Labour Research May 2013


Is it only a flash of recognition?

There has been a sharp decline in applications under the statutory procedure for union recognition, while the procedure has brought UK unions both challenges and opportunities. Sian Moore and Sonia McKay explain.

On 6 June 2000, a new statutory procedure was brought into effect in the UK which meant trade unions could seek recognition from employers for the purposes of collective bargaining over pay, hours and holidays.

The law, established under the Employment Relations Act 1999, provides that, where a union claim for recognition for collective bargaining purposes can’t be mutually resolved between the union and the employer, the union can have the matter determined by the Central Arbitration Committee (CAC).

This is the body that oversees the scheme and adjudicates on applications relating to the recognition (and de-recognition) of unions.

For a union to gain recognition, it must demonstrate to the CAC, either through membership levels or a ballot, that it has secured the support of the majority of the workers in a particular bargaining unit — the group of workers the union wants recognition for.

So has the procedure expanded opportunities for unions to extend their reach in UK workplaces? And what challenges have unions encountered in accessing the statutory recognition procedure and what can be learned from these?

In our forthcoming book, Statutory regulation and employment relations: the impact of statutory trade union recognition, we evaluate the impact of the recognition procedure from its inception in 2000 up to 2011-12.

And we examine the implications for the “voluntarist” approach to regulating industrial relations, while Sarah Veale, head of the TUC’s equality and employment rights department, provides the first account of the negotiations that led to the introduction of the procedure.

Level of applications to the Central Arbitration Committee

We found that the most notable trend is the sharp decline in applications under the statutory procedure since 2004.

By the end of the year 2011-12, there had been a total of 785 applications to the CAC. However, after peaking in 2001-02, the proportion subsequently dropped from an average of 89 applications a year in the first five years, to 54 in the five years to 2010-11.

One explanation for the relatively low number of applications, even in the initial period, is that since the aim of the law was to assist in promoting voluntary recognition agreements (outside of the law), the number of applications submitted to the CAC should not necessarily be seen as an indication of the procedure’s effectiveness.

Another is that unions have adopted a strategy to avoid a high level of failed applications that might jeopardise the future of the procedure. This meant that a number of more speculative claims were not submitted or were quickly withdrawn — 27% of applications were withdrawn after submission over the whole of the 10-year period under study.

Withdrawn claims

Withdrawal allows the union to resubmit, whereas having the claim declared invalid by the CAC excludes the union from submitting an application for the same bargaining unit for a three-year period.

In addition to withdrawals for technical reasons, the CAC recorded that in 2002-03, around 20% of applications were withdrawn at some stage because recognition had been reached on a voluntary — or “semi-voluntary” — basis. The latter is where the employer agrees to recognise the union after a recognition procedure has been initiated, but before the CAC has made a declaration or ordered a ballot. This figure has fluctuated — in 2011-12 it was 16%.

Before accepting any application, the CAC must decide if the application is admissible and applies a set of validity tests to determine whether this is the case. Among these are the major requirements that the union is able to demonstrate that it has at least 10% of the membership in the proposed bargaining unit, and that a majority of workers are likely to support recognition.

Rejected claims

CAC-published records show that between 2000 and 2011-12, 465 claims were accepted — with 17% rejected.

The principal cause of the application not being accepted was, in the CAC’s view, lack of evidence that a majority of workers was likely to be in favour of recognition. Approaching half (43%) of all rejections were on this ground.

The second reason for rejection (in 19% of cases) was the pre-existence of a collective agreement covering the bargaining unit.

One cause for concern has been the then TGWU union’s unsuccessful application for recognition for warehouse operatives and drivers at Asda’s depot in Falkirk.

In this case, the employer’s agreement with the GMB general union, limited to representation rights in disciplinary and grievance matters, was judged to pre-empt the TGWU’s claim.

The definition of collective bargaining applied by the CAC in this case — confined to facilities only — had represented a very real limitation on the scope and content of collective bargaining.

However, recently this has been challenged in the important new case of the Pharmacists’ Defence Association Union (PDAU) v Boots Management Services (TUR1/823/2012), which established that a union can only be trumped by a pre-existing collective agreement where it covers at least pay, hours and holidays.

Here, John Hendy QC successfully argued that a pre-existing agreement with what was deemed to be a trade union, the Boots Pharmacists Association, did not constitute an agreement for collective bargaining in the terms of Article 11 of the European Convention of Human Rights, following Demir v Turkey [2009] IRLR 766, since it expressly excluded bargaining on hours, pay and holiday.

Definition of the bargaining unit has significant impact on outcome

Another finding from our research is that the definition of the bargaining unit was a crucial component of any successful recognition application.

Between 2000 and 2011-12 there was agreement between the parties on over half (58%) of bargaining units, with the CAC determining the bargaining unit in the remainder.

Unions were far more likely to win recognition (through a ballot, automatic recognition or through a semi-voluntary agreement) where the CAC supported the union’s definition of the bargaining unit.

They were far less likely to be successful where the employer’s definition of the bargaining unit prevailed (leading to the application not being revalidated; the application being withdrawn because the union could not demonstrate majority support in the new bargaining unit; or the union losing a ballot).

And unions have experienced difficulties where the employer sought to include all or more sites in the company, as they generally can’t subsequently demonstrate sufficient support for recognition among the workers on the other sites included in the revised bargaining unit.

Over the period, the CAC declared recognition without a ballot in over three-quarters (77%) of the cases where a union originally had majority membership. But this still left around one in four cases where, despite majority membership, a ballot was ordered. In just over one in five of these cases, the ballot was called because the CAC had received evidence to doubt that a significant number of union members wanted the union to bargain on their behalf.

Outcome of recognition ballots

In over three-quarters of these cases, the union subsequently lost. In a third of them, the ballot was called because the union no longer had a majority in membership, in some cases because the numbers of workers in the bargaining unit had increased.

This means that membership levels (through changes in the bargaining unit) fell during the procedure and/or that union members abstained or even voted against recognition in the ballot.

Not only does this uncertainty encourage employers to intervene in the procedure to ensure that a ballot takes place, but this particular aspect of the rules raises questions about the ability of unions to sustain membership once they are in the procedure.

Overall, by 2011-12, the CAC had reported on the outcome of 203 recognition ballots held under the statutory procedure since 2000.

The statistics on ballots show that consistently, a third were lost, despite the fact that the union demonstrated majority support on application. Furthermore, in nearly one quarter of unsuccessful ballots, 50% or more of those voting had supported recognition.

This failure was due to the fact that the legal rules state that a union must not only achieve a majority in favour of recognition, it must also secure the support of at least 40% of the workers in the bargaining unit. In these cases, the union did not achieve the 40% threshold needed to secure recognition.

Only half accepted applications lead to statutory recogniton

Of the 465 applications accepted as valid by the CAC, recognition was awarded without a ballot in 106 cases (23% of all the cases accepted) while in another 126 (27%), recognition was awarded following a ballot.

So overall, just 50% of the applications that were accepted secured recognition through the procedure (excluding those withdrawn because the parties secured a semi-voluntary agreement). However, taking all of the original 785 applications into account, the proportion achieving statutory recognition fell to just under one in three (30%).

We argue that the design of the procedure encourages employers to contest claims, and evidence shows that union membership and support are volatile in the face of employer opposition. Moreover, employer capacity to stall the CAC process further corrodes union support.

While employers have not completely undermined the 2000 procedure, they have nevertheless been able to constrain union success. And although both of these aspects are a function of the procedure’s design, the willingness of the CAC to call a ballot where employers produce “evidence” to challenge union support for recognition has resulted in uncertain outcomes for unions.

Unfair practices provision inefective

Concerns about the operation of the 1999 Employment Relations Act, which established the procedure, led to the law being amended in 2004 through the introduction of an unfair practices provision. This provision banned intimidation by unions or employers to unduly influence workers in the bargaining unit during statutory recognition ballots.

Yet the fundamental inequity of the employment relationship continued to be ignored: in the eight years since the introduction of the provision, none of the cases brought under it, either by unions or employers, have been upheld.

Here, a party not only has to prove that there has been an unfair practice, but that it has affected the way that workers vote in the ballot. This obviously places a very high bar to a successful union complaint. What’s more, in raising such a complaint, a union would effectively be prolonging the procedure, something that generally works against it and in favour of the employer.

Crucially, the fact that the provision covers only the ballot period means unions have been unable to address systematic campaigns to defeat them in the workplace from the point at which the union requests recognition.

Impact of statutory procedure on voluntary recognition

Despite these considerable problems, a survey of unions which we undertook in 2010 as part of the background research for the book showed that the statutory procedure remained significant for many of them as they looked to reverse membership decline, with its use closely linked to organising strategies.

In the past five years, privatisation has encouraged public services unions to resort to the statutory procedure, reflecting the fragmentation of collective bargaining in the sector, and resulting in a number of contract-based bargaining units covering very small numbers of workers.

The survey found that the value of the recognition procedure continued to be in its shadow effect in encouraging voluntary agreements — although the number of recognition targets and the success rate for unions has reduced.

There is little published data on the number of voluntary recognition agreements concluded in the second five years of the procedure.

But what evidence there is (based on Acas figures), suggests that there has been a similar or even greater rate of decline here than has been the case for the statutory scheme.

The legal procedure has also shaped voluntary relationships between trade unions and employers, which may restrict the scope and depth of bargaining.

This is because unions also reported that the statutory process had influenced employers’ responses to recognition claims outside of the procedure, with employers demanding majority membership or a ballot before conceding voluntary recognition.

CAC applications unlikely to increase substantially

The operation of the statutory procedure may discourage unions from using it; but the decline in applications also reflects the capacity of the unions to organise. The operation, but more importantly, the design of the procedure places few restrictions on employer advantage in the workplace, while an examination of union strategies suggests unions’ limited capacity to generate new recognition claims.

Both are a function of the wider landscape of work and employment relations and its political and legal context. This interaction suggests that it’s unlikely that CAC applications will increase substantially in the future.

This feature is drawn from research by Sian Moore and Sonia McKay with Sarah Veale in Statutory regulation and employment relations: the impact of statutory trade union recognition; due to be published June 2013 by Palgrave Macmillan.

Sian Moore is professor of Work and Employment Relations at the Centre for Employment Studies Research at the University of the West of England. Sonia McKay is professor of European Socio-legal Studies at the Working Lives Research Institute at London Metropolitan University.