Gall Bjir99

British Journal of Industrial Relations 37:4 December 1999 0007-1080 pp. 601-614

Developments in Union Recognition and Derecognition in Britain, 1994-1998 Gregor Gall and Sonia McKay

Abstract This research note shows a marked change in the relative incidence of cases of derecognition and recognition in the period 1994-1998. It shows that the level of derecognition has fallen significantly in recent years while that of the signing of new recognition agreements has continued at its farmer level, so that on balance new recognition agreements clearly outnumber cases of derecognition. The context and reasons for this are explained by reference to developments in public policy, employer views and union practice. The results of derecognition and the prospects for union recognition are also examined.

1. Introduction Practitioner, media and academic interest in union recognition and derecognition in Britain remains unabated (see e.g. Claydon 1996; Disney et al. 1996; IRS 1997; LRD 1998, 1999; Wood 1997). This stems from the continuing debates over the extent of change in the contemporary employment relationship in terms of HRM, the 'new industrial relations', the decollectivization and individualization of industrial relations, and the view of derecognition as a weathervane for judging these. The debate on the appropriate procedures for recognition has been ongoing in the light of proposals from the Trades Union Congress (TUC 1995) and the prospect of and then the election of a Labour Party government in May 1997 with its plans for establishing statutory procedures for gaining union recognition (Department of Trade and Industry 1998). Previous studies of recognition and derecognition indicated a number of salient points. First, after a period of stability in the early to mid-1980s, the number of recognition agreements and the number of workers covered by them began to fall quite markedly from the late 1980s onwards (Gallie et al. Gregor Gall is in the Department of Management and Organization at the University of Stirling. Sonia McKay is with the Labour Research Department, London. X'-'' Blackwell Publishers Lid/London School of Economics 1999, Published by Blackwell Publishers Ltd, 108 Cowley Road, Oxford, OX4 lJF. UK and 350 Main Street. Maiden. MA 02148. USA.

602 British Journal of Industrial Relations 1996; Millward and Stevens 1986; Millward et al. 1992). Furthermore, the Labour Force Survey (Cully and Woodland 1997,1998) shows, as does the 1998 Workplace Employee Relations Survey (WERS) (Cully et al. 1998), further falls in the level of recognition. The WERS98 survey records the number of workplaces with union recognition falling from 53 per cent in 1990 to 45 per cent in 1998 (Cully et al. 1998; 15-16). Such reductions are likely to result both from the establishment of sizeable new workplaces that have no union recognition and from the continuing retrenchment in the size of, and closure of, unionized workplaces (cf. Millward 1994). Second, incidences of derecognition have increased from relatively small levels to become significantly greater than many anticipated (Gall and McKay 1994; Geroski et al. 1995; Gregg and Yates 1991; Gregg and Machin 1992; Millward et al. 1992). Third, the environment for unions to gain new recognition agreements has worsened from that existing in the early 1980s (ACAS1991). Given these contexts, in this research note we seek to build on our earlier study (Gall and McKay 1994), which covered the period from August 1998 to July 1994, by assessing and explaining developments in derecognition and recognition from the beginning of 1994 to the end of 1998. This shows that the previous pattern of cases of derecognition exceeding those of new recognition agreements has now been reversed. We then consider the sectoral distribution and nature of, and reasons for, derecognition and recognition. Finally we suggest explanations for such developments and consider the basis of workplace relations after derecognition and the future trajectory for union recognition.

2. Methodology The data for this research are derived from two sources: the Labour Research Department/Trades Union Congress (LRD/TUC 1995, 1996a, b, 1997a, b, 1998a, b) Trade Union Trends surveys, and a number of secondary sources such as trade union journals, ACAS Annual Reports, surveys by interest groups and newspaper and periodical reports. The Trends surveys covered on average 70 per cent of the TUC affiliates' membership (of 6.75 million in 1998 (TUC 1999)). The data generated from them are likely to be fairly inclusive on recognition and slightly less so on derecognition, given that unions are more willing to publicize their 'successes' than their 'failures'. None the less, compared with the ACAS data (generated exclusively from its casework), the Trends data set is likely to be more representative of developments. Inevitably, there are limitations with this type of data (Beaumont and Harris 1995; 391; Tuckman and Finnerty 1998: 452); but, equally, large-scale surveys such as WIRS (e.g. Millward et al. 1992), which considered recognition and derecognition, are not without their problems in terms of (intra- and inter-) consistency of data and findings (Claydon 1996; Kessler and Bayliss 1995; Marginson 1998a). © Blackwell Publishers Ltd/London School of Economics 1999.

Union Recognition and Derecognition in Britain 603 Therefore the main type of data on which this research note is based remains of much use and can supplement such wide-scale surveys by providing more immediate, detailed and circumstantial material over shorter time scales.

3. Developments in derecognition and recognition Table 1 shows the annual rate of reported cases of derecognition and recognition since 1989. This reveals that the annual rate of derecognition peaked in the early 1990s and declined thereafter, while the annual rate of reported cases of new recognition agreements roughly matched those of derecognition from 1989 to 1992 and then increased towards the end of the 1990s. Indeed, in each year since 1993 the number of cases of new

recognition agreements has exceeded those of derecognition. Table 1 indicates that over the period the overall rate of derecognition has slowed down while the rate of new recognition agreements has increased slightly, in total exceeding those cases of flcrua/derecognition by two hundred cases. This contrasts quite markedly with previous studies, showing the rate and extent of derecognition increasing (Claydon 1996; Gall and McKay 1994), and indicates that a reduction in the level of recognition over time does not preclude falling incidences of derecognition and increasing incidences of new recognition agreements within this general environment. Accepting the limitations of the first findings of WERS98 (Cully et al. 1998) vis-a-vis WIRS90, WERS98 reports that derecognition has continued at around a level similar to that recorded in WIRS90 while the occurrence of new recognition agreements was negligible. The reasons for the seeming disjuncture between WERS90 and our data are two-fold; the different TABLE 1 Reported Cases of Actual Derecognition and New Recognition Agreements, t989-1998'' Derecognilion t989 t990 t991 t992 t993 t994 t995 t996 1997 t998 Totals

Recognition

57 53

(5) (6)

58 49

68 79

(6)

76

(7) (6)

56 57

20 (5) 71 (5) 61 (7) 33 (2) 4 (1) 494 (50)

27 88 76 85 72 644

48

" Figures in brackets indicate those cases that were threatened but not carried out or resisted by the concerned work-forces. Sources: Gall and McKay (1994) for 1989-1993. LRD/TUC Trade Union Trends surveys and union journals for 1994 onwards. © Blagkwell Publishers Lid/London School of Economics 1999.

604 British Journal of Industrial Relations methodological tools employed (sampling as opposed to 'self-selection' of cases), and the difference in the time periods examined, whereby WERS98 examined only workplaces that were more than five years old, i.e. were established before 1993. Table 2 shows that the known number of employees covered by new recognition agreements exceeds those covered by derecognition by 110,490 to 41.308,^ a factor of 2.5:1. The average size of companies/organizations or groups of employees covered by recognition and dereeognition is remarkably similar, at 350 for derecognition and 390 for recognition (cf. Gall and McKay 1994: 446). While cases of derecognition were evenly balanced between those of full and partial types, cases of new full recognition agreements outnumber those of partial recognition by 6 to 1, TABLE 2 Types of and Numbers Covered by Reported Derecognition (Actual) and New Recognition Agreements, 1994-1998 No. of cases Derecogniiion 1994 1995 19% 1997 1998 Total Recognition 1994 1995 1996 1997

15 66 54 31 3 169 27 88 76 85

1998

72

Total

348

Workers covered {from known no. ofca.ses)

3,800 15,931 16.851 4.362

Known type of derecogniiion

364

07) (3)

13 full, 28 full, 25 full, 15 full. 2 full

41,308

(114)

87 full,

61 partial

9,520 27,404 26,377 15.994 31,195 110.490

(14) (64) (64) (61)

6 full 74 full. 61 full, 56 full, 64 full, 261 full,

11 partial 13 partial 10 partial 6 partial 40 partial

(6) (42) (46)

(53) (257)

i 33 23 4

partial partial partial partial

Sources: LRD/TUC Trade Union Trends surveys and union journals for 1994 onwards.

The validity of the data in Tables 1 and 2 is weakened, given that the main source of data, the Trends surveys, eovers the period from the beginning of 1995 to the end of 1998 and does not cover derecognition for 1998. Thus, other more limited secondary sources had to be relied on for the years 1994 and 1998. However, data from the ACAS Annual Reports (1989-1998) suggests that the figures for 1994 are likely to be generally representative (although the ACAS data would appear to be weaker with regard to derecognition). The ACAS data itself shows that in the period 1994-1998 there were 30 cases of derecognition, of which all but 8 were for full derecognition, while there were 264 cases of new recognition agreements, of whieh 193 were for full recognition (see Table 3). However, such a sharp imbalance of dereeognition to recognition in the ACAS casework is related to derecognition being less likely, and recognition requests more © Blackwell Publishers Ltd/London School of Economics 1999,

Union Recognition and Derecogniiion in Britain 605 likely, to be contested by the unions concerned. In cases of derecognition, largely for reasons of low membership, the unions have been less keen to spend resources trying to boost membership where they believe this is unlikely to happen. However, where they have a high and rising membership they have been willing to assert their 'right' to be recognized on the basis of representing a majority of workers. In the absence of Trends survey data on derecognition for 1998, newspaper and union journal reports had to be relied on. Although producing little data, this is not seen as unrepresentative, given the move away from derecognition (see below). TABLE 3 Recognilion Claims involving ACAS, 1989-1998

No. of

claims

1989 1990 1991 1992 1993

136 159

1994 1995 1996 1997 1998

87 100 109 94 125

Total

1200

174

122 94

No.

Success

of successful claims

(%)

31 38 58 26

23 24 33 21 38

n/a

37 51 60 57 50

50 57 57 60 69

36 32 51 65 54 62 453

Total

Of these. % full collective bargaining rights

% partial rights

70 71 51 71

30 29 49

29 n/a 50 43 43 40

31

numbers covered by new agreements 4800 3500" 3500" 3000

n/a n/a n/a n/a n/a n/a

38

" Authors' estimate (from Gall and McKay 1994:444), Source: ACAS Annual Reports.

Together, these data would suggest something of a softening in employers' more recent attitudes towards union recognition. In contrast to earlier reports and research (e.g. ACAS 1991; Kessler and Bayliss 1995). the environment for gaining new recognition agreements now seems relatively more favourable. Thus, although the number of claims for recognition through ACAS in the 1990s is around half of that in the mid-1980s, the proportion of successful claims has increased significantly over the same period (see Table 3). This is paralleled by an increase in the number of respondents in both the later Trends survey (LRD/TUC 1996b, 1997a, b, 1998a, b) and the Indusirial Relations Surveys (Dibb Lupton Broomhead 1996: 6; Dibb Lupton AUsop 1997: 11,15) reporting a decline in employer resistance to granting recognition. Such a trend is also confirmed by other surveys of personnel managers (Institute of Management 1998; People Management, 16 April 1998; Personnel Today, 3 September 1996, 5 June 1997; Wood 1997: 53). The overall trend for growth in recognition agreements is matched by a © Btackwell Publishers Ltd/London School of Economics 1999.

606 British Journal of Industrial Relations disinclination among employers to derecognize trade unions. Again, growing numbers of respondents to the Trends, Industrial Relations and Personnel Today surveys (LRDATUC 1996b, 1997a, b, 1998a; Dibb Lupton Broomhead 1996: 7; Dibb Lupton AUsop 1997: 11, 15; Personnel Today, 3 September 1996, 5 June 1997) indicate that derecognition was less likely. This may confound some conventional wisdom (e.g. Claydon 1996: 170), which anticipates either a continuation of the previous rate of derecognition or an upward trend in it. as well as a continuation of the decline in the signing of new recognition agreements as a result of the maintenance of a hostile environment for trade unionism. Comparing the sectoral distribution of derecognition between 1988 and 1994 (see Gall and McKay 1994) with that of 1994-1998, Figure 1 shows that it is still concentrated in a relatively smai! number of sectors but nevertheless in a greater number than was the case earlier. Thus, while derecognition remained relatively common in transportation, oil refining and the printed media, it has become quite noticeable in the food and drink sector and in further education. Some of the derecognition in these areas (other than further education) is counterbalanced to some extent by new recognition agreements. What is particularly noticeable is that new recognition agreements are being signed not only in manufacturing but also in the private service sector (cf. Millward et al. 1992). For example, the IDS (1998: 11) survey of 120 call centres with 32,000 staff reports that over half of centres (rather than of employers) have granted union recognition.

FIGURE 1 Sectoral Distribution of Cases of Derecognition and New Recognition Agreements. 1984-1998 Note: Not all cases were identifiable by sector. 1IS Wedu I com^unicslio'i

_j;;u

^

^

IP"

CoPili'ijCtiDn

loods tnginccrin^ & irenicies

laNm neagnltton agiMniada CnKni«ii*n,«,aiB.i-..i,o-.

Energy 1 watar tuDDly

© Blackwell Publishers Ltd/London School of Economics 1999.

JDOifieegnlllon

Union Recognition and Derecognition in Britain 607 largely through the extension of existing agreements {cf. Millward 1994: 21). Furthermore, unions are beginning to have some success in gaining recognition agreements with traditionally non-union subcontracting service companies working in both public and private sectors. This may be because some companies believe that union recognition helps combat their perceived image of low pay, autocratic management practices and poor conditions, which has led to lower staff commitment and productivity and higher labour turnover (IRS 1998; LRD 1996: 30). Examples of subcontracting companies working in the public sector under compulsory competitive tendering which have signed recognition deals are ServiceTeam, Onyx, Ecovert, Brophy, Compass and SITA (GB). Deals covering catering, cleaning and environmental services have been signed with the public-sector union Unison and the GMB general union. The Communication Workers' Union and the Banking, Insurance and Finance Union have also had some success in gaining recognition agreements with employment agencies that provide temporary workers to British Telecom's call centres and the high street banks. The reported cases of derecognition from 1994 to 1998 indicate a lower incidence of cases of complete derecognition (full derecognition across all grades of employees and all sections of the organization) and of derecognition across the whole company for a specific grade where the employer acted in a single action compared with the period 1988-1994 (see Gall and McKay 1994). Moreover, the cases from 1994-1998 give little indication of cumulative or associated actions, whether of a purposive/strategic or an opportunist /reactive manner. Therefore, while there are some instances of derecognition covering one group of employees being followed by derecognition covering others in the same company or organizations (e.g. Allied Domecq drinks and spirits and Newsquest newspapers), the majority do not appear to follow any identifiable pattern, for they are single actions carried out by individual companies or organizations in unrelated actions (cf. Claydon 1989,1996). Commensurate with this, grade-specific derecognition remains the most common type. While there have been attempts by unions to gain recognition through the extension of existing recognition to other sites, such as at Fullarton Computer Industries (Iron and Steel Trades Confederation) or P&O Stena Line (Rail, Maritime and Transport Union), there are only a small number of reported successful cases (cf. Millward 1994: 24) although call centres represent a notable exception in this regard. Thus, in common with derecognition, new cases of recognition agreements tend to be found where there has been no recognition before in the company/organization or for that type of employee group. Figure 2 reports the types of staff affected by derecognition and new recognition agreements. It shows that, as with earlier research (Gall and McKay 1994), manual workers too are affected by derecognition, but to a lesser extent than non-manual, professional and managerial staffs. However, this has to be seen in the context of the fall in derecognition (D Blackwell Publishers Ltd/London School of Economics 1999.

608

British Journal of Industrial Relations .

HGURE 2 Known Types of Staff Affected by Derecognition and Recognition Agreements, 1984-1998 Note: 'Non-manual workers' refers to office workers below the level of supervisors; •professionar refers to research and scientific staffs

and the growth in new recognition agreements, particularly for manual workers. Looking at the known reasons for derecognition (Figure 3), the relative importance of several factors has changed since the late 1980s/early 1990s (cf. Gall and McKay 1994). Less prominence is given to reasons of Mow or falling levels of union membership', 'move to single union/fewer unions', 'resulting from privatization/deregulation' and 'part of a wider dispute', while 'change in ownership' and 'relocation or reorganization' receive a much higher rating. This indicates a number of factors. First, there is a reduction in the scope for derecognition as fewer privatizations have recently taken place; second, where derecognition does still occur, it is less likely the result of employers seeking to benefit from weak and business-type unionism (falling membership and moves to single/fewer union deals). These may reflect both a move towards 'partnership' and the ability of stronger areas of union organization to pre-empt or resist bids for derecognition. Figure 4 shows the known reasons for granting union recognition. It indicates that some employers are not hostile to direct approaches for recognition, while others may need to be convinced by some sort of campaign, including disruptive work-time meetings, indicative ballots and demonstrations (cf. Millward 1994: 24). However, none of this imphes that ® Blackwell Publishers Ltd/London School of Economics 1999.

Union Recognition and Derecognition in Britain 609 FIGURE 3 Known Reasons for DeTecognilion (Actual), 1994-1998 Note: From 108 cases. More than one response is possible.

Move la pcnonil c

Pmabutionftlireguiatiiin

RBkicatlan/raorganiulian

FIGURE 4 Known Reasons for and Contexts of the Granting of New Union Recognition Agreements, 1994-1998 Note: From 190 cases. More than one response is possible. ch by Hnplov>r ClianB* In Min«f«hip

Dy union to employer

RKog

RgcnjRment £imDaign/nM in m«fnb»:snip

© Blackwell Publishers Ltd/London School of Economics 1999.

610 British Journal of Industrial Relations those unions that gained recognition had a simple majority in membership. A small number of instances are recorded ('following change in ownership' and 'approach by employer to union') in which the employer's attitude is important in initiating and facilitating recognition.

4. Explaining developments in derecognition and recognition

When considering the reasons for the discernible shift from derecognition to recognition, a number of explanations are possible. First, the number of companies that are likely to derecognize is becoming fewer as those that have sought to derecognize have already done so; thus, the pool of possible 'derecognizers', whether 'strategists' or 'opportunists', has been reduced over time. Second, the notions of employers working with unions in 'partnership' are becoming more widespread, indicating that pragmatic 'pluralism' rather than 'unitarism' within employer circles may be experiencing something of a revival. Borne of a 'business case' for trade unionism, many trade unions report through the Trends surveys that some employers appear to be turning away from non-union collectivized and individualized strategies for conducting and regulating the employment relationship and towards using trade unions, for a number of reasons. These employers have recognized the costs and difficulties of organizing employee relations through works councils and performance-related pay and the ease, cheapness and legitimacy of the joint regulation that union recognition can afford. Often the vehicle for this change has been a change in ownership or a change in management personnel {LRD/TUC 1998a, b). Third, the uncertainty from the mid-1990s over the future direction of government employment policy may have led some employers to wait to see what would happen before making any changes with regard to union recognition. This 'wait and see' strategy continued into 1998 while lobbying to influence the content of the proposed legislation on Fairness at Work went on. However, once it became clear that there would be some statutory mechanism for gaining recognition, this assisted in the creation of a change in the climate of industrial relations, with other companies seeking to adapt to and endorse rather than oppose or ignore union organization. Fourth, the move from the 'servicing' to the 'organizing' model of trade unionism (Blyton and Turnbull 1998) under the relaunch of the TUC with the 'New Unionism' project in 1995 has led many unions to put increasing resources into trade union membership and recognition campaigns. Such targeted high-profile campaigns with dedicated organizers have resulted in trade unions more successfully requesting recognition and resisting derecognition from stronger bases than before. Consequently it is increasingly the case that persistent and bold union campaigns can secure recognition from unwilling employers. Finally, many non-union employers are reducing their opposition to recognition as a result of the growing juridification of employee relations © Blackwell Publishers Ltd/London School of Economics 1999.

Union Recognition and Derecognition in Britain 611 through the implementation of European Union Directives into national law. Although the stress is on 'worker representatives', recognized unions are regarded as helpful in dealing with the more complex situation employers find themselves in. These combined reasons explain why the overall level and incidence of derecognition remains relatively limited, and highlight why many employers still choose to recognize unions (see Marchington and Wilkinson 1996).

5. The workplace after derecognition A considerable number of employers have established non-union institutional forms of work-force representation, communication and consultation shortly before, during or after derecognition. These may take the form of company/staff/employec regional or work councils/forums/committees. Others have employed much more circumscribed and employer-dependent methods of representation and communication, such as focus groups, attitude surveys, team briefings and suggestion schemes. However, many instances of derecognition have not been followed by the establishment of any institutional mechanism of representation, communication or consultation, and this raises the prospect of the 'bleak house' scenario (Guest 1995). With regard to the determination of pay, there is some evidence that derecognition is associated with the introduction of individual performance-related pay (Heery 1997a, b; Gall 1995). None the less, derecognition, by and large, has not led to deunionization (see Gall 1995, 1998a, b). In some cases there is evidence that the level of union membership dips immediately after derecognition in response to the realization by the work-force that retention of union membership is regarded in a hostile manner by the employer and then returns to its former level a year or two later, e.g. at the National Maritime Museum or Newsquest Newspapers. However, sustained employer opposition to union organization (activists, meetings, activities and information dissemination) means that the retention of membership remains a largely passive and semi-collective, if not individual, phenomenon.

6. Future prospects for recognition The later Trade Union Trends surveys report unions engaging in recognition campaigns covering over 300 companies and organizations employing around 235,000 known workers (LRD/TUC 1997a, b, 1998a, b).^ Assuming that recognition would be gained under a statutory procedure by the achievement of a simple majority plus one in membership among the workforce, in around a third of these campaigns recognition would be granted (LRD/TUC 1997a, b, 1998a, b). These cover some 46,000 workers. In another 20 per cent of campaigns, where membership density is between 40 © Blackwell Publishers Ltd/London School of Economics 1999,

612 British Journal of Industrial Relations and 50 per cent, recognition would probably be achieved, since the prospect of securing rights of representation and consultation is likely to persuade others to join so that the 50 per cent-plus-one threshold could be reached (LRDATUC 1997a, b, 1998a, b). This group covers some 10,000 workers. Marginson {1998b) calculated from the WERS data that 2 per cent of workplaces in Britain have union membership above 50 per cent but no union recognition, and the Confederation of British Industry found from its 1999 Employment Trends Survey, covering 830 companies with 2.4 million workers, that around 60 per cent of businesses employing 5000 or more workers and around 50 per cent of businesses employing between 500 and 50(K) workers expected a claim for recognition in some part of their operations (Financial Times, 13 May 1999). Taken together with survey evidence that employers may be less hostile to granting recognition, we may gain some indication of the potential for a growth in union recognition, particularly in the context of survey evidence (Dibb Lupton Allsop 1997: 11,16; LRDATUC 1997a, b, 1998a; Wood 1997; 49), pointing to a continuing decline in derecognition. Further research should be able to identify whether this potential "bridgehead' into non-unionism has had the ability to start reversing the years of decline in trade union recognition. Final version accepted 24 May 1999.

Acknowledgement Thanks are due to Sian Moore of LRD for constructing the graphics.

Notes 1. The data from the Trade Union Trends surveys (LRD/TUC 1995. 1996a, b. 1997a, b, 1998a, b) has been altered to exclude cases of new recognition agreements where this reflected a change in reeognition and/or bargaining structures such as cases in the National Health Service and further education. Furthermore, the figures for numbers of workers covered by recognition agreements does not include the Unison-Compass national framework agreement, which deals with general terms and conditions of employment but not pay for around 50,(K)0 workers, because to date only in the north-east of England has a comprehensive agreement comprising all terms and conditions been negotiated. 2. Further campaigns, numbering around 60 and covering some 25,000 workers and not recorded in the Trends surveys, are known of from union journals.

References ACAS (1989-99). Annual Reports. London: ACAS. © Blackwell Publishers Ltd/London School of Economics 1999.

Union Recognition and Derecognition in Britain 613 Beaumont, P. and Harris, R. (1995). 'Union de-recognition and declining union density in Britain'. Industrial and Labour Relations Review.. 48: 379-402. Blyton, P. and Turnbull, P. (1998). The Dynamics of Employee Relations, 2nd edn. London: Macmillan. Claydon, T. (1989). "Union derecognition in Britain in the 1980s'. British Journal of Industrial Relations, 27: 214-24. (1996). 'Union derecognition: a re-examination'. In I. Beardwell (ed.). Contemporary Industrial Relations: A Critical Analysis. Oxford: Oxford University Press, pp. 151-74. Cully. M. and Woodland, S. (1997). 'Trade union membership and recognition'. Labour Market Trends, June: 231-9. and (1998). 'Trade union membership and recognition 1996-97'. Labour Market Trends, July: 353-64. O'Reilly, A., Millward, N., Forth, J., Woodland, S., Dix. G. and Bryson, A. (1998). The 1998 Workplace Employee Relations Survey: First Findings. London: DTI/ESRC/ACAS/PSI. Department of Trade and Industry (DTI) (1998). Fairness al Work. London: HMSO. Dibb Lupton Allsop (1997). The Industrial Relations Survey 1997. London: Dibb Lupton Allsop. Dibb Lupton Broomhead (1996). The Industrial Relations Survey 1996. London: Dibb Lupton Broomhead. Disney, R., Gosling, A. and Machin, S. (1996). 'What has happened to union recognition in Britain?' Economica, 63:1-18. Gall, G. (1995). 'HRM in practice: a case study revisited'. Personnel Review, 24(2); 44-56. (1998a). 'Resisting the rise of the non-union employer: the case of the press workers'. Capital and Cla.ss, 64: 43-61. (1998b). 'The changing social relations of production: union derecognition and industrial relations in the UK magazine industry'. Industrial Relations Journal, 29: 151-61. and McKay, S. (1994). 'Trade union derecognition in Britain, 1988-1994'. British Journal of Industrial Relations, 32: 433-48. Gallie, D., Penn, R. and Rose, M. (eds.) (1996). Trade Unionism in Recession. Oxford: Oxford University Press. Geroski, P., Gregg, P. and Desjonqueres. T. (1995). 'Did the retreat of UK trade unionism accelerate during the 1990-1993 recession?' British Journal of Industrial Relations, 33: 35-54. Gregg. P. and Machin, S. (1992). 'Union, the demise of the closed shop and wage growth in the 1980s'. Oxford Bulletin of Economics and Statistics, 54: 53-72. and Yates. A. (1991). 'Changes in trade union wage setting arrangements in the 1980s'. British Journal of Industrial Relations, 29: 361-76. Guest, D. (1995). 'Human resource management, trade unions and industrial relations'. In J. Storey (ed.). Human Resource Management: A Critical Text. London: Routledge, pp. 110^1. Heery, E. (1997a). 'Performance-related pay and trade union de-recognition'. Employee Relations, 19: 208-221. (1997b). 'Performance-related pay and trade union membership'. Employee Relations, 19; 430-42. © Blackwell Publishers Ltd/London School of Economics 1999.

614

British Journal of Industrial Relations

Income Data Services (IDS) (1998). Pay and Conditions in Call Centres. London: IDS. Industrial Relations Services (IRS) (1997). 'Recognizing the unions'. IRS Employment Trends, no. 641: 12-16. (1998). 'Union recognition for temporary agency workers'. IRS Employment Trends, no. 659: 4-8. Institute of Management (1998). 'Quick Reaction Survey'. London: Institute of Management, 23 June. Kessler, S. and Bayliss, F. (1995). Contemporary British Industrial Relations, 2nd edn. London: Macmillan. Labour Research Department (LRD) (1996). Defending Recognition. London: LRD. (1998). Will 'Fairness' Work? LRD's Guide to the Employment White Paper. London: LRD. (1999). The Employment Relations Bill: A Trade Union Analysis. London: LRD. and TUC (1995). Trade Union Trends. London: LRD/TUC. (1996a). Trade Union Trends. London: LRD/TUC. (1996b). Trade Union Trends. London: LRD/TUC. (1997a). Trade Union Trends. London: LRD/TUC. (1997b). Trade Union Trends. London: LRDATUC. (1998a). Trade Union Trends: Focus on Recognition. London: LRD/ TUC. (1998b). Trade Union Trends: Focus on Recognition. London: LRD/ TUC. Marchington, M. and Wilkinson, A. (1996). Core Personnel and Development. London: IPD. Marginson, P. (1998a). 'The survey tradition in British industrial relations research: an assessment of the contribution of large-scale workplace and enterprise surveys'. British Joumai of Industrial Relations, 36: 361-88. (1998b). '[UK] Industrial relations under New Labour: an update*. European Industrial Relations Observatory. EIRO Online, Dublin, December. Millward. N. (1994). The New Industrial Relations? London: PSL and Stevens, M. (1986). British Workplace Industrial Relations 1980-1984: the DE/ESRC/PSl/ACAS Surveys. Aldershot: Gower. Smart, D. and Hawes. W. (1992). Workplace Industrial Relations in Transition. Aldershot: Dartmouth. TUC (1995). Your Voice at Work. London: Trades Union Congress. (1999). 7999 Directory. London: Trades Union Congress. Tuckman, A. and Finnerty, C. (1998). 'Individual contracts, collective bargaining and trade unionism; a case for the union voice'. Personnel Review, 27: 448-59. Wood, S. (1997). Statutory Union Recognition, Research Report no. 17. London: Institute of Personnel and Development.

© Blackwell Publishers Ltd/London Schooi of Economics 1999.