levels (Times, June 25, 2002) and Dibb Lupton Alsop (2002) also reported a sharp fall in the number of agreements signed and the number of employers approached in 2001¿2002 compared to 2000¿2001.
     This downward trajectory suggests that unions have used up the majority of their "good" cases and are experiencing difficulties in quickly producing a further batch to go forward with. Thus, unions, having scored easy victories by using up this pool, are now facing a more difficult challenge„i.e. "harder nuts to crack" by virtue of opposition from employers or employee indifference. Worryingly for the unions, in light of the obstacles and difficulties of gaining recognition through ERA, they may become more reliant on it for gaining recognition as their ability to gain recognition voluntarily declines. The proportion of direct and indirect CAC recognition agreements has risen from 1 percent of all agreements in 2000 to 6 percent in 2001 to 14 percent in 2002 and 20 percent in 2003. Further, doubts might be cast over the limited extent of campaigning and the location of these campaigns, compared to the task of achieving significant increases in union recognition. Unions have further financial, organizational, and personnel resources available, but it seems unlikely that they will, like their U.S. counterparts, substantially increase the proportion (e.g., to 30 percent of expenditure) of these devoted to organizing new workers and new areas. Another might concern efficacy. With 60,000 workers ostensibly eligible for automatic recognition, this could be seen as both a strength (getting to that point) and a weakness (employers refusing to grant recognition). And, of top of this, 90 percent of workers covered by campaigns have yet to reach the 50 percent threshold.
     Setting aside the inability to see thirty-five years ahead, the likely conditions under which gradual, unspectacular and continuous growth could take place can, nonetheless, be discussed. These conditions concern union resources, employer response, the legal and public policy context, and the balance and interaction between them.
     We have already noted the relative paucity of union resources. Unions are generally underresourced relative to the task of substantially increasing their membership and reach (i.e., recognition and influence with employers). Unable to fund, by whatever means and for whatever reasons, further dedicated FTOs or recruiters or other FTOs, the absence of a large milieu of lay activists who are predisposed, often through left-wing ideologies, to make the personal sacrifices necessary to unionize workplaces from within or without is marked. Whether these activists are union members or community or student activists matters much less than their absence and willingness to expend their time and effort. It remains unclear whether the financial return from recent growth in members will outweigh the costs of recruiting and servicing them to allow further resources to be put into recognition campaigns. Even if it is, the amounts may not be substantial.
     This kind of weakness may incline unions that do not enthusiastically endorse explicit partnership to adopt this approach when campaigning for recognition employers and, in particular, with multisite employers. The utility of the partnership approach here is that it more easily allows a union with relatively low but hard gained membership presence throughout the employing organization to make a successful pitch for recognition. Thus, "organizing for partnership" involves attempting to make the leap from weak union presence to recognition, not through relying on union strength (with density serving as a proxy), but by offering to "add value" to the employer's organization. More explicitly than is normally the case, the national union promotes and offers the "business case" for trade unionism, seeking to become a credible and valued partner in the eyes of the employer. This holds out the prospect of a relatively quicker, less difficult, less expensive, and more controllable process by which the national union can gain recognition, compared to the alternative„the long, hard, and sometimes unsuccessful slog of building up membership across sites or within large organizations where workforce turnover, limited resources, and employer hostility frustrate efforts. Other than the tiny handful of cases in which an employer wants to avoid a particular union, the Achilles heel in the strategy of organizing for partnership is two-fold. The degree of compulsion upon the employer is limited because of limited membership strength and arguments of mutualism are not likely to be receptive to anti-union employers. Moreover, the agreement then signed might be a heavily constrained one.
     The next issue to examine is whether employers without recognition are "non-union" or "anti-union." Large swathes of unorganized workers are within organizations with very little or no previous contact with, or experience of, unions. They are consequently more costly to organize, in regard of time and resources. Those in non-union organizations may be said to have no or little demand for trade unionism, whatever latent demand being satisfied or dissipated through deliberate or unconscious union substitution. By contrast, those in anti-union organizations may be said to have their demand suppressed. "Anti-union" can also refer to employers who respond to an actual, potential or hypothetical union threats and non-union to those employing organizations that have merely never been "troubled" by workforce propensity to unionize. Although employing organizations may display both elements of such dichotomous phenomena and practices across space and time, the simon-pure versions have significant implications for the costs of organizing for national unions in the context of resource limitation whereby organizing approaches in non-union organizations will meet polite workforce indifference rather than workforce and employer hostility, whereas organizing approaches in anti-union organizations will be met with fear and apprehension on the part of the workforce as a result of management actions.
     It may, therefore, be the case that non-union companies are, relatively speaking, the most cost-effective targets if workers there can be cognitively liberated so that they realize they have grievances (assuming they do) and these are best resolved through unionization and union recognition. Costeffectiveness is judged not just to involve union outlay (financial and personnel resources) measured against returns, assessed by union density and achievement of recognition, over time (volume by return over time), but also additional costs imposed upon union personnel and campaigns by employers. Greater relative cost-effectiveness is envisaged under non-unionism because this type of employer style is predicated neither upon union suppression, where additional costs are imposed on union membership/internal activity, nor upon the price of entry to the workplace/ access to the workforce being raised. The difficulty is, however, in finding workers in non-union organization that can act as "proselytizers" and overcoming the apparent contentment. By contrast, workers in anti-union organizations are more likely to have grievances but lack the resources, capability, and willingness to resolve these through unionization/recognition for fear of punitive measures.
     An important assumption in these assessments, however, is that a nonunion employer does not become an anti-union employer, and this may be erroneous. An employer prepared to spend resources on one method to avoid trade unionism may be prepared to do so on another involving suppression. Unions will find the answers to these conundrums only by testing the water in each particular case. With this information to hand, decisions on the utility of beginning or continuing recognition campaigns can be made; however, this may not produce positive union public relations material„ "Union X refused to help us organize because they said it was too expensive!"„ and bring about internal union controversy. Either way, the costs of organizing and winning recognition among non- and anti-employers is high but not necessarily uniform across the binary divide.
     The outcomes of applications to the CAC so far have been relatively favorable toward the unions and not a haven for resistant anti-union employers (see Younson 2002). By the end of 2003, 330 applications had been made under Schedule 1. Of them, 90 were withdrawn before admissibility decisions, 212 were accepted, 38 were rejected, and 42 were resubmitted after withdrawal. Of accepted applications, thirty-three were withdrawn before bargaining unit determination. In bargaining unit determination, fifty-six were pro-union, fourteen were pro-employer, and nineteen were not contentious. Five were withdrawn after this, with seventeen passing and five failing validity tests. Examining final outcomes, sixty-eight have resulted in voluntary recognition agreements, forty in automatic awards through membership audit and forty-nine ballots have resulted in recognition. Furthermore, the CAC's decisions have been reaffirmed in the two cases of judicial review so far, and automatic awards have been made in the face of employer demands for ballots and with the 40 percent threshold only just being met. Among the "scalps" are well-known companies such as Honda, MTV, Virgin and Easyjet airways, and the Bristol Evening Post, owned by the "non-union friendly" Daily Mail and General Trust. The salient point here is not only that this trend would have to continue, but do so on a larger scale. This is related, in part, to whether the statutory provisions are revised under the promised review.

Question 4: To What Extent Has the Nature of the Statutory Union Recognition Disinclined Unions to Use It?
     The number of applications so far has been relatively limited in terms of the number of new recognition agreements and campaigns for recognition. This exists for a number of reasons related to the nature of the statutory provisions; the various support thresholds, their complexity, and the bar on reapplication inter alia. Other reasons also exist such as union preference for voluntary deals per se and the ability to use the shadow effect of the ERA. In a broader sense, however, the limited usage also reflects the restraints to union organizing capacity, in terms of cases where employer opposition has been met but membership density of between 10 percent and 50 percent exists. This raises the question of whether a less complex, less lengthy, and thus more favorable statutory procedure would dramatically increase the propensity of unions to use it. This is a pertinent issue because if unions are to increase the number of voluntary agreements to the level needed to make a significant impact on industrial relations as well as successfully tackle the "anti-" and opposed to the "non-union" employers, important changes will be needed to the statutory procedure to strengthen it in terms of its demonstration and shadow effects. The unions have prioritized the removal of the fewer than twenty-one workers employed bar and the requirement in ballots to have at least 40 percent of all those entitled to vote voting for recognition. The Labour government promised the unions a review of the ERA. The review, however, did not concede any significant ground to the unions on these issues, merely making suggestions concerning easier access for unions to the workforce during the balloting period and the possibility of an "unfair labor practice." The government took this position because it believes that the system is working well, and there are no major problems. It is backed by the employers who oppose any further "pro-union" measures.
     In any case, although union criticism of the union recognition provisions may be not only justified and significant, it probably the case that more pertinent factor is union organizing capacity in terms of ability and resources. Greater union strength in this regard would facilitate further recognition gains, making the unhelpful complexion of the ERA recognition provisions less important. What is also important to recognize is that the ERA's unhelpful complexion could be ameliorated by union-friendly public policies and governmental attitudes making unions credible social partners and therefore facilitating a more conducive environment for securing voluntary agreements by influencing employers (contrast Bain and Price 1983, 19). At present and into the foreseeable future, however, there are quite obvious parameters to this how far this process can and will go.

Question 5: Can a "Virtuous Upward Spiral" Be Created between Union Recognition and Union Recruitment?
     Recognition is difficult to achieve without substantial or majority membership and these are difficult to achieve without prior recognition, creating a "Catch 22" situation and suggesting that a virtuous upward and self-reinforcing spiral is a long way off being created. Bain and Price (1983, 18¿19) express this as "the greater degree of union recognition the more likely workers are to join unions and remain in them. Union recognition and union growth are mutually dependent, however, because the degree to which employers are prepared to recognize unions is at least partly dependent upon their membership strength. Hence union recognition and union growth combine together in a virtuous circle of cause and effect." This focuses attention on whether the demonstration and shadow effects of the ERA's recognition provisions are sufficiently strong enough to "square the circle" to produce an enormous increase in union recognition coverage, in light of employer hostility and their strategies of suppression and substitution.
     Examining Britain's two previous statutory recognition procedures may give some indication of the potential for the ERA to do this given the continuing culture and practice of voluntarism within British industrial relations. However, cognizance needs to be taken of the different contexts in which they operated vis-ö-vis union strength and public policy. The first was the Commission on Industrial Relations (CIR). Established in 1969 as a Royal commission, it was charged from 1969 to 1971 with helping to resolve recognition claims but without any mandatory powers of enforcement. It did so in thirteen cases, recommending recognition in ten, of which five were not accepted by the employer. From 1971 to 1974, under the Industrial Relations Act 1971, the CIR's powers were extended by way of certain sanctions, although it still sought to encourage voluntary agreements. In this form, it dealt with twentyseven references, making eighteen recommendations for recognition, of which four were not accepted by the employer. All but five of these eighteen cases concerned white-collar workers (involving approximately 75,000 workers). The relatively small number of cases and the preponderance of whitecollar workers reflect a number of factors. The initiative for the CIR and its recognition functions did not spring from the union movement but rather governments (Conservative and Labour) and some employers (see the Donovan Report 1968). The unions felt sufficiently strong at this point not to require legal or statutory assistance. Furthermore, they feared legal intervention, seeing it as a stalking horse for unwanted intervention in other areas of industrial relations. Indeed, the majority was not legally entitled to use the CIR as they chose not to register under the IRA because they opposed the act's provisions; however, the mechanisms were used by registered unions in areas of traditional weakness„namely, the service sector (retail, finance primarily) and among white-collar employees in manufacturing operations (Kessler 1995; Kessler and Palmer 1996).
     The experience of ACAS between 1976 and 1980 under the Employment Protection Act 1975 (EPA) is more substantial. Under the EPA, ACAS was charged with providing voluntary conciliation where requested in recognition disputes. Under a referral by a union, ACAS was obligated to examine the issue, seek to resolve it by conciliation, and, failing that, make further inquiries and prepare a written report setting out its findings and any recommendation for recognition. The act provided for, in the event of employer noncompliance, further conciliation, but, where this failed, ACAS could make a legally binding award. ACAS dealt with 1,610 references for recognition. In 1,115 cases (70 percent) voluntary settlements were reached, in which 518 cases the union was fully or partly successful in securing recognition. ACAS (1981, 32, 65) calculated that this method brought 64,000 new workers under recognition (with 16,000 through final recommendations) while in the same period some 77,000 new workers were brought under some form of recognition by the more long-standing voluntary route using ACAS; however, in only about a third of the statutory awards was recognition achieved (ACAS 1981, 92).
     Although this represented a longer and more substantial experience than that of the CIR, the gains to recognition were similarly relatively limited. Again, unions were not desperate to use the means available as membership and recognition coverage (by establishment and workers) continued to grow in the period, although more use was made of references in manufacturing, probably reflecting use by the previously nonregistered unions. It may be the case that the presence of the procedure, threat of recourse, and actual usage had a much wider, if unquantifiable, impact (Dickens and Bain 1986, 94). ACAS (1981, 100) concluded, after the statutory mechanisms were abolished in 1980 by the Thatcher government, they "had no more than a marginal impact except in one or two sectors of industry" and had "clearly failed to achieve a major breakthrough that had been envisaged by their early advocates." The overall generally supportive tone of public policy toward unionism, of which ACAS was part, however, exerted a stronger indirect impact on employer attitudes and actions than direct regulatory mechanisms (Bain and Price 1983, 20¿21; Dickens and Bain 1986).
     It should thus be clear that the greatest impact of statutory recognition provisions for unions derives from their demonstration and shadow effects, rather than the quantity of their usage per se. Consequently, the most important measure is not the total number of cases where statutory recognition has been granted and how many workers are covered by these awards. As Metcalf (2001, 18) points out, the volume of cases and union wins needed to make a substantial and positive direct impact on the existing coverage of recognition would far exceed what the CAC is capable of dealing with. The most important aspect is the "message" sent out by overall case outcomes. Simply put, a large volume of cases with a majority of unambiguous union wins in the various determinations (i.e., not just final awards) are needed to demonstrate to employers that there is little to be gained from refusing voluntary recognition where a statutory application will then be forthcoming. Thus, when approached by unions with a credible presence, most employers will conclude that signing a voluntary agreement is the most appropriate action to take. It now looks possible, however, the trade unions will become increasingly reliant on the statutory provisions to secure further recognition agreements by making a relatively small number of applications because they have few secure cases to apply with in an effort to deal with the prevalent tendency of employer opposition„that is, "the hard nuts."
     The experience of the CIR and ACAS suggests that it would be rather na¥ve to expect the CAC mechanisms to offer a strong prospect of "squaring the circle" of achieving enormous increases in relative short space of time. The appropriate levels of volume and high success rate of applications are unlikely to emerge to create the shadow and demonstration effects. Moreover, even if these did exist„and this is unlikely„it is doubtful whether their impacts would be sufficiently strong to have the desired outcome. This situation arises because although unions maybe relatively more willing to use the CAC (notwithstanding its obstacles for them) by virtue of their general weakness, their general weakness itself means the shadow and demonstration effects of the ERA are weaker than they might otherwise have been. This points to the outcomes of CAC applications operating in the current circumstances as being unable to help deliver the size of increase in both recognition coverage and union density. Only if public policy on other matters with regard to trade union influence and legitimacy was dramatically and positively changed could this situation be reversed. There are, however, no signs that this is likely.

Possible Answers and Possible Solutions?

     The simplest conclusion to draw is that there is an overwhelming need for unions not only to keep battling away but to battle harder, longer and in many more places viewed from whatever standpoint. In sum, they need to put much greater financial and human resources into recruiting, organizing, and servicing so that the resource gains of new members can be used to recruit and organize additional members and that potential members can thus see around them, in a meaningful way, the benefits of union membership through "strength in numbers." But this is mere exhortation, and an unrealistic and unrealizable exhortation at that. We need, therefore, to more concretely further examine the likely conditions for growth. The conditions which might produce the "great leap forward" for the trade unions on the recognition front are a much higher level of industrial struggle and class struggle (with the removal of existing restrictions on industrial action) and the emergence of a wider milieu of politically motivated union activists. This would mean an upswing in successful industrial struggle and worker mobilization in Britain of the scale of type witnessed in 1910¿1920 or 1968¿1974 and the development on a layer of workplace militants akin to what surrounded the Communist Party in the period 1950¿1980. The utility of these conditions is predicated on their ability to constitute mobilization of workers' collective power. Of course, this is not unproblematic, not least because it is unclear where resurgence may come from and because employers may respond in a punitive manner as a result of the higher costs and infringements upon their managerial prerogative consequent upon resurgence. If the "great leap forward" does not transpire, much less ambitious growth becomes the order of the day. Thus, something parallel to the TUC's target of a million extra members over the next five to ten years appears more realistic.

References

ACAS. 1981, 1991, 1995¿1999, 2001. Annual Report. London: ACAS.

Bain, G., and R. Price. 1983. "Union Growth: Dimensions, Determinants, and Destiny." In G. Bain, ed., Industrial Relations in Britain. Oxford: Blackwell, pp. 3¿33.

Brook, K. 2002, 2003. "Trade Union Membership." Labour Market Trends (July).

Confederation of British Industry. 1998 Employment Trends Survey. London: Confederation of British Industry.

Cully, M., S. Woodland, A. O'Reilly, and G. Dix. 1999. Britain at Work: As Depicted by the 1998 Workplace Industrial Relations Survey. London: Routledge.

Dickens, L., and G. Bain. 1986. "A Duty to Bargain? Recognition and Information Disclosure." In R. Lewis, ed., Labour Law in Britain. Oxford: Blackwell, pp. 80¿108.

Kessler, S. 1995. "Recognition: CIR and ACAS experience." Employee Relations, Vol. 17, no. 6, pp. 52¿66.

Kessler, S., and G. Palmer. 1996. "The Commission on Industrial Relations in Britain, 1969¿74: A Retrospective and Prospective Evaluation." Employee Relations, Vol. 18, no. 4, pp. 6¿96.

LRD/TUC. 1996-2004. Trade Union Trends: Focus on Union Recognition. London: TUC.

Machin, S. 2000a. "How Are the Mighty Fallen." Centre Piece, Vol. 5, no. 2, pp. 28¿30.

Machin, S. 2000b. "Union Decline in Britain." British Journal of Industrial Relations, Vol. 38, no. 4, pp. 631¿46.

Metcalf, D. 2001. "British Union: Dissolution or Resurgence Revisited." Discussion paper 493, Centre for Economic Performance, London School of Economics.

Millward, N., A. Bryson, and J. Forth. 2000 All Change at Work? British Employment Relations 1980¿1998, as Portrayed by the Workplace Industrial Relations Survey Series. London: Routledge.

.